Hugging Face's logo

Jjzzzz
/
finetuned_bge-small-en-v1.5

Sentence Similarity
sentence-transformers
TensorBoard
Safetensors
bert
feature-extraction
text-embeddings-inference
Model card Files
xet
Metrics Community
Jjzzzz commited on
Commit
7b997ff
·
verified ·
1 Parent(s): 0c0dd9d

Upload folder using huggingface_hub

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. 1_Pooling/config.json +10 -0
  2. README.md +144 -0
  3. checkpoint-1000/1_Pooling/config.json +10 -0
  4. checkpoint-1000/README.md +144 -0
  5. checkpoint-1000/config.json +31 -0
  6. checkpoint-1000/config_sentence_transformers.json +10 -0
  7. checkpoint-1000/model.safetensors +3 -0
  8. checkpoint-1000/modules.json +20 -0
  9. checkpoint-1000/optimizer.pt +3 -0
  10. checkpoint-1000/rng_state.pth +3 -0
  11. checkpoint-1000/scheduler.pt +3 -0
  12. checkpoint-1000/sentence_bert_config.json +4 -0
  13. checkpoint-1000/special_tokens_map.json +37 -0
  14. checkpoint-1000/tokenizer.json +0 -0
  15. checkpoint-1000/tokenizer_config.json +57 -0
  16. checkpoint-1000/trainer_state.json +733 -0
  17. checkpoint-1000/training_args.bin +3 -0
  18. checkpoint-1000/vocab.txt +0 -0
  19. checkpoint-2000/1_Pooling/config.json +10 -0
  20. checkpoint-2000/README.md +144 -0
  21. checkpoint-2000/config.json +31 -0
  22. checkpoint-2000/config_sentence_transformers.json +10 -0
  23. checkpoint-2000/model.safetensors +3 -0
  24. checkpoint-2000/modules.json +20 -0
  25. checkpoint-2000/optimizer.pt +3 -0
  26. checkpoint-2000/rng_state.pth +3 -0
  27. checkpoint-2000/scheduler.pt +3 -0
  28. checkpoint-2000/sentence_bert_config.json +4 -0
  29. checkpoint-2000/special_tokens_map.json +37 -0
  30. checkpoint-2000/tokenizer.json +0 -0
  31. checkpoint-2000/tokenizer_config.json +57 -0
  32. checkpoint-2000/trainer_state.json +1433 -0
  33. checkpoint-2000/training_args.bin +3 -0
  34. checkpoint-2000/vocab.txt +0 -0
  35. checkpoint-3000/1_Pooling/config.json +10 -0
  36. checkpoint-3000/README.md +144 -0
  37. checkpoint-3000/config.json +31 -0
  38. checkpoint-3000/config_sentence_transformers.json +10 -0
  39. checkpoint-3000/model.safetensors +3 -0
  40. checkpoint-3000/modules.json +20 -0
  41. checkpoint-3000/optimizer.pt +3 -0
  42. checkpoint-3000/rng_state.pth +3 -0
  43. checkpoint-3000/scheduler.pt +3 -0
  44. checkpoint-3000/sentence_bert_config.json +4 -0
  45. checkpoint-3000/special_tokens_map.json +37 -0
  46. checkpoint-3000/tokenizer.json +0 -0
  47. checkpoint-3000/tokenizer_config.json +57 -0
  48. checkpoint-3000/trainer_state.json +2133 -0
  49. checkpoint-3000/training_args.bin +3 -0
  50. checkpoint-3000/vocab.txt +0 -0
1_Pooling/config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "word_embedding_dimension": 384,
3
+ "pooling_mode_cls_token": true,
4
+ "pooling_mode_mean_tokens": false,
5
+ "pooling_mode_max_tokens": false,
6
+ "pooling_mode_mean_sqrt_len_tokens": false,
7
+ "pooling_mode_weightedmean_tokens": false,
8
+ "pooling_mode_lasttoken": false,
9
+ "include_prompt": true
10
+ }
README.md ADDED
@@ -0,0 +1,144 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: []
3
+ library_name: sentence-transformers
4
+ tags:
5
+ - sentence-transformers
6
+ - sentence-similarity
7
+ - feature-extraction
8
+ datasets: []
9
+ widget: []
10
+ pipeline_tag: sentence-similarity
11
+ ---
12
+
13
+ # SentenceTransformer
14
+
15
+ This is a [sentence-transformers](https://www.SBERT.net) model trained. It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
16
+
17
+ ## Model Details
18
+
19
+ ### Model Description
20
+ - **Model Type:** Sentence Transformer
21
+ <!-- - **Base model:** [Unknown](https://huggingface.co/unknown) -->
22
+ - **Maximum Sequence Length:** 512 tokens
23
+ - **Output Dimensionality:** 384 tokens
24
+ - **Similarity Function:** Cosine Similarity
25
+ <!-- - **Training Dataset:** Unknown -->
26
+ <!-- - **Language:** Unknown -->
27
+ <!-- - **License:** Unknown -->
28
+
29
+ ### Model Sources
30
+
31
+ - **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
32
+ - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
33
+ - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
34
+
35
+ ### Full Model Architecture
36
+
37
+ ```
38
+ SentenceTransformer(
39
+ (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel
40
+ (1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
41
+ (2): Normalize()
42
+ )
43
+ ```
44
+
45
+ ## Usage
46
+
47
+ ### Direct Usage (Sentence Transformers)
48
+
49
+ First install the Sentence Transformers library:
50
+
51
+ ```bash
52
+ pip install -U sentence-transformers
53
+ ```
54
+
55
+ Then you can load this model and run inference.
56
+ ```python
57
+ from sentence_transformers import SentenceTransformer
58
+
59
+ # Download from the 🤗 Hub
60
+ model = SentenceTransformer("sentence_transformers_model_id")
61
+ # Run inference
62
+ sentences = [
63
+ 'The weather is lovely today.',
64
+ "It's so sunny outside!",
65
+ 'He drove to the stadium.',
66
+ ]
67
+ embeddings = model.encode(sentences)
68
+ print(embeddings.shape)
69
+ # [3, 384]
70
+
71
+ # Get the similarity scores for the embeddings
72
+ similarities = model.similarity(embeddings, embeddings)
73
+ print(similarities.shape)
74
+ # [3, 3]
75
+ ```
76
+
77
+ <!--
78
+ ### Direct Usage (Transformers)
79
+
80
+ <details><summary>Click to see the direct usage in Transformers</summary>
81
+
82
+ </details>
83
+ -->
84
+
85
+ <!--
86
+ ### Downstream Usage (Sentence Transformers)
87
+
88
+ You can finetune this model on your own dataset.
89
+
90
+ <details><summary>Click to expand</summary>
91
+
92
+ </details>
93
+ -->
94
+
95
+ <!--
96
+ ### Out-of-Scope Use
97
+
98
+ *List how the model may foreseeably be misused and address what users ought not to do with the model.*
99
+ -->
100
+
101
+ <!--
102
+ ## Bias, Risks and Limitations
103
+
104
+ *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
105
+ -->
106
+
107
+ <!--
108
+ ### Recommendations
109
+
110
+ *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
111
+ -->
112
+
113
+ ## Training Details
114
+
115
+ ### Framework Versions
116
+ - Python: 3.11.7
117
+ - Sentence Transformers: 3.0.1
118
+ - Transformers: 4.41.1
119
+ - PyTorch: 2.3.1+cu121
120
+ - Accelerate: 0.30.1
121
+ - Datasets: 2.19.1
122
+ - Tokenizers: 0.19.1
123
+
124
+ ## Citation
125
+
126
+ ### BibTeX
127
+
128
+ <!--
129
+ ## Glossary
130
+
131
+ *Clearly define terms in order to be accessible across audiences.*
132
+ -->
133
+
134
+ <!--
135
+ ## Model Card Authors
136
+
137
+ *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
138
+ -->
139
+
140
+ <!--
141
+ ## Model Card Contact
142
+
143
+ *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
144
+ -->
checkpoint-1000/1_Pooling/config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "word_embedding_dimension": 384,
3
+ "pooling_mode_cls_token": true,
4
+ "pooling_mode_mean_tokens": false,
5
+ "pooling_mode_max_tokens": false,
6
+ "pooling_mode_mean_sqrt_len_tokens": false,
7
+ "pooling_mode_weightedmean_tokens": false,
8
+ "pooling_mode_lasttoken": false,
9
+ "include_prompt": true
10
+ }
checkpoint-1000/README.md ADDED
@@ -0,0 +1,144 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: []
3
+ library_name: sentence-transformers
4
+ tags:
5
+ - sentence-transformers
6
+ - sentence-similarity
7
+ - feature-extraction
8
+ datasets: []
9
+ widget: []
10
+ pipeline_tag: sentence-similarity
11
+ ---
12
+
13
+ # SentenceTransformer
14
+
15
+ This is a [sentence-transformers](https://www.SBERT.net) model trained. It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
16
+
17
+ ## Model Details
18
+
19
+ ### Model Description
20
+ - **Model Type:** Sentence Transformer
21
+ <!-- - **Base model:** [Unknown](https://huggingface.co/unknown) -->
22
+ - **Maximum Sequence Length:** 512 tokens
23
+ - **Output Dimensionality:** 384 tokens
24
+ - **Similarity Function:** Cosine Similarity
25
+ <!-- - **Training Dataset:** Unknown -->
26
+ <!-- - **Language:** Unknown -->
27
+ <!-- - **License:** Unknown -->
28
+
29
+ ### Model Sources
30
+
31
+ - **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
32
+ - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
33
+ - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
34
+
35
+ ### Full Model Architecture
36
+
37
+ ```
38
+ SentenceTransformer(
39
+ (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel
40
+ (1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
41
+ (2): Normalize()
42
+ )
43
+ ```
44
+
45
+ ## Usage
46
+
47
+ ### Direct Usage (Sentence Transformers)
48
+
49
+ First install the Sentence Transformers library:
50
+
51
+ ```bash
52
+ pip install -U sentence-transformers
53
+ ```
54
+
55
+ Then you can load this model and run inference.
56
+ ```python
57
+ from sentence_transformers import SentenceTransformer
58
+
59
+ # Download from the 🤗 Hub
60
+ model = SentenceTransformer("sentence_transformers_model_id")
61
+ # Run inference
62
+ sentences = [
63
+ 'The weather is lovely today.',
64
+ "It's so sunny outside!",
65
+ 'He drove to the stadium.',
66
+ ]
67
+ embeddings = model.encode(sentences)
68
+ print(embeddings.shape)
69
+ # [3, 384]
70
+
71
+ # Get the similarity scores for the embeddings
72
+ similarities = model.similarity(embeddings, embeddings)
73
+ print(similarities.shape)
74
+ # [3, 3]
75
+ ```
76
+
77
+ <!--
78
+ ### Direct Usage (Transformers)
79
+
80
+ <details><summary>Click to see the direct usage in Transformers</summary>
81
+
82
+ </details>
83
+ -->
84
+
85
+ <!--
86
+ ### Downstream Usage (Sentence Transformers)
87
+
88
+ You can finetune this model on your own dataset.
89
+
90
+ <details><summary>Click to expand</summary>
91
+
92
+ </details>
93
+ -->
94
+
95
+ <!--
96
+ ### Out-of-Scope Use
97
+
98
+ *List how the model may foreseeably be misused and address what users ought not to do with the model.*
99
+ -->
100
+
101
+ <!--
102
+ ## Bias, Risks and Limitations
103
+
104
+ *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
105
+ -->
106
+
107
+ <!--
108
+ ### Recommendations
109
+
110
+ *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
111
+ -->
112
+
113
+ ## Training Details
114
+
115
+ ### Framework Versions
116
+ - Python: 3.11.7
117
+ - Sentence Transformers: 3.0.1
118
+ - Transformers: 4.41.1
119
+ - PyTorch: 2.3.1+cu121
120
+ - Accelerate: 0.30.1
121
+ - Datasets: 2.19.1
122
+ - Tokenizers: 0.19.1
123
+
124
+ ## Citation
125
+
126
+ ### BibTeX
127
+
128
+ <!--
129
+ ## Glossary
130
+
131
+ *Clearly define terms in order to be accessible across audiences.*
132
+ -->
133
+
134
+ <!--
135
+ ## Model Card Authors
136
+
137
+ *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
138
+ -->
139
+
140
+ <!--
141
+ ## Model Card Contact
142
+
143
+ *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
144
+ -->
checkpoint-1000/config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "finetuned_bge-small-en/checkpoint-1000",
3
+ "architectures": [
4
+ "BertModel"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 384,
11
+ "id2label": {
12
+ "0": "LABEL_0"
13
+ },
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 1536,
16
+ "label2id": {
17
+ "LABEL_0": 0
18
+ },
19
+ "layer_norm_eps": 1e-12,
20
+ "max_position_embeddings": 512,
21
+ "model_type": "bert",
22
+ "num_attention_heads": 12,
23
+ "num_hidden_layers": 12,
24
+ "pad_token_id": 0,
25
+ "position_embedding_type": "absolute",
26
+ "torch_dtype": "float32",
27
+ "transformers_version": "4.41.1",
28
+ "type_vocab_size": 2,
29
+ "use_cache": true,
30
+ "vocab_size": 30522
31
+ }
checkpoint-1000/config_sentence_transformers.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "__version__": {
3
+ "sentence_transformers": "3.0.1",
4
+ "transformers": "4.41.1",
5
+ "pytorch": "2.3.1+cu121"
6
+ },
7
+ "prompts": {},
8
+ "default_prompt_name": null,
9
+ "similarity_fn_name": null
10
+ }
checkpoint-1000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7a0b6575d5ad404c2975ad1dae11b8399239efa01fecc6e79d313db51eeb77ee
3
+ size 133462128
checkpoint-1000/modules.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "idx": 0,
4
+ "name": "0",
5
+ "path": "",
6
+ "type": "sentence_transformers.models.Transformer"
7
+ },
8
+ {
9
+ "idx": 1,
10
+ "name": "1",
11
+ "path": "1_Pooling",
12
+ "type": "sentence_transformers.models.Pooling"
13
+ },
14
+ {
15
+ "idx": 2,
16
+ "name": "2",
17
+ "path": "2_Normalize",
18
+ "type": "sentence_transformers.models.Normalize"
19
+ }
20
+ ]
checkpoint-1000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ea5f63b0f5d7a1f0818494ed76cf981154fb6b9b1b27ca02b227912be013cc47
3
+ size 265862074
checkpoint-1000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cd6aa4a5218ea0900cd51a461b3b79c0f263d6df39ee94cb99a6ba5a126cb262
3
+ size 14244
checkpoint-1000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5d4cfd452c221ec9df01151a258f429464ca32c658fabd86bf6b4b0fe117415b
3
+ size 1064
checkpoint-1000/sentence_bert_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "max_seq_length": 512,
3
+ "do_lower_case": false
4
+ }
checkpoint-1000/special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
checkpoint-1000/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1000/tokenizer_config.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "100": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "101": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "102": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "103": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": true,
45
+ "cls_token": "[CLS]",
46
+ "do_basic_tokenize": true,
47
+ "do_lower_case": true,
48
+ "mask_token": "[MASK]",
49
+ "model_max_length": 512,
50
+ "never_split": null,
51
+ "pad_token": "[PAD]",
52
+ "sep_token": "[SEP]",
53
+ "strip_accents": null,
54
+ "tokenize_chinese_chars": true,
55
+ "tokenizer_class": "BertTokenizer",
56
+ "unk_token": "[UNK]"
57
+ }
checkpoint-1000/trainer_state.json ADDED
@@ -0,0 +1,733 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 4.081632653061225,
5
+ "eval_steps": 500,
6
+ "global_step": 1000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.04081632653061224,
13
+ "grad_norm": 7.114395618438721,
14
+ "learning_rate": 9.981632653061225e-06,
15
+ "loss": 0.7362,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.08163265306122448,
20
+ "grad_norm": 11.572301864624023,
21
+ "learning_rate": 9.961224489795919e-06,
22
+ "loss": 0.8729,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.12244897959183673,
27
+ "grad_norm": 9.383491516113281,
28
+ "learning_rate": 9.940816326530614e-06,
29
+ "loss": 0.773,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.16326530612244897,
34
+ "grad_norm": 7.83120059967041,
35
+ "learning_rate": 9.920408163265307e-06,
36
+ "loss": 0.7817,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.20408163265306123,
41
+ "grad_norm": 10.92087173461914,
42
+ "learning_rate": 9.9e-06,
43
+ "loss": 0.6256,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.24489795918367346,
48
+ "grad_norm": 3.8826725482940674,
49
+ "learning_rate": 9.879591836734695e-06,
50
+ "loss": 0.5759,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.2857142857142857,
55
+ "grad_norm": 11.15483283996582,
56
+ "learning_rate": 9.859183673469388e-06,
57
+ "loss": 0.7333,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.32653061224489793,
62
+ "grad_norm": 11.470726013183594,
63
+ "learning_rate": 9.838775510204083e-06,
64
+ "loss": 0.5943,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.3673469387755102,
69
+ "grad_norm": 13.159674644470215,
70
+ "learning_rate": 9.818367346938777e-06,
71
+ "loss": 0.7804,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.40816326530612246,
76
+ "grad_norm": 9.58558464050293,
77
+ "learning_rate": 9.79795918367347e-06,
78
+ "loss": 0.6491,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.4489795918367347,
83
+ "grad_norm": 9.653897285461426,
84
+ "learning_rate": 9.777551020408163e-06,
85
+ "loss": 0.5919,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.4897959183673469,
90
+ "grad_norm": 8.117432594299316,
91
+ "learning_rate": 9.757142857142858e-06,
92
+ "loss": 0.4571,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.5306122448979592,
97
+ "grad_norm": 6.9328460693359375,
98
+ "learning_rate": 9.736734693877551e-06,
99
+ "loss": 0.6597,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.5714285714285714,
104
+ "grad_norm": 7.962501049041748,
105
+ "learning_rate": 9.716326530612246e-06,
106
+ "loss": 0.5132,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.6122448979591837,
111
+ "grad_norm": 10.508763313293457,
112
+ "learning_rate": 9.69591836734694e-06,
113
+ "loss": 0.6893,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.6530612244897959,
118
+ "grad_norm": 7.637253761291504,
119
+ "learning_rate": 9.675510204081635e-06,
120
+ "loss": 0.6142,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.6938775510204082,
125
+ "grad_norm": 10.0332670211792,
126
+ "learning_rate": 9.655102040816328e-06,
127
+ "loss": 0.582,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.7346938775510204,
132
+ "grad_norm": 8.150875091552734,
133
+ "learning_rate": 9.634693877551021e-06,
134
+ "loss": 0.477,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.7755102040816326,
139
+ "grad_norm": 10.330913543701172,
140
+ "learning_rate": 9.614285714285714e-06,
141
+ "loss": 0.5916,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.8163265306122449,
146
+ "grad_norm": 11.654999732971191,
147
+ "learning_rate": 9.593877551020408e-06,
148
+ "loss": 0.6236,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.8571428571428571,
153
+ "grad_norm": 8.048078536987305,
154
