Hugging Face's logo

trongg
/
Finetune_hackathon_BKAI

Safetensors
Model card Files Community
trongg commited on
Commit
d77563d
•
1 Parent(s): 025db00

Upload folder using huggingface_hub

Browse files
Files changed (14) hide show
  1. .gitattributes +5 -0
  2. checkpoint-22560/colbert_linear.pt +3 -0
  3. checkpoint-22560/config.json +28 -0
  4. checkpoint-22560/global_step22560/mp_rank_00_model_states.pt +3 -0
  5. checkpoint-22560/latest +1 -0
  6. checkpoint-22560/model.safetensors +3 -0
  7. checkpoint-22560/rng_state.pth +0 -0
  8. checkpoint-22560/sentencepiece.bpe.model +3 -0
  9. checkpoint-22560/sparse_linear.pt +0 -0
  10. checkpoint-22560/special_tokens_map.json +51 -0
  11. checkpoint-22560/tokenizer.json +3 -0
  12. checkpoint-22560/tokenizer_config.json +55 -0
  13. checkpoint-22560/trainer_state.json +3190 -0
  14. checkpoint-22560/training_args.bin +0 -0
.gitattributes CHANGED
@@ -26,3 +26,8 @@ checkpoint-21250/global_step21250/mp_rank_00_model_states.pt filter=lfs diff=lfs
26
  checkpoint-21250/model.safetensors filter=lfs diff=lfs merge=lfs -text
27
  checkpoint-21250/sentencepiece.bpe.model filter=lfs diff=lfs merge=lfs -text
28
  checkpoint-21250/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
26
  checkpoint-21250/model.safetensors filter=lfs diff=lfs merge=lfs -text
27
  checkpoint-21250/sentencepiece.bpe.model filter=lfs diff=lfs merge=lfs -text
28
  checkpoint-21250/tokenizer.json filter=lfs diff=lfs merge=lfs -text
29
+ checkpoint-22560/colbert_linear.pt filter=lfs diff=lfs merge=lfs -text
30
+ checkpoint-22560/global_step22560/mp_rank_00_model_states.pt filter=lfs diff=lfs merge=lfs -text
31
+ checkpoint-22560/model.safetensors filter=lfs diff=lfs merge=lfs -text
32
+ checkpoint-22560/sentencepiece.bpe.model filter=lfs diff=lfs merge=lfs -text
33
+ checkpoint-22560/tokenizer.json filter=lfs diff=lfs merge=lfs -text
checkpoint-22560/colbert_linear.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d58c27f9c1fc56b2aa4501169a077c39a90421817e365cf6cf9a3cc2b6042826
3
+ size 2100674
checkpoint-22560/config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/root/.cache/huggingface/hub/models--BAAI--bge-m3/snapshots/5617a9f61b028005a4858fdac845db406aefb181",
3
+ "architectures": [
4
+ "XLMRobertaModel"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "eos_token_id": 2,
10
+ "hidden_act": "gelu",
11
+ "hidden_dropout_prob": 0.1,
12
+ "hidden_size": 1024,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 4096,
15
+ "layer_norm_eps": 1e-05,
16
+ "max_position_embeddings": 8194,
17
+ "model_type": "xlm-roberta",
18
+ "num_attention_heads": 16,
19
+ "num_hidden_layers": 24,
20
+ "output_past": true,
21
+ "pad_token_id": 1,
22
+ "position_embedding_type": "absolute",
23
+ "torch_dtype": "float16",
24
+ "transformers_version": "4.44.2",
25
+ "type_vocab_size": 1,
26
+ "use_cache": true,
27
+ "vocab_size": 250002
28
+ }
checkpoint-22560/global_step22560/mp_rank_00_model_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8923ea3d372caef94bc643f74f16971002a41a2c43f80be918d1c3c772fa2616
3
+ size 7963362556
checkpoint-22560/latest ADDED
@@ -0,0 +1 @@
 
 
1
+ global_step22560
checkpoint-22560/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:84d9b553c6e31a7c33522be0c0c107aac242bf273348ffa1a15f617ebda4a096
3
+ size 1135554344
checkpoint-22560/rng_state.pth ADDED
Binary file (14.2 kB). View file
 
checkpoint-22560/sentencepiece.bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
3
+ size 5069051
checkpoint-22560/sparse_linear.pt ADDED
Binary file (3.52 kB). View file
 
checkpoint-22560/special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "<s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "</s>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "<mask>",
25
+ "lstrip": true,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "<pad>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "</s>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "<unk>",
46
+ "lstrip": false,
47
+ "normalized": false,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
checkpoint-22560/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33cd99e33ce09bdd8a6136fddfe90a1c47f85bafedf7309d0eecc19012d43586
3
+ size 17082897
checkpoint-22560/tokenizer_config.json ADDED
@@ -0,0 +1,55 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "250001": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "<s>",
45
+ "clean_up_tokenization_spaces": true,
46
+ "cls_token": "<s>",
47
+ "eos_token": "</s>",
48
+ "mask_token": "<mask>",
49
+ "model_max_length": 8192,
50
+ "pad_token": "<pad>",
51
+ "sep_token": "</s>",
52
+ "sp_model_kwargs": {},
53
+ "tokenizer_class": "XLMRobertaTokenizer",
54
+ "unk_token": "<unk>"
55
+ }
checkpoint-22560/trainer_state.json ADDED
@@ -0,0 +1,3190 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.0,
5
+ "eval_steps": 500,
6
+ "global_step": 22560,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.004432624113475178,
13
+ "grad_norm": 0.009513450788341307,
14
+ "learning_rate": 5.066502241635949e-06,
15
+ "loss": 0.0353,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 0.008865248226950355,
20
+ "grad_norm": 0.12959164414811705,
21
+ "learning_rate": 5.964204463534375e-06,
22
+ "loss": 0.0209,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 0.013297872340425532,
27
+ "grad_norm": 0.03079899259156006,
28
+ "learning_rate": 6.489326600159014e-06,
29
+ "loss": 0.0114,
30
+ "step": 150
31
+ },
32
+ {
33
+ "epoch": 0.01773049645390071,
34
+ "grad_norm": 1.904334998115993,
35
+ "learning_rate": 6.855414889666249e-06,
36
+ "loss": 0.0426,
37
+ "step": 200
38
+ },
39
+ {
40
+ "epoch": 0.022163120567375887,
41
+ "grad_norm": 0.00029657300723797526,
42
+ "learning_rate": 7.14571142073327e-06,
43
+ "loss": 0.0092,
44
+ "step": 250
45
+ },
46
+ {
47
+ "epoch": 0.026595744680851064,
48
+ "grad_norm": 0.5635451567630495,
49
+ "learning_rate": 7.382704575824005e-06,
50
+ "loss": 0.0207,
51
+ "step": 300
52
+ },
53
+ {
54
+ "epoch": 0.03102836879432624,
55
+ "grad_norm": 17.174480153841266,
56
+ "learning_rate": 7.582965380729319e-06,
57
+ "loss": 0.0128,
58
+ "step": 350
59
+ },
60
+ {
61
+ "epoch": 0.03546099290780142,
62
+ "grad_norm": 0.0029856714585610065,
63
+ "learning_rate": 7.756367076993413e-06,
64
+ "loss": 0.0087,
65
+ "step": 400
66
+ },
67
+ {
68
+ "epoch": 0.0398936170212766,
69
+ "grad_norm": 46.59324697917607,
70
+ "learning_rate": 7.909269732702365e-06,
71
+ "loss": 0.0358,
72
+ "step": 450
73
+ },
74
+ {
75
+ "epoch": 0.044326241134751775,
76
+ "grad_norm": 0.011038539916815544,
77
+ "learning_rate": 8.046011658635775e-06,
78
+ "loss": 0.0099,
79
+ "step": 500
80
+ },
81
+ {
82
+ "epoch": 0.04875886524822695,
83
+ "grad_norm": 0.06122707900561642,
84
+ "learning_rate": 8.169684802472282e-06,
85
+ "loss": 0.0257,
86
+ "step": 550
87
+ },
88
+ {
89
+ "epoch": 0.05319148936170213,
90
+ "grad_norm": 11.17528630173658,
91
+ "learning_rate": 8.28257072561721e-06,
92
+ "loss": 0.0399,
93
+ "step": 600
94
+ },
95
+ {
96
+ "epoch": 0.057624113475177305,
97
+ "grad_norm": 0.004434274436946811,
98
+ "learning_rate": 8.386401186603816e-06,
99
+ "loss": 0.0106,
100
+ "step": 650
101
+ },
102
+ {
103
+ "epoch": 0.06205673758865248,
104
+ "grad_norm": 0.00022486652288303773,
105
+ "learning_rate": 8.482521734002021e-06,
106
+ "loss": 0.0175,
107
+ "step": 700
108
+ },
109
+ {
110
+ "epoch": 0.06648936170212766,
111
+ "grad_norm": 0.08252216933057055,
112
+ "learning_rate": 8.571998643735778e-06,
113
+ "loss": 0.0122,
114
+ "step": 750
115
+ },
116
+ {
117
+ "epoch": 0.07092198581560284,
118
+ "grad_norm": 38.71346721678791,
119
+ "learning_rate": 8.655691228400912e-06,
120
+ "loss": 0.0107,
121
+ "step": 800
122
+ },
123
+ {
124
+ "epoch": 0.07535460992907801,
125
+ "grad_norm": 0.016399532174579123,
126
+ "learning_rate": 8.734302159716567e-06,
127
+ "loss": 0.0272,
128
+ "step": 850
129
+ },
130
+ {
131
+ "epoch": 0.0797872340425532,
132
+ "grad_norm": 13.96210364288869,
133
+ "learning_rate": 8.808413368821617e-06,
134
+ "loss": 0.0162,
135
+ "step": 900
136
+ },
137
+ {
138
+ "epoch": 0.08421985815602837,
139
+ "grad_norm": 20.616657581291367,
140
+ "learning_rate": 8.878512216644468e-06,
141
+ "loss": 0.018,
142
+ "step": 950
143
+ },
144
+ {
145
+ "epoch": 0.08865248226950355,
146
+ "grad_norm": 8.554096191036654,
147
+ "learning_rate": 8.945010936770532e-06,
148
+ "loss": 0.0122,
149
+ "step": 1000
150
+ },
151
+ {
152
+ "epoch": 0.09308510638297872,
153
+ "grad_norm": 9.002995310481221,
154
+ "learning_rate": 9.008261325009621e-06,
155
+ "loss": 0.004,
156
+ "step": 1050
157
+ },
158
+ {
159
+ "epoch": 0.0975177304964539,
160
+ "grad_norm": 0.04332552525052799,
161
+ "learning_rate": 9.068566005374956e-06,
162
+ "loss": 0.0049,
163
+ "step": 1100
164
+ },
165
+ {
166
+ "epoch": 0.10195035460992907,
167
+ "grad_norm": 0.566818534495066,
168
+ "learning_rate": 9.126187187473155e-06,
169
+ "loss": 0.0348,
170
+ "step": 1150
171
+ },
172
+ {
173
+ "epoch": 0.10638297872340426,
174
+ "grad_norm": 2.2102759352121244,
175
+ "learning_rate": 9.181353557126703e-06,
176
+ "loss": 0.0157,
177
+ "step": 1200
178
+ },
179
+ {
180
+ "epoch": 0.11081560283687943,
181
+ "grad_norm": 0.0007883033335850071,
182
+ "learning_rate": 9.233228424991629e-06,
183
+ "loss": 0.0153,
184
+ "step": 1250
185
+ },
186
+ {
187
+ "epoch": 0.11524822695035461,
188
+ "grad_norm": 8.962657174789879,
189
+ "learning_rate": 9.28410340850224e-06,
190
+ "loss": 0.011,
191
+ "step": 1300
192
+ },
193
+ {
194
+ "epoch": 0.1196808510638298,
195
+ "grad_norm": 0.08478258585114734,
196
+ "learning_rate": 9.333055212266742e-06,
197
+ "loss": 0.0193,
198
+ "step": 1350
199
+ },
200
+ {
201
+ "epoch": 0.12411347517730496,
202
+ "grad_norm": 5.126269299634291,
203
+ "learning_rate": 9.380223955900445e-06,
204
+ "loss": 0.0224,
205
+ "step": 1400
206
+ },
207
+ {
208
+ "epoch": 0.12854609929078015,
209
+ "grad_norm": 0.001630326745394445,
210
+ "learning_rate": 9.425734982425631e-06,
211
+ "loss": 0.0198,
212
+ "step": 1450
213
+ },
214
+ {
215
+ "epoch": 0.13297872340425532,
216
+ "grad_norm": 31.3251284414765,
217
+ "learning_rate": 9.469700865634203e-06,
218
+ "loss": 0.0201,
219
+ "step": 1500
220
+ },
221
+ {
222
+ "epoch": 0.1374113475177305,
223
+ "grad_norm": 0.003720201610220148,
224
+ "learning_rate": 9.512223087748166e-06,
225
+ "loss": 0.0386,
226
+ "step": 1550
227
+ },
228
+ {
229
+ "epoch": 0.14184397163120568,
230
+ "grad_norm": 1.4908536371789232,
231
+ "learning_rate": 9.553393450299339e-06,
232
+ "loss": 0.0155,
233
+ "step": 1600
234
+ },
235
+ {
236
+ "epoch": 0.14627659574468085,
237
+ "grad_norm": 0.18557111682138577,
238
+ "learning_rate": 9.593295267571064e-06,
239
+ "loss": 0.0132,
240
+ "step": 1650
241
+ },
242
+ {
243
+ "epoch": 0.15070921985815602,
244
+ "grad_norm": 0.0012351164899243579,
245
+ "learning_rate": 9.632004381614992e-06,
246
+ "loss": 0.0169,
247
+ "step": 1700
248
+ },
249
+ {
250
+ "epoch": 0.1551418439716312,
251
+ "grad_norm": 4.390950581164127,
252
+ "learning_rate": 9.669590029926173e-06,
253
+ "loss": 0.0188,
254
+ "step": 1750
255
+ },
256
+ {
257
+ "epoch": 0.1595744680851064,
258
+ "grad_norm": 1.5823201233659507,
259
+ "learning_rate": 9.706115590720043e-06,
260
+ "loss": 0.0176,
261
+ "step": 1800
262
+ },
263
+ {
264
+ "epoch": 0.16400709219858156,
265
+ "grad_norm": 0.6041784257950227,
266
+ "learning_rate": 9.741639225963296e-06,
267
+ "loss": 0.0076,
268
+ "step": 1850
269
+ },
270
+ {
271
+ "epoch": 0.16843971631205673,
272
+ "grad_norm": 8.096633927567137,
273
+ "learning_rate": 9.776214438542894e-06,
274
+ "loss": 0.0254,
275
+ "step": 1900
276
+ },
277
+ {
278
+ "epoch": 0.17287234042553193,
279
+ "grad_norm": 0.001381384244099819,
280
+ "learning_rate": 9.80989055697396e-06,
281
+ "loss": 0.021,
282
+ "step": 1950
283
+ },
284
+ {
285
+ "epoch": 0.1773049645390071,
286
+ "grad_norm": 0.6173202999510908,
287
+ "learning_rate": 9.842713158668956e-06,
288
+ "loss": 0.0153,
289
+ "step": 2000
290
+ },
291
+ {
292
+ "epoch": 0.18173758865248227,
293
+ "grad_norm": 0.4702304667158159,
294
+ "learning_rate": 9.874724440882673e-06,
295
+ "loss": 0.0322,
296
+ "step": 2050
297
+ },
298
+ {
299
+ "epoch": 0.18617021276595744,
300
+ "grad_norm": 0.007418267836546937,
301
+ "learning_rate": 9.905963546908046e-06,
302
+ "loss": 0.0189,
303
+ "step": 2100
304
+ },
305
+ {
306
+ "epoch": 0.19060283687943264,
307
+ "grad_norm": 0.040611583128721894,
308
+ "learning_rate": 9.936466853848526e-06,
309
+ "loss": 0.004,
310
+ "step": 2150
311
+ },
312
+ {
313
+ "epoch": 0.1950354609929078,
314
+ "grad_norm": 35.80087991837194,
315
+ "learning_rate": 9.96626822727338e-06,
316
+ "loss": 0.022,
317
+ "step": 2200
318
+ },
319
+ {
320
+ "epoch": 0.19946808510638298,
321
+ "grad_norm": 0.44489231698014525,
322
+ "learning_rate": 9.995399247226628e-06,
323
+ "loss": 0.0165,
324
+ "step": 2250
325
+ },
326
+ {
327
+ "epoch": 0.20390070921985815,
328
+ "grad_norm": 0.0002357955980587179,
329
+ "learning_rate": 9.980299448384557e-06,
330
+ "loss": 0.0091,
331
+ "step": 2300
332
+ },
333
+ {
334
+ "epoch": 0.20833333333333334,
335
+ "grad_norm": 0.011156109382490509,
336
+ "learning_rate": 9.955673758865249e-06,
337
+ "loss": 0.0293,
338
+ "step": 2350
339
+ },
340
+ {
341
+ "epoch": 0.2127659574468085,
342
+ "grad_norm": 0.010237311021590336,
343
+ "learning_rate": 9.931048069345942e-06,
344
+ "loss": 0.0267,
345
+ "step": 2400
346
+ },
347
+ {
348
+ "epoch": 0.21719858156028368,
349
+ "grad_norm": 0.1907181340319684,
350
+ "learning_rate": 9.906422379826635e-06,
351
+ "loss": 0.0035,
352
+ "step": 2450
353
+ },
354
+ {
355
+ "epoch": 0.22163120567375885,
356
+ "grad_norm": 7.244482111036572,
357
+ "learning_rate": 9.881796690307329e-06,
358
+ "loss": 0.0129,
359
+ "step": 2500
360
+ },
361
+ {
362
+ "epoch": 0.22606382978723405,
363
+ "grad_norm": 2.614333142067009,
364
+ "learning_rate": 9.857171000788024e-06,
365
+ "loss": 0.0176,
366
+ "step": 2550
367
+ },
368
+ {
369
+ "epoch": 0.23049645390070922,
370
+ "grad_norm": 0.19026758190049964,
371
+ "learning_rate": 9.832545311268717e-06,
372
+ "loss": 0.0118,
373
+ "step": 2600
374
+ },
375
+ {
376
+ "epoch": 0.2349290780141844,
377
+ "grad_norm": 1.926354870277716,
378
+ "learning_rate": 9.80791962174941e-06,
379
+ "loss": 0.0113,
380
+ "step": 2650
381
+ },
382
+ {
383
+ "epoch": 0.2393617021276596,
384
+ "grad_norm": 1.3390029353713434e-05,
385
+ "learning_rate": 9.783293932230104e-06,
386
+ "loss": 0.0095,
387
+ "step": 2700
388
+ },
389
+ {
390
+ "epoch": 0.24379432624113476,
391
+ "grad_norm": 0.25506782394102356,
392
+ "learning_rate": 9.758668242710797e-06,
393
+ "loss": 0.0235,
394
+ "step": 2750
395
+ },
396
+ {
397
+ "epoch": 0.24822695035460993,
398
+ "grad_norm": 0.0011667263533171556,
399
+ "learning_rate": 9.73404255319149e-06,
400
+ "loss": 0.01,
401
+ "step": 2800
402
+ },
403
+ {
404
+ "epoch": 0.2526595744680851,
405
+ "grad_norm": 2.676783820787547,
406
+ "learning_rate": 9.709416863672184e-06,
407
+ "loss": 0.0305,
408
+ "step": 2850
409
+ },
410
+ {
411
+ "epoch": 0.2570921985815603,
412
+ "grad_norm": 0.0019099837220493063,
413
+ "learning_rate": 9.684791174152877e-06,
414
+ "loss": 0.0246,
415
+ "step": 2900
416
+ },
417
+ {
418
+ "epoch": 0.26152482269503546,
419
+ "grad_norm": 0.6898702380926142,
420
+ "learning_rate": 9.66016548463357e-06,
421
+ "loss": 0.0211,
422
+ "step": 2950
423
+ },
424
+ {
425
+ "epoch": 0.26595744680851063,
426
+ "grad_norm": 0.0006438124789484562,
427
+ "learning_rate": 9.635539795114264e-06,
428
+ "loss": 0.008,
429
+ "step": 3000
430
+ },
431
+ {
432
+ "epoch": 0.2703900709219858,
433
+ "grad_norm": 3.819700570925564e-05,
434
+ "learning_rate": 9.610914105594957e-06,
435
+ "loss": 0.0199,
436
+ "step": 3050
437
+ },
438
+ {
439
+ "epoch": 0.274822695035461,
440
+ "grad_norm": 0.019487722143147885,
441
+ "learning_rate": 9.58628841607565e-06,
442
+ "loss": 0.0126,
443
+ "step": 3100
444
+ },
445
+ {
446
+ "epoch": 0.27925531914893614,
447
+ "grad_norm": 0.49075883570668455,
448
+ "learning_rate": 9.561662726556344e-06,
449
+ "loss": 0.0114,
450
+ "step": 3150
451
+ },
452
+ {
453
+ "epoch": 0.28368794326241137,
454
+ "grad_norm": 0.661703227107662,
455
+ "learning_rate": 9.537037037037037e-06,
456
+ "loss": 0.0081,
457
+ "step": 3200
458
+ },
459
+ {
460
+ "epoch": 0.28812056737588654,
461
+ "grad_norm": 0.00012406562558187035,
462
+ "learning_rate": 9.512411347517732e-06,
463
+ "loss": 0.0154,
464
+ "step": 3250
465
+ },
466
+ {
467
+ "epoch": 0.2925531914893617,
468
+ "grad_norm": 10.456271349514031,
469
+ "learning_rate": 9.487785657998426e-06,
470
+ "loss": 0.0056,
471
+ "step": 3300
472
+ },
473
+ {
474
+ "epoch": 0.2969858156028369,
475
+ "grad_norm": 34.88954914653437,
476
+ "learning_rate": 9.463159968479119e-06,
477
+ "loss": 0.0215,
478
+ "step": 3350