+ "learning_rate": 9.573469387755103e-06,
155
+ "loss": 0.6142,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.8979591836734694,
160
+ "grad_norm": 9.869592666625977,
161
+ "learning_rate": 9.553061224489798e-06,
162
+ "loss": 0.625,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.9387755102040817,
167
+ "grad_norm": 8.321409225463867,
168
+ "learning_rate": 9.532653061224491e-06,
169
+ "loss": 0.5767,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.9795918367346939,
174
+ "grad_norm": 7.6769256591796875,
175
+ "learning_rate": 9.512244897959184e-06,
176
+ "loss": 0.5134,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 1.0204081632653061,
181
+ "grad_norm": 9.609123229980469,
182
+ "learning_rate": 9.491836734693877e-06,
183
+ "loss": 0.5868,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 1.0612244897959184,
188
+ "grad_norm": 9.19683837890625,
189
+ "learning_rate": 9.471428571428572e-06,
190
+ "loss": 0.5215,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 1.1020408163265305,
195
+ "grad_norm": 7.328164577484131,
196
+ "learning_rate": 9.451020408163266e-06,
197
+ "loss": 0.5422,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 1.1428571428571428,
202
+ "grad_norm": 6.913904190063477,
203
+ "learning_rate": 9.430612244897959e-06,
204
+ "loss": 0.5214,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 1.183673469387755,
209
+ "grad_norm": 9.28811264038086,
210
+ "learning_rate": 9.410204081632654e-06,
211
+ "loss": 0.5319,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 1.2244897959183674,
216
+ "grad_norm": 9.132966041564941,
217
+ "learning_rate": 9.389795918367349e-06,
218
+ "loss": 0.5581,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 1.2653061224489797,
223
+ "grad_norm": 6.9722065925598145,
224
+ "learning_rate": 9.369387755102042e-06,
225
+ "loss": 0.431,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 1.306122448979592,
230
+ "grad_norm": 5.06177282333374,
231
+ "learning_rate": 9.348979591836736e-06,
232
+ "loss": 0.4583,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 1.346938775510204,
237
+ "grad_norm": 7.732840538024902,
238
+ "learning_rate": 9.328571428571429e-06,
239
+ "loss": 0.4194,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 1.3877551020408163,
244
+ "grad_norm": 8.94101333618164,
245
+ "learning_rate": 9.308163265306122e-06,
246
+ "loss": 0.4519,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 1.4285714285714286,
251
+ "grad_norm": 7.5437750816345215,
252
+ "learning_rate": 9.287755102040817e-06,
253
+ "loss": 0.5095,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 1.469387755102041,
258
+ "grad_norm": 5.702700138092041,
259
+ "learning_rate": 9.26734693877551e-06,
260
+ "loss": 0.3936,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 1.510204081632653,
265
+ "grad_norm": 9.153871536254883,
266
+ "learning_rate": 9.246938775510205e-06,
267
+ "loss": 0.4566,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 1.5510204081632653,
272
+ "grad_norm": 13.249794006347656,
273
+ "learning_rate": 9.226530612244899e-06,
274
+ "loss": 0.5216,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 1.5918367346938775,
279
+ "grad_norm": 7.065913200378418,
280
+ "learning_rate": 9.206122448979594e-06,
281
+ "loss": 0.4562,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 1.6326530612244898,
286
+ "grad_norm": 7.559301853179932,
287
+ "learning_rate": 9.185714285714287e-06,
288
+ "loss": 0.3883,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 1.6734693877551021,
293
+ "grad_norm": 12.103629112243652,
294
+ "learning_rate": 9.16530612244898e-06,
295
+ "loss": 0.4149,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 1.7142857142857144,
300
+ "grad_norm": 7.9720072746276855,
301
+ "learning_rate": 9.144897959183673e-06,
302
+ "loss": 0.4718,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 1.7551020408163265,
307
+ "grad_norm": 4.845782279968262,
308
+ "learning_rate": 9.124489795918368e-06,
309
+ "loss": 0.4304,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 1.7959183673469388,
314
+ "grad_norm": 6.954368591308594,
315
+ "learning_rate": 9.104081632653062e-06,
316
+ "loss": 0.3436,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 1.836734693877551,
321
+ "grad_norm": 4.751299858093262,
322
+ "learning_rate": 9.083673469387757e-06,
323
+ "loss": 0.4366,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 1.8775510204081631,
328
+ "grad_norm": 6.507364273071289,
329
+ "learning_rate": 9.06326530612245e-06,
330
+ "loss": 0.5794,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 1.9183673469387754,
335
+ "grad_norm": 8.891802787780762,
336
+ "learning_rate": 9.042857142857143e-06,
337
+ "loss": 0.4616,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 1.9591836734693877,
342
+ "grad_norm": 10.056327819824219,
343
+ "learning_rate": 9.022448979591838e-06,
344
+ "loss": 0.4946,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 2.0,
349
+ "grad_norm": 7.899660110473633,
350
+ "learning_rate": 9.002040816326531e-06,
351
+ "loss": 0.4437,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 2.0408163265306123,
356
+ "grad_norm": 6.761326313018799,
357
+ "learning_rate": 8.981632653061225e-06,
358
+ "loss": 0.4303,
359
+ "step": 500
360
+ },
361
+ {
362
+ "epoch": 2.0816326530612246,
363
+ "grad_norm": 8.639615058898926,
364
+ "learning_rate": 8.96122448979592e-06,
365
+ "loss": 0.3267,
366
+ "step": 510
367
+ },
368
+ {
369
+ "epoch": 2.122448979591837,
370
+ "grad_norm": 7.710758209228516,
371
+ "learning_rate": 8.940816326530613e-06,
372
+ "loss": 0.3559,
373
+ "step": 520
374
+ },
375
+ {
376
+ "epoch": 2.163265306122449,
377
+ "grad_norm": 6.812905311584473,
378
+ "learning_rate": 8.920408163265308e-06,
379
+ "loss": 0.4761,
380
+ "step": 530
381
+ },
382
+ {
383
+ "epoch": 2.204081632653061,
384
+ "grad_norm": 7.2431511878967285,
385
+ "learning_rate": 8.900000000000001e-06,
386
+ "loss": 0.405,
387
+ "step": 540
388
+ },
389
+ {
390
+ "epoch": 2.2448979591836733,
391
+ "grad_norm": 7.230724811553955,
392
+ "learning_rate": 8.879591836734694e-06,
393
+ "loss": 0.3638,
394
+ "step": 550
395
+ },
396
+ {
397
+ "epoch": 2.2857142857142856,
398
+ "grad_norm": 9.520208358764648,
399
+ "learning_rate": 8.859183673469388e-06,
400
+ "loss": 0.3473,
401
+ "step": 560
402
+ },
403
+ {
404
+ "epoch": 2.326530612244898,
405
+ "grad_norm": 7.048585414886475,
406
+ "learning_rate": 8.838775510204083e-06,
407
+ "loss": 0.3652,
408
+ "step": 570
409
+ },
410
+ {
411
+ "epoch": 2.36734693877551,
412
+ "grad_norm": 6.979404449462891,
413
+ "learning_rate": 8.818367346938776e-06,
414
+ "loss": 0.3855,
415
+ "step": 580
416
+ },
417
+ {
418
+ "epoch": 2.4081632653061225,
419
+ "grad_norm": 3.765305280685425,
420
+ "learning_rate": 8.797959183673471e-06,
421
+ "loss": 0.3452,
422
+ "step": 590
423
+ },
424
+ {
425
+ "epoch": 2.4489795918367347,
426
+ "grad_norm": 10.533697128295898,
427
+ "learning_rate": 8.777551020408164e-06,
428
+ "loss": 0.3874,
429
+ "step": 600
430
+ },
431
+ {
432
+ "epoch": 2.489795918367347,
433
+ "grad_norm": 8.108145713806152,
434
+ "learning_rate": 8.757142857142858e-06,
435
+ "loss": 0.3695,
436
+ "step": 610
437
+ },
438
+ {
439
+ "epoch": 2.5306122448979593,
440
+ "grad_norm": 7.947360992431641,
441
+ "learning_rate": 8.736734693877552e-06,
442
+ "loss": 0.408,
443
+ "step": 620
444
+ },
445
+ {
446
+ "epoch": 2.571428571428571,
447
+ "grad_norm": 7.8081374168396,
448
+ "learning_rate": 8.716326530612246e-06,
449
+ "loss": 0.4059,
450
+ "step": 630
451
+ },
452
+ {
453
+ "epoch": 2.612244897959184,
454
+ "grad_norm": 8.579155921936035,
455
+ "learning_rate": 8.695918367346939e-06,
456
+ "loss": 0.3934,
457
+ "step": 640
458
+ },
459
+ {
460
+ "epoch": 2.6530612244897958,
461
+ "grad_norm": 6.4387712478637695,
462
+ "learning_rate": 8.675510204081632e-06,
463
+ "loss": 0.4256,
464
+ "step": 650
465
+ },
466
+ {
467
+ "epoch": 2.693877551020408,
468
+ "grad_norm": 8.415692329406738,
469
+ "learning_rate": 8.655102040816327e-06,
470
+ "loss": 0.3453,
471
+ "step": 660
472
+ },
473
+ {
474
+ "epoch": 2.7346938775510203,
475
+ "grad_norm": 8.50904369354248,
476
+ "learning_rate": 8.63469387755102e-06,
477
+ "loss": 0.4766,
478
+ "step": 670
479
+ },
480
+ {
481
+ "epoch": 2.7755102040816326,
482
+ "grad_norm": 4.662519931793213,
483
+ "learning_rate": 8.614285714285716e-06,
484
+ "loss": 0.3973,
485
+ "step": 680
486
+ },
487
+ {
488
+ "epoch": 2.816326530612245,
489
+ "grad_norm": 6.288435935974121,
490
+ "learning_rate": 8.593877551020409e-06,
491
+ "loss": 0.4408,
492
+ "step": 690
493
+ },
494
+ {
495
+ "epoch": 2.857142857142857,
496
+ "grad_norm": 6.625838279724121,
497
+ "learning_rate": 8.573469387755102e-06,
498
+ "loss": 0.2908,
499
+ "step": 700
500
+ },
501
+ {
502
+ "epoch": 2.8979591836734695,
503
+ "grad_norm": 8.510032653808594,
504
+ "learning_rate": 8.553061224489797e-06,
505
+ "loss": 0.3813,
506
+ "step": 710
507
+ },
508
+ {
509
+ "epoch": 2.938775510204082,
510
+ "grad_norm": 11.82463264465332,
511
+ "learning_rate": 8.53265306122449e-06,
512
+ "loss": 0.4352,
513
+ "step": 720
514
+ },
515
+ {
516
+ "epoch": 2.979591836734694,
517
+ "grad_norm": 8.821819305419922,
518
+ "learning_rate": 8.512244897959184e-06,
519
+ "loss": 0.4318,
520
+ "step": 730
521
+ },
522
+ {
523
+ "epoch": 3.020408163265306,
524
+ "grad_norm": 8.010713577270508,
525
+ "learning_rate": 8.491836734693879e-06,
526
+ "loss": 0.2323,
527
+ "step": 740
528
+ },
529
+ {
530
+ "epoch": 3.061224489795918,
531
+ "grad_norm": 9.03991985321045,
532
+ "learning_rate": 8.471428571428572e-06,
533
+ "loss": 0.3603,
534
+ "step": 750
535
+ },
536
+ {
537
+ "epoch": 3.1020408163265305,
538
+ "grad_norm": 10.94204044342041,
539
+ "learning_rate": 8.451020408163267e-06,
540
+ "loss": 0.3576,
541
+ "step": 760
542
+ },
543
+ {
544
+ "epoch": 3.142857142857143,
545
+ "grad_norm": 7.89410924911499,
546
+ "learning_rate": 8.43061224489796e-06,
547
+ "loss": 0.2851,
548
+ "step": 770
549
+ },
550
+ {
551
+ "epoch": 3.183673469387755,
552
+ "grad_norm": 6.53656005859375,
553
+ "learning_rate": 8.410204081632653e-06,
554
+ "loss": 0.318,
555
+ "step": 780
556
+ },
557
+ {
558
+ "epoch": 3.2244897959183674,
559
+ "grad_norm": 6.487284183502197,
560
+ "learning_rate": 8.389795918367347e-06,
561
+ "loss": 0.317,
562
+ "step": 790
563
+ },
564
+ {
565
+ "epoch": 3.2653061224489797,
566
+ "grad_norm": 6.947931289672852,
567
+ "learning_rate": 8.369387755102042e-06,
568
+ "loss": 0.2879,
569
+ "step": 800
570
+ },
571
+ {
572
+ "epoch": 3.306122448979592,
573
+ "grad_norm": 4.166048526763916,
574
+ "learning_rate": 8.348979591836735e-06,
575
+ "loss": 0.3392,
576
+ "step": 810
577
+ },
578
+ {
579
+ "epoch": 3.3469387755102042,
580
+ "grad_norm": 9.974846839904785,
581
+ "learning_rate": 8.32857142857143e-06,
582
+ "loss": 0.3663,
583
+ "step": 820
584
+ },
585
+ {
586
+ "epoch": 3.387755102040816,
587
+ "grad_norm": 9.668428421020508,
588
+ "learning_rate": 8.308163265306123e-06,
589
+ "loss": 0.3212,
590
+ "step": 830
591
+ },
592
+ {
593
+ "epoch": 3.4285714285714284,
594
+ "grad_norm": 11.81507396697998,
595
+ "learning_rate": 8.287755102040816e-06,
596
+ "loss": 0.3241,
597
+ "step": 840
598
+ },
599
+ {
600
+ "epoch": 3.4693877551020407,
601
+ "grad_norm": 13.690321922302246,
602
+ "learning_rate": 8.267346938775511e-06,
603
+ "loss": 0.4535,
604
+ "step": 850
605
+ },
606
+ {
607
+ "epoch": 3.510204081632653,
608
+ "grad_norm": 11.042778968811035,
609
+ "learning_rate": 8.246938775510205e-06,
610
+ "loss": 0.3826,
611
+ "step": 860
612
+ },
613
+ {
614
+ "epoch": 3.5510204081632653,
615
+ "grad_norm": 8.57719612121582,
616
+ "learning_rate": 8.226530612244898e-06,
617
+ "loss": 0.3905,
618
+ "step": 870
619
+ },
620
+ {
621
+ "epoch": 3.5918367346938775,
622
+ "grad_norm": 7.843425750732422,
623
+ "learning_rate": 8.206122448979591e-06,
624
+ "loss": 0.3125,
625
+ "step": 880
626
+ },
627
+ {
628
+ "epoch": 3.63265306122449,
629
+ "grad_norm": 5.9236931800842285,
630
+ "learning_rate": 8.185714285714286e-06,
631
+ "loss": 0.3512,
632
+ "step": 890
633
+ },
634
+ {
635
+ "epoch": 3.673469387755102,
636
+ "grad_norm": 8.213603973388672,
637
+ "learning_rate": 8.165306122448981e-06,
638
+ "loss": 0.4094,
639
+ "step": 900
640
+ },
641
+ {
642
+ "epoch": 3.7142857142857144,
643
+ "grad_norm": 3.8083949089050293,
644
+ "learning_rate": 8.144897959183674e-06,
645
+ "loss": 0.2751,
646
+ "step": 910
647
+ },
648
+ {
649
+ "epoch": 3.7551020408163263,
650
+ "grad_norm": 12.339240074157715,
651
+ "learning_rate": 8.124489795918368e-06,
652
+ "loss": 0.3296,
653
+ "step": 920
654
+ },
655
+ {
656
+ "epoch": 3.795918367346939,
657
+ "grad_norm": 9.532052040100098,
658
+ "learning_rate": 8.104081632653061e-06,
659
+ "loss": 0.3033,
660
+ "step": 930
661
+ },
662
+ {
663
+ "epoch": 3.836734693877551,
664
+ "grad_norm": 6.307032108306885,
665
+ "learning_rate": 8.083673469387756e-06,
666
+ "loss": 0.3765,
667
+ "step": 940
668
+ },
669
+ {
670
+ "epoch": 3.877551020408163,
671
+ "grad_norm": 7.3003010749816895,
672
+ "learning_rate": 8.06326530612245e-06,
673
+ "loss": 0.2161,
674
+ "step": 950
675
+ },
676
+ {
677
+ "epoch": 3.9183673469387754,
678
+ "grad_norm": 7.6572699546813965,
679
+ "learning_rate": 8.042857142857143e-06,
680
+ "loss": 0.2886,
681
+ "step": 960
682
+ },
683
+ {
684
+ "epoch": 3.9591836734693877,
685
+ "grad_norm": 6.745776176452637,
686
+ "learning_rate": 8.022448979591838e-06,
687
+ "loss": 0.3376,
688
+ "step": 970
689
+ },
690
+ {
691
+ "epoch": 4.0,
692
+ "grad_norm": 10.482270240783691,
693
+ "learning_rate": 8.002040816326533e-06,
694
+ "loss": 0.2657,
695
+ "step": 980
696
+ },
697
+ {
698
+ "epoch": 4.040816326530612,
699
+ "grad_norm": 6.213717460632324,
700
+ "learning_rate": 7.981632653061226e-06,
701
+ "loss": 0.2596,
702
+ "step": 990
703
+ },
704
+ {
705
+ "epoch": 4.081632653061225,
706
+ "grad_norm": 10.256094932556152,
707
+ "learning_rate": 7.961224489795919e-06,
708
+ "loss": 0.2338,
709
+ "step": 1000
710
+ }
711
+ ],
712
+ "logging_steps": 10,
713
+ "max_steps": 4900,
714
+ "num_input_tokens_seen": 0,
715
+ "num_train_epochs": 20,
716
+ "save_steps": 1000,
717
+ "stateful_callbacks": {
718
+ "TrainerControl": {
719
+ "args": {
720
+ "should_epoch_stop": false,
721
+ "should_evaluate": false,
722
+ "should_log": false,
723
+ "should_save": true,
724
+ "should_training_stop": false
725
+ },
726
+ "attributes": {}
727
+ }
728
+ },
729
+ "total_flos": 0.0,
730
+ "train_batch_size": 30,
731
+ "trial_name": null,
732
+ "trial_params": null
733
+ }
checkpoint-1000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c4dbbc953b5f35ee9087cab8da231051761999ae7db38443ee8d71e351fa7db6
3
+ size 5304
checkpoint-1000/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-2000/1_Pooling/config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "word_embedding_dimension": 384,
3
+ "pooling_mode_cls_token": true,
4
+ "pooling_mode_mean_tokens": false,
5
+ "pooling_mode_max_tokens": false,
6
+ "pooling_mode_mean_sqrt_len_tokens": false,
7
+ "pooling_mode_weightedmean_tokens": false,
8
+ "pooling_mode_lasttoken": false,
9
+ "include_prompt": true
10
+ }
checkpoint-2000/README.md ADDED
@@ -0,0 +1,144 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: []
3
+ library_name: sentence-transformers
4
+ tags:
5
+ - sentence-transformers
6
+ - sentence-similarity
7
+ - feature-extraction
8
+ datasets: []
9
+ widget: []
10
+ pipeline_tag: sentence-similarity
11
+ ---
12
+
13
+ # SentenceTransformer
14
+
15
+ This is a [sentence-transformers](https://www.SBERT.net) model trained. It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
16
+
17
+ ## Model Details
18
+
19
+ ### Model Description
20
+ - **Model Type:** Sentence Transformer
21
+ <!-- - **Base model:** [Unknown](https://huggingface.co/unknown) -->
22
+ - **Maximum Sequence Length:** 512 tokens
23
+ - **Output Dimensionality:** 384 tokens
24
+ - **Similarity Function:** Cosine Similarity
25
+ <!-- - **Training Dataset:** Unknown -->
26
+ <!-- - **Language:** Unknown -->
27
+ <!-- - **License:** Unknown -->
28
+
29
+ ### Model Sources
30
+
31
+ - **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
32
+ - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
33
+ - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
34
+
35
+ ### Full Model Architecture
36
+
37
+ ```
38
+ SentenceTransformer(
39
+ (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel
40
+ (1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
41
+ (2): Normalize()
42
+ )
43
+ ```
44
+
45
+ ## Usage
46
+
47
+ ### Direct Usage (Sentence Transformers)
48
+
49
+ First install the Sentence Transformers library:
50
+
51
+ ```bash
52
+ pip install -U sentence-transformers
53
+ ```
54
+
55
+ Then you can load this model and run inference.
56
+ ```python
57
+ from sentence_transformers import SentenceTransformer
58
+
59
+ # Download from the 🤗 Hub
60
+ model = SentenceTransformer("sentence_transformers_model_id")
61
+ # Run inference
62
+ sentences = [
63
+ 'The weather is lovely today.',
64
+ "It's so sunny outside!",
65
+ 'He drove to the stadium.',
66
+ ]
67
+ embeddings = model.encode(sentences)
68
+ print(embeddings.shape)
69
+ # [3, 384]
70
+
71
+ # Get the similarity scores for the embeddings
72
+ similarities = model.similarity(embeddings, embeddings)
73
+ print(similarities.shape)
74
+ # [3, 3]
75
+ ```
76
+
77
+ <!--
78
+ ### Direct Usage (Transformers)
79
+
80
+ <details><summary>Click to see the direct usage in Transformers</summary>
81
+
82
+ </details>
83
+ -->
84
+
85
+ <!--
86
+ ### Downstream Usage (Sentence Transformers)
87
+
88
+ You can finetune this model on your own dataset.