479
+ },
480
+ {
481
+ "epoch": 0.30141843971631205,
482
+ "grad_norm": 0.0037401951660654617,
483
+ "learning_rate": 9.43853427895981e-06,
484
+ "loss": 0.0073,
485
+ "step": 3400
486
+ },
487
+ {
488
+ "epoch": 0.3058510638297872,
489
+ "grad_norm": 1.9770145427070573,
490
+ "learning_rate": 9.413908589440504e-06,
491
+ "loss": 0.026,
492
+ "step": 3450
493
+ },
494
+ {
495
+ "epoch": 0.3102836879432624,
496
+ "grad_norm": 0.0005171843885202088,
497
+ "learning_rate": 9.389282899921197e-06,
498
+ "loss": 0.0054,
499
+ "step": 3500
500
+ },
501
+ {
502
+ "epoch": 0.3147163120567376,
503
+ "grad_norm": 6.671945039975605,
504
+ "learning_rate": 9.364657210401892e-06,
505
+ "loss": 0.0193,
506
+ "step": 3550
507
+ },
508
+ {
509
+ "epoch": 0.3191489361702128,
510
+ "grad_norm": 24.199369765973504,
511
+ "learning_rate": 9.340031520882586e-06,
512
+ "loss": 0.0253,
513
+ "step": 3600
514
+ },
515
+ {
516
+ "epoch": 0.32358156028368795,
517
+ "grad_norm": 0.03237450123162466,
518
+ "learning_rate": 9.31540583136328e-06,
519
+ "loss": 0.0108,
520
+ "step": 3650
521
+ },
522
+ {
523
+ "epoch": 0.3280141843971631,
524
+ "grad_norm": 0.06582091922650166,
525
+ "learning_rate": 9.290780141843973e-06,
526
+ "loss": 0.0052,
527
+ "step": 3700
528
+ },
529
+ {
530
+ "epoch": 0.3324468085106383,
531
+ "grad_norm": 1.1046639468259907,
532
+ "learning_rate": 9.266154452324666e-06,
533
+ "loss": 0.0077,
534
+ "step": 3750
535
+ },
536
+ {
537
+ "epoch": 0.33687943262411346,
538
+ "grad_norm": 0.12685978465854558,
539
+ "learning_rate": 9.24152876280536e-06,
540
+ "loss": 0.0055,
541
+ "step": 3800
542
+ },
543
+ {
544
+ "epoch": 0.34131205673758863,
545
+ "grad_norm": 0.001176875548815465,
546
+ "learning_rate": 9.216903073286053e-06,
547
+ "loss": 0.0116,
548
+ "step": 3850
549
+ },
550
+ {
551
+ "epoch": 0.34574468085106386,
552
+ "grad_norm": 19.93923722594081,
553
+ "learning_rate": 9.192277383766746e-06,
554
+ "loss": 0.0339,
555
+ "step": 3900
556
+ },
557
+ {
558
+ "epoch": 0.350177304964539,
559
+ "grad_norm": 0.5136555426144171,
560
+ "learning_rate": 9.16765169424744e-06,
561
+ "loss": 0.0109,
562
+ "step": 3950
563
+ },
564
+ {
565
+ "epoch": 0.3546099290780142,
566
+ "grad_norm": 29.577148937276096,
567
+ "learning_rate": 9.143026004728133e-06,
568
+ "loss": 0.0195,
569
+ "step": 4000
570
+ },
571
+ {
572
+ "epoch": 0.35904255319148937,
573
+ "grad_norm": 5.471652062023114,
574
+ "learning_rate": 9.118400315208826e-06,
575
+ "loss": 0.0075,
576
+ "step": 4050
577
+ },
578
+ {
579
+ "epoch": 0.36347517730496454,
580
+ "grad_norm": 0.0039179745195827426,
581
+ "learning_rate": 9.09377462568952e-06,
582
+ "loss": 0.0203,
583
+ "step": 4100
584
+ },
585
+ {
586
+ "epoch": 0.3679078014184397,
587
+ "grad_norm": 30.214203138409303,
588
+ "learning_rate": 9.069148936170213e-06,
589
+ "loss": 0.0323,
590
+ "step": 4150
591
+ },
592
+ {
593
+ "epoch": 0.3723404255319149,
594
+ "grad_norm": 1.0344365911211837,
595
+ "learning_rate": 9.044523246650908e-06,
596
+ "loss": 0.0252,
597
+ "step": 4200
598
+ },
599
+ {
600
+ "epoch": 0.37677304964539005,
601
+ "grad_norm": 0.6080095692433664,
602
+ "learning_rate": 9.019897557131601e-06,
603
+ "loss": 0.0065,
604
+ "step": 4250
605
+ },
606
+ {
607
+ "epoch": 0.38120567375886527,
608
+ "grad_norm": 1.132001494443736,
609
+ "learning_rate": 8.995271867612294e-06,
610
+ "loss": 0.0137,
611
+ "step": 4300
612
+ },
613
+ {
614
+ "epoch": 0.38563829787234044,
615
+ "grad_norm": 0.6281310332658542,
616
+ "learning_rate": 8.971138691883373e-06,
617
+ "loss": 0.0086,
618
+ "step": 4350
619
+ },
620
+ {
621
+ "epoch": 0.3900709219858156,
622
+ "grad_norm": 2.5248461085388904e-05,
623
+ "learning_rate": 8.946513002364066e-06,
624
+ "loss": 0.0015,
625
+ "step": 4400
626
+ },
627
+ {
628
+ "epoch": 0.3945035460992908,
629
+ "grad_norm": 0.010001162867900915,
630
+ "learning_rate": 8.921887312844761e-06,
631
+ "loss": 0.0174,
632
+ "step": 4450
633
+ },
634
+ {
635
+ "epoch": 0.39893617021276595,
636
+ "grad_norm": 0.3111641106006767,
637
+ "learning_rate": 8.897261623325454e-06,
638
+ "loss": 0.0174,
639
+ "step": 4500
640
+ },
641
+ {
642
+ "epoch": 0.4033687943262411,
643
+ "grad_norm": 0.022375831871911778,
644
+ "learning_rate": 8.872635933806148e-06,
645
+ "loss": 0.0115,
646
+ "step": 4550
647
+ },
648
+ {
649
+ "epoch": 0.4078014184397163,
650
+ "grad_norm": 0.08396303659869327,
651
+ "learning_rate": 8.848010244286841e-06,
652
+ "loss": 0.0136,
653
+ "step": 4600
654
+ },
655
+ {
656
+ "epoch": 0.4122340425531915,
657
+ "grad_norm": 5.854304446661381,
658
+ "learning_rate": 8.823384554767534e-06,
659
+ "loss": 0.0042,
660
+ "step": 4650
661
+ },
662
+ {
663
+ "epoch": 0.4166666666666667,
664
+ "grad_norm": 16.679488984474407,
665
+ "learning_rate": 8.798758865248228e-06,
666
+ "loss": 0.0102,
667
+ "step": 4700
668
+ },
669
+ {
670
+ "epoch": 0.42109929078014185,
671
+ "grad_norm": 0.6379686502927935,
672
+ "learning_rate": 8.774625689519308e-06,
673
+ "loss": 0.019,
674
+ "step": 4750
675
+ },
676
+ {
677
+ "epoch": 0.425531914893617,
678
+ "grad_norm": 0.05658909698246918,
679
+ "learning_rate": 8.750000000000001e-06,
680
+ "loss": 0.0178,
681
+ "step": 4800
682
+ },
683
+ {
684
+ "epoch": 0.4299645390070922,
685
+ "grad_norm": 0.36425987836236606,
686
+ "learning_rate": 8.725374310480694e-06,
687
+ "loss": 0.0212,
688
+ "step": 4850
689
+ },
690
+ {
691
+ "epoch": 0.43439716312056736,
692
+ "grad_norm": 2.0754475949590696,
693
+ "learning_rate": 8.700748620961388e-06,
694
+ "loss": 0.0077,
695
+ "step": 4900
696
+ },
697
+ {
698
+ "epoch": 0.43882978723404253,
699
+ "grad_norm": 0.012181818674097341,
700
+ "learning_rate": 8.676122931442081e-06,
701
+ "loss": 0.0125,
702
+ "step": 4950
703
+ },
704
+ {
705
+ "epoch": 0.4432624113475177,
706
+ "grad_norm": 0.00018403879761046281,
707
+ "learning_rate": 8.651497241922774e-06,
708
+ "loss": 0.0188,
709
+ "step": 5000
710
+ },
711
+ {
712
+ "epoch": 0.44769503546099293,
713
+ "grad_norm": 0.004407047372502831,
714
+ "learning_rate": 8.626871552403468e-06,
715
+ "loss": 0.0073,
716
+ "step": 5050
717
+ },
718
+ {
719
+ "epoch": 0.4521276595744681,
720
+ "grad_norm": 0.012230911143282817,
721
+ "learning_rate": 8.602245862884161e-06,
722
+ "loss": 0.008,
723
+ "step": 5100
724
+ },
725
+ {
726
+ "epoch": 0.45656028368794327,
727
+ "grad_norm": 0.9921871395560984,
728
+ "learning_rate": 8.577620173364854e-06,
729
+ "loss": 0.0168,
730
+ "step": 5150
731
+ },
732
+ {
733
+ "epoch": 0.46099290780141844,
734
+ "grad_norm": 0.1285429154497177,
735
+ "learning_rate": 8.552994483845548e-06,
736
+ "loss": 0.0122,
737
+ "step": 5200
738
+ },
739
+ {
740
+ "epoch": 0.4654255319148936,
741
+ "grad_norm": 3.308588561647349,
742
+ "learning_rate": 8.528861308116628e-06,
743
+ "loss": 0.0158,
744
+ "step": 5250
745
+ },
746
+ {
747
+ "epoch": 0.4698581560283688,
748
+ "grad_norm": 2.8184778898385976,
749
+ "learning_rate": 8.504235618597323e-06,
750
+ "loss": 0.0237,
751
+ "step": 5300
752
+ },
753
+ {
754
+ "epoch": 0.47429078014184395,
755
+ "grad_norm": 0.47692861720548735,
756
+ "learning_rate": 8.479609929078016e-06,
757
+ "loss": 0.0181,
758
+ "step": 5350
759
+ },
760
+ {
761
+ "epoch": 0.4787234042553192,
762
+ "grad_norm": 0.002116990082837995,
763
+ "learning_rate": 8.454984239558708e-06,
764
+ "loss": 0.0235,
765
+ "step": 5400
766
+ },
767
+ {
768
+ "epoch": 0.48315602836879434,
769
+ "grad_norm": 0.016230203235315883,
770
+ "learning_rate": 8.430358550039401e-06,
771
+ "loss": 0.0123,
772
+ "step": 5450
773
+ },
774
+ {
775
+ "epoch": 0.4875886524822695,
776
+ "grad_norm": 0.10798463365752625,
777
+ "learning_rate": 8.405732860520094e-06,
778
+ "loss": 0.0202,
779
+ "step": 5500
780
+ },
781
+ {
782
+ "epoch": 0.4920212765957447,
783
+ "grad_norm": 0.00038212180284718085,
784
+ "learning_rate": 8.381107171000788e-06,
785
+ "loss": 0.0049,
786
+ "step": 5550
787
+ },
788
+ {
789
+ "epoch": 0.49645390070921985,
790
+ "grad_norm": 0.0022870323047176084,
791
+ "learning_rate": 8.356481481481483e-06,
792
+ "loss": 0.0061,
793
+ "step": 5600
794
+ },
795
+ {
796
+ "epoch": 0.500886524822695,
797
+ "grad_norm": 0.4205002106769229,
798
+ "learning_rate": 8.331855791962176e-06,
799
+ "loss": 0.0018,
800
+ "step": 5650
801
+ },
802
+ {
803
+ "epoch": 0.5053191489361702,
804
+ "grad_norm": 0.0003576863161645832,
805
+ "learning_rate": 8.30723010244287e-06,
806
+ "loss": 0.01,
807
+ "step": 5700
808
+ },
809
+ {
810
+ "epoch": 0.5097517730496454,
811
+ "grad_norm": 27.715734272602834,
812
+ "learning_rate": 8.282604412923563e-06,
813
+ "loss": 0.0173,
814
+ "step": 5750
815
+ },
816
+ {
817
+ "epoch": 0.5141843971631206,
818
+ "grad_norm": 0.0029419354875175013,
819
+ "learning_rate": 8.257978723404256e-06,
820
+ "loss": 0.0067,
821
+ "step": 5800
822
+ },
823
+ {
824
+ "epoch": 0.5186170212765957,
825
+ "grad_norm": 27.610931266268892,
826
+ "learning_rate": 8.23335303388495e-06,
827
+ "loss": 0.0116,
828
+ "step": 5850
829
+ },
830
+ {
831
+ "epoch": 0.5230496453900709,
832
+ "grad_norm": 0.004511848611750885,
833
+ "learning_rate": 8.208727344365643e-06,
834
+ "loss": 0.0115,
835
+ "step": 5900
836
+ },
837
+ {
838
+ "epoch": 0.5274822695035462,
839
+ "grad_norm": 0.5244401751220326,
840
+ "learning_rate": 8.184101654846336e-06,
841
+ "loss": 0.0025,
842
+ "step": 5950
843
+ },
844
+ {
845
+ "epoch": 0.5319148936170213,
846
+ "grad_norm": 0.0075670950741103474,
847
+ "learning_rate": 8.15947596532703e-06,
848
+ "loss": 0.0164,
849
+ "step": 6000
850
+ },
851
+ {
852
+ "epoch": 0.5363475177304965,
853
+ "grad_norm": 0.06038816285079991,
854
+ "learning_rate": 8.134850275807723e-06,
855
+ "loss": 0.0079,
856
+ "step": 6050
857
+ },
858
+ {
859
+ "epoch": 0.5407801418439716,
860
+ "grad_norm": 0.0007768875893699433,
861
+ "learning_rate": 8.110224586288416e-06,
862
+ "loss": 0.0052,
863
+ "step": 6100
864
+ },
865
+ {
866
+ "epoch": 0.5452127659574468,
867
+ "grad_norm": 0.2909181418775317,
868
+ "learning_rate": 8.08559889676911e-06,
869
+ "loss": 0.0032,
870
+ "step": 6150
871
+ },
872
+ {
873
+ "epoch": 0.549645390070922,
874
+ "grad_norm": 6.463507993140611,
875
+ "learning_rate": 8.060973207249803e-06,
876
+ "loss": 0.0098,
877
+ "step": 6200
878
+ },
879
+ {
880
+ "epoch": 0.5540780141843972,
881
+ "grad_norm": 0.26740235451618943,
882
+ "learning_rate": 8.036347517730498e-06,
883
+ "loss": 0.0144,
884
+ "step": 6250
885
+ },
886
+ {
887
+ "epoch": 0.5585106382978723,
888
+ "grad_norm": 0.00010577380374890247,
889
+ "learning_rate": 8.011721828211191e-06,
890
+ "loss": 0.0156,
891
+ "step": 6300
892
+ },
893
+ {
894
+ "epoch": 0.5629432624113475,
895
+ "grad_norm": 0.14691208386477084,
896
+ "learning_rate": 7.987096138691885e-06,
897
+ "loss": 0.0152,
898
+ "step": 6350
899
+ },
900
+ {
901
+ "epoch": 0.5673758865248227,
902
+ "grad_norm": 0.07499554653157833,
903
+ "learning_rate": 7.962470449172578e-06,
904
+ "loss": 0.0282,
905
+ "step": 6400
906
+ },
907
+ {
908
+ "epoch": 0.5718085106382979,
909
+ "grad_norm": 0.005398079692845853,
910
+ "learning_rate": 7.93784475965327e-06,
911
+ "loss": 0.0193,
912
+ "step": 6450
913
+ },
914
+ {
915
+ "epoch": 0.5762411347517731,
916
+ "grad_norm": 0.8004218807047324,
917
+ "learning_rate": 7.913219070133963e-06,
918
+ "loss": 0.0059,
919
+ "step": 6500
920
+ },
921
+ {
922
+ "epoch": 0.5806737588652482,
923
+ "grad_norm": 0.0018961263364736887,
924
+ "learning_rate": 7.888593380614658e-06,
925
+ "loss": 0.0192,
926
+ "step": 6550
927
+ },
928
+ {
929
+ "epoch": 0.5851063829787234,
930
+ "grad_norm": 7.167523295933813,
931
+ "learning_rate": 7.863967691095352e-06,
932
+ "loss": 0.0026,
933
+ "step": 6600
934
+ },
935
+ {
936
+ "epoch": 0.5895390070921985,
937
+ "grad_norm": 0.2545664640109103,
938
+ "learning_rate": 7.839342001576045e-06,
939
+ "loss": 0.0178,
940
+ "step": 6650
941
+ },
942
+ {
943
+ "epoch": 0.5939716312056738,
944
+ "grad_norm": 0.0003677116767933612,
945
+ "learning_rate": 7.814716312056738e-06,
946
+ "loss": 0.0127,
947
+ "step": 6700
948
+ },
949
+ {
950
+ "epoch": 0.598404255319149,
951
+ "grad_norm": 1.449028640163906,
952
+ "learning_rate": 7.790090622537432e-06,
953
+ "loss": 0.0353,
954
+ "step": 6750
955
+ },
956
+ {
957
+ "epoch": 0.6028368794326241,
958
+ "grad_norm": 0.018076926628577003,
959
+ "learning_rate": 7.765464933018125e-06,
960
+ "loss": 0.0022,
961
+ "step": 6800
962
+ },
963
+ {
964
+ "epoch": 0.6072695035460993,
965
+ "grad_norm": 1.314844716480283,
966
+ "learning_rate": 7.740839243498818e-06,
967
+ "loss": 0.0125,
968
+ "step": 6850
969
+ },
970
+ {
971
+ "epoch": 0.6117021276595744,
972
+ "grad_norm": 0.05713668183450516,
973
+ "learning_rate": 7.716213553979512e-06,
974
+ "loss": 0.0103,
975
+ "step": 6900
976
+ },
977
+ {
978
+ "epoch": 0.6161347517730497,
979
+ "grad_norm": 0.00012171886896694071,
980
+ "learning_rate": 7.691587864460207e-06,
981
+ "loss": 0.0411,
982
+ "step": 6950
983
+ },
984
+ {
985
+ "epoch": 0.6205673758865248,
986
+ "grad_norm": 0.005050200509222459,
987
+ "learning_rate": 7.666962174940898e-06,
988
+ "loss": 0.0088,
989
+ "step": 7000
990
+ },
991
+ {
992
+ "epoch": 0.625,
993
+ "grad_norm": 0.044231465953618015,
994
+ "learning_rate": 7.642336485421592e-06,
995
+ "loss": 0.0121,
996
+ "step": 7050
997
+ },
998
+ {
999
+ "epoch": 0.6294326241134752,
1000
+ "grad_norm": 0.002970509653416846,
1001
+ "learning_rate": 7.617710795902286e-06,
1002
+ "loss": 0.0115,
1003
+ "step": 7100
1004
+ },
1005
+ {
1006
+ "epoch": 0.6338652482269503,
1007
+ "grad_norm": 0.061580769377440274,
1008
+ "learning_rate": 7.593085106382979e-06,
1009
+ "loss": 0.0074,
1010
+ "step": 7150
1011
+ },
1012
+ {
1013
+ "epoch": 0.6382978723404256,
1014
+ "grad_norm": 0.0030036827619670152,
1015
+ "learning_rate": 7.568459416863673e-06,
1016
+ "loss": 0.0075,
1017
+ "step": 7200
1018
+ },
1019
+ {
1020
+ "epoch": 0.6427304964539007,
1021
+ "grad_norm": 0.12485293215547061,
1022
+ "learning_rate": 7.543833727344366e-06,
1023
+ "loss": 0.0112,
1024
+ "step": 7250
1025
+ },
1026
+ {
1027
+ "epoch": 0.6471631205673759,
1028
+ "grad_norm": 0.001231009728533709,
1029
+ "learning_rate": 7.519208037825059e-06,
1030
+ "loss": 0.0068,
1031
+ "step": 7300
1032
+ },
1033
+ {
1034
+ "epoch": 0.651595744680851,
1035
+ "grad_norm": 1.066825644497329,
1036
+ "learning_rate": 7.494582348305754e-06,
1037
+ "loss": 0.0219,
1038
+ "step": 7350
1039
+ },
1040
+ {
1041
+ "epoch": 0.6560283687943262,
1042
+ "grad_norm": 0.18183803085941427,
1043
+ "learning_rate": 7.469956658786447e-06,
1044
+ "loss": 0.0052,
1045
+ "step": 7400
1046
+ },
1047
+ {
1048
+ "epoch": 0.6604609929078015,
1049
+ "grad_norm": 7.176261541067922,
1050
+ "learning_rate": 7.44533096926714e-06,
1051
+ "loss": 0.012,
1052
+ "step": 7450
1053
+ },
1054
+ {
1055
+ "epoch": 0.6648936170212766,
1056
+ "grad_norm": 0.08470951895549413,
1057
+ "learning_rate": 7.420705279747834e-06,
1058
+ "loss": 0.005,
1059
+ "step": 7500
1060
+ },
1061
+ {
1062
+ "epoch": 0.6693262411347518,
1063
+ "grad_norm": 0.43564519071924557,
1064
+ "learning_rate": 7.396079590228526e-06,
1065
+ "loss": 0.0214,
1066
+ "step": 7550
1067
+ },
1068
+ {
1069
+ "epoch": 0.6737588652482269,
1070
+ "grad_norm": 0.006827701243280031,
1071
+ "learning_rate": 7.37145390070922e-06,
1072
+ "loss": 0.0382,
1073
+ "step": 7600
1074
+ },
1075
+ {
1076
+ "epoch": 0.6781914893617021,
1077
+ "grad_norm": 0.0006360372069413289,
1078
+ "learning_rate": 7.346828211189914e-06,
1079
+ "loss": 0.0058,
1080
+ "step": 7650
1081
+ },
1082
+ {
1083
+ "epoch": 0.6826241134751773,
1084
+ "grad_norm": 0.011063982123596255,
1085
+ "learning_rate": 7.322202521670607e-06,
1086
+ "loss": 0.0074,
1087
+ "step": 7700
1088
+ },
1089
+ {
1090
+ "epoch": 0.6870567375886525,
1091
+ "grad_norm": 0.8991727722769103,
1092
+ "learning_rate": 7.2975768321513005e-06,
1093
+ "loss": 0.0108,
1094
+ "step": 7750
1095
+ },
1096
+ {
1097
+ "epoch": 0.6914893617021277,
1098
+ "grad_norm": 0.012017251234112784,
1099
+ "learning_rate": 7.272951142631995e-06,
1100
+ "loss": 0.0127,
1101
+ "step": 7800
1102
+ },
1103
+ {
1104
+ "epoch": 0.6959219858156028,
1105
+ "grad_norm": 0.19374738160794408,
1106
+ "learning_rate": 7.248325453112688e-06,
1107
+ "loss": 0.0035,
1108
+ "step": 7850
1109
+ },
1110
+ {
1111
+ "epoch": 0.700354609929078,
1112
+ "grad_norm": 2.341510058377949,
1113
+ "learning_rate": 7.223699763593381e-06,
1114
+ "loss": 0.0317,
1115
+ "step": 7900
1116
+ },
1117
+ {
1118
+ "epoch": 0.7047872340425532,
1119
+ "grad_norm": 5.84240259775226e-06,
1120
+ "learning_rate": 7.199074074074075e-06,
1121
+ "loss": 0.0205,
1122
+ "step": 7950
1123
+ },
1124
+ {
1125
+ "epoch": 0.7092198581560284,
1126
+ "grad_norm": 0.0022186038540356347,
1127
+ "learning_rate": 7.174448384554769e-06,
1128
+ "loss": 0.0206,
1129
+ "step": 8000
1130
+ },
1131
+ {
1132
+ "epoch": 0.7136524822695035,