89
+
90
+ <details><summary>Click to expand</summary>
91
+
92
+ </details>
93
+ -->
94
+
95
+ <!--
96
+ ### Out-of-Scope Use
97
+
98
+ *List how the model may foreseeably be misused and address what users ought not to do with the model.*
99
+ -->
100
+
101
+ <!--
102
+ ## Bias, Risks and Limitations
103
+
104
+ *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
105
+ -->
106
+
107
+ <!--
108
+ ### Recommendations
109
+
110
+ *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
111
+ -->
112
+
113
+ ## Training Details
114
+
115
+ ### Framework Versions
116
+ - Python: 3.11.7
117
+ - Sentence Transformers: 3.0.1
118
+ - Transformers: 4.41.1
119
+ - PyTorch: 2.3.1+cu121
120
+ - Accelerate: 0.30.1
121
+ - Datasets: 2.19.1
122
+ - Tokenizers: 0.19.1
123
+
124
+ ## Citation
125
+
126
+ ### BibTeX
127
+
128
+ <!--
129
+ ## Glossary
130
+
131
+ *Clearly define terms in order to be accessible across audiences.*
132
+ -->
133
+
134
+ <!--
135
+ ## Model Card Authors
136
+
137
+ *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
138
+ -->
139
+
140
+ <!--
141
+ ## Model Card Contact
142
+
143
+ *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
144
+ -->
checkpoint-2000/config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "finetuned_bge-small-en/checkpoint-2000",
3
+ "architectures": [
4
+ "BertModel"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 384,
11
+ "id2label": {
12
+ "0": "LABEL_0"
13
+ },
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 1536,
16
+ "label2id": {
17
+ "LABEL_0": 0
18
+ },
19
+ "layer_norm_eps": 1e-12,
20
+ "max_position_embeddings": 512,
21
+ "model_type": "bert",
22
+ "num_attention_heads": 12,
23
+ "num_hidden_layers": 12,
24
+ "pad_token_id": 0,
25
+ "position_embedding_type": "absolute",
26
+ "torch_dtype": "float32",
27
+ "transformers_version": "4.41.1",
28
+ "type_vocab_size": 2,
29
+ "use_cache": true,
30
+ "vocab_size": 30522
31
+ }
checkpoint-2000/config_sentence_transformers.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "__version__": {
3
+ "sentence_transformers": "3.0.1",
4
+ "transformers": "4.41.1",
5
+ "pytorch": "2.3.1+cu121"
6
+ },
7
+ "prompts": {},
8
+ "default_prompt_name": null,
9
+ "similarity_fn_name": null
10
+ }
checkpoint-2000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd43e85b9e44062c6f87ccdc22e6e38d7a6e4ec35bf027c27a78f93f84ea8d5d
3
+ size 133462128
checkpoint-2000/modules.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "idx": 0,
4
+ "name": "0",
5
+ "path": "",
6
+ "type": "sentence_transformers.models.Transformer"
7
+ },
8
+ {
9
+ "idx": 1,
10
+ "name": "1",
11
+ "path": "1_Pooling",
12
+ "type": "sentence_transformers.models.Pooling"
13
+ },
14
+ {
15
+ "idx": 2,
16
+ "name": "2",
17
+ "path": "2_Normalize",
18
+ "type": "sentence_transformers.models.Normalize"
19
+ }
20
+ ]
checkpoint-2000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:76dfb41871552f30a89086c30f05da480916e3e14317abe91c9f80c3a33449c1
3
+ size 265862074
checkpoint-2000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:13eba22d0db4d08542ed30abb043b8fb91ed7eec91a0e4eb2dfc7e1449d51a62
3
+ size 14244
checkpoint-2000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:596247e860dbf4a62f36b957ab6d136bde0c00b1e94a15f0db4f466907e5ce26
3
+ size 1064
checkpoint-2000/sentence_bert_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "max_seq_length": 512,
3
+ "do_lower_case": false
4
+ }
checkpoint-2000/special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
checkpoint-2000/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-2000/tokenizer_config.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "100": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "101": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "102": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "103": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": true,
45
+ "cls_token": "[CLS]",
46
+ "do_basic_tokenize": true,
47
+ "do_lower_case": true,
48
+ "mask_token": "[MASK]",
49
+ "model_max_length": 512,
50
+ "never_split": null,
51
+ "pad_token": "[PAD]",
52
+ "sep_token": "[SEP]",
53
+ "strip_accents": null,
54
+ "tokenize_chinese_chars": true,
55
+ "tokenizer_class": "BertTokenizer",
56
+ "unk_token": "[UNK]"
57
+ }
checkpoint-2000/trainer_state.json ADDED
@@ -0,0 +1,1433 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 8.16326530612245,
5
+ "eval_steps": 500,
6
+ "global_step": 2000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.04081632653061224,
13
+ "grad_norm": 7.114395618438721,
14
+ "learning_rate": 9.981632653061225e-06,
15
+ "loss": 0.7362,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.08163265306122448,
20
+ "grad_norm": 11.572301864624023,
21
+ "learning_rate": 9.961224489795919e-06,
22
+ "loss": 0.8729,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.12244897959183673,
27
+ "grad_norm": 9.383491516113281,
28
+ "learning_rate": 9.940816326530614e-06,
29
+ "loss": 0.773,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.16326530612244897,
34
+ "grad_norm": 7.83120059967041,
35
+ "learning_rate": 9.920408163265307e-06,
36
+ "loss": 0.7817,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.20408163265306123,
41
+ "grad_norm": 10.92087173461914,
42
+ "learning_rate": 9.9e-06,
43
+ "loss": 0.6256,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.24489795918367346,
48
+ "grad_norm": 3.8826725482940674,
49
+ "learning_rate": 9.879591836734695e-06,
50
+ "loss": 0.5759,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.2857142857142857,
55
+ "grad_norm": 11.15483283996582,
56
+ "learning_rate": 9.859183673469388e-06,
57
+ "loss": 0.7333,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.32653061224489793,
62
+ "grad_norm": 11.470726013183594,
63
+ "learning_rate": 9.838775510204083e-06,
64
+ "loss": 0.5943,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.3673469387755102,
69
+ "grad_norm": 13.159674644470215,
70
+ "learning_rate": 9.818367346938777e-06,
71
+ "loss": 0.7804,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.40816326530612246,
76
+ "grad_norm": 9.58558464050293,
77
+ "learning_rate": 9.79795918367347e-06,
78
+ "loss": 0.6491,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.4489795918367347,
83
+ "grad_norm": 9.653897285461426,
84
+ "learning_rate": 9.777551020408163e-06,
85
+ "loss": 0.5919,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.4897959183673469,
90
+ "grad_norm": 8.117432594299316,
91
+ "learning_rate": 9.757142857142858e-06,
92
+ "loss": 0.4571,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.5306122448979592,
97
+ "grad_norm": 6.9328460693359375,
98
+ "learning_rate": 9.736734693877551e-06,
99
+ "loss": 0.6597,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.5714285714285714,
104
+ "grad_norm": 7.962501049041748,
105
+ "learning_rate": 9.716326530612246e-06,
106
+ "loss": 0.5132,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.6122448979591837,
111
+ "grad_norm": 10.508763313293457,
112
+ "learning_rate": 9.69591836734694e-06,
113
+ "loss": 0.6893,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.6530612244897959,
118
+ "grad_norm": 7.637253761291504,
119
+ "learning_rate": 9.675510204081635e-06,
120
+ "loss": 0.6142,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.6938775510204082,
125
+ "grad_norm": 10.0332670211792,
126
+ "learning_rate": 9.655102040816328e-06,
127
+ "loss": 0.582,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.7346938775510204,
132
+ "grad_norm": 8.150875091552734,
133
+ "learning_rate": 9.634693877551021e-06,
134
+ "loss": 0.477,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.7755102040816326,
139
+ "grad_norm": 10.330913543701172,
140
+ "learning_rate": 9.614285714285714e-06,
141
+ "loss": 0.5916,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.8163265306122449,
146
+ "grad_norm": 11.654999732971191,
147
+ "learning_rate": 9.593877551020408e-06,
148
+ "loss": 0.6236,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.8571428571428571,
153
+ "grad_norm": 8.048078536987305,
154
+ "learning_rate": 9.573469387755103e-06,
155
+ "loss": 0.6142,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.8979591836734694,
160
+ "grad_norm": 9.869592666625977,
161
+ "learning_rate": 9.553061224489798e-06,
162
+ "loss": 0.625,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.9387755102040817,
167
+ "grad_norm": 8.321409225463867,
168
+ "learning_rate": 9.532653061224491e-06,
169
+ "loss": 0.5767,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.9795918367346939,
174
+ "grad_norm": 7.6769256591796875,
175
+ "learning_rate": 9.512244897959184e-06,
176
+ "loss": 0.5134,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 1.0204081632653061,
181
+ "grad_norm": 9.609123229980469,
182
+ "learning_rate": 9.491836734693877e-06,
183
+ "loss": 0.5868,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 1.0612244897959184,
188
+ "grad_norm": 9.19683837890625,
189
+ "learning_rate": 9.471428571428572e-06,
190
+ "loss": 0.5215,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 1.1020408163265305,
195
+ "grad_norm": 7.328164577484131,
196
+ "learning_rate": 9.451020408163266e-06,
197
+ "loss": 0.5422,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 1.1428571428571428,
202
+ "grad_norm": 6.913904190063477,
203
+ "learning_rate": 9.430612244897959e-06,
204
+ "loss": 0.5214,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 1.183673469387755,
209
+ "grad_norm": 9.28811264038086,
210
+ "learning_rate": 9.410204081632654e-06,
211
+ "loss": 0.5319,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 1.2244897959183674,
216
+ "grad_norm": 9.132966041564941,
217
+ "learning_rate": 9.389795918367349e-06,
218
+ "loss": 0.5581,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 1.2653061224489797,
223
+ "grad_norm": 6.9722065925598145,
224
+ "learning_rate": 9.369387755102042e-06,
225
+ "loss": 0.431,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 1.306122448979592,
230
+ "grad_norm": 5.06177282333374,
231
+ "learning_rate": 9.348979591836736e-06,
232
+ "loss": 0.4583,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 1.346938775510204,
237
+ "grad_norm": 7.732840538024902,
238
+ "learning_rate": 9.328571428571429e-06,
239
+ "loss": 0.4194,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 1.3877551020408163,
244
+ "grad_norm": 8.94101333618164,
245
+ "learning_rate": 9.308163265306122e-06,
246
+ "loss": 0.4519,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 1.4285714285714286,
251
+ "grad_norm": 7.5437750816345215,
252
+ "learning_rate": 9.287755102040817e-06,
253
+ "loss": 0.5095,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 1.469387755102041,
258
+ "grad_norm": 5.702700138092041,
259
+ "learning_rate": 9.26734693877551e-06,
260
+ "loss": 0.3936,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 1.510204081632653,
265
+ "grad_norm": 9.153871536254883,
266
+ "learning_rate": 9.246938775510205e-06,
267
+ "loss": 0.4566,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 1.5510204081632653,
272
+ "grad_norm": 13.249794006347656,
273
+ "learning_rate": 9.226530612244899e-06,
274
+ "loss": 0.5216,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 1.5918367346938775,
279
+ "grad_norm": 7.065913200378418,
280
+ "learning_rate": 9.206122448979594e-06,
281
+ "loss": 0.4562,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 1.6326530612244898,
286
+ "grad_norm": 7.559301853179932,
287
+ "learning_rate": 9.185714285714287e-06,
288
+ "loss": 0.3883,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 1.6734693877551021,
293
+ "grad_norm": 12.103629112243652,
294
+ "learning_rate": 9.16530612244898e-06,
295
+ "loss": 0.4149,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 1.7142857142857144,
300
+ "grad_norm": 7.9720072746276855,
301
+ "learning_rate": 9.144897959183673e-06,
302
+ "loss": 0.4718,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 1.7551020408163265,
307
+ "grad_norm": 4.845782279968262,
308
+ "learning_rate": 9.124489795918368e-06,
309
+ "loss": 0.4304,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 1.7959183673469388,
314
+ "grad_norm": 6.954368591308594,
315
+ "learning_rate": 9.104081632653062e-06,
316
+ "loss": 0.3436,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 1.836734693877551,
321
+ "grad_norm": 4.751299858093262,
322
+ "learning_rate": 9.083673469387757e-06,
323
+ "loss": 0.4366,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 1.8775510204081631,
328
+ "grad_norm": 6.507364273071289,
329
+ "learning_rate": 9.06326530612245e-06,
330
+ "loss": 0.5794,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 1.9183673469387754,
335
+ "grad_norm": 8.891802787780762,
336
+ "learning_rate": 9.042857142857143e-06,
337
+ "loss": 0.4616,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 1.9591836734693877,
342
+ "grad_norm": 10.056327819824219,
343
+ "learning_rate": 9.022448979591838e-06,
344
+ "loss": 0.4946,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 2.0,
349
+ "grad_norm": 7.899660110473633,
350
+ "learning_rate": 9.002040816326531e-06,
351
+ "loss": 0.4437,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 2.0408163265306123,
356
+ "grad_norm": 6.761326313018799,
357
+ "learning_rate": 8.981632653061225e-06,
358
+ "loss": 0.4303,
359
+ "step": 500
360
+ },
361
+ {
362
+ "epoch": 2.0816326530612246,
363
+ "grad_norm": 8.639615058898926,
364
+ "learning_rate": 8.96122448979592e-06,
365
+ "loss": 0.3267,
366
+ "step": 510
367
+ },
368
+ {
369
+ "epoch": 2.122448979591837,
370
+ "grad_norm": 7.710758209228516,
371
+ "learning_rate": 8.940816326530613e-06,
372
+ "loss": 0.3559,
373
+ "step": 520
374
+ },
375
+ {
376
+ "epoch": 2.163265306122449,
377
+ "grad_norm": 6.812905311584473,
378
+ "learning_rate": 8.920408163265308e-06,
379
+ "loss": 0.4761,
380
+ "step": 530
381
+ },
382
+ {
383
+ "epoch": 2.204081632653061,
384
+ "grad_norm": 7.2431511878967285,
385
+ "learning_rate": 8.900000000000001e-06,
386
+ "loss": 0.405,
387
+ "step": 540
388
+ },
389
+ {
390
+ "epoch": 2.2448979591836733,
391
+ "grad_norm": 7.230724811553955,
392
+ "learning_rate": 8.879591836734694e-06,
393
+ "loss": 0.3638,
394
+ "step": 550
395
+ },
396
+ {
397
+ "epoch": 2.2857142857142856,
398
+ "grad_norm": 9.520208358764648,
399
+ "learning_rate": 8.859183673469388e-06,
400
+ "loss": 0.3473,
401
+ "step": 560
402
+ },
403
+ {
404
+ "epoch": 2.326530612244898,
405
+ "grad_norm": 7.048585414886475,
406
+ "learning_rate": 8.838775510204083e-06,
407
+ "loss": 0.3652,
408
+ "step": 570
409
+ },
410
+ {
411
+ "epoch": 2.36734693877551,
412
+ "grad_norm": 6.979404449462891,
413
+ "learning_rate": 8.818367346938776e-06,
414
+ "loss": 0.3855,
415
+ "step": 580
416
+ },
417
+ {
418
+ "epoch": 2.4081632653061225,
419
+ "grad_norm": 3.765305280685425,
420
+ "learning_rate": 8.797959183673471e-06,
421
+ "loss": 0.3452,
422
+ "step": 590
423
+ },
424
+ {
425
+ "epoch": 2.4489795918367347,
426
+ "grad_norm": 10.533697128295898,
427
+ "learning_rate": 8.777551020408164e-06,
428
+ "loss": 0.3874,
429
+ "step": 600
430
+ },
431
+ {
432
+ "epoch": 2.489795918367347,
433
+ "grad_norm": 8.108145713806152,
434
+ "learning_rate": 8.757142857142858e-06,
435
+ "loss": 0.3695,
436
+ "step": 610
437
+ },
438
+ {
439
+ "epoch": 2.5306122448979593,
440
+ "grad_norm": 7.947360992431641,
441
+ "learning_rate": 8.736734693877552e-06,
442
+ "loss": 0.408,
443
+ "step": 620
444
+ },
445
+ {
446
+ "epoch": 2.571428571428571,
447
+ "grad_norm": 7.8081374168396,
448
+ "learning_rate": 8.716326530612246e-06,
449
+ "loss": 0.4059,
450
+ "step": 630
451
+ },
452
+ {
453
+ "epoch": 2.612244897959184,
454
+ "grad_norm": 8.579155921936035,
455
+ "learning_rate": 8.695918367346939e-06,
456
+ "loss": 0.3934,
457
+ "step": 640
458
+ },
459
+ {
460
+ "epoch": 2.6530612244897958,
461
+ "grad_norm": 6.4387712478637695,
462
+ "learning_rate": 8.675510204081632e-06,
463
+ "loss": 0.4256,
464
+ "step": 650
465
+ },
466
+ {
467
+ "epoch": 2.693877551020408,
468
+ "grad_norm": 8.415692329406738,
469
+ "learning_rate": 8.655102040816327e-06,
470
+ "loss": 0.3453,
471
+ "step": 660
472
+ },
473
+ {
474
+ "epoch": 2.7346938775510203,
475
+ "grad_norm": 8.50904369354248,
476
+ "learning_rate": 8.63469387755102e-06,
477
+ "loss": 0.4766,
478
+ "step": 670
479
+ },
480
+ {
481
+ "epoch": 2.7755102040816326,
482
+ "grad_norm": 4.662519931793213,
483
+ "learning_rate": 8.614285714285716e-06,
484
+ "loss": 0.3973,
485
+ "step": 680
486
+ },
487
+ {
488
+ "epoch": 2.816326530612245,
489
+ "grad_norm": 6.288435935974121,
490
+ "learning_rate": 8.593877551020409e-06,
491
+ "loss": 0.4408,
492
+ "step": 690
493
+ },
494
+ {
495
+ "epoch": 2.857142857142857,
496
+ "grad_norm": 6.625838279724121,
497
+ "learning_rate": 8.573469387755102e-06,
498
+ "loss": 0.2908,
499
+ "step": 700
500
+ },
501
+ {
502
+ "epoch": 2.8979591836734695,
503
+ "grad_norm": 8.510032653808594,
504
+ "learning_rate": 8.553061224489797e-06,
505
+ "loss": 0.3813,
506
+ "step": 710
507
+ },
508
+ {
509
+ "epoch": 2.938775510204082,
510
+ "grad_norm": 11.82463264465332,
511
+ "learning_rate": 8.53265306122449e-06,
512
+ "loss": 0.4352,
513
+ "step": 720