1133
+ "grad_norm": 2.3575055124131414,
1134
+ "learning_rate": 7.149822695035462e-06,
1135
+ "loss": 0.0047,
1136
+ "step": 8050
1137
+ },
1138
+ {
1139
+ "epoch": 0.7180851063829787,
1140
+ "grad_norm": 0.01023308841827913,
1141
+ "learning_rate": 7.125197005516155e-06,
1142
+ "loss": 0.0293,
1143
+ "step": 8100
1144
+ },
1145
+ {
1146
+ "epoch": 0.7225177304964538,
1147
+ "grad_norm": 0.006391574218526128,
1148
+ "learning_rate": 7.100571315996848e-06,
1149
+ "loss": 0.0137,
1150
+ "step": 8150
1151
+ },
1152
+ {
1153
+ "epoch": 0.7269503546099291,
1154
+ "grad_norm": 1.7478784234914295,
1155
+ "learning_rate": 7.0759456264775415e-06,
1156
+ "loss": 0.0144,
1157
+ "step": 8200
1158
+ },
1159
+ {
1160
+ "epoch": 0.7313829787234043,
1161
+ "grad_norm": 0.4480580469017403,
1162
+ "learning_rate": 7.051319936958235e-06,
1163
+ "loss": 0.0236,
1164
+ "step": 8250
1165
+ },
1166
+ {
1167
+ "epoch": 0.7358156028368794,
1168
+ "grad_norm": 0.0001810108372971621,
1169
+ "learning_rate": 7.0271867612293155e-06,
1170
+ "loss": 0.0162,
1171
+ "step": 8300
1172
+ },
1173
+ {
1174
+ "epoch": 0.7402482269503546,
1175
+ "grad_norm": 0.004944937344540902,
1176
+ "learning_rate": 7.002561071710009e-06,
1177
+ "loss": 0.0045,
1178
+ "step": 8350
1179
+ },
1180
+ {
1181
+ "epoch": 0.7446808510638298,
1182
+ "grad_norm": 0.009033810427587684,
1183
+ "learning_rate": 6.977935382190701e-06,
1184
+ "loss": 0.0021,
1185
+ "step": 8400
1186
+ },
1187
+ {
1188
+ "epoch": 0.749113475177305,
1189
+ "grad_norm": 0.0007266417865651897,
1190
+ "learning_rate": 6.953309692671395e-06,
1191
+ "loss": 0.0028,
1192
+ "step": 8450
1193
+ },
1194
+ {
1195
+ "epoch": 0.7535460992907801,
1196
+ "grad_norm": 0.5472428310965101,
1197
+ "learning_rate": 6.928684003152089e-06,
1198
+ "loss": 0.0072,
1199
+ "step": 8500
1200
+ },
1201
+ {
1202
+ "epoch": 0.7579787234042553,
1203
+ "grad_norm": 0.40243503312801704,
1204
+ "learning_rate": 6.904058313632782e-06,
1205
+ "loss": 0.015,
1206
+ "step": 8550
1207
+ },
1208
+ {
1209
+ "epoch": 0.7624113475177305,
1210
+ "grad_norm": 0.0047908998463532726,
1211
+ "learning_rate": 6.879432624113476e-06,
1212
+ "loss": 0.0071,
1213
+ "step": 8600
1214
+ },
1215
+ {
1216
+ "epoch": 0.7668439716312057,
1217
+ "grad_norm": 0.0009125120021038514,
1218
+ "learning_rate": 6.854806934594169e-06,
1219
+ "loss": 0.0037,
1220
+ "step": 8650
1221
+ },
1222
+ {
1223
+ "epoch": 0.7712765957446809,
1224
+ "grad_norm": 0.08131393086136933,
1225
+ "learning_rate": 6.830181245074863e-06,
1226
+ "loss": 0.0047,
1227
+ "step": 8700
1228
+ },
1229
+ {
1230
+ "epoch": 0.775709219858156,
1231
+ "grad_norm": 0.0006308892137283422,
1232
+ "learning_rate": 6.8055555555555566e-06,
1233
+ "loss": 0.014,
1234
+ "step": 8750
1235
+ },
1236
+ {
1237
+ "epoch": 0.7801418439716312,
1238
+ "grad_norm": 0.0454133683781099,
1239
+ "learning_rate": 6.78092986603625e-06,
1240
+ "loss": 0.0143,
1241
+ "step": 8800
1242
+ },
1243
+ {
1244
+ "epoch": 0.7845744680851063,
1245
+ "grad_norm": 24.069321019826646,
1246
+ "learning_rate": 6.756304176516943e-06,
1247
+ "loss": 0.0069,
1248
+ "step": 8850
1249
+ },
1250
+ {
1251
+ "epoch": 0.7890070921985816,
1252
+ "grad_norm": 0.018348520956196407,
1253
+ "learning_rate": 6.731678486997636e-06,
1254
+ "loss": 0.008,
1255
+ "step": 8900
1256
+ },
1257
+ {
1258
+ "epoch": 0.7934397163120568,
1259
+ "grad_norm": 0.045503745490795604,
1260
+ "learning_rate": 6.707052797478329e-06,
1261
+ "loss": 0.0038,
1262
+ "step": 8950
1263
+ },
1264
+ {
1265
+ "epoch": 0.7978723404255319,
1266
+ "grad_norm": 0.2684381266709852,
1267
+ "learning_rate": 6.682427107959023e-06,
1268
+ "loss": 0.0026,
1269
+ "step": 9000
1270
+ },
1271
+ {
1272
+ "epoch": 0.8023049645390071,
1273
+ "grad_norm": 0.11829149058897026,
1274
+ "learning_rate": 6.657801418439717e-06,
1275
+ "loss": 0.0233,
1276
+ "step": 9050
1277
+ },
1278
+ {
1279
+ "epoch": 0.8067375886524822,
1280
+ "grad_norm": 0.16600459567065376,
1281
+ "learning_rate": 6.63317572892041e-06,
1282
+ "loss": 0.0149,
1283
+ "step": 9100
1284
+ },
1285
+ {
1286
+ "epoch": 0.8111702127659575,
1287
+ "grad_norm": 0.0005939769820284411,
1288
+ "learning_rate": 6.608550039401103e-06,
1289
+ "loss": 0.0232,
1290
+ "step": 9150
1291
+ },
1292
+ {
1293
+ "epoch": 0.8156028368794326,
1294
+ "grad_norm": 4.742621865403474,
1295
+ "learning_rate": 6.583924349881798e-06,
1296
+ "loss": 0.003,
1297
+ "step": 9200
1298
+ },
1299
+ {
1300
+ "epoch": 0.8200354609929078,
1301
+ "grad_norm": 0.0022694228849504507,
1302
+ "learning_rate": 6.559298660362491e-06,
1303
+ "loss": 0.0088,
1304
+ "step": 9250
1305
+ },
1306
+ {
1307
+ "epoch": 0.824468085106383,
1308
+ "grad_norm": 0.01267761477851963,
1309
+ "learning_rate": 6.534672970843184e-06,
1310
+ "loss": 0.0042,
1311
+ "step": 9300
1312
+ },
1313
+ {
1314
+ "epoch": 0.8289007092198581,
1315
+ "grad_norm": 6.0142319714349085e-05,
1316
+ "learning_rate": 6.510047281323878e-06,
1317
+ "loss": 0.0036,
1318
+ "step": 9350
1319
+ },
1320
+ {
1321
+ "epoch": 0.8333333333333334,
1322
+ "grad_norm": 0.029585797478583992,
1323
+ "learning_rate": 6.485914105594957e-06,
1324
+ "loss": 0.0089,
1325
+ "step": 9400
1326
+ },
1327
+ {
1328
+ "epoch": 0.8377659574468085,
1329
+ "grad_norm": 0.11418292860246061,
1330
+ "learning_rate": 6.461288416075651e-06,
1331
+ "loss": 0.0023,
1332
+ "step": 9450
1333
+ },
1334
+ {
1335
+ "epoch": 0.8421985815602837,
1336
+ "grad_norm": 0.49982011062420273,
1337
+ "learning_rate": 6.436662726556344e-06,
1338
+ "loss": 0.0044,
1339
+ "step": 9500
1340
+ },
1341
+ {
1342
+ "epoch": 0.8466312056737588,
1343
+ "grad_norm": 18.307836570245335,
1344
+ "learning_rate": 6.4120370370370375e-06,
1345
+ "loss": 0.0169,
1346
+ "step": 9550
1347
+ },
1348
+ {
1349
+ "epoch": 0.851063829787234,
1350
+ "grad_norm": 0.001667923555688216,
1351
+ "learning_rate": 6.387411347517731e-06,
1352
+ "loss": 0.0072,
1353
+ "step": 9600
1354
+ },
1355
+ {
1356
+ "epoch": 0.8554964539007093,
1357
+ "grad_norm": 0.07738383812395583,
1358
+ "learning_rate": 6.362785657998425e-06,
1359
+ "loss": 0.0088,
1360
+ "step": 9650
1361
+ },
1362
+ {
1363
+ "epoch": 0.8599290780141844,
1364
+ "grad_norm": 4.3471602630804025,
1365
+ "learning_rate": 6.3381599684791185e-06,
1366
+ "loss": 0.0092,
1367
+ "step": 9700
1368
+ },
1369
+ {
1370
+ "epoch": 0.8643617021276596,
1371
+ "grad_norm": 0.0011124382494966075,
1372
+ "learning_rate": 6.313534278959811e-06,
1373
+ "loss": 0.0053,
1374
+ "step": 9750
1375
+ },
1376
+ {
1377
+ "epoch": 0.8687943262411347,
1378
+ "grad_norm": 0.19437101421378722,
1379
+ "learning_rate": 6.288908589440504e-06,
1380
+ "loss": 0.0044,
1381
+ "step": 9800
1382
+ },
1383
+ {
1384
+ "epoch": 0.87322695035461,
1385
+ "grad_norm": 0.00870818269634644,
1386
+ "learning_rate": 6.264282899921198e-06,
1387
+ "loss": 0.0059,
1388
+ "step": 9850
1389
+ },
1390
+ {
1391
+ "epoch": 0.8776595744680851,
1392
+ "grad_norm": 0.42404259032680214,
1393
+ "learning_rate": 6.239657210401892e-06,
1394
+ "loss": 0.0141,
1395
+ "step": 9900
1396
+ },
1397
+ {
1398
+ "epoch": 0.8820921985815603,
1399
+ "grad_norm": 0.000391160720835517,
1400
+ "learning_rate": 6.215031520882585e-06,
1401
+ "loss": 0.0144,
1402
+ "step": 9950
1403
+ },
1404
+ {
1405
+ "epoch": 0.8865248226950354,
1406
+ "grad_norm": 0.07186189200249775,
1407
+ "learning_rate": 6.190405831363279e-06,
1408
+ "loss": 0.0087,
1409
+ "step": 10000
1410
+ },
1411
+ {
1412
+ "epoch": 0.8909574468085106,
1413
+ "grad_norm": 9.1428649766071,
1414
+ "learning_rate": 6.165780141843972e-06,
1415
+ "loss": 0.0141,
1416
+ "step": 10050
1417
+ },
1418
+ {
1419
+ "epoch": 0.8953900709219859,
1420
+ "grad_norm": 0.28709998082991656,
1421
+ "learning_rate": 6.141154452324666e-06,
1422
+ "loss": 0.0161,
1423
+ "step": 10100
1424
+ },
1425
+ {
1426
+ "epoch": 0.899822695035461,
1427
+ "grad_norm": 0.0012613353256104856,
1428
+ "learning_rate": 6.1165287628053595e-06,
1429
+ "loss": 0.0083,
1430
+ "step": 10150
1431
+ },
1432
+ {
1433
+ "epoch": 0.9042553191489362,
1434
+ "grad_norm": 0.00043243530861486094,
1435
+ "learning_rate": 6.091903073286053e-06,
1436
+ "loss": 0.0062,
1437
+ "step": 10200
1438
+ },
1439
+ {
1440
+ "epoch": 0.9086879432624113,
1441
+ "grad_norm": 0.0012885774516513972,
1442
+ "learning_rate": 6.067277383766746e-06,
1443
+ "loss": 0.0032,
1444
+ "step": 10250
1445
+ },
1446
+ {
1447
+ "epoch": 0.9131205673758865,
1448
+ "grad_norm": 0.13567630623602833,
1449
+ "learning_rate": 6.042651694247439e-06,
1450
+ "loss": 0.0047,
1451
+ "step": 10300
1452
+ },
1453
+ {
1454
+ "epoch": 0.9175531914893617,
1455
+ "grad_norm": 0.001507828461742134,
1456
+ "learning_rate": 6.018026004728132e-06,
1457
+ "loss": 0.0111,
1458
+ "step": 10350
1459
+ },
1460
+ {
1461
+ "epoch": 0.9219858156028369,
1462
+ "grad_norm": 28.02482049835188,
1463
+ "learning_rate": 5.993400315208826e-06,
1464
+ "loss": 0.013,
1465
+ "step": 10400
1466
+ },
1467
+ {
1468
+ "epoch": 0.9264184397163121,
1469
+ "grad_norm": 0.0005712245953407916,
1470
+ "learning_rate": 5.96877462568952e-06,
1471
+ "loss": 0.0074,
1472
+ "step": 10450
1473
+ },
1474
+ {
1475
+ "epoch": 0.9308510638297872,
1476
+ "grad_norm": 0.6880719016963764,
1477
+ "learning_rate": 5.944148936170213e-06,
1478
+ "loss": 0.0118,
1479
+ "step": 10500
1480
+ },
1481
+ {
1482
+ "epoch": 0.9352836879432624,
1483
+ "grad_norm": 0.0004923493172913049,
1484
+ "learning_rate": 5.919523246650906e-06,
1485
+ "loss": 0.0285,
1486
+ "step": 10550
1487
+ },
1488
+ {
1489
+ "epoch": 0.9397163120567376,
1490
+ "grad_norm": 0.0005338033588183963,
1491
+ "learning_rate": 5.8948975571316006e-06,
1492
+ "loss": 0.0171,
1493
+ "step": 10600
1494
+ },
1495
+ {
1496
+ "epoch": 0.9441489361702128,
1497
+ "grad_norm": 0.01769753140101656,
1498
+ "learning_rate": 5.8707643814026795e-06,
1499
+ "loss": 0.0077,
1500
+ "step": 10650
1501
+ },
1502
+ {
1503
+ "epoch": 0.9485815602836879,
1504
+ "grad_norm": 12.31398706121417,
1505
+ "learning_rate": 5.846138691883373e-06,
1506
+ "loss": 0.0098,
1507
+ "step": 10700
1508
+ },
1509
+ {
1510
+ "epoch": 0.9530141843971631,
1511
+ "grad_norm": 1.086003198625972,
1512
+ "learning_rate": 5.821513002364066e-06,
1513
+ "loss": 0.0099,
1514
+ "step": 10750
1515
+ },
1516
+ {
1517
+ "epoch": 0.9574468085106383,
1518
+ "grad_norm": 2.675868747143798,
1519
+ "learning_rate": 5.7968873128447604e-06,
1520
+ "loss": 0.0134,
1521
+ "step": 10800
1522
+ },
1523
+ {
1524
+ "epoch": 0.9618794326241135,
1525
+ "grad_norm": 0.0004962731547164742,
1526
+ "learning_rate": 5.772261623325454e-06,
1527
+ "loss": 0.0231,
1528
+ "step": 10850
1529
+ },
1530
+ {
1531
+ "epoch": 0.9663120567375887,
1532
+ "grad_norm": 0.11931707651459274,
1533
+ "learning_rate": 5.747635933806147e-06,
1534
+ "loss": 0.0031,
1535
+ "step": 10900
1536
+ },
1537
+ {
1538
+ "epoch": 0.9707446808510638,
1539
+ "grad_norm": 0.027268096455214143,
1540
+ "learning_rate": 5.7230102442868405e-06,
1541
+ "loss": 0.0114,
1542
+ "step": 10950
1543
+ },
1544
+ {
1545
+ "epoch": 0.975177304964539,
1546
+ "grad_norm": 0.00014594318076532122,
1547
+ "learning_rate": 5.698384554767534e-06,
1548
+ "loss": 0.0049,
1549
+ "step": 11000
1550
+ },
1551
+ {
1552
+ "epoch": 0.9796099290780141,
1553
+ "grad_norm": 20.25133896156815,
1554
+ "learning_rate": 5.673758865248228e-06,
1555
+ "loss": 0.0114,
1556
+ "step": 11050
1557
+ },
1558
+ {
1559
+ "epoch": 0.9840425531914894,
1560
+ "grad_norm": 6.232025239063323,
1561
+ "learning_rate": 5.649133175728921e-06,
1562
+ "loss": 0.0115,
1563
+ "step": 11100
1564
+ },
1565
+ {
1566
+ "epoch": 0.9884751773049646,
1567
+ "grad_norm": 19.42076091686018,
1568
+ "learning_rate": 5.624507486209614e-06,
1569
+ "loss": 0.0084,
1570
+ "step": 11150
1571
+ },
1572
+ {
1573
+ "epoch": 0.9929078014184397,
1574
+ "grad_norm": 1.603272060616664,
1575
+ "learning_rate": 5.599881796690307e-06,
1576
+ "loss": 0.008,
1577
+ "step": 11200
1578
+ },
1579
+ {
1580
+ "epoch": 0.9973404255319149,
1581
+ "grad_norm": 1.0847716076156708,
1582
+ "learning_rate": 5.575256107171001e-06,
1583
+ "loss": 0.0114,
1584
+ "step": 11250
1585
+ },
1586
+ {
1587
+ "epoch": 1.00177304964539,
1588
+ "grad_norm": 2.3633525863666684e-05,
1589
+ "learning_rate": 5.550630417651695e-06,
1590
+ "loss": 0.0043,
1591
+ "step": 11300
1592
+ },
1593
+ {
1594
+ "epoch": 1.0062056737588652,
1595
+ "grad_norm": 0.013210380349044053,
1596
+ "learning_rate": 5.526004728132388e-06,
1597
+ "loss": 0.0164,
1598
+ "step": 11350
1599
+ },
1600
+ {
1601
+ "epoch": 1.0106382978723405,
1602
+ "grad_norm": 12.943589109031448,
1603
+ "learning_rate": 5.5013790386130815e-06,
1604
+ "loss": 0.0028,
1605
+ "step": 11400
1606
+ },
1607
+ {
1608
+ "epoch": 1.0150709219858156,
1609
+ "grad_norm": 0.24737069186236824,
1610
+ "learning_rate": 5.476753349093775e-06,
1611
+ "loss": 0.0342,
1612
+ "step": 11450
1613
+ },
1614
+ {
1615
+ "epoch": 1.0195035460992907,
1616
+ "grad_norm": 8.077551697802006,
1617
+ "learning_rate": 5.452127659574469e-06,
1618
+ "loss": 0.0069,
1619
+ "step": 11500
1620
+ },
1621
+ {
1622
+ "epoch": 1.023936170212766,
1623
+ "grad_norm": 21.99730319053338,
1624
+ "learning_rate": 5.4275019700551625e-06,
1625
+ "loss": 0.0077,
1626
+ "step": 11550
1627
+ },
1628
+ {
1629
+ "epoch": 1.0283687943262412,
1630
+ "grad_norm": 0.012495400608913515,
1631
+ "learning_rate": 5.402876280535856e-06,
1632
+ "loss": 0.0082,
1633
+ "step": 11600
1634
+ },
1635
+ {
1636
+ "epoch": 1.0328014184397163,
1637
+ "grad_norm": 23.172233223798802,
1638
+ "learning_rate": 5.378250591016549e-06,
1639
+ "loss": 0.0134,
1640
+ "step": 11650
1641
+ },
1642
+ {
1643
+ "epoch": 1.0372340425531914,
1644
+ "grad_norm": 0.00010344838148590977,
1645
+ "learning_rate": 5.354117415287628e-06,
1646
+ "loss": 0.0077,
1647
+ "step": 11700
1648
+ },
1649
+ {
1650
+ "epoch": 1.0416666666666667,
1651
+ "grad_norm": 0.026233233489346697,
1652
+ "learning_rate": 5.329491725768322e-06,
1653
+ "loss": 0.0064,
1654
+ "step": 11750
1655
+ },
1656
+ {
1657
+ "epoch": 1.0460992907801419,
1658
+ "grad_norm": 0.2523227021500782,
1659
+ "learning_rate": 5.304866036249016e-06,
1660
+ "loss": 0.0051,
1661
+ "step": 11800
1662
+ },
1663
+ {
1664
+ "epoch": 1.050531914893617,
1665
+ "grad_norm": 0.0036641836500692503,
1666
+ "learning_rate": 5.280240346729709e-06,
1667
+ "loss": 0.0096,
1668
+ "step": 11850
1669
+ },
1670
+ {
1671
+ "epoch": 1.0549645390070923,
1672
+ "grad_norm": 0.08388736022952638,
1673
+ "learning_rate": 5.255614657210402e-06,
1674
+ "loss": 0.0146,
1675
+ "step": 11900
1676
+ },
1677
+ {
1678
+ "epoch": 1.0593971631205674,
1679
+ "grad_norm": 0.7881169460771477,
1680
+ "learning_rate": 5.230988967691097e-06,
1681
+ "loss": 0.0048,
1682
+ "step": 11950
1683
+ },
1684
+ {
1685
+ "epoch": 1.0638297872340425,
1686
+ "grad_norm": 0.001005287973612562,
1687
+ "learning_rate": 5.206363278171789e-06,
1688
+ "loss": 0.0059,
1689
+ "step": 12000
1690
+ },
1691
+ {
1692
+ "epoch": 1.0682624113475176,
1693
+ "grad_norm": 3.0530614403128533,
1694
+ "learning_rate": 5.1817375886524825e-06,
1695
+ "loss": 0.0031,
1696
+ "step": 12050
1697
+ },
1698
+ {
1699
+ "epoch": 1.072695035460993,
1700
+ "grad_norm": 0.005283256557374208,
1701
+ "learning_rate": 5.157111899133176e-06,
1702
+ "loss": 0.0121,
1703
+ "step": 12100
1704
+ },
1705
+ {
1706
+ "epoch": 1.077127659574468,
1707
+ "grad_norm": 0.005667942165875843,
1708
+ "learning_rate": 5.132486209613869e-06,
1709
+ "loss": 0.0058,
1710
+ "step": 12150
1711
+ },
1712
+ {
1713
+ "epoch": 1.0815602836879432,
1714
+ "grad_norm": 0.012103555924136282,
1715
+ "learning_rate": 5.107860520094563e-06,
1716
+ "loss": 0.0022,
1717
+ "step": 12200
1718
+ },
1719
+ {
1720
+ "epoch": 1.0859929078014185,
1721
+ "grad_norm": 0.0015274237471755395,
1722
+ "learning_rate": 5.083234830575257e-06,
1723
+ "loss": 0.0043,
1724
+ "step": 12250
1725
+ },
1726
+ {
1727
+ "epoch": 1.0904255319148937,
1728
+ "grad_norm": 0.0017057462918492119,
1729
+ "learning_rate": 5.05860914105595e-06,
1730
+ "loss": 0.0055,
1731
+ "step": 12300
1732
+ },
1733
+ {
1734
+ "epoch": 1.0948581560283688,
1735
+ "grad_norm": 0.0678202349342981,
1736
+ "learning_rate": 5.0339834515366434e-06,
1737
+ "loss": 0.0208,
1738
+ "step": 12350
1739
+ },
1740
+ {
1741
+ "epoch": 1.099290780141844,
1742
+ "grad_norm": 5.820093588584535,
1743
+ "learning_rate": 5.009357762017338e-06,
1744
+ "loss": 0.0071,
1745
+ "step": 12400
1746
+ },
1747
+ {
1748
+ "epoch": 1.1037234042553192,
1749
+ "grad_norm": 0.0002967902068188711,
1750
+ "learning_rate": 4.98473207249803e-06,
1751
+ "loss": 0.0279,
1752
+ "step": 12450
1753
+ },
1754
+ {
1755
+ "epoch": 1.1081560283687943,
1756
+ "grad_norm": 0.00016505633598070985,
1757
+ "learning_rate": 4.9601063829787235e-06,
1758
+ "loss": 0.0079,
1759
+ "step": 12500
1760
+ },
1761
+ {
1762
+ "epoch": 1.1125886524822695,
1763