514
+ },
515
+ {
516
+ "epoch": 2.979591836734694,
517
+ "grad_norm": 8.821819305419922,
518
+ "learning_rate": 8.512244897959184e-06,
519
+ "loss": 0.4318,
520
+ "step": 730
521
+ },
522
+ {
523
+ "epoch": 3.020408163265306,
524
+ "grad_norm": 8.010713577270508,
525
+ "learning_rate": 8.491836734693879e-06,
526
+ "loss": 0.2323,
527
+ "step": 740
528
+ },
529
+ {
530
+ "epoch": 3.061224489795918,
531
+ "grad_norm": 9.03991985321045,
532
+ "learning_rate": 8.471428571428572e-06,
533
+ "loss": 0.3603,
534
+ "step": 750
535
+ },
536
+ {
537
+ "epoch": 3.1020408163265305,
538
+ "grad_norm": 10.94204044342041,
539
+ "learning_rate": 8.451020408163267e-06,
540
+ "loss": 0.3576,
541
+ "step": 760
542
+ },
543
+ {
544
+ "epoch": 3.142857142857143,
545
+ "grad_norm": 7.89410924911499,
546
+ "learning_rate": 8.43061224489796e-06,
547
+ "loss": 0.2851,
548
+ "step": 770
549
+ },
550
+ {
551
+ "epoch": 3.183673469387755,
552
+ "grad_norm": 6.53656005859375,
553
+ "learning_rate": 8.410204081632653e-06,
554
+ "loss": 0.318,
555
+ "step": 780
556
+ },
557
+ {
558
+ "epoch": 3.2244897959183674,
559
+ "grad_norm": 6.487284183502197,
560
+ "learning_rate": 8.389795918367347e-06,
561
+ "loss": 0.317,
562
+ "step": 790
563
+ },
564
+ {
565
+ "epoch": 3.2653061224489797,
566
+ "grad_norm": 6.947931289672852,
567
+ "learning_rate": 8.369387755102042e-06,
568
+ "loss": 0.2879,
569
+ "step": 800
570
+ },
571
+ {
572
+ "epoch": 3.306122448979592,
573
+ "grad_norm": 4.166048526763916,
574
+ "learning_rate": 8.348979591836735e-06,
575
+ "loss": 0.3392,
576
+ "step": 810
577
+ },
578
+ {
579
+ "epoch": 3.3469387755102042,
580
+ "grad_norm": 9.974846839904785,
581
+ "learning_rate": 8.32857142857143e-06,
582
+ "loss": 0.3663,
583
+ "step": 820
584
+ },
585
+ {
586
+ "epoch": 3.387755102040816,
587
+ "grad_norm": 9.668428421020508,
588
+ "learning_rate": 8.308163265306123e-06,
589
+ "loss": 0.3212,
590
+ "step": 830
591
+ },
592
+ {
593
+ "epoch": 3.4285714285714284,
594
+ "grad_norm": 11.81507396697998,
595
+ "learning_rate": 8.287755102040816e-06,
596
+ "loss": 0.3241,
597
+ "step": 840
598
+ },
599
+ {
600
+ "epoch": 3.4693877551020407,
601
+ "grad_norm": 13.690321922302246,
602
+ "learning_rate": 8.267346938775511e-06,
603
+ "loss": 0.4535,
604
+ "step": 850
605
+ },
606
+ {
607
+ "epoch": 3.510204081632653,
608
+ "grad_norm": 11.042778968811035,
609
+ "learning_rate": 8.246938775510205e-06,
610
+ "loss": 0.3826,
611
+ "step": 860
612
+ },
613
+ {
614
+ "epoch": 3.5510204081632653,
615
+ "grad_norm": 8.57719612121582,
616
+ "learning_rate": 8.226530612244898e-06,
617
+ "loss": 0.3905,
618
+ "step": 870
619
+ },
620
+ {
621
+ "epoch": 3.5918367346938775,
622
+ "grad_norm": 7.843425750732422,
623
+ "learning_rate": 8.206122448979591e-06,
624
+ "loss": 0.3125,
625
+ "step": 880
626
+ },
627
+ {
628
+ "epoch": 3.63265306122449,
629
+ "grad_norm": 5.9236931800842285,
630
+ "learning_rate": 8.185714285714286e-06,
631
+ "loss": 0.3512,
632
+ "step": 890
633
+ },
634
+ {
635
+ "epoch": 3.673469387755102,
636
+ "grad_norm": 8.213603973388672,
637
+ "learning_rate": 8.165306122448981e-06,
638
+ "loss": 0.4094,
639
+ "step": 900
640
+ },
641
+ {
642
+ "epoch": 3.7142857142857144,
643
+ "grad_norm": 3.8083949089050293,
644
+ "learning_rate": 8.144897959183674e-06,
645
+ "loss": 0.2751,
646
+ "step": 910
647
+ },
648
+ {
649
+ "epoch": 3.7551020408163263,
650
+ "grad_norm": 12.339240074157715,
651
+ "learning_rate": 8.124489795918368e-06,
652
+ "loss": 0.3296,
653
+ "step": 920
654
+ },
655
+ {
656
+ "epoch": 3.795918367346939,
657
+ "grad_norm": 9.532052040100098,
658
+ "learning_rate": 8.104081632653061e-06,
659
+ "loss": 0.3033,
660
+ "step": 930
661
+ },
662
+ {
663
+ "epoch": 3.836734693877551,
664
+ "grad_norm": 6.307032108306885,
665
+ "learning_rate": 8.083673469387756e-06,
666
+ "loss": 0.3765,
667
+ "step": 940
668
+ },
669
+ {
670
+ "epoch": 3.877551020408163,
671
+ "grad_norm": 7.3003010749816895,
672
+ "learning_rate": 8.06326530612245e-06,
673
+ "loss": 0.2161,
674
+ "step": 950
675
+ },
676
+ {
677
+ "epoch": 3.9183673469387754,
678
+ "grad_norm": 7.6572699546813965,
679
+ "learning_rate": 8.042857142857143e-06,
680
+ "loss": 0.2886,
681
+ "step": 960
682
+ },
683
+ {
684
+ "epoch": 3.9591836734693877,
685
+ "grad_norm": 6.745776176452637,
686
+ "learning_rate": 8.022448979591838e-06,
687
+ "loss": 0.3376,
688
+ "step": 970
689
+ },
690
+ {
691
+ "epoch": 4.0,
692
+ "grad_norm": 10.482270240783691,
693
+ "learning_rate": 8.002040816326533e-06,
694
+ "loss": 0.2657,
695
+ "step": 980
696
+ },
697
+ {
698
+ "epoch": 4.040816326530612,
699
+ "grad_norm": 6.213717460632324,
700
+ "learning_rate": 7.981632653061226e-06,
701
+ "loss": 0.2596,
702
+ "step": 990
703
+ },
704
+ {
705
+ "epoch": 4.081632653061225,
706
+ "grad_norm": 10.256094932556152,
707
+ "learning_rate": 7.961224489795919e-06,
708
+ "loss": 0.2338,
709
+ "step": 1000
710
+ },
711
+ {
712
+ "epoch": 4.122448979591836,
713
+ "grad_norm": 8.640167236328125,
714
+ "learning_rate": 7.940816326530612e-06,
715
+ "loss": 0.2936,
716
+ "step": 1010
717
+ },
718
+ {
719
+ "epoch": 4.163265306122449,
720
+ "grad_norm": 6.750300407409668,
721
+ "learning_rate": 7.920408163265306e-06,
722
+ "loss": 0.3511,
723
+ "step": 1020
724
+ },
725
+ {
726
+ "epoch": 4.204081632653061,
727
+ "grad_norm": 9.488007545471191,
728
+ "learning_rate": 7.9e-06,
729
+ "loss": 0.2913,
730
+ "step": 1030
731
+ },
732
+ {
733
+ "epoch": 4.244897959183674,
734
+ "grad_norm": 4.2671003341674805,
735
+ "learning_rate": 7.879591836734694e-06,
736
+ "loss": 0.2768,
737
+ "step": 1040
738
+ },
739
+ {
740
+ "epoch": 4.285714285714286,
741
+ "grad_norm": 5.375782489776611,
742
+ "learning_rate": 7.859183673469389e-06,
743
+ "loss": 0.274,
744
+ "step": 1050
745
+ },
746
+ {
747
+ "epoch": 4.326530612244898,
748
+ "grad_norm": 10.316515922546387,
749
+ "learning_rate": 7.838775510204082e-06,
750
+ "loss": 0.2373,
751
+ "step": 1060
752
+ },
753
+ {
754
+ "epoch": 4.36734693877551,
755
+ "grad_norm": 6.733712673187256,
756
+ "learning_rate": 7.818367346938777e-06,
757
+ "loss": 0.32,
758
+ "step": 1070
759
+ },
760
+ {
761
+ "epoch": 4.408163265306122,
762
+ "grad_norm": 6.432683944702148,
763
+ "learning_rate": 7.79795918367347e-06,
764
+ "loss": 0.3149,
765
+ "step": 1080
766
+ },
767
+ {
768
+ "epoch": 4.448979591836735,
769
+ "grad_norm": 8.520115852355957,
770
+ "learning_rate": 7.777551020408164e-06,
771
+ "loss": 0.3369,
772
+ "step": 1090
773
+ },
774
+ {
775
+ "epoch": 4.489795918367347,
776
+ "grad_norm": 3.52677845954895,
777
+ "learning_rate": 7.757142857142857e-06,
778
+ "loss": 0.2584,
779
+ "step": 1100
780
+ },
781
+ {
782
+ "epoch": 4.530612244897959,
783
+ "grad_norm": 9.642569541931152,
784
+ "learning_rate": 7.736734693877552e-06,
785
+ "loss": 0.3076,
786
+ "step": 1110
787
+ },
788
+ {
789
+ "epoch": 4.571428571428571,
790
+ "grad_norm": 11.233070373535156,
791
+ "learning_rate": 7.716326530612245e-06,
792
+ "loss": 0.3263,
793
+ "step": 1120
794
+ },
795
+ {
796
+ "epoch": 4.612244897959184,
797
+ "grad_norm": 7.919038772583008,
798
+ "learning_rate": 7.69591836734694e-06,
799
+ "loss": 0.3152,
800
+ "step": 1130
801
+ },
802
+ {
803
+ "epoch": 4.653061224489796,
804
+ "grad_norm": 7.116144180297852,
805
+ "learning_rate": 7.675510204081633e-06,
806
+ "loss": 0.2809,
807
+ "step": 1140
808
+ },
809
+ {
810
+ "epoch": 4.6938775510204085,
811
+ "grad_norm": 6.757585525512695,
812
+ "learning_rate": 7.655102040816327e-06,
813
+ "loss": 0.2642,
814
+ "step": 1150
815
+ },
816
+ {
817
+ "epoch": 4.73469387755102,
818
+ "grad_norm": 9.032824516296387,
819
+ "learning_rate": 7.634693877551022e-06,
820
+ "loss": 0.3546,
821
+ "step": 1160
822
+ },
823
+ {
824
+ "epoch": 4.775510204081632,
825
+ "grad_norm": 7.837385654449463,
826
+ "learning_rate": 7.614285714285715e-06,
827
+ "loss": 0.3196,
828
+ "step": 1170
829
+ },
830
+ {
831
+ "epoch": 4.816326530612245,
832
+ "grad_norm": 5.3134846687316895,
833
+ "learning_rate": 7.593877551020409e-06,
834
+ "loss": 0.2068,
835
+ "step": 1180
836
+ },
837
+ {
838
+ "epoch": 4.857142857142857,
839
+ "grad_norm": 8.036781311035156,
840
+ "learning_rate": 7.573469387755102e-06,
841
+ "loss": 0.24,
842
+ "step": 1190
843
+ },
844
+ {
845
+ "epoch": 4.8979591836734695,
846
+ "grad_norm": 12.13339900970459,
847
+ "learning_rate": 7.5530612244897965e-06,
848
+ "loss": 0.2795,
849
+ "step": 1200
850
+ },
851
+ {
852
+ "epoch": 4.938775510204081,
853
+ "grad_norm": 5.253749370574951,
854
+ "learning_rate": 7.532653061224491e-06,
855
+ "loss": 0.2667,
856
+ "step": 1210
857
+ },
858
+ {
859
+ "epoch": 4.979591836734694,
860
+ "grad_norm": 7.921318531036377,
861
+ "learning_rate": 7.512244897959185e-06,
862
+ "loss": 0.3218,
863
+ "step": 1220
864
+ },
865
+ {
866
+ "epoch": 5.020408163265306,
867
+ "grad_norm": 9.056379318237305,
868
+ "learning_rate": 7.491836734693878e-06,
869
+ "loss": 0.2324,
870
+ "step": 1230
871
+ },
872
+ {
873
+ "epoch": 5.061224489795919,
874
+ "grad_norm": 5.0674920082092285,
875
+ "learning_rate": 7.471428571428571e-06,
876
+ "loss": 0.3063,
877
+ "step": 1240
878
+ },
879
+ {
880
+ "epoch": 5.1020408163265305,
881
+ "grad_norm": 12.247352600097656,
882
+ "learning_rate": 7.451020408163266e-06,
883
+ "loss": 0.3469,
884
+ "step": 1250
885
+ },
886
+ {
887
+ "epoch": 5.142857142857143,
888
+ "grad_norm": 7.278679847717285,
889
+ "learning_rate": 7.43061224489796e-06,
890
+ "loss": 0.2851,
891
+ "step": 1260
892
+ },
893
+ {
894
+ "epoch": 5.183673469387755,
895
+ "grad_norm": 11.491589546203613,
896
+ "learning_rate": 7.410204081632654e-06,
897
+ "loss": 0.245,
898
+ "step": 1270
899
+ },
900
+ {
901
+ "epoch": 5.224489795918367,
902
+ "grad_norm": 8.652173042297363,
903
+ "learning_rate": 7.389795918367347e-06,
904
+ "loss": 0.2328,
905
+ "step": 1280
906
+ },
907
+ {
908
+ "epoch": 5.26530612244898,
909
+ "grad_norm": 7.953402519226074,
910
+ "learning_rate": 7.369387755102041e-06,
911
+ "loss": 0.2784,
912
+ "step": 1290
913
+ },
914
+ {
915
+ "epoch": 5.3061224489795915,
916
+ "grad_norm": 6.98084831237793,
917
+ "learning_rate": 7.348979591836736e-06,
918
+ "loss": 0.2379,
919
+ "step": 1300
920
+ },
921
+ {
922
+ "epoch": 5.346938775510204,
923
+ "grad_norm": 7.400093078613281,
924
+ "learning_rate": 7.328571428571429e-06,
925
+ "loss": 0.2057,
926
+ "step": 1310
927
+ },
928
+ {
929
+ "epoch": 5.387755102040816,
930
+ "grad_norm": 4.131850719451904,
931
+ "learning_rate": 7.308163265306123e-06,
932
+ "loss": 0.2792,
933
+ "step": 1320
934
+ },
935
+ {
936
+ "epoch": 5.428571428571429,
937
+ "grad_norm": 10.25373363494873,
938
+ "learning_rate": 7.287755102040817e-06,
939
+ "loss": 0.2642,
940
+ "step": 1330
941
+ },
942
+ {
943
+ "epoch": 5.469387755102041,
944
+ "grad_norm": 5.801217079162598,
945
+ "learning_rate": 7.267346938775512e-06,
946
+ "loss": 0.2301,
947
+ "step": 1340
948
+ },
949
+ {
950
+ "epoch": 5.510204081632653,
951
+ "grad_norm": 3.3520116806030273,
952
+ "learning_rate": 7.246938775510205e-06,
953
+ "loss": 0.2728,
954
+ "step": 1350
955
+ },
956
+ {
957
+ "epoch": 5.551020408163265,
958
+ "grad_norm": 5.335054874420166,
959
+ "learning_rate": 7.226530612244898e-06,
960
+ "loss": 0.2812,
961
+ "step": 1360
962
+ },
963
+ {
964
+ "epoch": 5.591836734693878,
965
+ "grad_norm": 4.713162899017334,
966
+ "learning_rate": 7.206122448979592e-06,
967
+ "loss": 0.221,
968
+ "step": 1370
969
+ },
970
+ {
971
+ "epoch": 5.63265306122449,
972
+ "grad_norm": 8.641826629638672,
973
+ "learning_rate": 7.185714285714286e-06,
974
+ "loss": 0.2204,
975
+ "step": 1380
976
+ },
977
+ {
978
+ "epoch": 5.673469387755102,
979
+ "grad_norm": 4.286067485809326,
980
+ "learning_rate": 7.165306122448981e-06,
981
+ "loss": 0.2446,
982
+ "step": 1390
983
+ },
984
+ {
985
+ "epoch": 5.714285714285714,
986
+ "grad_norm": 7.071314811706543,
987
+ "learning_rate": 7.144897959183674e-06,
988
+ "loss": 0.1788,
989
+ "step": 1400
990
+ },
991
+ {
992
+ "epoch": 5.755102040816326,
993
+ "grad_norm": 9.792560577392578,
994
+ "learning_rate": 7.124489795918368e-06,
995
+ "loss": 0.272,
996
+ "step": 1410
997
+ },
998
+ {
999
+ "epoch": 5.795918367346939,
1000
+ "grad_norm": 6.726722240447998,
1001
+ "learning_rate": 7.104081632653061e-06,
1002
+ "loss": 0.18,
1003
+ "step": 1420
1004
+ },
1005
+ {
1006
+ "epoch": 5.836734693877551,
1007
+ "grad_norm": 7.378534317016602,
1008
+ "learning_rate": 7.083673469387755e-06,
1009
+ "loss": 0.2308,
1010
+ "step": 1430
1011
+ },
1012
+ {
1013
+ "epoch": 5.877551020408164,
1014
+ "grad_norm": 10.169758796691895,
1015
+ "learning_rate": 7.0632653061224495e-06,
1016
+ "loss": 0.296,
1017
+ "step": 1440
1018
+ },
1019
+ {
1020
+ "epoch": 5.918367346938775,
1021
+ "grad_norm": 5.440324306488037,
1022
+ "learning_rate": 7.042857142857144e-06,
1023
+ "loss": 0.2725,
1024
+ "step": 1450
1025
+ },
1026
+ {
1027
+ "epoch": 5.959183673469388,
1028
+ "grad_norm": 8.175464630126953,
1029
+ "learning_rate": 7.022448979591837e-06,
1030
+ "loss": 0.2251,
1031
+ "step": 1460
1032
+ },
1033
+ {
1034
+ "epoch": 6.0,
1035
+ "grad_norm": 6.7653398513793945,
1036
+ "learning_rate": 7.002040816326531e-06,
1037
+ "loss": 0.3411,
1038
+ "step": 1470
1039
+ },
1040
+ {
1041
+ "epoch": 6.040816326530612,
1042
+ "grad_norm": 6.0152716636657715,
1043
+ "learning_rate": 6.981632653061225e-06,
1044
+ "loss": 0.2071,
1045
+ "step": 1480
1046
+ },
1047
+ {
1048
+ "epoch": 6.081632653061225,
1049
+ "grad_norm": 8.066520690917969,
1050
+ "learning_rate": 6.961224489795919e-06,
1051
+ "loss": 0.1639,
1052
+ "step": 1490
1053
+ },
1054
+ {
1055
+ "epoch": 6.122448979591836,
1056
+ "grad_norm": 4.027646541595459,
1057
+ "learning_rate": 6.940816326530613e-06,
1058
+ "loss": 0.2069,
1059
+ "step": 1500
1060
+ },
1061
+ {
1062
+ "epoch": 6.163265306122449,
1063
+ "grad_norm": 5.61140251159668,
1064
+ "learning_rate": 6.920408163265307e-06,
1065
+ "loss": 0.2413,
1066
+ "step": 1510
1067
+ },
1068
+ {
1069
+ "epoch": 6.204081632653061,
1070
+ "grad_norm": 9.809159278869629,
1071
+ "learning_rate": 6.9e-06,
1072
+ "loss": 0.26,
1073
+ "step": 1520
1074
+ },
1075
+ {
1076
+ "epoch": 6.244897959183674,
1077
+ "grad_norm": 6.755568504333496,
1078
+ "learning_rate": 6.879591836734695e-06,
1079
+ "loss": 0.1965,
1080
+ "step": 1530
1081
+ },
1082
+ {
1083
+ "epoch": 6.285714285714286,
1084
+ "grad_norm": 4.21774435043335,
1085
+ "learning_rate": 6.859183673469388e-06,
1086
+ "loss": 0.1959,
1087
+ "step": 1540
1088
+ },
1089
+ {
1090
+ "epoch": 6.326530612244898,
1091
+ "grad_norm": 5.166352272033691,
1092
+ "learning_rate": 6.838775510204082e-06,
1093
+ "loss": 0.1941,
1094
+ "step": 1550
1095
+ },
1096
+ {
1097
+ "epoch": 6.36734693877551,
1098
+ "grad_norm": 10.346336364746094,
1099
+ "learning_rate": 6.818367346938776e-06,
1100
+ "loss": 0.2278,
1101
+ "step": 1560
1102
+ },
1103
+ {
1104
+ "epoch": 6.408163265306122,
1105
+ "grad_norm": 7.703672409057617,
1106
+ "learning_rate": 6.797959183673471e-06,
1107
+ "loss": 0.2854,
1108
+ "step": 1570
1109
+ },
1110
+ {
1111
+ "epoch": 6.448979591836735,
1112
+ "grad_norm": 9.366389274597168,
1113
+ "learning_rate": 6.777551020408164e-06,
1114
+ "loss": 0.2229,
1115
+ "step": 1580
1116
+ },
1117
+ {
1118
+ "epoch": 6.489795918367347,
1119
+ "grad_norm": 7.013561248779297,
1120
+ "learning_rate": 6.757142857142858e-06,
1121
+ "loss": 0.2732,
1122
+ "step": 1590
1123
+ },
1124
+ {
1125
+ "epoch": 6.530612244897959,
1126
+ "grad_norm": 5.82119083404541,
1127
+ "learning_rate": 6.736734693877551e-06,
1128
+ "loss": 0.2389,
1129
+ "step": 1600
1130
+ },
1131
+ {
1132
+ "epoch": 6.571428571428571,
1133
+ "grad_norm": 8.112947463989258,
1134
+ "learning_rate": 6.716326530612245e-06,
1135
+ "loss": 0.193,
1136
+ "step": 1610
1137
+ },
1138
+ {
1139
+ "epoch": 6.612244897959184,
1140
+ "grad_norm": 8.238506317138672,
1141
+ "learning_rate": 6.6959183673469396e-06,
1142
+ "loss": 0.2344,
1143
+ "step": 1620
1144
+ },
1145
+ {
1146
+ "epoch": 6.653061224489796,
1147
+ "grad_norm": 6.531697750091553,
1148
+ "learning_rate": 6.675510204081634e-06,
1149
+ "loss": 0.2146,
1150
+ "step": 1630
1151
+ },
1152
+ {
1153
+ "epoch": 6.6938775510204085,
1154
+ "grad_norm": 6.525634288787842,
1155
+ "learning_rate": 6.655102040816327e-06,
1156
+ "loss": 0.1993,
1157
+ "step": 1640
1158
+ },
1159
+ {
1160
+ "epoch": 6.73469387755102,
1161
+ "grad_norm": 8.847986221313477,
1162
+ "learning_rate": 6.63469387755102e-06,
1163
+ "loss": 0.2884,
1164
+ "step": 1650
1165
+ },
1166
+ {
1167
+ "epoch": 6.775510204081632,
1168
+ "grad_norm": 9.006918907165527,
1169
+ "learning_rate": 6.614285714285715e-06,
1170
+ "loss": 0.2173,
1171
+ "step": 1660
1172
+ },
1173
+ {
1174
+ "epoch": 6.816326530612245,
1175
+ "grad_norm": 9.229476928710938,
1176
+ "learning_rate": 6.593877551020409e-06,
1177
+ "loss": 0.201,
1178
+ "step": 1670
1179
+ },
1180
+ {
1181
+ "epoch": 6.857142857142857,
1182
+ "grad_norm": 5.367541313171387,
1183
+ "learning_rate": 6.573469387755103e-06,
1184
+ "loss": 0.2505,
1185
+ "step": 1680
1186
+ },
1187
+ {
1188
+ "epoch": 6.8979591836734695,
1189
+ "grad_norm": 7.771108150482178,
1190
+ "learning_rate": 6.553061224489796e-06,