+ "grad_norm": 8.265879214198836e-05,
1764
+ "learning_rate": 4.935480693459418e-06,
1765
+ "loss": 0.0119,
1766
+ "step": 12550
1767
+ },
1768
+ {
1769
+ "epoch": 1.1170212765957448,
1770
+ "grad_norm": 5.33596270040032,
1771
+ "learning_rate": 4.910855003940111e-06,
1772
+ "loss": 0.0079,
1773
+ "step": 12600
1774
+ },
1775
+ {
1776
+ "epoch": 1.12145390070922,
1777
+ "grad_norm": 0.004774018567874852,
1778
+ "learning_rate": 4.886229314420804e-06,
1779
+ "loss": 0.0127,
1780
+ "step": 12650
1781
+ },
1782
+ {
1783
+ "epoch": 1.125886524822695,
1784
+ "grad_norm": 0.04021199029597082,
1785
+ "learning_rate": 4.861603624901498e-06,
1786
+ "loss": 0.0139,
1787
+ "step": 12700
1788
+ },
1789
+ {
1790
+ "epoch": 1.1303191489361701,
1791
+ "grad_norm": 11.357779746350145,
1792
+ "learning_rate": 4.836977935382191e-06,
1793
+ "loss": 0.0019,
1794
+ "step": 12750
1795
+ },
1796
+ {
1797
+ "epoch": 1.1347517730496455,
1798
+ "grad_norm": 0.0002728474474517395,
1799
+ "learning_rate": 4.8123522458628845e-06,
1800
+ "loss": 0.0054,
1801
+ "step": 12800
1802
+ },
1803
+ {
1804
+ "epoch": 1.1391843971631206,
1805
+ "grad_norm": 0.0015642870272121851,
1806
+ "learning_rate": 4.787726556343578e-06,
1807
+ "loss": 0.0021,
1808
+ "step": 12850
1809
+ },
1810
+ {
1811
+ "epoch": 1.1436170212765957,
1812
+ "grad_norm": 0.19008505520492977,
1813
+ "learning_rate": 4.763100866824271e-06,
1814
+ "loss": 0.0012,
1815
+ "step": 12900
1816
+ },
1817
+ {
1818
+ "epoch": 1.148049645390071,
1819
+ "grad_norm": 5.237984668093131,
1820
+ "learning_rate": 4.7384751773049646e-06,
1821
+ "loss": 0.0028,
1822
+ "step": 12950
1823
+ },
1824
+ {
1825
+ "epoch": 1.1524822695035462,
1826
+ "grad_norm": 0.004434596646202779,
1827
+ "learning_rate": 4.713849487785658e-06,
1828
+ "loss": 0.0043,
1829
+ "step": 13000
1830
+ },
1831
+ {
1832
+ "epoch": 1.1569148936170213,
1833
+ "grad_norm": 0.005018009875278763,
1834
+ "learning_rate": 4.689716312056738e-06,
1835
+ "loss": 0.0142,
1836
+ "step": 13050
1837
+ },
1838
+ {
1839
+ "epoch": 1.1613475177304964,
1840
+ "grad_norm": 0.0002677238680507841,
1841
+ "learning_rate": 4.665090622537432e-06,
1842
+ "loss": 0.0026,
1843
+ "step": 13100
1844
+ },
1845
+ {
1846
+ "epoch": 1.1657801418439717,
1847
+ "grad_norm": 0.000470140748192606,
1848
+ "learning_rate": 4.640464933018125e-06,
1849
+ "loss": 0.0092,
1850
+ "step": 13150
1851
+ },
1852
+ {
1853
+ "epoch": 1.1702127659574468,
1854
+ "grad_norm": 0.0018058005755851082,
1855
+ "learning_rate": 4.615839243498818e-06,
1856
+ "loss": 0.0023,
1857
+ "step": 13200
1858
+ },
1859
+ {
1860
+ "epoch": 1.174645390070922,
1861
+ "grad_norm": 0.04680604382760653,
1862
+ "learning_rate": 4.591213553979512e-06,
1863
+ "loss": 0.0058,
1864
+ "step": 13250
1865
+ },
1866
+ {
1867
+ "epoch": 1.1790780141843973,
1868
+ "grad_norm": 0.044852503825975776,
1869
+ "learning_rate": 4.566587864460205e-06,
1870
+ "loss": 0.0039,
1871
+ "step": 13300
1872
+ },
1873
+ {
1874
+ "epoch": 1.1835106382978724,
1875
+ "grad_norm": 11.036633612390176,
1876
+ "learning_rate": 4.541962174940899e-06,
1877
+ "loss": 0.0022,
1878
+ "step": 13350
1879
+ },
1880
+ {
1881
+ "epoch": 1.1879432624113475,
1882
+ "grad_norm": 0.09774370568985054,
1883
+ "learning_rate": 4.517336485421592e-06,
1884
+ "loss": 0.0024,
1885
+ "step": 13400
1886
+ },
1887
+ {
1888
+ "epoch": 1.1923758865248226,
1889
+ "grad_norm": 0.0004694848374047449,
1890
+ "learning_rate": 4.492710795902286e-06,
1891
+ "loss": 0.0088,
1892
+ "step": 13450
1893
+ },
1894
+ {
1895
+ "epoch": 1.196808510638298,
1896
+ "grad_norm": 0.023902851668442845,
1897
+ "learning_rate": 4.468085106382979e-06,
1898
+ "loss": 0.0114,
1899
+ "step": 13500
1900
+ },
1901
+ {
1902
+ "epoch": 1.201241134751773,
1903
+ "grad_norm": 0.05315316107757854,
1904
+ "learning_rate": 4.443459416863672e-06,
1905
+ "loss": 0.0156,
1906
+ "step": 13550
1907
+ },
1908
+ {
1909
+ "epoch": 1.2056737588652482,
1910
+ "grad_norm": 1.358682872795046,
1911
+ "learning_rate": 4.418833727344366e-06,
1912
+ "loss": 0.0035,
1913
+ "step": 13600
1914
+ },
1915
+ {
1916
+ "epoch": 1.2101063829787235,
1917
+ "grad_norm": 0.004028994195015635,
1918
+ "learning_rate": 4.39420803782506e-06,
1919
+ "loss": 0.0078,
1920
+ "step": 13650
1921
+ },
1922
+ {
1923
+ "epoch": 1.2145390070921986,
1924
+ "grad_norm": 2.60025465642063e-05,
1925
+ "learning_rate": 4.369582348305753e-06,
1926
+ "loss": 0.0031,
1927
+ "step": 13700
1928
+ },
1929
+ {
1930
+ "epoch": 1.2189716312056738,
1931
+ "grad_norm": 0.15044035027572922,
1932
+ "learning_rate": 4.344956658786446e-06,
1933
+ "loss": 0.0073,
1934
+ "step": 13750
1935
+ },
1936
+ {
1937
+ "epoch": 1.2234042553191489,
1938
+ "grad_norm": 0.029395537343344148,
1939
+ "learning_rate": 4.32033096926714e-06,
1940
+ "loss": 0.0092,
1941
+ "step": 13800
1942
+ },
1943
+ {
1944
+ "epoch": 1.2278368794326242,
1945
+ "grad_norm": 0.00013265795388455031,
1946
+ "learning_rate": 4.295705279747833e-06,
1947
+ "loss": 0.002,
1948
+ "step": 13850
1949
+ },
1950
+ {
1951
+ "epoch": 1.2322695035460993,
1952
+ "grad_norm": 4.134334983257184,
1953
+ "learning_rate": 4.2710795902285265e-06,
1954
+ "loss": 0.0059,
1955
+ "step": 13900
1956
+ },
1957
+ {
1958
+ "epoch": 1.2367021276595744,
1959
+ "grad_norm": 2.4040766603197826e-05,
1960
+ "learning_rate": 4.246453900709221e-06,
1961
+ "loss": 0.009,
1962
+ "step": 13950
1963
+ },
1964
+ {
1965
+ "epoch": 1.2411347517730495,
1966
+ "grad_norm": 5.73501558270658e-05,
1967
+ "learning_rate": 4.2223207249803e-06,
1968
+ "loss": 0.0334,
1969
+ "step": 14000
1970
+ },
1971
+ {
1972
+ "epoch": 1.2455673758865249,
1973
+ "grad_norm": 0.0006514599750282896,
1974
+ "learning_rate": 4.197695035460993e-06,
1975
+ "loss": 0.004,
1976
+ "step": 14050
1977
+ },
1978
+ {
1979
+ "epoch": 1.25,
1980
+ "grad_norm": 5.3366246965865256e-05,
1981
+ "learning_rate": 4.173069345941686e-06,
1982
+ "loss": 0.006,
1983
+ "step": 14100
1984
+ },
1985
+ {
1986
+ "epoch": 1.2544326241134751,
1987
+ "grad_norm": 0.031237656606814084,
1988
+ "learning_rate": 4.1484436564223805e-06,
1989
+ "loss": 0.0057,
1990
+ "step": 14150
1991
+ },
1992
+ {
1993
+ "epoch": 1.2588652482269502,
1994
+ "grad_norm": 0.3490867651221054,
1995
+ "learning_rate": 4.123817966903074e-06,
1996
+ "loss": 0.004,
1997
+ "step": 14200
1998
+ },
1999
+ {
2000
+ "epoch": 1.2632978723404256,
2001
+ "grad_norm": 0.003986709752659374,
2002
+ "learning_rate": 4.099192277383767e-06,
2003
+ "loss": 0.009,
2004
+ "step": 14250
2005
+ },
2006
+ {
2007
+ "epoch": 1.2677304964539007,
2008
+ "grad_norm": 0.041880938463403676,
2009
+ "learning_rate": 4.074566587864461e-06,
2010
+ "loss": 0.0018,
2011
+ "step": 14300
2012
+ },
2013
+ {
2014
+ "epoch": 1.272163120567376,
2015
+ "grad_norm": 0.13043894056028277,
2016
+ "learning_rate": 4.049940898345154e-06,
2017
+ "loss": 0.0089,
2018
+ "step": 14350
2019
+ },
2020
+ {
2021
+ "epoch": 1.2765957446808511,
2022
+ "grad_norm": 20.525739931994153,
2023
+ "learning_rate": 4.025315208825847e-06,
2024
+ "loss": 0.0038,
2025
+ "step": 14400
2026
+ },
2027
+ {
2028
+ "epoch": 1.2810283687943262,
2029
+ "grad_norm": 0.03957694418665241,
2030
+ "learning_rate": 4.000689519306541e-06,
2031
+ "loss": 0.0047,
2032
+ "step": 14450
2033
+ },
2034
+ {
2035
+ "epoch": 1.2854609929078014,
2036
+ "grad_norm": 9.270907908158689e-06,
2037
+ "learning_rate": 3.976063829787235e-06,
2038
+ "loss": 0.0085,
2039
+ "step": 14500
2040
+ },
2041
+ {
2042
+ "epoch": 1.2898936170212765,
2043
+ "grad_norm": 0.019093848993790852,
2044
+ "learning_rate": 3.951438140267928e-06,
2045
+ "loss": 0.0073,
2046
+ "step": 14550
2047
+ },
2048
+ {
2049
+ "epoch": 1.2943262411347518,
2050
+ "grad_norm": 0.000544650767309452,
2051
+ "learning_rate": 3.926812450748621e-06,
2052
+ "loss": 0.004,
2053
+ "step": 14600
2054
+ },
2055
+ {
2056
+ "epoch": 1.298758865248227,
2057
+ "grad_norm": 8.507651438455978e-05,
2058
+ "learning_rate": 3.902186761229315e-06,
2059
+ "loss": 0.0149,
2060
+ "step": 14650
2061
+ },
2062
+ {
2063
+ "epoch": 1.3031914893617023,
2064
+ "grad_norm": 0.00015750838935733676,
2065
+ "learning_rate": 3.877561071710008e-06,
2066
+ "loss": 0.0024,
2067
+ "step": 14700
2068
+ },
2069
+ {
2070
+ "epoch": 1.3076241134751774,
2071
+ "grad_norm": 2.1216331063148686,
2072
+ "learning_rate": 3.852935382190702e-06,
2073
+ "loss": 0.0027,
2074
+ "step": 14750
2075
+ },
2076
+ {
2077
+ "epoch": 1.3120567375886525,
2078