1191
+ "loss": 0.2225,
1192
+ "step": 1690
1193
+ },
1194
+ {
1195
+ "epoch": 6.938775510204081,
1196
+ "grad_norm": 5.306410789489746,
1197
+ "learning_rate": 6.53265306122449e-06,
1198
+ "loss": 0.2549,
1199
+ "step": 1700
1200
+ },
1201
+ {
1202
+ "epoch": 6.979591836734694,
1203
+ "grad_norm": 4.359670162200928,
1204
+ "learning_rate": 6.512244897959185e-06,
1205
+ "loss": 0.2585,
1206
+ "step": 1710
1207
+ },
1208
+ {
1209
+ "epoch": 7.020408163265306,
1210
+ "grad_norm": 4.528923034667969,
1211
+ "learning_rate": 6.491836734693878e-06,
1212
+ "loss": 0.2489,
1213
+ "step": 1720
1214
+ },
1215
+ {
1216
+ "epoch": 7.061224489795919,
1217
+ "grad_norm": 4.760287761688232,
1218
+ "learning_rate": 6.4714285714285715e-06,
1219
+ "loss": 0.1955,
1220
+ "step": 1730
1221
+ },
1222
+ {
1223
+ "epoch": 7.1020408163265305,
1224
+ "grad_norm": 9.205543518066406,
1225
+ "learning_rate": 6.451020408163266e-06,
1226
+ "loss": 0.2283,
1227
+ "step": 1740
1228
+ },
1229
+ {
1230
+ "epoch": 7.142857142857143,
1231
+ "grad_norm": 5.910974025726318,
1232
+ "learning_rate": 6.43061224489796e-06,
1233
+ "loss": 0.2308,
1234
+ "step": 1750
1235
+ },
1236
+ {
1237
+ "epoch": 7.183673469387755,
1238
+ "grad_norm": 5.042090892791748,
1239
+ "learning_rate": 6.410204081632654e-06,
1240
+ "loss": 0.1883,
1241
+ "step": 1760
1242
+ },
1243
+ {
1244
+ "epoch": 7.224489795918367,
1245
+ "grad_norm": 5.0842742919921875,
1246
+ "learning_rate": 6.389795918367347e-06,
1247
+ "loss": 0.1942,
1248
+ "step": 1770
1249
+ },
1250
+ {
1251
+ "epoch": 7.26530612244898,
1252
+ "grad_norm": 9.219313621520996,
1253
+ "learning_rate": 6.369387755102041e-06,
1254
+ "loss": 0.1946,
1255
+ "step": 1780
1256
+ },
1257
+ {
1258
+ "epoch": 7.3061224489795915,
1259
+ "grad_norm": 7.5656208992004395,
1260
+ "learning_rate": 6.348979591836735e-06,
1261
+ "loss": 0.2016,
1262
+ "step": 1790
1263
+ },
1264
+ {
1265
+ "epoch": 7.346938775510204,
1266
+ "grad_norm": 5.8213653564453125,
1267
+ "learning_rate": 6.3285714285714296e-06,
1268
+ "loss": 0.2808,
1269
+ "step": 1800
1270
+ },
1271
+ {
1272
+ "epoch": 7.387755102040816,
1273
+ "grad_norm": 8.004778861999512,
1274
+ "learning_rate": 6.308163265306123e-06,
1275
+ "loss": 0.2393,
1276
+ "step": 1810
1277
+ },
1278
+ {
1279
+ "epoch": 7.428571428571429,
1280
+ "grad_norm": 7.5001139640808105,
1281
+ "learning_rate": 6.287755102040817e-06,
1282
+ "loss": 0.2007,
1283
+ "step": 1820
1284
+ },
1285
+ {
1286
+ "epoch": 7.469387755102041,
1287
+ "grad_norm": 9.618229866027832,
1288
+ "learning_rate": 6.26734693877551e-06,
1289
+ "loss": 0.2464,
1290
+ "step": 1830
1291
+ },
1292
+ {
1293
+ "epoch": 7.510204081632653,
1294
+ "grad_norm": 7.257756233215332,
1295
+ "learning_rate": 6.246938775510205e-06,
1296
+ "loss": 0.1692,
1297
+ "step": 1840
1298
+ },
1299
+ {
1300
+ "epoch": 7.551020408163265,
1301
+ "grad_norm": 7.658279895782471,
1302
+ "learning_rate": 6.2265306122448985e-06,
1303
+ "loss": 0.2367,
1304
+ "step": 1850
1305
+ },
1306
+ {
1307
+ "epoch": 7.591836734693878,
1308
+ "grad_norm": 6.590469837188721,
1309
+ "learning_rate": 6.206122448979593e-06,
1310
+ "loss": 0.2508,
1311
+ "step": 1860
1312
+ },
1313
+ {
1314
+ "epoch": 7.63265306122449,
1315
+ "grad_norm": 8.601705551147461,
1316
+ "learning_rate": 6.185714285714286e-06,
1317
+ "loss": 0.2168,
1318
+ "step": 1870
1319
+ },
1320
+ {
1321
+ "epoch": 7.673469387755102,
1322
+ "grad_norm": 6.144942283630371,
1323
+ "learning_rate": 6.16530612244898e-06,
1324
+ "loss": 0.2308,
1325
+ "step": 1880
1326
+ },
1327
+ {
1328
+ "epoch": 7.714285714285714,
1329
+ "grad_norm": 3.256690502166748,
1330
+ "learning_rate": 6.144897959183674e-06,
1331
+ "loss": 0.1791,
1332
+ "step": 1890
1333
+ },
1334
+ {
1335
+ "epoch": 7.755102040816326,
1336
+ "grad_norm": 6.810645580291748,
1337
+ "learning_rate": 6.124489795918368e-06,
1338
+ "loss": 0.2335,
1339
+ "step": 1900
1340
+ },
1341
+ {
1342
+ "epoch": 7.795918367346939,
1343
+ "grad_norm": 3.4259018898010254,
1344
+ "learning_rate": 6.1040816326530616e-06,
1345
+ "loss": 0.212,
1346
+ "step": 1910
1347
+ },
1348
+ {
1349
+ "epoch": 7.836734693877551,
1350
+ "grad_norm": 8.353039741516113,
1351
+ "learning_rate": 6.083673469387756e-06,
1352
+ "loss": 0.2312,
1353
+ "step": 1920
1354
+ },
1355
+ {
1356
+ "epoch": 7.877551020408164,
1357
+ "grad_norm": 5.548733711242676,
1358
+ "learning_rate": 6.06326530612245e-06,
1359
+ "loss": 0.2666,
1360
+ "step": 1930
1361
+ },
1362
+ {
1363
+ "epoch": 7.918367346938775,
1364
+ "grad_norm": 8.386053085327148,
1365
+ "learning_rate": 6.042857142857144e-06,
1366
+ "loss": 0.2543,
1367
+ "step": 1940
1368
+ },
1369
+ {
1370
+ "epoch": 7.959183673469388,
1371
+ "grad_norm": 7.863219261169434,
1372
+ "learning_rate": 6.022448979591837e-06,
1373
+ "loss": 0.2171,
1374
+ "step": 1950
1375
+ },
1376
+ {
1377
+ "epoch": 8.0,
1378
+ "grad_norm": 5.328917503356934,
1379
+ "learning_rate": 6.0020408163265305e-06,
1380
+ "loss": 0.1592,
1381
+ "step": 1960
1382
+ },
1383
+ {
1384
+ "epoch": 8.040816326530612,
1385
+ "grad_norm": 11.456307411193848,
1386
+ "learning_rate": 5.981632653061225e-06,
1387
+ "loss": 0.2345,
1388
+ "step": 1970
1389
+ },
1390
+ {
1391
+ "epoch": 8.081632653061224,
1392
+ "grad_norm": 3.9219276905059814,
1393
+ "learning_rate": 5.96122448979592e-06,
1394
+ "loss": 0.228,
1395
+ "step": 1980
1396
+ },
1397
+ {
1398
+ "epoch": 8.122448979591837,
1399
+ "grad_norm": 7.155372142791748,
1400
+ "learning_rate": 5.940816326530613e-06,
1401
+ "loss": 0.1879,
1402
+ "step": 1990
1403
+ },
1404
+ {
1405
+ "epoch": 8.16326530612245,
1406
+ "grad_norm": 8.530526161193848,
1407
+ "learning_rate": 5.920408163265306e-06,
1408
+ "loss": 0.1808,
1409
+ "step": 2000
1410
+ }
1411
+ ],
1412
+ "logging_steps": 10,
1413
+ "max_steps": 4900,
1414
+ "num_input_tokens_seen": 0,
1415
+ "num_train_epochs": 20,
1416
+ "save_steps": 1000,
1417
+ "stateful_callbacks": {
1418
+ "TrainerControl": {
1419
+ "args": {
1420
+ "should_epoch_stop": false,
1421
+ "should_evaluate": false,
1422
+ "should_log": false,
1423
+ "should_save": true,
1424
+ "should_training_stop": false
1425
+ },
1426
+ "attributes": {}
1427
+ }
1428
+ },
1429
+ "total_flos": 0.0,
1430
+ "train_batch_size": 30,
1431
+ "trial_name": null,
1432
+ "trial_params": null
1433
+ }
checkpoint-2000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c4dbbc953b5f35ee9087cab8da231051761999ae7db38443ee8d71e351fa7db6
3
+ size 5304
checkpoint-2000/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-3000/1_Pooling/config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "word_embedding_dimension": 384,
3
+ "pooling_mode_cls_token": true,
4
+ "pooling_mode_mean_tokens": false,
5
+ "pooling_mode_max_tokens": false,
6
+ "pooling_mode_mean_sqrt_len_tokens": false,
7
+ "pooling_mode_weightedmean_tokens": false,
8
+ "pooling_mode_lasttoken": false,
9
+ "include_prompt": true
10
+ }
checkpoint-3000/README.md ADDED
@@ -0,0 +1,144 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: []
3
+ library_name: sentence-transformers
4
+ tags:
5
+ - sentence-transformers
6
+ - sentence-similarity
7
+ - feature-extraction
8
+ datasets: []
9
+ widget: []
10
+ pipeline_tag: sentence-similarity
11
+ ---
12
+
13
+ # SentenceTransformer
14
+
15
+ This is a [sentence-transformers](https://www.SBERT.net) model trained. It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
16
+
17
+ ## Model Details
18
+
19
+ ### Model Description
20
+ - **Model Type:** Sentence Transformer
21
+ <!-- - **Base model:** [Unknown](https://huggingface.co/unknown) -->
22
+ - **Maximum Sequence Length:** 512 tokens
23
+ - **Output Dimensionality:** 384 tokens
24
+ - **Similarity Function:** Cosine Similarity
25
+ <!-- - **Training Dataset:** Unknown -->
26
+ <!-- - **Language:** Unknown -->
27
+ <!-- - **License:** Unknown -->
28
+
29
+ ### Model Sources
30
+
31
+ - **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
32
+ - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
33
+ - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
34
+
35
+ ### Full Model Architecture
36
+
37
+ ```
38
+ SentenceTransformer(
39
+ (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel
40
+ (1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
41
+ (2): Normalize()
42
+ )
43
+ ```
44
+
45
+ ## Usage
46
+
47
+ ### Direct Usage (Sentence Transformers)
48
+
49
+ First install the Sentence Transformers library:
50
+
51
+ ```bash
52
+ pip install -U sentence-transformers
53
+ ```
54
+
55
+ Then you can load this model and run inference.
56
+ ```python
57
+ from sentence_transformers import SentenceTransformer
58
+
59
+ # Download from the 🤗 Hub
60
+ model = SentenceTransformer("sentence_transformers_model_id")
61
+ # Run inference
62
+ sentences = [
63
+ 'The weather is lovely today.',
64
+ "It's so sunny outside!",
65
+ 'He drove to the stadium.',
66
+ ]
67
+ embeddings = model.encode(sentences)
68
+ print(embeddings.shape)
69
+ # [3, 384]
70
+
71
+ # Get the similarity scores for the embeddings
72
+ similarities = model.similarity(embeddings, embeddings)
73
+ print(similarities.shape)
74
+ # [3, 3]
75
+ ```
76
+
77
+ <!--
78
+ ### Direct Usage (Transformers)
79
+
80
+ <details><summary>Click to see the direct usage in Transformers</summary>
81
+
82
+ </details>
83
+ -->
84
+
85
+ <!--
86
+ ### Downstream Usage (Sentence Transformers)
87
+
88
+ You can finetune this model on your own dataset.
89
+
90
+ <details><summary>Click to expand</summary>
91
+
92
+ </details>
93
+ -->
94
+
95
+ <!--
96
+ ### Out-of-Scope Use
97
+
98
+ *List how the model may foreseeably be misused and address what users ought not to do with the model.*
99
+ -->
100
+
101
+ <!--
102
+ ## Bias, Risks and Limitations
103
+
104
+ *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
105
+ -->
106
+
107
+ <!--
108
+ ### Recommendations
109
+
110
+ *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
111
+ -->
112
+
113
+ ## Training Details
114
+
115
+ ### Framework Versions
116
+ - Python: 3.11.7
117
+ - Sentence Transformers: 3.0.1
118
+ - Transformers: 4.41.1
119
+ - PyTorch: 2.3.1+cu121
120
+ - Accelerate: 0.30.1
121
+ - Datasets: 2.19.1
122
+ - Tokenizers: 0.19.1
123
+
124
+ ## Citation
125
+
126
+ ### BibTeX
127
+
128
+ <!--
129
+ ## Glossary
130
+
131
+ *Clearly define terms in order to be accessible across audiences.*
132
+ -->
133
+
134
+ <!--
135
+ ## Model Card Authors
136
+
137
+ *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
138
+ -->
139
+
140
+ <!--
141
+ ## Model Card Contact
142
+
143
+ *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
144
+ -->
checkpoint-3000/config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "finetuned_bge-small-en/checkpoint-3000",
3
+ "architectures": [
4
+ "BertModel"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 384,
11
+ "id2label": {
12
+ "0": "LABEL_0"
13
+ },
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 1536,
16
+ "label2id": {
17
+ "LABEL_0": 0
18
+ },
19
+ "layer_norm_eps": 1e-12,
20
+ "max_position_embeddings": 512,
21
+ "model_type": "bert",
22
+ "num_attention_heads": 12,
23
+ "num_hidden_layers": 12,
24
+ "pad_token_id": 0,
25
+ "position_embedding_type": "absolute",
26
+ "torch_dtype": "float32",
27
+ "transformers_version": "4.41.1",
28
+ "type_vocab_size": 2,
29
+ "use_cache": true,
30
+ "vocab_size": 30522
31
+ }
checkpoint-3000/config_sentence_transformers.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "__version__": {
3
+ "sentence_transformers": "3.0.1",
4
+ "transformers": "4.41.1",
5
+ "pytorch": "2.3.1+cu121"
6
+ },
7
+ "prompts": {},
8
+ "default_prompt_name": null,
9
+ "similarity_fn_name": null
10
+ }
checkpoint-3000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:11bd87ee5aa1cc8f767165ed6c4d41f1b8438ba5fbe20204c8bb9881cf08f9ba
3
+ size 133462128
checkpoint-3000/modules.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "idx": 0,
4
+ "name": "0",
5
+ "path": "",
6
+ "type": "sentence_transformers.models.Transformer"
7
+ },
8
+ {
9
+ "idx": 1,
10
+ "name": "1",
11
+ "path": "1_Pooling",
12
+ "type": "sentence_transformers.models.Pooling"
13
+ },
14
+ {
15
+ "idx": 2,
16
+ "name": "2",
17
+ "path": "2_Normalize",
18
+ "type": "sentence_transformers.models.Normalize"
19
+ }
20
+ ]
checkpoint-3000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:afd268f1efe6856700ac5323732cf206bbf6e27f7a8e1ca5d5dbfc9cec4ed0fa
3
+ size 265862074
checkpoint-3000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:944a9f94d2aa5a5a522b522b7451949e4695f0a5ce9c3381104aff95e4253724
3
+ size 14244
checkpoint-3000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d30509ba731da478647e60fd0049fa59b260e9aa7b56a621962ec1f34e84caa0
3
+ size 1064
checkpoint-3000/sentence_bert_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "max_seq_length": 512,
3
+ "do_lower_case": false
4
+ }
checkpoint-3000/special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
checkpoint-3000/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-3000/tokenizer_config.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "100": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "101": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "102": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "103": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": true,
45
+ "cls_token": "[CLS]",
46
+ "do_basic_tokenize": true,
47
+ "do_lower_case": true,
48
+ "mask_token": "[MASK]",
49
+ "model_max_length": 512,
50
+ "never_split": null,
51
+ "pad_token": "[PAD]",
52
+ "sep_token": "[SEP]",
53
+ "strip_accents": null,
54
+ "tokenize_chinese_chars": true,
55
+ "tokenizer_class": "BertTokenizer",
56
+ "unk_token": "[UNK]"
57
+ }
checkpoint-3000/trainer_state.json ADDED
@@ -0,0 +1,2133 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 12.244897959183673,
5
+ "eval_steps": 500,
6
+ "global_step": 3000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.04081632653061224,
13
+ "grad_norm": 7.114395618438721,
14
+ "learning_rate": 9.981632653061225e-06,
15
+ "loss": 0.7362,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.08163265306122448,
20
+ "grad_norm": 11.572301864624023,
21
+ "learning_rate": 9.961224489795919e-06,
22
+ "loss": 0.8729,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.12244897959183673,
27
+ "grad_norm": 9.383491516113281,
28
+ "learning_rate": 9.940816326530614e-06,
29
+ "loss": 0.773,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.16326530612244897,
34
+ "grad_norm": 7.83120059967041,
35
+ "learning_rate": 9.920408163265307e-06,
36
+ "loss": 0.7817,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.20408163265306123,
41
+ "grad_norm": 10.92087173461914,
42
+ "learning_rate": 9.9e-06,
43
+ "loss": 0.6256,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.24489795918367346,
48
+ "grad_norm": 3.8826725482940674,
49
+ "learning_rate": 9.879591836734695e-06,
50
+ "loss": 0.5759,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.2857142857142857,
55
+ "grad_norm": 11.15483283996582,
56
+ "learning_rate": 9.859183673469388e-06,
57
+ "loss": 0.7333,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.32653061224489793,
62
+ "grad_norm": 11.470726013183594,
63
+ "learning_rate": 9.838775510204083e-06,
64
+ "loss": 0.5943,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.3673469387755102,
69
+ "grad_norm": 13.159674644470215,
70
+ "learning_rate": 9.818367346938777e-06,
71
+ "loss": 0.7804,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.40816326530612246,
76
+ "grad_norm": 9.58558464050293,
77
+ "learning_rate": 9.79795918367347e-06,
78
+ "loss": 0.6491,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.4489795918367347,
83
+ "grad_norm": 9.653897285461426,
84
+ "learning_rate": 9.777551020408163e-06,
85
+ "loss": 0.5919,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.4897959183673469,
90
+ "grad_norm": 8.117432594299316,
91
+ "learning_rate": 9.757142857142858e-06,
92
+ "loss": 0.4571,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.5306122448979592,
97
+ "grad_norm": 6.9328460693359375,
98
+ "learning_rate": 9.736734693877551e-06,
99
+ "loss": 0.6597,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.5714285714285714,
104
+ "grad_norm": 7.962501049041748,
105
+ "learning_rate": 9.716326530612246e-06,
106
+ "loss": 0.5132,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.6122448979591837,
111
+ "grad_norm": 10.508763313293457,
112
+ "learning_rate": 9.69591836734694e-06,
113
+ "loss": 0.6893,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.6530612244897959,
118
+ "grad_norm": 7.637253761291504,
119
+ "learning_rate": 9.675510204081635e-06,
120
+ "loss": 0.6142,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.6938775510204082,
125
+ "grad_norm": 10.0332670211792,
126
+ "learning_rate": 9.655102040816328e-06,
127
+ "loss": 0.582,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.7346938775510204,
132
+ "grad_norm": 8.150875091552734,
133
+ "learning_rate": 9.634693877551021e-06,
134
+ "loss": 0.477,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.7755102040816326,
139
+ "grad_norm": 10.330913543701172,
140
+ "learning_rate": 9.614285714285714e-06,
141
+ "loss": 0.5916,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.8163265306122449,
146
+ "grad_norm": 11.654999732971191,
147
+ "learning_rate": 9.593877551020408e-06,
148
+ "loss": 0.6236,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.8571428571428571,
153
+ "grad_norm": 8.048078536987305,
154
+ "learning_rate": 9.573469387755103e-06,
155
+ "loss": 0.6142,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.8979591836734694,
160
+ "grad_norm": 9.869592666625977,
161
+ "learning_rate": 9.553061224489798e-06,
162
+ "loss": 0.625,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.9387755102040817,
167
+ "grad_norm": 8.321409225463867,
168
+ "learning_rate": 9.532653061224491e-06,
169
+ "loss": 0.5767,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.9795918367346939,
174
+ "grad_norm": 7.6769256591796875,
175
+ "learning_rate": 9.512244897959184e-06,
176
+ "loss": 0.5134,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 1.0204081632653061,
181
+ "grad_norm": 9.609123229980469,
182
+ "learning_rate": 9.491836734693877e-06,
183
+ "loss": 0.5868,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 1.0612244897959184,
188
+ "grad_norm": 9.19683837890625,
189
+ "learning_rate": 9.471428571428572e-06,
190
+ "loss": 0.5215,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 1.1020408163265305,
195
+ "grad_norm": 7.328164577484131,
196
+ "learning_rate": 9.451020408163266e-06,
197
+ "loss": 0.5422,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 1.1428571428571428,
202
+ "grad_norm": 6.913904190063477,
203
+ "learning_rate": 9.430612244897959e-06,
204
+ "loss": 0.5214,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 1.183673469387755,
209
+ "grad_norm": 9.28811264038086,
210
+ "learning_rate": 9.410204081632654e-06,
211
+ "loss": 0.5319,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 1.2244897959183674,
216
+ "grad_norm": 9.132966041564941,
217
+ "learning_rate": 9.389795918367349e-06,
218
+ "loss": 0.5581,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 1.2653061224489797,
223
+ "grad_norm": 6.9722065925598145,
224
+ "learning_rate": 9.369387755102042e-06,
225
+ "loss": 0.431,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 1.306122448979592,
230
+ "grad_norm": 5.06177282333374,
231
+ "learning_rate": 9.348979591836736e-06,
232
+ "loss": 0.4583,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 1.346938775510204,
237
+ "grad_norm": 7.732840538024902,
238
+ "learning_rate": 9.328571428571429e-06,
239
+ "loss": 0.4194,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 1.3877551020408163,
244
+ "grad_norm": 8.94101333618164,
245
+ "learning_rate": 9.308163265306122e-06,
246
+ "loss": 0.4519,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 1.4285714285714286,
251
+ "grad_norm": 7.5437750816345215,
252
+ "learning_rate": 9.287755102040817e-06,
253
+ "loss": 0.5095,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 1.469387755102041,
258
+ "grad_norm": 5.702700138092041,
259
+ "learning_rate": 9.26734693877551e-06,
260
+ "loss": 0.3936,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 1.510204081632653,
265
+ "grad_norm": 9.153871536254883,
266
+ "learning_rate": 9.246938775510205e-06,
267
+ "loss": 0.4566,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 1.5510204081632653,
272
+ "grad_norm": 13.249794006347656,
273
+ "learning_rate": 9.226530612244899e-06,
274
+ "loss": 0.5216,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 1.5918367346938775,
279
+ "grad_norm": 7.065913200378418,
280
+ "learning_rate": 9.206122448979594e-06,
281
+ "loss": 0.4562,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 1.6326530612244898,
286
+ "grad_norm": 7.559301853179932,
287
+ "learning_rate": 9.185714285714287e-06,
288
+ "loss": 0.3883,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 1.6734693877551021,
293
+ "grad_norm": 12.103629112243652,
294
+ "learning_rate": 9.16530612244898e-06,
295
+ "loss": 0.4149,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 1.7142857142857144,
300
+ "grad_norm": 7.9720072746276855,
301
+ "learning_rate": 9.144897959183673e-06,
302
+ "loss": 0.4718,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 1.7551020408163265,
307
+ "grad_norm": 4.845782279968262,
308
+ "learning_rate": 9.124489795918368e-06,
309
+ "loss": 0.4304,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 1.7959183673469388,
314
+ "grad_norm": 6.954368591308594,
315
+ "learning_rate": 9.104081632653062e-06,
316
+ "loss": 0.3436,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 1.836734693877551,
321
+ "grad_norm": 4.751299858093262,
322
+ "learning_rate": 9.083673469387757e-06,
323
+ "loss": 0.4366,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 1.8775510204081631,
328
+ "grad_norm": 6.507364273071289,
329
+ "learning_rate": 9.06326530612245e-06,
330
+ "loss": 0.5794,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 1.9183673469387754,
335
+ "grad_norm": 8.891802787780762,
336
+ "learning_rate": 9.042857142857143e-06,
337
+ "loss": 0.4616,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 1.9591836734693877,
342
+ "grad_norm": 10.056327819824219,
343
+ "learning_rate": 9.022448979591838e-06,
344
+ "loss": 0.4946,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 2.0,
349
+ "grad_norm": 7.899660110473633,
350
+ "learning_rate": 9.002040816326531e-06,
351
+ "loss": 0.4437,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 2.0408163265306123,
356
+ "grad_norm": 6.761326313018799,
357
+ "learning_rate": 8.981632653061225e-06,
358
+ "loss": 0.4303,
359
+ "step": 500
360
+ },
361
+ {
362
+ "epoch": 2.0816326530612246,
363
+ "grad_norm": 8.639615058898926,
364
+ "learning_rate": 8.96122448979592e-06,
365
+ "loss": 0.3267,
366
+ "step": 510
367
+ },
368
+ {
369
+ "epoch": 2.122448979591837,
370
+ "grad_norm": 7.710758209228516,
371
+ "learning_rate": 8.940816326530613e-06,
372
+ "loss": 0.3559,
373
+ "step": 520
374
+ },
375
+ {
376
+ "epoch": 2.163265306122449,
377
+ "grad_norm": 6.812905311584473,
378
+ "learning_rate": 8.920408163265308e-06,
379
+ "loss": 0.4761,
380
+ "step": 530
381
+ },
382
+ {
383
+ "epoch": 2.204081632653061,
384
+ "grad_norm": 7.2431511878967285,
385
+ "learning_rate": 8.900000000000001e-06,
386
+ "loss": 0.405,
387
+ "step": 540
388
+ },
389
+ {
390
+ "epoch": 2.2448979591836733,
391
+ "grad_norm": 7.230724811553955,
392
+ "learning_rate": 8.879591836734694e-06,
393
+ "loss": 0.3638,
394
+ "step": 550
395
+ },
396
+ {
397
+ "epoch": 2.2857142857142856,
398
+ "grad_norm": 9.520208358764648,
399
+ "learning_rate": 8.859183673469388e-06,
400
+ "loss": 0.3473,
401
+ "step": 560
402
+ },
403
+ {
404
+ "epoch": 2.326530612244898,
405
+ "grad_norm": 7.048585414886475,
406
+ "learning_rate": 8.838775510204083e-06,
407
+ "loss": 0.3652,
408
+ "step": 570
409
+ },
410
+ {
411
+ "epoch": 2.36734693877551,
412
+ "grad_norm": 6.979404449462891,
413
+ "learning_rate": 8.818367346938776e-06,
414
+ "loss": 0.3855,
415
+ "step": 580
416
+ },
417
+ {
418
+ "epoch": 2.4081632653061225,
419
+ "grad_norm": 3.765305280685425,
420
+ "learning_rate": 8.797959183673471e-06,
421
+ "loss": 0.3452,
422
+ "step": 590
423
+ },
424
+ {
425
+ "epoch": 2.4489795918367347,
426
+ "grad_norm": 10.533697128295898,
427
+ "learning_rate": 8.777551020408164e-06,
428
+ "loss": 0.3874,
429
+ "step": 600
430
+ },
431
+ {
432
+ "epoch": 2.489795918367347,
433
+ "grad_norm": 8.108145713806152,
434
+ "learning_rate": 8.757142857142858e-06,
435
+ "loss": 0.3695,
436
+ "step": 610
437
+ },
438
+ {
439
+ "epoch": 2.5306122448979593,
440
+ "grad_norm": 7.947360992431641,
441
+ "learning_rate": 8.736734693877552e-06,
442
+ "loss": 0.408,
443
+ "step": 620
444
+ },
445
+ {
446
+ "epoch": 2.571428571428571,
447
+ "grad_norm": 7.8081374168396,
448
+ "learning_rate": 8.716326530612246e-06,
449
+ "loss": 0.4059,
450
+ "step": 630
451
+ },
452
+ {
453
+ "epoch": 2.612244897959184,
454
+ "grad_norm": 8.579155921936035,
455
+ "learning_rate": 8.695918367346939e-06,
456
+ "loss": 0.3934,
457
+ "step": 640
458
+ },
459
+ {
460
+ "epoch": 2.6530612244897958,
461
+ "grad_norm": 6.4387712478637695,
462
+ "learning_rate": 8.675510204081632e-06,
463
+ "loss": 0.4256,
464
+ "step": 650
465
+ },
466
+ {
467
+ "epoch": 2.693877551020408,
468
+ "grad_norm": 8.415692329406738,
469
+ "learning_rate": 8.655102040816327e-06,
470
+ "loss": 0.3453,
471
+ "step": 660
472
+ },
473
+ {
474
+ "epoch": 2.7346938775510203,
475
+ "grad_norm": 8.50904369354248,
476
+ "learning_rate": 8.63469387755102e-06,
477
+ "loss": 0.4766,
478
+ "step": 670
479
+ },
480
+ {
481
+ "epoch": 2.7755102040816326,
482
+ "grad_norm": 4.662519931793213,
483
+ "learning_rate": 8.614285714285716e-06,
484
+ "loss": 0.3973,
485
+ "step": 680
486
+ },
487
+ {
488
+ "epoch": 2.816326530612245,
489
+ "grad_norm": 6.288435935974121,
490
+ "learning_rate": 8.593877551020409e-06,
491
+ "loss": 0.4408,
492
+ "step": 690
493
+ },
494
+ {
495
+ "epoch": 2.857142857142857,
496
+ "grad_norm": 6.625838279724121,
497
+ "learning_rate": 8.573469387755102e-06,
498
+ "loss": 0.2908,
499
+ "step": 700
500
+ },
501
+ {
502
+ "epoch": 2.8979591836734695,
503
+ "grad_norm": 8.510032653808594,
504
+ "learning_rate": 8.553061224489797e-06,
505
+ "loss": 0.3813,
506
+ "step": 710
507
+ },
508
+ {
509
+ "epoch": 2.938775510204082,
510
+ "grad_norm": 11.82463264465332,
511
+ "learning_rate": 8.53265306122449e-06,
512
+ "loss": 0.4352,
513
+ "step": 720
514
+ },
515
+ {
516
+ "epoch": 2.979591836734694,
517
+ "grad_norm": 8.821819305419922,
518
+ "learning_rate": 8.512244897959184e-06,
519
+ "loss": 0.4318,
520
+ "step": 730
521
+ },
522
+ {
523
+ "epoch": 3.020408163265306,
524
+ "grad_norm": 8.010713577270508,
525
+ "learning_rate": 8.491836734693879e-06,
526
+ "loss": 0.2323,
527
+ "step": 740
528
+ },
529
+ {
530
+ "epoch": 3.061224489795918,
531
+ "grad_norm": 9.03991985321045,
532
+ "learning_rate": 8.471428571428572e-06,
533
+ "loss": 0.3603,
534
+ "step": 750
535
+ },
536
+ {
537
+ "epoch": 3.1020408163265305,
538
+ "grad_norm": 10.94204044342041,
539
+ "learning_rate": 8.451020408163267e-06,
540
+ "loss": 0.3576,
541
+ "step": 760
542
+ },
543
+ {
544
+ "epoch": 3.142857142857143,
545
+ "grad_norm": 7.89410924911499,
546
+ "learning_rate": 8.43061224489796e-06,
547
+ "loss": 0.2851,
548
+ "step": 770
549
+ },
550
+ {
551
+ "epoch": 3.183673469387755,
552
+ "grad_norm": 6.53656005859375,
553
+ "learning_rate": 8.410204081632653e-06,
554
+ "loss": 0.318,
555
+ "step": 780
556
+ },
557
+ {
558
+ "epoch": 3.2244897959183674,
559
+ "grad_norm": 6.487284183502197,
560
+ "learning_rate": 8.389795918367347e-06,
561
+ "loss": 0.317,
562
+ "step": 790
563
+ },
564
+ {
565
+ "epoch": 3.2653061224489797,
566
+ "grad_norm": 6.947931289672852,
567
+ "learning_rate": 8.369387755102042e-06,
568
+ "loss": 0.2879,
569
+ "step": 800
570
+ },
571
+ {
572
+ "epoch": 3.306122448979592,
573
+ "grad_norm": 4.166048526763916,
574
+ "learning_rate": 8.348979591836735e-06,
575
+ "loss": 0.3392,
576
+ "step": 810
577
+ },
578
+ {
579
+ "epoch": 3.3469387755102042,
580
+ "grad_norm": 9.974846839904785,
581
+ "learning_rate": 8.32857142857143e-06,
582
+ "loss": 0.3663,
583
+ "step": 820
584
+ },
585
+ {
586
+ "epoch": 3.387755102040816,
587
+ "grad_norm": 9.668428421020508,
588
+ "learning_rate": 8.308163265306123e-06,
589
+ "loss": 0.3212,
590
+ "step": 830
591
+ },
592
+ {
593
+ "epoch": 3.4285714285714284,
594
+ "grad_norm": 11.81507396697998,
595
+ "learning_rate": 8.287755102040816e-06,
596
+ "loss": 0.3241,
597
+ "step": 840
598
+ },
599
+ {
600
+ "epoch": 3.4693877551020407,
601
+ "grad_norm": 13.690321922302246,
602
+ "learning_rate": 8.267346938775511e-06,
603
+ "loss": 0.4535,
604
+ "step": 850
605
+ },
606
+ {
607
+ "epoch": 3.510204081632653,
608
+ "grad_norm": 11.042778968811035,
609
+ "learning_rate": 8.246938775510205e-06,
610
+ "loss": 0.3826,
611
+ "step": 860
612
+ },
613
+ {
614
+ "epoch": 3.5510204081632653,
615
+ "grad_norm": 8.57719612121582,
616
+ "learning_rate": 8.226530612244898e-06,
617
+ "loss": 0.3905,
618
+ "step": 870
619
+ },
620
+ {
621
+ "epoch": 3.5918367346938775,
622
+ "grad_norm": 7.843425750732422,
623
+ "learning_rate": 8.206122448979591e-06,
624
+ "loss": 0.3125,
625
+ "step": 880
626
+ },
627
+ {
628
+ "epoch": 3.63265306122449,
629
+ "grad_norm": 5.9236931800842285,
630
+ "learning_rate": 8.185714285714286e-06,
631
+ "loss": 0.3512,
632
+ "step": 890
633
+ },
634
+ {
635
+ "epoch": 3.673469387755102,
636
+ "grad_norm": 8.213603973388672,
637
+ "learning_rate": 8.165306122448981e-06,
638
+ "loss": 0.4094,
639
+ "step": 900
640
+ },
641
+ {
642
+ "epoch": 3.7142857142857144,
643
+ "grad_norm": 3.8083949089050293,
644
+ "learning_rate": 8.144897959183674e-06,
645
+ "loss": 0.2751,
646
+ "step": 910
647
+ },
648
+ {
649
+ "epoch": 3.7551020408163263,
650
+ "grad_norm": 12.339240074157715,
651
+ "learning_rate": 8.124489795918368e-06,
652
+ "loss": 0.3296,
653
+ "step": 920
654
+ },
655
+ {
656
+ "epoch": 3.795918367346939,
657
+ "grad_norm": 9.532052040100098,
658
+ "learning_rate": 8.104081632653061e-06,
659
+ "loss": 0.3033,
660
+ "step": 930
661
+ },
662
+ {
663
+ "epoch": 3.836734693877551,
664
+ "grad_norm": 6.307032108306885,
665
+ "learning_rate": 8.083673469387756e-06,
666
+ "loss": 0.3765,
667
+ "step": 940
668
+ },
669
+ {
670
+ "epoch": 3.877551020408163,
671
+ "grad_norm": 7.3003010749816895,
672
+ "learning_rate": 8.06326530612245e-06,
673
+ "loss": 0.2161,
674
+ "step": 950
675
+ },
676
+ {
677
+ "epoch": 3.9183673469387754,
678
+ "grad_norm": 7.6572699546813965,
679
+ "learning_rate": 8.042857142857143e-06,
680
+ "loss": 0.2886,
681
+ "step": 960
682
+ },
683
+ {
684
+ "epoch": 3.9591836734693877,
685
+ "grad_norm": 6.745776176452637,
686
+ "learning_rate": 8.022448979591838e-06,
687
+ "loss": 0.3376,
688
+ "step": 970
689
+ },
690
+ {
691
+ "epoch": 4.0,
692
+ "grad_norm": 10.482270240783691,
693
+ "learning_rate": 8.002040816326533e-06,
694
+ "loss": 0.2657,
695
+ "step": 980
696
+ },
697
+ {
698
+ "epoch": 4.040816326530612,
699
+ "grad_norm": 6.213717460632324,
700
+ "learning_rate": 7.981632653061226e-06,
701
+ "loss": 0.2596,
702
+ "step": 990
703
+ },
704
+ {
705
+ "epoch": 4.081632653061225,
706
+ "grad_norm": 10.256094932556152,
707
+ "learning_rate": 7.961224489795919e-06,
708
+ "loss": 0.2338,
709
+ "step": 1000
710
+ },
711
+ {
712
+ "epoch": 4.122448979591836,
713
+ "grad_norm": 8.640167236328125,
714
+ "learning_rate": 7.940816326530612e-06,
715
+ "loss": 0.2936,
716
+ "step": 1010
717
+ },
718
+ {
719
+ "epoch": 4.163265306122449,
720
+ "grad_norm": 6.750300407409668,
721
+ "learning_rate": 7.920408163265306e-06,
722
+ "loss": 0.3511,
723
+ "step": 1020
724
+ },
725
+ {
726
+ "epoch": 4.204081632653061,
727
+ "grad_norm": 9.488007545471191,
728
+ "learning_rate": 7.9e-06,
729
+ "loss": 0.2913,
730
+ "step": 1030
731
+ },
732
+ {
733
+ "epoch": 4.244897959183674,
734
+ "grad_norm": 4.2671003341674805,
735
+ "learning_rate": 7.879591836734694e-06,
736
+ "loss": 0.2768,
737
+ "step": 1040
738
+ },
739
+ {
740
+ "epoch": 4.285714285714286,
741
+ "grad_norm": 5.375782489776611,
742
+ "learning_rate": 7.859183673469389e-06,
743
+ "loss": 0.274,
744
+ "step": 1050
745
+ },
746
+ {
747
+ "epoch": 4.326530612244898,
748
+ "grad_norm": 10.316515922546387,
749
+ "learning_rate": 7.838775510204082e-06,
750
+ "loss": 0.2373,
751
+ "step": 1060
752
+ },
753
+ {
754
+ "epoch": 4.36734693877551,
755
+ "grad_norm": 6.733712673187256,
756
+ "learning_rate": 7.818367346938777e-06,
757
+ "loss": 0.32,
758
+ "step": 1070
759
+ },
760
+ {
761
+ "epoch": 4.408163265306122,
762
+ "grad_norm": 6.432683944702148,
763
+ "learning_rate": 7.79795918367347e-06,
764
+ "loss": 0.3149,
765
+ "step": 1080
766
+ },
767
+ {
768
+ "epoch": 4.448979591836735,
769
+ "grad_norm": 8.520115852355957,
770
+ "learning_rate": 7.777551020408164e-06,
771
+ "loss": 0.3369,
772
+ "step": 1090
773
+ },
774
+ {
775
+ "epoch": 4.489795918367347,
776
+ "grad_norm": 3.52677845954895,
777
+ "learning_rate": 7.757142857142857e-06,
778
+ "loss": 0.2584,
779
+ "step": 1100
780
+ },
781
+ {
782
+ "epoch": 4.530612244897959,
783
+ "grad_norm": 9.642569541931152,
784
+ "learning_rate": 7.736734693877552e-06,
785
+ "loss": 0.3076,
786
+ "step": 1110
787
+ },
788
+ {
789
+ "epoch": 4.571428571428571,
790
+ "grad_norm": 11.233070373535156,
791
+ "learning_rate": 7.716326530612245e-06,
792
+ "loss": 0.3263,
793
+ "step": 1120
794
+ },
795
+ {
796
+ "epoch": 4.612244897959184,
797
+ "grad_norm": 7.919038772583008,
798
+ "learning_rate": 7.69591836734694e-06,
799
+ "loss": 0.3152,
800
+ "step": 1130
801
+ },
802
+ {
803
+ "epoch": 4.653061224489796,
804
+ "grad_norm": 7.116144180297852,
805
+ "learning_rate": 7.675510204081633e-06,
806
+ "loss": 0.2809,
807
+ "step": 1140
808
+ },
809
+ {
810
+ "epoch": 4.6938775510204085,
811
+ "grad_norm": 6.757585525512695,
812
+ "learning_rate": 7.655102040816327e-06,
813
+ "loss": 0.2642,
814
+ "step": 1150
815
+ },
816
+ {
817
+ "epoch": 4.73469387755102,
818
+ "grad_norm": 9.032824516296387,
819
+ "learning_rate": 7.634693877551022e-06,
820
+ "loss": 0.3546,
821
+ "step": 1160
822
+ },
823
+ {
824
+ "epoch": 4.775510204081632,
825
+ "grad_norm": 7.837385654449463,
826
+ "learning_rate": 7.614285714285715e-06,
827
+ "loss": 0.3196,
828
+ "step": 1170
829
+ },
830
+ {
831
+ "epoch": 4.816326530612245,
832
+ "grad_norm": 5.3134846687316895,
833
+ "learning_rate": 7.593877551020409e-06,
834
+ "loss": 0.2068,
835
+ "step": 1180
836
+ },
837
+ {
838
+ "epoch": 4.857142857142857,
839
+ "grad_norm": 8.036781311035156,
840
+ "learning_rate": 7.573469387755102e-06,
841
+ "loss": 0.24,
842
+ "step": 1190
843
+ },
844
+ {
845
+ "epoch": 4.8979591836734695,
846
+ "grad_norm": 12.13339900970459,
847
+ "learning_rate": 7.5530612244897965e-06,
848
+ "loss": 0.2795,
849
+ "step": 1200
850
+ },
851
+ {
852
+ "epoch": 4.938775510204081,
853
+ "grad_norm": 5.253749370574951,
854
+ "learning_rate": 7.532653061224491e-06,
855
+ "loss": 0.2667,
856
+ "step": 1210
857
+ },
858
+ {
859
+ "epoch": 4.979591836734694,
860
+ "grad_norm": 7.921318531036377,
861
+ "learning_rate": 7.512244897959185e-06,
862
+ "loss": 0.3218,
863
+ "step": 1220
864
+ },
865
+ {
866
+ "epoch": 5.020408163265306,
867
+ "grad_norm": 9.056379318237305,
868
+ "learning_rate": 7.491836734693878e-06,
869
+ "loss": 0.2324,
870
+ "step": 1230
871
+ },
872
+ {
873
+ "epoch": 5.061224489795919,
874
+ "grad_norm": 5.0674920082092285,
875
+ "learning_rate": 7.471428571428571e-06,
876
+ "loss": 0.3063,
877
+ "step": 1240
878
+ },
879
+ {
880
+ "epoch": 5.1020408163265305,
881
+ "grad_norm": 12.247352600097656,
882
+ "learning_rate": 7.451020408163266e-06,
883
+ "loss": 0.3469,
884
+ "step": 1250
885
+ },
886
+ {
887
+ "epoch": 5.142857142857143,
888
+ "grad_norm": 7.278679847717285,
889
+ "learning_rate": 7.43061224489796e-06,
890
+ "loss": 0.2851,
891
+ "step": 1260
892
+ },
893
+ {
894
+ "epoch": 5.183673469387755,
895
+ "grad_norm": 11.491589546203613,
896
+ "learning_rate": 7.410204081632654e-06,
897
+ "loss": 0.245,
898
+ "step": 1270
899
+ },
900
+ {
901
+ "epoch": 5.224489795918367,
902
+ "grad_norm": 8.652173042297363,
903
+ "learning_rate": 7.389795918367347e-06,
904
+ "loss": 0.2328,
905
+ "step": 1280
906
+ },
907
+ {
908
+ "epoch": 5.26530612244898,
909
+ "grad_norm": 7.953402519226074,
910
+ "learning_rate": 7.369387755102041e-06,
911
+ "loss": 0.2784,
912
+ "step": 1290
913
+ },
914
+ {
915
+ "epoch": 5.3061224489795915,
916
+ "grad_norm": 6.98084831237793,
917
+ "learning_rate": 7.348979591836736e-06,
918
+ "loss": 0.2379,
919
+ "step": 1300
920
+ },
921
+ {
922
+ "epoch": 5.346938775510204,
923
+ "grad_norm": 7.400093078613281,
924
+ "learning_rate": 7.328571428571429e-06,
925
+ "loss": 0.2057,
926
+ "step": 1310
927
+ },
928
+ {
929
+ "epoch": 5.387755102040816,
930
+ "grad_norm": 4.131850719451904,
931
+ "learning_rate": 7.308163265306123e-06,
932
+ "loss": 0.2792,
933
+ "step": 1320
934
+ },
935
+ {
936
+ "epoch": 5.428571428571429,
937
+ "grad_norm": 10.25373363494873,
938
+ "learning_rate": 7.287755102040817e-06,
939
+ "loss": 0.2642,
940
+ "step": 1330
941
+ },
942
+ {
943
+ "epoch": 5.469387755102041,
944
+ "grad_norm": 5.801217079162598,
945
+ "learning_rate": 7.267346938775512e-06,
946
+ "loss": 0.2301,
947
+ "step": 1340
948
+ },
949
+ {
950
+ "epoch": 5.510204081632653,
951
+ "grad_norm": 3.3520116806030273,
952
+ "learning_rate": 7.246938775510205e-06,
953
+ "loss": 0.2728,
954
+ "step": 1350
955
+ },
956
+ {
957
+ "epoch": 5.551020408163265,
958
+ "grad_norm": 5.335054874420166,
959
+ "learning_rate": 7.226530612244898e-06,
960
+ "loss": 0.2812,
961
+ "step": 1360
962
+ },
963
+ {
964
+ "epoch": 5.591836734693878,
965
+ "grad_norm": 4.713162899017334,
966
+ "learning_rate": 7.206122448979592e-06,
967
+ "loss": 0.221,
968
+ "step": 1370
969
+ },
970
+ {
971
+ "epoch": 5.63265306122449,
972
+ "grad_norm": 8.641826629638672,
973
+ "learning_rate": 7.185714285714286e-06,
974
+ "loss": 0.2204,
975
+ "step": 1380
976
+ },
977
+ {
978
+ "epoch": 5.673469387755102,
979
+ "grad_norm": 4.286067485809326,
980
+ "learning_rate": 7.165306122448981e-06,
981
+ "loss": 0.2446,
982
+ "step": 1390
983
+ },
984
+ {
985
+ "epoch": 5.714285714285714,
986
+ "grad_norm": 7.071314811706543,
987
+ "learning_rate": 7.144897959183674e-06,
988
+ "loss": 0.1788,
989
+ "step": 1400
990
+ },
991
+ {
992
+ "epoch": 5.755102040816326,
993
+ "grad_norm": 9.792560577392578,
994
+ "learning_rate": 7.124489795918368e-06,
995
+ "loss": 0.272,
996
+ "step": 1410
997
+ },
998
+ {
999
+ "epoch": 5.795918367346939,
1000
+ "grad_norm": 6.726722240447998,
1001
+ "learning_rate": 7.104081632653061e-06,
1002
+ "loss": 0.18,
1003
+ "step": 1420
1004
+ },
1005
+ {
1006
+ "epoch": 5.836734693877551,
1007
+ "grad_norm": 7.378534317016602,
1008
+ "learning_rate": 7.083673469387755e-06,
1009
+ "loss": 0.2308,
1010
+ "step": 1430
1011
+ },
1012
+ {
1013
+ "epoch": 5.877551020408164,
1014
+ "grad_norm": 10.169758796691895,
1015
+ "learning_rate": 7.0632653061224495e-06,
1016
+ "loss": 0.296,
1017
+ "step": 1440
1018
+ },
1019
+ {
1020
+ "epoch": 5.918367346938775,
1021
+ "grad_norm": 5.440324306488037,
1022
+ "learning_rate": 7.042857142857144e-06,
1023
+ "loss": 0.2725,
1024
+ "step": 1450
1025
+ },
1026
+ {
1027
+ "epoch": 5.959183673469388,
1028
+ "grad_norm": 8.175464630126953,
1029
+ "learning_rate": 7.022448979591837e-06,
1030
+ "loss": 0.2251,
1031
+ "step": 1460
1032
+ },
1033
+ {
1034
+ "epoch": 6.0,
1035
+ "grad_norm": 6.7653398513793945,
1036
+ "learning_rate": 7.002040816326531e-06,
1037
+ "loss": 0.3411,
1038
+ "step": 1470
1039
+ },
1040
+ {
1041
+ "epoch": 6.040816326530612,
1042
+ "grad_norm": 6.0152716636657715,
1043
+ "learning_rate": 6.981632653061225e-06,
1044
+ "loss": 0.2071,
1045
+ "step": 1480
1046
+ },
1047
+ {
1048
+ "epoch": 6.081632653061225,
1049
+ "grad_norm": 8.066520690917969,
1050
+ "learning_rate": 6.961224489795919e-06,
1051
+ "loss": 0.1639,
1052
+ "step": 1490
1053
+ },
1054
+ {
1055
+ "epoch": 6.122448979591836,
1056
+ "grad_norm": 4.027646541595459,
1057
+ "learning_rate": 6.940816326530613e-06,
1058
+ "loss": 0.2069,
1059
+ "step": 1500
1060
+ },
1061
+ {
1062
+ "epoch": 6.163265306122449,
1063
+ "grad_norm": 5.61140251159668,
1064
+ "learning_rate": 6.920408163265307e-06,
1065
+ "loss": 0.2413,
1066
+ "step": 1510
1067
+ },
1068
+ {
1069
+ "epoch": 6.204081632653061,
1070
+ "grad_norm": 9.809159278869629,
1071
+ "learning_rate": 6.9e-06,
1072
+ "loss": 0.26,
1073
+ "step": 1520
1074
+ },
1075
+ {
1076
+ "epoch": 6.244897959183674,
1077
+ "grad_norm": 6.755568504333496,
1078
+ "learning_rate": 6.879591836734695e-06,
1079
+ "loss": 0.1965,
1080
+ "step": 1530
1081
+ },
1082
+ {
1083
+ "epoch": 6.285714285714286,
1084
+ "grad_norm": 4.21774435043335,
1085
+ "learning_rate": 6.859183673469388e-06,
1086
+ "loss": 0.1959,
1087
+ "step": 1540
1088
+ },
1089
+ {
1090
+ "epoch": 6.326530612244898,
1091
+ "grad_norm": 5.166352272033691,
1092
+ "learning_rate": 6.838775510204082e-06,
1093
+ "loss": 0.1941,
1094
+ "step": 1550
1095
+ },
1096
+ {
1097
+ "epoch": 6.36734693877551,
1098
+ "grad_norm": 10.346336364746094,
1099
+ "learning_rate": 6.818367346938776e-06,
1100
+ "loss": 0.2278,
1101
+ "step": 1560
1102
+ },
1103
+ {
1104
+ "epoch": 6.408163265306122,
1105
+ "grad_norm": 7.703672409057617,
1106
+ "learning_rate": 6.797959183673471e-06,
1107
+ "loss": 0.2854,
1108
+ "step": 1570
1109
+ },
1110
+ {
1111
+ "epoch": 6.448979591836735,
1112
+ "grad_norm": 9.366389274597168,
1113
+ "learning_rate": 6.777551020408164e-06,
1114
+ "loss": 0.2229,
1115
+ "step": 1580
1116
+ },
1117
+ {
1118
+ "epoch": 6.489795918367347,
1119
+ "grad_norm": 7.013561248779297,
1120
+ "learning_rate": 6.757142857142858e-06,
1121
+ "loss": 0.2732,
1122
+ "step": 1590
1123
+ },
1124
+ {
1125
+ "epoch": 6.530612244897959,
1126
+ "grad_norm": 5.82119083404541,
1127
+ "learning_rate": 6.736734693877551e-06,
1128
+ "loss": 0.2389,
1129
+ "step": 1600
1130
+ },
1131
+ {
1132
+ "epoch": 6.571428571428571,
1133
+ "grad_norm": 8.112947463989258,
1134
+ "learning_rate": 6.716326530612245e-06,
1135
+ "loss": 0.193,
1136
+ "step": 1610
1137
+ },
1138
+ {
1139
+ "epoch": 6.612244897959184,
1140
+ "grad_norm": 8.238506317138672,
1141
+ "learning_rate": 6.6959183673469396e-06,
1142
+ "loss": 0.2344,
1143
+ "step": 1620
1144
+ },
1145
+ {
1146
+ "epoch": 6.653061224489796,
1147
+ "grad_norm": 6.531697750091553,
1148
+ "learning_rate": 6.675510204081634e-06,
1149
+ "loss": 0.2146,
1150
+ "step": 1630
1151
+ },
1152
+ {
1153
+ "epoch": 6.6938775510204085,
1154
+ "grad_norm": 6.525634288787842,
1155
+ "learning_rate": 6.655102040816327e-06,
1156
+ "loss": 0.1993,
1157
+ "step": 1640
1158
+ },
1159
+ {
1160
+ "epoch": 6.73469387755102,
1161
+ "grad_norm": 8.847986221313477,
1162
+ "learning_rate": 6.63469387755102e-06,
1163
+ "loss": 0.2884,
1164
+ "step": 1650
1165
+ },
1166
+ {
1167
+ "epoch": 6.775510204081632,
1168
+ "grad_norm": 9.006918907165527,
1169
+ "learning_rate": 6.614285714285715e-06,
1170
+ "loss": 0.2173,
1171
+ "step": 1660
1172
+ },
1173
+ {
1174
+ "epoch": 6.816326530612245,
1175
+ "grad_norm": 9.229476928710938,
1176
+ "learning_rate": 6.593877551020409e-06,
1177
+ "loss": 0.201,
1178
+ "step": 1670
1179
+ },
1180
+ {
1181
+ "epoch": 6.857142857142857,
1182
+ "grad_norm": 5.367541313171387,
1183
+ "learning_rate": 6.573469387755103e-06,
1184
+ "loss": 0.2505,
1185
+ "step": 1680
1186
+ },
1187
+ {
1188
+ "epoch": 6.8979591836734695,
1189
+ "grad_norm": 7.771108150482178,
1190
+ "learning_rate": 6.553061224489796e-06,
1191
+ "loss": 0.2225,
1192
+ "step": 1690
1193
+ },
1194
+ {
1195
+ "epoch": 6.938775510204081,
1196
+ "grad_norm": 5.306410789489746,
1197
+ "learning_rate": 6.53265306122449e-06,
1198
+ "loss": 0.2549,
1199
+ "step": 1700
1200
+ },
1201
+ {
1202
+ "epoch": 6.979591836734694,
1203
+ "grad_norm": 4.359670162200928,
1204
+ "learning_rate": 6.512244897959185e-06,
1205
+ "loss": 0.2585,
1206
+ "step": 1710
1207
+ },
1208
+ {
1209
+ "epoch": 7.020408163265306,
1210
+ "grad_norm": 4.528923034667969,
1211
+ "learning_rate": 6.491836734693878e-06,
1212
+ "loss": 0.2489,
1213
+ "step": 1720
1214
+ },
1215
+ {
1216
+ "epoch": 7.061224489795919,
1217
+ "grad_norm": 4.760287761688232,
1218
+ "learning_rate": 6.4714285714285715e-06,
1219
+ "loss": 0.1955,
1220
+ "step": 1730
1221
+ },
1222
+ {
1223
+ "epoch": 7.1020408163265305,
1224
+ "grad_norm": 9.205543518066406,
1225
+ "learning_rate": 6.451020408163266e-06,
1226
+ "loss": 0.2283,
1227
+ "step": 1740
1228
+ },
1229
+ {
1230
+ "epoch": 7.142857142857143,
1231
+ "grad_norm": 5.910974025726318,
1232
+ "learning_rate": 6.43061224489796e-06,
1233
+ "loss": 0.2308,
1234
+ "step": 1750
1235
+ },
1236
+ {
1237
+ "epoch": 7.183673469387755,
1238
+ "grad_norm": 5.042090892791748,
1239
+ "learning_rate": 6.410204081632654e-06,
1240
+ "loss": 0.1883,
1241
+ "step": 1760
1242
+ },
1243
+ {
1244
+ "epoch": 7.224489795918367,
1245
+ "grad_norm": 5.0842742919921875,
1246
+ "learning_rate": 6.389795918367347e-06,
1247
+ "loss": 0.1942,
1248
+ "step": 1770
1249
+ },
1250
+ {
1251
+ "epoch": 7.26530612244898,
1252
+ "grad_norm": 9.219313621520996,
1253
+ "learning_rate": 6.369387755102041e-06,
1254
+ "loss": 0.1946,
1255
+ "step": 1780
1256
+ },
1257
+ {
1258
+ "epoch": 7.3061224489795915,
1259
+ "grad_norm": 7.5656208992004395,
1260
+ "learning_rate": 6.348979591836735e-06,
1261
+ "loss": 0.2016,
1262
+ "step": 1790
1263
+ },
1264
+ {
1265
+ "epoch": 7.346938775510204,
1266
+ "grad_norm": 5.8213653564453125,
1267
+ "learning_rate": 6.3285714285714296e-06,
1268
+ "loss": 0.2808,
1269
+ "step": 1800
1270
+ },
1271
+ {
1272
+ "epoch": 7.387755102040816,
1273
+ "grad_norm": 8.004778861999512,
1274
+ "learning_rate": 6.308163265306123e-06,
1275
+ "loss": 0.2393,
1276
+ "step": 1810
1277
+ },
1278
+ {
1279
+ "epoch": 7.428571428571429,
1280
+ "grad_norm": 7.5001139640808105,
1281
+ "learning_rate": 6.287755102040817e-06,
1282
+ "loss": 0.2007,
1283
+ "step": 1820
1284
+ },
1285
+ {
1286
+ "epoch": 7.469387755102041,
1287
+ "grad_norm": 9.618229866027832,
1288
+ "learning_rate": 6.26734693877551e-06,
1289
+ "loss": 0.2464,
1290
+ "step": 1830
1291
+ },
1292
+ {
1293
+ "epoch": 7.510204081632653,
1294
+ "grad_norm": 7.257756233215332,
1295
+ "learning_rate": 6.246938775510205e-06,
1296
+ "loss": 0.1692,
1297
+ "step": 1840
1298
+ },
1299
+ {
1300
+ "epoch": 7.551020408163265,
1301
+ "grad_norm": 7.658279895782471,
1302
+ "learning_rate": 6.2265306122448985e-06,
1303
+ "loss": 0.2367,
1304
+ "step": 1850
1305
+ },
1306
+ {
1307
+ "epoch": 7.591836734693878,
1308
+ "grad_norm": 6.590469837188721,
1309
+ "learning_rate": 6.206122448979593e-06,
1310
+ "loss": 0.2508,
1311
+ "step": 1860
1312
+ },
1313
+ {
1314
+ "epoch": 7.63265306122449,
1315
+ "grad_norm": 8.601705551147461,
1316
+ "learning_rate": 6.185714285714286e-06,
1317
+ "loss": 0.2168,
1318
+ "step": 1870
1319
+ },
1320
+ {
1321
+ "epoch": 7.673469387755102,
1322
+ "grad_norm": 6.144942283630371,
1323
+ "learning_rate": 6.16530612244898e-06,
1324
+ "loss": 0.2308,
1325
+ "step": 1880
1326
+ },
1327
+ {
1328
+ "epoch": 7.714285714285714,
1329
+ "grad_norm": 3.256690502166748,
1330
+ "learning_rate": 6.144897959183674e-06,
1331
+ "loss": 0.1791,
1332
+ "step": 1890
1333
+ },
1334
+ {
1335
+ "epoch": 7.755102040816326,
1336
+ "grad_norm": 6.810645580291748,
1337
+ "learning_rate": 6.124489795918368e-06,
1338
+ "loss": 0.2335,
1339
+ "step": 1900
1340
+ },
1341
+ {
1342
+ "epoch": 7.795918367346939,
1343
+ "grad_norm": 3.4259018898010254,
1344
+ "learning_rate": 6.1040816326530616e-06,
1345
+ "loss": 0.212,
1346
+ "step": 1910
1347
+ },
1348
+ {
1349
+ "epoch": 7.836734693877551,
1350
+ "grad_norm": 8.353039741516113,
1351
+ "learning_rate": 6.083673469387756e-06,
1352
+ "loss": 0.2312,
1353
+ "step": 1920
1354
+ },
1355
+ {
1356
+ "epoch": 7.877551020408164,
1357
+ "grad_norm": 5.548733711242676,
1358
+ "learning_rate": 6.06326530612245e-06,
1359
+ "loss": 0.2666,
1360
+ "step": 1930
1361
+ },
1362
+ {
1363
+ "epoch": 7.918367346938775,
1364
+ "grad_norm": 8.386053085327148,
1365
+ "learning_rate": 6.042857142857144e-06,
1366
+ "loss": 0.2543,
1367
+ "step": 1940
1368
+ },
1369
+ {
1370
+ "epoch": 7.959183673469388,
1371
+ "grad_norm": 7.863219261169434,
1372
+ "learning_rate": 6.022448979591837e-06,
1373
+ "loss": 0.2171,
1374
+ "step": 1950
1375
+ },
1376
+ {
1377
+ "epoch": 8.0,
1378
+ "grad_norm": 5.328917503356934,
1379
+ "learning_rate": 6.0020408163265305e-06,
1380
+ "loss": 0.1592,
1381
+ "step": 1960
1382
+ },
1383
+ {
1384
+ "epoch": 8.040816326530612,
1385
+ "grad_norm": 11.456307411193848,
1386
+ "learning_rate": 5.981632653061225e-06,
1387
+ "loss": 0.2345,
1388
+ "step": 1970
1389
+ },
1390
+ {
1391
+ "epoch": 8.081632653061224,
1392
+ "grad_norm": 3.9219276905059814,
1393
+ "learning_rate": 5.96122448979592e-06,
1394
+ "loss": 0.228,
1395
+ "step": 1980
1396
+ },
1397
+ {
1398
+ "epoch": 8.122448979591837,
1399
+ "grad_norm": 7.155372142791748,
1400
+ "learning_rate": 5.940816326530613e-06,
1401
+ "loss": 0.1879,
1402
+ "step": 1990
1403
+ },
1404
+ {
1405
+ "epoch": 8.16326530612245,
1406
+ "grad_norm": 8.530526161193848,
1407
+ "learning_rate": 5.920408163265306e-06,
1408
+ "loss": 0.1808,
1409
+ "step": 2000
1410
+ },
1411
+ {
1412
+ "epoch": 8.204081632653061,
1413
+ "grad_norm": 10.660292625427246,
1414
+ "learning_rate": 5.902040816326531e-06,
1415
+ "loss": 0.2545,
1416
+ "step": 2010
1417
+ },
1418
+ {
1419
+ "epoch": 8.244897959183673,
1420
+ "grad_norm": 2.4285929203033447,
1421
+ "learning_rate": 5.881632653061225e-06,
1422
+ "loss": 0.1697,
1423
+ "step": 2020
1424
+ },
1425
+ {
1426
+ "epoch": 8.285714285714286,
1427
+ "grad_norm": 7.390512943267822,
1428
+ "learning_rate": 5.861224489795919e-06,
1429
+ "loss": 0.1868,
1430
+ "step": 2030
1431
+ },
1432
+ {
1433
+ "epoch": 8.326530612244898,
1434
+ "grad_norm": 6.263368129730225,
1435
+ "learning_rate": 5.840816326530613e-06,
1436
+ "loss": 0.2508,
1437
+ "step": 2040
1438
+ },
1439
+ {
1440
+ "epoch": 8.36734693877551,
1441
+ "grad_norm": 0.5456896424293518,
1442
+ "learning_rate": 5.820408163265306e-06,
1443
+ "loss": 0.1737,
1444
+ "step": 2050
1445
+ },
1446
+ {
1447
+ "epoch": 8.408163265306122,
1448
+ "grad_norm": 9.859989166259766,
1449
+ "learning_rate": 5.8e-06,
1450
+ "loss": 0.2154,
1451
+ "step": 2060
1452
+ },
1453
+ {
1454
+ "epoch": 8.448979591836734,
1455
+ "grad_norm": 7.002127647399902,
1456
+ "learning_rate": 5.7795918367346945e-06,
1457
+ "loss": 0.178,
1458
+ "step": 2070
1459
+ },
1460
+ {
1461
+ "epoch": 8.489795918367347,
1462
+ "grad_norm": 11.321815490722656,
1463
+ "learning_rate": 5.759183673469389e-06,
1464
+ "loss": 0.2147,
1465
+ "step": 2080
1466
+ },
1467
+ {
1468
+ "epoch": 8.53061224489796,
1469
+ "grad_norm": 4.573301315307617,
1470
+ "learning_rate": 5.738775510204082e-06,
1471
+ "loss": 0.2011,
1472
+ "step": 2090
1473
+ },
1474
+ {
1475
+ "epoch": 8.571428571428571,
1476
+ "grad_norm": 5.418411731719971,
1477
+ "learning_rate": 5.718367346938776e-06,
1478
+ "loss": 0.1873,
1479
+ "step": 2100
1480
+ },
1481
+ {
1482
+ "epoch": 8.612244897959183,
1483
+ "grad_norm": 5.087385177612305,
1484
+ "learning_rate": 5.697959183673469e-06,
1485
+ "loss": 0.2519,
1486
+ "step": 2110
1487
+ },
1488
+ {
1489
+ "epoch": 8.653061224489797,
1490
+ "grad_norm": 5.11081600189209,
1491
+ "learning_rate": 5.677551020408164e-06,
1492
+ "loss": 0.1733,
1493
+ "step": 2120
1494
+ },
1495
+ {
1496
+ "epoch": 8.693877551020408,
1497
+ "grad_norm": 9.11174201965332,
1498
+ "learning_rate": 5.6571428571428576e-06,
1499
+ "loss": 0.2496,
1500
+ "step": 2130
1501
+ },
1502
+ {
1503
+ "epoch": 8.73469387755102,
1504
+ "grad_norm": 7.031070709228516,
1505
+ "learning_rate": 5.636734693877552e-06,
1506
+ "loss": 0.2239,
1507
+ "step": 2140
1508
+ },
1509
+ {
1510
+ "epoch": 8.775510204081632,
1511
+ "grad_norm": 7.756526947021484,
1512
+ "learning_rate": 5.616326530612245e-06,
1513
+ "loss": 0.1974,
1514
+ "step": 2150
1515
+ },
1516
+ {
1517
+ "epoch": 8.816326530612244,
1518
+ "grad_norm": 7.697716236114502,
1519
+ "learning_rate": 5.59591836734694e-06,
1520
+ "loss": 0.258,
1521
+ "step": 2160
1522
+ },
1523
+ {
1524
+ "epoch": 8.857142857142858,
1525
+ "grad_norm": 8.356952667236328,
1526
+ "learning_rate": 5.575510204081633e-06,
1527
+ "loss": 0.1936,
1528
+ "step": 2170
1529
+ },
1530
+ {
1531
+ "epoch": 8.89795918367347,
1532
+ "grad_norm": 7.454901695251465,
1533
+ "learning_rate": 5.555102040816327e-06,
1534
+ "loss": 0.1504,
1535
+ "step": 2180
1536
+ },
1537
+ {
1538
+ "epoch": 8.938775510204081,
1539
+ "grad_norm": 4.652848720550537,
1540
+ "learning_rate": 5.534693877551021e-06,
1541
+ "loss": 0.1793,
1542
+ "step": 2190
1543
+ },
1544
+ {
1545
+ "epoch": 8.979591836734693,
1546
+ "grad_norm": 6.558859348297119,
1547
+ "learning_rate": 5.514285714285714e-06,
1548
+ "loss": 0.1974,
1549
+ "step": 2200
1550
+ },
1551
+ {
1552
+ "epoch": 9.020408163265307,
1553
+ "grad_norm": 7.869504928588867,
1554
+ "learning_rate": 5.493877551020409e-06,
1555
+ "loss": 0.2128,
1556
+ "step": 2210
1557
+ },
1558
+ {
1559
+ "epoch": 9.061224489795919,
1560
+ "grad_norm": 8.646539688110352,
1561
+ "learning_rate": 5.473469387755103e-06,
1562
+ "loss": 0.1755,
1563
+ "step": 2220
1564
+ },
1565
+ {
1566
+ "epoch": 9.10204081632653,
1567
+ "grad_norm": 7.58027458190918,
1568
+ "learning_rate": 5.453061224489796e-06,
1569
+ "loss": 0.1701,
1570
+ "step": 2230
1571
+ },
1572
+ {
1573
+ "epoch": 9.142857142857142,
1574
+ "grad_norm": 8.236135482788086,
1575
+ "learning_rate": 5.4326530612244895e-06,
1576
+ "loss": 0.219,
1577
+ "step": 2240
1578
+ },
1579
+ {
1580
+ "epoch": 9.183673469387756,
1581
+ "grad_norm": 6.239899635314941,
1582
+ "learning_rate": 5.4122448979591845e-06,
1583
+ "loss": 0.1846,
1584
+ "step": 2250
1585
+ },
1586
+ {
1587
+ "epoch": 9.224489795918368,
1588
+ "grad_norm": 5.651657581329346,
1589
+ "learning_rate": 5.391836734693879e-06,
1590
+ "loss": 0.237,
1591
+ "step": 2260
1592
+ },
1593
+ {
1594
+ "epoch": 9.26530612244898,
1595
+ "grad_norm": 6.521792411804199,
1596
+ "learning_rate": 5.371428571428572e-06,
1597
+ "loss": 0.2158,
1598
+ "step": 2270
1599
+ },
1600
+ {
1601
+ "epoch": 9.306122448979592,
1602
+ "grad_norm": 12.538895606994629,
1603
+ "learning_rate": 5.351020408163265e-06,
1604
+ "loss": 0.2103,
1605
+ "step": 2280
1606
+ },
1607
+ {
1608
+ "epoch": 9.346938775510203,
1609
+ "grad_norm": 6.341105937957764,
1610
+ "learning_rate": 5.330612244897959e-06,
1611
+ "loss": 0.1468,
1612
+ "step": 2290
1613
+ },
1614
+ {
1615
+ "epoch": 9.387755102040817,
1616
+ "grad_norm": 8.830240249633789,
1617
+ "learning_rate": 5.310204081632654e-06,
1618
+ "loss": 0.165,
1619
+ "step": 2300
1620
+ },
1621
+ {
1622
+ "epoch": 9.428571428571429,
1623
+ "grad_norm": 7.01543664932251,
1624
+ "learning_rate": 5.2897959183673476e-06,
1625
+ "loss": 0.1742,
1626
+ "step": 2310
1627
+ },
1628
+ {
1629
+ "epoch": 9.46938775510204,
1630
+ "grad_norm": 7.823669910430908,
1631
+ "learning_rate": 5.269387755102041e-06,
1632
+ "loss": 0.2414,
1633
+ "step": 2320
1634
+ },
1635
+ {
1636
+ "epoch": 9.510204081632653,
1637
+ "grad_norm": 7.9082770347595215,
1638
+ "learning_rate": 5.248979591836735e-06,
1639
+ "loss": 0.1961,
1640
+ "step": 2330
1641
+ },
1642
+ {
1643
+ "epoch": 9.551020408163264,
1644
+ "grad_norm": 8.552570343017578,
1645
+ "learning_rate": 5.22857142857143e-06,
1646
+ "loss": 0.1862,
1647
+ "step": 2340
1648
+ },
1649
+ {
1650
+ "epoch": 9.591836734693878,
1651
+ "grad_norm": 10.67483901977539,
1652
+ "learning_rate": 5.208163265306123e-06,
1653
+ "loss": 0.1773,
1654
+ "step": 2350
1655
+ },
1656
+ {
1657
+ "epoch": 9.63265306122449,
1658
+ "grad_norm": 5.287507057189941,
1659
+ "learning_rate": 5.1877551020408165e-06,
1660
+ "loss": 0.2165,
1661
+ "step": 2360
1662
+ },
1663
+ {
1664
+ "epoch": 9.673469387755102,
1665
+ "grad_norm": 3.9037272930145264,
1666
+ "learning_rate": 5.167346938775511e-06,
1667
+ "loss": 0.1598,
1668
+ "step": 2370
1669
+ },
1670
+ {
1671
+ "epoch": 9.714285714285714,
1672
+ "grad_norm": 2.465554714202881,
1673
+ "learning_rate": 5.146938775510204e-06,
1674
+ "loss": 0.1398,
1675
+ "step": 2380
1676
+ },
1677
+ {
1678
+ "epoch": 9.755102040816327,
1679
+ "grad_norm": 4.5391950607299805,
1680
+ "learning_rate": 5.126530612244899e-06,
1681
+ "loss": 0.1642,
1682
+ "step": 2390
1683
+ },
1684
+ {
1685
+ "epoch": 9.795918367346939,
1686
+ "grad_norm": 9.52216911315918,
1687
+ "learning_rate": 5.106122448979592e-06,
1688
+ "loss": 0.2317,
1689
+ "step": 2400
1690
+ },
1691
+ {
1692
+ "epoch": 9.83673469387755,
1693
+ "grad_norm": 7.413267135620117,
1694
+ "learning_rate": 5.085714285714286e-06,
1695
+ "loss": 0.2339,
1696
+ "step": 2410
1697
+ },
1698
+ {
1699
+ "epoch": 9.877551020408163,
1700
+ "grad_norm": 11.318732261657715,
1701
+ "learning_rate": 5.0653061224489795e-06,
1702
+ "loss": 0.2225,
1703
+ "step": 2420
1704
+ },
1705
+ {
1706
+ "epoch": 9.918367346938776,
1707
+ "grad_norm": 11.242213249206543,
1708
+ "learning_rate": 5.0448979591836745e-06,
1709
+ "loss": 0.3078,
1710
+ "step": 2430
1711
+ },
1712
+ {
1713
+ "epoch": 9.959183673469388,
1714
+ "grad_norm": 9.637420654296875,
1715
+ "learning_rate": 5.024489795918368e-06,
1716
+ "loss": 0.1856,
1717
+ "step": 2440
1718
+ },
1719
+ {
1720
+ "epoch": 10.0,
1721
+ "grad_norm": 10.665837287902832,
1722
+ "learning_rate": 5.004081632653062e-06,
1723
+ "loss": 0.2125,
1724
+ "step": 2450
1725
+ },
1726
+ {
1727
+ "epoch": 10.040816326530612,
1728
+ "grad_norm": 5.397568702697754,
1729
+ "learning_rate": 4.983673469387755e-06,
1730
+ "loss": 0.1722,
1731
+ "step": 2460
1732
+ },
1733
+ {
1734
+ "epoch": 10.081632653061224,
1735
+ "grad_norm": 5.377045631408691,
1736
+ "learning_rate": 4.963265306122449e-06,
1737
+ "loss": 0.1958,
1738
+ "step": 2470
1739
+ },
1740
+ {
1741
+ "epoch": 10.122448979591837,
1742
+ "grad_norm": 7.449840068817139,
1743
+ "learning_rate": 4.9428571428571435e-06,
1744
+ "loss": 0.1832,
1745
+ "step": 2480
1746
+ },
1747
+ {
1748
+ "epoch": 10.16326530612245,
1749
+ "grad_norm": 7.062196731567383,
1750
+ "learning_rate": 4.9224489795918376e-06,
1751
+ "loss": 0.1913,
1752
+ "step": 2490
1753
+ },
1754
+ {
1755
+ "epoch": 10.204081632653061,
1756
+ "grad_norm": 6.084071636199951,
1757
+ "learning_rate": 4.902040816326531e-06,
1758
+ "loss": 0.1187,
1759
+ "step": 2500
1760
+ },
1761
+ {
1762
+ "epoch": 10.244897959183673,
1763
+ "grad_norm": 4.809638023376465,
1764
+ "learning_rate": 4.881632653061225e-06,
1765
+ "loss": 0.1405,
1766
+ "step": 2510
1767
+ },
1768
+ {
1769
+ "epoch": 10.285714285714286,
1770
+ "grad_norm": 4.271588325500488,
1771
+ "learning_rate": 4.861224489795919e-06,
1772
+ "loss": 0.1357,
1773
+ "step": 2520
1774
+ },
1775
+ {
1776
+ "epoch": 10.326530612244898,
1777
+ "grad_norm": 8.498083114624023,
1778
+ "learning_rate": 4.840816326530612e-06,
1779
+ "loss": 0.2201,
1780
+ "step": 2530
1781
+ },
1782
+ {
1783
+ "epoch": 10.36734693877551,
1784
+ "grad_norm": 7.665253162384033,
1785
+ "learning_rate": 4.8204081632653065e-06,
1786
+ "loss": 0.1959,
1787
+ "step": 2540
1788
+ },
1789
+ {
1790
+ "epoch": 10.408163265306122,
1791
+ "grad_norm": 5.3332743644714355,
1792
+ "learning_rate": 4.800000000000001e-06,
1793
+ "loss": 0.1956,
1794
+ "step": 2550
1795
+ },
1796
+ {
1797
+ "epoch": 10.448979591836734,
1798
+ "grad_norm": 4.948959827423096,
1799
+ "learning_rate": 4.779591836734695e-06,
1800
+ "loss": 0.2092,
1801
+ "step": 2560
1802
+ },
1803
+ {
1804
+ "epoch": 10.489795918367347,
1805
+ "grad_norm": 7.226635456085205,
1806
+ "learning_rate": 4.759183673469388e-06,
1807
+ "loss": 0.1584,
1808
+ "step": 2570
1809
+ },
1810
+ {
1811
+ "epoch": 10.53061224489796,
1812
+ "grad_norm": 4.212419509887695,
1813
+ "learning_rate": 4.738775510204082e-06,
1814
+ "loss": 0.1476,
1815
+ "step": 2580
1816
+ },
1817
+ {
1818
+ "epoch": 10.571428571428571,
1819
+ "grad_norm": 7.545459270477295,
1820
+ "learning_rate": 4.7183673469387754e-06,
1821
+ "loss": 0.2103,
1822
+ "step": 2590
1823
+ },
1824
+ {
1825
+ "epoch": 10.612244897959183,
1826
+ "grad_norm": 1.853903889656067,
1827
+ "learning_rate": 4.69795918367347e-06,
1828
+ "loss": 0.2105,
1829
+ "step": 2600
1830
+ },
1831
+ {
1832
+ "epoch": 10.653061224489797,
1833
+ "grad_norm": 4.872255802154541,
1834
+ "learning_rate": 4.677551020408164e-06,
1835
+ "loss": 0.1979,
1836
+ "step": 2610
1837
+ },
1838
+ {
1839
+ "epoch": 10.693877551020408,
1840
+ "grad_norm": 8.121048927307129,
1841
+ "learning_rate": 4.657142857142857e-06,
1842
+ "loss": 0.1904,
1843
+ "step": 2620
1844
+ },
1845
+ {
1846
+ "epoch": 10.73469387755102,
1847
+ "grad_norm": 6.789455890655518,
1848
+ "learning_rate": 4.636734693877551e-06,
1849
+ "loss": 0.231,
1850
+ "step": 2630
1851
+ },
1852
+ {
1853
+ "epoch": 10.775510204081632,
1854
+ "grad_norm": 6.744252681732178,
1855
+ "learning_rate": 4.616326530612245e-06,
1856
+ "loss": 0.1028,
1857
+ "step": 2640
1858
+ },
1859
+ {
1860
+ "epoch": 10.816326530612244,
1861
+ "grad_norm": 1.3881915807724,
1862
+ "learning_rate": 4.595918367346939e-06,
1863
+ "loss": 0.1611,
1864
+ "step": 2650
1865
+ },
1866
+ {
1867
+ "epoch": 10.857142857142858,
1868
+ "grad_norm": 4.106182098388672,
1869
+ "learning_rate": 4.575510204081633e-06,
1870
+ "loss": 0.1721,
1871
+ "step": 2660
1872
+ },
1873
+ {
1874
+ "epoch": 10.89795918367347,
1875
+ "grad_norm": 6.046643257141113,
1876
+ "learning_rate": 4.555102040816327e-06,
1877
+ "loss": 0.2162,
1878
+ "step": 2670
1879
+ },
1880
+ {
1881
+ "epoch": 10.938775510204081,
1882
+ "grad_norm": 2.952366352081299,
1883
+ "learning_rate": 4.534693877551021e-06,
1884
+ "loss": 0.1941,
1885
+ "step": 2680
1886
+ },
1887
+ {
1888
+ "epoch": 10.979591836734693,
1889
+ "grad_norm": 5.679470539093018,
1890
+ "learning_rate": 4.514285714285714e-06,
1891
+ "loss": 0.1322,
1892
+ "step": 2690
1893
+ },
1894
+ {
1895
+ "epoch": 11.020408163265307,
1896
+ "grad_norm": 4.781472206115723,
1897
+ "learning_rate": 4.493877551020408e-06,
1898
+ "loss": 0.1756,
1899
+ "step": 2700
1900
+ },
1901
+ {
1902
+ "epoch": 11.061224489795919,
1903
+ "grad_norm": 4.453323841094971,
1904
+ "learning_rate": 4.473469387755102e-06,
1905
+ "loss": 0.1652,
1906
+ "step": 2710
1907
+ },
1908
+ {
1909
+ "epoch": 11.10204081632653,
1910
+ "grad_norm": 6.313192367553711,
1911
+ "learning_rate": 4.4530612244897965e-06,
1912
+ "loss": 0.2071,
1913
+ "step": 2720
1914
+ },
1915
+ {
1916
+ "epoch": 11.142857142857142,
1917
+ "grad_norm": 3.620354413986206,
1918
+ "learning_rate": 4.43265306122449e-06,
1919
+ "loss": 0.2166,
1920
+ "step": 2730
1921
+ },
1922
+ {
1923
+ "epoch": 11.183673469387756,
1924
+ "grad_norm": 5.309145450592041,
1925
+ "learning_rate": 4.412244897959184e-06,
1926
+ "loss": 0.1787,
1927
+ "step": 2740
1928
+ },
1929
+ {
1930
+ "epoch": 11.224489795918368,
1931
+ "grad_norm": 5.886495590209961,
1932
+ "learning_rate": 4.391836734693878e-06,
1933
+ "loss": 0.1699,
1934
+ "step": 2750
1935
+ },
1936
+ {
1937
+ "epoch": 11.26530612244898,
1938
+ "grad_norm": 6.391637325286865,
1939
+ "learning_rate": 4.371428571428572e-06,
1940
+ "loss": 0.1928,
1941
+ "step": 2760
1942
+ },
1943
+ {
1944
+ "epoch": 11.306122448979592,
1945
+ "grad_norm": 9.88926887512207,
1946
+ "learning_rate": 4.3510204081632654e-06,
1947
+ "loss": 0.2344,
1948
+ "step": 2770
1949
+ },
1950
+ {
1951
+ "epoch": 11.346938775510203,
1952
+ "grad_norm": 4.097028732299805,
1953
+ "learning_rate": 4.3306122448979596e-06,
1954
+ "loss": 0.1457,
1955
+ "step": 2780
1956
+ },
1957
+ {
1958
+ "epoch": 11.387755102040817,
1959
+ "grad_norm": 6.217812538146973,
1960
+ "learning_rate": 4.310204081632654e-06,
1961
+ "loss": 0.1773,
1962
+ "step": 2790
1963
+ },
1964
+ {
1965
+ "epoch": 11.428571428571429,
1966
+ "grad_norm": 6.636881351470947,
1967
+ "learning_rate": 4.289795918367347e-06,
1968
+ "loss": 0.1977,
1969
+ "step": 2800
1970
+ },
1971
+ {
1972
+ "epoch": 11.46938775510204,
1973
+ "grad_norm": 4.406877517700195,
1974
+ "learning_rate": 4.269387755102041e-06,
1975
+ "loss": 0.1331,
1976
+ "step": 2810
1977
+ },
1978
+ {
1979
+ "epoch": 11.510204081632653,
1980
+ "grad_norm": 7.739985466003418,
1981
+ "learning_rate": 4.248979591836735e-06,
1982
+ "loss": 0.1624,
1983
+ "step": 2820
1984
+ },
1985
+ {
1986
+ "epoch": 11.551020408163264,
1987
+ "grad_norm": 8.672730445861816,
1988
+ "learning_rate": 4.228571428571429e-06,
1989
+ "loss": 0.1435,
1990
+ "step": 2830
1991
+ },
1992
+ {
1993
+ "epoch": 11.591836734693878,
1994
+ "grad_norm": 8.240572929382324,
1995
+ "learning_rate": 4.208163265306123e-06,
1996
+ "loss": 0.1864,
1997
+ "step": 2840
1998
+ },
1999
+ {
2000
+ "epoch": 11.63265306122449,
2001
+ "grad_norm": 6.257174968719482,
2002
+ "learning_rate": 4.187755102040817e-06,
2003
+ "loss": 0.2008,
2004
+ "step": 2850
2005
+ },
2006
+ {
2007
+ "epoch": 11.673469387755102,
2008
+ "grad_norm": 7.619112014770508,
2009
+ "learning_rate": 4.167346938775511e-06,
2010
+ "loss": 0.1529,
2011
+ "step": 2860
2012
+ },
2013
+ {
2014
+ "epoch": 11.714285714285714,
2015
+ "grad_norm": 3.644754648208618,
2016
+ "learning_rate": 4.146938775510204e-06,
2017
+ "loss": 0.1434,
2018
+ "step": 2870
2019
+ },
2020
+ {
2021
+ "epoch": 11.755102040816327,
2022
+ "grad_norm": 3.984260082244873,
2023
+ "learning_rate": 4.126530612244898e-06,
2024
+ "loss": 0.1327,
2025
+ "step": 2880
2026
+ },
2027
+ {
2028
+ "epoch": 11.795918367346939,
2029
+ "grad_norm": 7.851540565490723,
2030
+ "learning_rate": 4.106122448979592e-06,
2031
+ "loss": 0.1709,
2032
+ "step": 2890
2033
+ },
2034
+ {
2035
+ "epoch": 11.83673469387755,
2036
+ "grad_norm": 2.9527790546417236,
2037
+ "learning_rate": 4.0857142857142865e-06,
2038
+ "loss": 0.1487,
2039
+ "step": 2900
2040
+ },
2041
+ {
2042
+ "epoch": 11.877551020408163,
2043
+ "grad_norm": 3.5094189643859863,
2044
+ "learning_rate": 4.06530612244898e-06,
2045
+ "loss": 0.1733,
2046
+ "step": 2910
2047
+ },
2048
+ {
2049
+ "epoch": 11.918367346938776,
2050
+ "grad_norm": 8.27623176574707,
2051
+ "learning_rate": 4.044897959183674e-06,
2052
+ "loss": 0.2207,
2053
+ "step": 2920
2054
+ },
2055
+ {
2056
+ "epoch": 11.959183673469388,
2057
+ "grad_norm": 6.333447456359863,
2058
+ "learning_rate": 4.024489795918368e-06,
2059
+ "loss": 0.1912,
2060
+ "step": 2930
2061
+ },
2062
+ {
2063
+ "epoch": 12.0,
2064
+ "grad_norm": 3.803018569946289,
2065
+ "learning_rate": 4.004081632653062e-06,
2066
+ "loss": 0.1969,
2067
+ "step": 2940
2068
+ },
2069
+ {
2070
+ "epoch": 12.040816326530612,
2071
+ "grad_norm": 7.993055820465088,
2072
+ "learning_rate": 3.9836734693877555e-06,
2073
+ "loss": 0.1708,
2074
+ "step": 2950
2075
+ },
2076
+ {
2077
+ "epoch": 12.081632653061224,
2078
+ "grad_norm": 3.1507952213287354,
2079
+ "learning_rate": 3.963265306122449e-06,
2080
+ "loss": 0.0929,
2081
+ "step": 2960
2082
+ },
2083
+ {
2084
+ "epoch": 12.122448979591837,
2085
+ "grad_norm": 3.0786514282226562,
2086
+ "learning_rate": 3.942857142857143e-06,
2087
+ "loss": 0.1392,
2088
+ "step": 2970
2089
+ },
2090
+ {
2091
+ "epoch": 12.16326530612245,
2092
+ "grad_norm": 12.369327545166016,
2093
+ "learning_rate": 3.922448979591837e-06,
2094
+ "loss": 0.1887,
2095
+ "step": 2980
2096
+ },
2097
+ {
2098
+ "epoch": 12.204081632653061,
2099
+ "grad_norm": 11.310943603515625,
2100
+ "learning_rate": 3.902040816326531e-06,
2101
+ "loss": 0.2074,
2102
+ "step": 2990
2103
+ },
2104
+ {
2105
+ "epoch": 12.244897959183673,
2106
+ "grad_norm": 8.117951393127441,
2107
+ "learning_rate": 3.881632653061224e-06,
2108
+ "loss": 0.2066,
2109
+ "step": 3000
2110
+ }
2111
+ ],
2112
+ "logging_steps": 10,
2113
+ "max_steps": 4900,
2114
+ "num_input_tokens_seen": 0,
2115
+ "num_train_epochs": 20,
2116
+ "save_steps": 1000,
2117
+ "stateful_callbacks": {
2118
+ "TrainerControl": {
2119
+ "args": {
2120
+ "should_epoch_stop": false,
2121
+ "should_evaluate": false,
2122
+ "should_log": false,
2123
+ "should_save": true,
2124
+ "should_training_stop": false
2125
+ },
2126
+ "attributes": {}
2127
+ }
2128
+ },
2129
+ "total_flos": 0.0,
2130
+ "train_batch_size": 30,
2131
+ "trial_name": null,
2132
+ "trial_params": null
2133
+ }
checkpoint-3000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c4dbbc953b5f35ee9087cab8da231051761999ae7db38443ee8d71e351fa7db6
3
+ size 5304
checkpoint-3000/vocab.txt ADDED
The diff for this file is too large to render. See raw diff