+ "grad_norm": 0.0865583346196146,
2079
+ "learning_rate": 3.828309692671395e-06,
2080
+ "loss": 0.0073,
2081
+ "step": 14800
2082
+ },
2083
+ {
2084
+ "epoch": 1.3164893617021276,
2085
+ "grad_norm": 0.010228577051675729,
2086
+ "learning_rate": 3.8036840031520884e-06,
2087
+ "loss": 0.0127,
2088
+ "step": 14850
2089
+ },
2090
+ {
2091
+ "epoch": 1.3209219858156027,
2092
+ "grad_norm": 0.0005275434137609081,
2093
+ "learning_rate": 3.7790583136327817e-06,
2094
+ "loss": 0.0016,
2095
+ "step": 14900
2096
+ },
2097
+ {
2098
+ "epoch": 1.325354609929078,
2099
+ "grad_norm": 0.0030741880563327027,
2100
+ "learning_rate": 3.7544326241134755e-06,
2101
+ "loss": 0.0111,
2102
+ "step": 14950
2103
+ },
2104
+ {
2105
+ "epoch": 1.3297872340425532,
2106
+ "grad_norm": 0.4322702812359092,
2107
+ "learning_rate": 3.729806934594169e-06,
2108
+ "loss": 0.0108,
2109
+ "step": 15000
2110
+ },
2111
+ {
2112
+ "epoch": 1.3342198581560285,
2113
+ "grad_norm": 16.680559293669063,
2114
+ "learning_rate": 3.7051812450748626e-06,
2115
+ "loss": 0.0082,
2116
+ "step": 15050
2117
+ },
2118
+ {
2119
+ "epoch": 1.3386524822695036,
2120
+ "grad_norm": 0.0004440540435326648,
2121
+ "learning_rate": 3.680555555555556e-06,
2122
+ "loss": 0.008,
2123
+ "step": 15100
2124
+ },
2125
+ {
2126
+ "epoch": 1.3430851063829787,
2127
+ "grad_norm": 0.0004015706606442299,
2128
+ "learning_rate": 3.655929866036249e-06,
2129
+ "loss": 0.003,
2130
+ "step": 15150
2131
+ },
2132
+ {
2133
+ "epoch": 1.3475177304964538,
2134
+ "grad_norm": 0.00018125937118334472,
2135
+ "learning_rate": 3.6313041765169427e-06,
2136
+ "loss": 0.0009,
2137
+ "step": 15200
2138
+ },
2139
+ {
2140
+ "epoch": 1.351950354609929,
2141
+ "grad_norm": 0.07848257548289153,
2142
+ "learning_rate": 3.606678486997636e-06,
2143
+ "loss": 0.002,
2144
+ "step": 15250
2145
+ },
2146
+ {
2147
+ "epoch": 1.3563829787234043,
2148
+ "grad_norm": 0.023025301609373643,
2149
+ "learning_rate": 3.58205279747833e-06,
2150
+ "loss": 0.0031,
2151
+ "step": 15300
2152
+ },
2153
+ {
2154
+ "epoch": 1.3608156028368794,
2155
+ "grad_norm": 4.517888013582396,
2156
+ "learning_rate": 3.557427107959023e-06,
2157
+ "loss": 0.004,
2158
+ "step": 15350
2159
+ },
2160
+ {
2161
+ "epoch": 1.3652482269503547,
2162
+ "grad_norm": 1.2658270686339672,
2163
+ "learning_rate": 3.5328014184397165e-06,
2164
+ "loss": 0.0079,
2165
+ "step": 15400
2166
+ },
2167
+ {
2168
+ "epoch": 1.3696808510638299,
2169
+ "grad_norm": 0.004071845591775745,
2170
+ "learning_rate": 3.50817572892041e-06,
2171
+ "loss": 0.0055,
2172
+ "step": 15450
2173
+ },
2174
+ {
2175
+ "epoch": 1.374113475177305,
2176
+ "grad_norm": 0.0002310397513515275,
2177
+ "learning_rate": 3.4835500394011037e-06,
2178
+ "loss": 0.0034,
2179
+ "step": 15500
2180
+ },
2181
+ {
2182
+ "epoch": 1.37854609929078,
2183
+ "grad_norm": 0.019889125511476966,
2184
+ "learning_rate": 3.458924349881797e-06,
2185
+ "loss": 0.0089,
2186
+ "step": 15550
2187
+ },
2188
+ {
2189
+ "epoch": 1.3829787234042552,
2190
+ "grad_norm": 0.12069528532648284,
2191
+ "learning_rate": 3.4342986603624904e-06,
2192
+ "loss": 0.0103,
2193
+ "step": 15600
2194
+ },
2195
+ {
2196
+ "epoch": 1.3874113475177305,
2197
+ "grad_norm": 2.221152066469194,
2198
+ "learning_rate": 3.409672970843184e-06,
2199
+ "loss": 0.0115,
2200
+ "step": 15650
2201
+ },
2202
+ {
2203
+ "epoch": 1.3918439716312057,
2204
+ "grad_norm": 0.21584041208425908,
2205
+ "learning_rate": 3.385047281323877e-06,
2206
+ "loss": 0.0024,
2207
+ "step": 15700
2208
+ },
2209
+ {
2210
+ "epoch": 1.3962765957446808,
2211
+ "grad_norm": 0.07962326798958183,
2212
+ "learning_rate": 3.360421591804571e-06,
2213
+ "loss": 0.0004,
2214
+ "step": 15750
2215
+ },
2216
+ {
2217
+ "epoch": 1.400709219858156,
2218
+ "grad_norm": 0.0002935001100762825,
2219
+ "learning_rate": 3.3357959022852642e-06,
2220
+ "loss": 0.0108,
2221
+ "step": 15800
2222
+ },
2223
+ {
2224
+ "epoch": 1.4051418439716312,
2225
+ "grad_norm": 9.58127799729887,
2226
+ "learning_rate": 3.311170212765958e-06,
2227
+ "loss": 0.0013,
2228
+ "step": 15850
2229
+ },
2230
+ {
2231
+ "epoch": 1.4095744680851063,
2232
+ "grad_norm": 0.013975459324715883,
2233
+ "learning_rate": 3.2865445232466514e-06,
2234
+ "loss": 0.006,
2235
+ "step": 15900
2236
+ },
2237
+ {
2238
+ "epoch": 1.4140070921985815,
2239
+ "grad_norm": 0.0003136739495490558,
2240
+ "learning_rate": 3.2619188337273443e-06,
2241
+ "loss": 0.0067,
2242
+ "step": 15950
2243
+ },
2244
+ {
2245
+ "epoch": 1.4184397163120568,
2246
+ "grad_norm": 0.23057539945365355,
2247
+ "learning_rate": 3.237293144208038e-06,
2248
+ "loss": 0.0029,
2249
+ "step": 16000
2250
+ },
2251
+ {
2252
+ "epoch": 1.422872340425532,
2253
+ "grad_norm": 0.11630128906398461,
2254
+ "learning_rate": 3.2126674546887314e-06,
2255
+ "loss": 0.0011,
2256
+ "step": 16050
2257
+ },
2258
+ {
2259
+ "epoch": 1.427304964539007,
2260
+ "grad_norm": 1.9025801682092172,
2261
+ "learning_rate": 3.1880417651694252e-06,
2262
+ "loss": 0.0008,
2263
+ "step": 16100
2264
+ },
2265
+ {
2266
+ "epoch": 1.4317375886524824,
2267
+ "grad_norm": 0.9957743413096815,
2268
+ "learning_rate": 3.1634160756501186e-06,
2269
+ "loss": 0.0006,
2270
+ "step": 16150
2271
+ },
2272
+ {
2273
+ "epoch": 1.4361702127659575,
2274
+ "grad_norm": 4.523729944492489e-05,
2275
+ "learning_rate": 3.1392828999211984e-06,
2276
+ "loss": 0.0064,
2277
+ "step": 16200
2278
+ },
2279
+ {
2280
+ "epoch": 1.4406028368794326,
2281
+ "grad_norm": 0.017807506139140707,
2282
+ "learning_rate": 3.1146572104018913e-06,
2283
+ "loss": 0.0028,
2284
+ "step": 16250
2285
+ },
2286
+ {
2287
+ "epoch": 1.4450354609929077,
2288
+ "grad_norm": 0.2936267584110255,
2289
+ "learning_rate": 3.0900315208825847e-06,
2290
+ "loss": 0.0218,
2291
+ "step": 16300
2292
+ },
2293
+ {
2294
+ "epoch": 1.449468085106383,
2295
+ "grad_norm": 2.201825023255903e-06,
2296
+ "learning_rate": 3.0654058313632784e-06,
2297
+ "loss": 0.0089,
2298
+ "step": 16350
2299
+ },
2300
+ {
2301
+ "epoch": 1.4539007092198581,
2302
+ "grad_norm": 0.0020967040078515953,
2303
+ "learning_rate": 3.040780141843972e-06,
2304
+ "loss": 0.0011,
2305
+ "step": 16400
2306
+ },
2307
+ {
2308
+ "epoch": 1.4583333333333333,
2309
+ "grad_norm": 0.0035587730480887963,
2310
+ "learning_rate": 3.0161544523246656e-06,
2311
+ "loss": 0.0061,
2312
+ "step": 16450
2313
+ },
2314
+ {
2315
+ "epoch": 1.4627659574468086,
2316
+ "grad_norm": 0.00011908233856549432,
2317
+ "learning_rate": 2.9915287628053585e-06,
2318
+ "loss": 0.0104,
2319
+ "step": 16500
2320
+ },
2321
+ {
2322
+ "epoch": 1.4671985815602837,
2323
+ "grad_norm": 1.358266707504535,
2324
+ "learning_rate": 2.9669030732860523e-06,
2325
+ "loss": 0.0146,
2326
+ "step": 16550
2327
+ },
2328
+ {
2329
+ "epoch": 1.4716312056737588,
2330
+ "grad_norm": 0.0005173968272110369,
2331
+ "learning_rate": 2.9422773837667456e-06,
2332
+ "loss": 0.0072,
2333
+ "step": 16600
2334
+ },
2335
+ {
2336
+ "epoch": 1.476063829787234,
2337
+ "grad_norm": 5.0037899912089685,
2338
+ "learning_rate": 2.917651694247439e-06,
2339
+ "loss": 0.0077,
2340
+ "step": 16650
2341
+ },
2342
+ {
2343
+ "epoch": 1.4804964539007093,
2344
+ "grad_norm": 0.00036722887470731584,
2345
+ "learning_rate": 2.8930260047281328e-06,
2346
+ "loss": 0.022,
2347
+ "step": 16700
2348
+ },
2349
+ {
2350
+ "epoch": 1.4849290780141844,
2351
+ "grad_norm": 0.002660326684541599,
2352
+ "learning_rate": 2.868400315208826e-06,
2353
+ "loss": 0.0052,
2354
+ "step": 16750
2355
+ },
2356
+ {
2357
+ "epoch": 1.4893617021276595,
2358
+ "grad_norm": 0.004956960764106037,
2359
+ "learning_rate": 2.8437746256895195e-06,
2360
+ "loss": 0.0072,
2361
+ "step": 16800
2362
+ },
2363
+ {
2364
+ "epoch": 1.4937943262411348,
2365
+ "grad_norm": 1.0478071930407054,
2366
+ "learning_rate": 2.819148936170213e-06,
2367
+ "loss": 0.0037,
2368
+ "step": 16850
2369
+ },
2370
+ {
2371
+ "epoch": 1.49822695035461,
2372
+ "grad_norm": 0.032938488132694524,
2373
+ "learning_rate": 2.7945232466509066e-06,
2374
+ "loss": 0.003,
2375
+ "step": 16900
2376
+ },
2377
+ {
2378
+ "epoch": 1.502659574468085,
2379
+ "grad_norm": 0.04903975876178212,
2380
+ "learning_rate": 2.7698975571316e-06,
2381
+ "loss": 0.0061,
2382
+ "step": 16950
2383
+ },
2384
+ {
2385
+ "epoch": 1.5070921985815602,
2386
+ "grad_norm": 0.3263491645553027,
2387
+ "learning_rate": 2.7452718676122938e-06,
2388
+ "loss": 0.019,
2389
+ "step": 17000
2390
+ },
2391
+ {
2392
+ "epoch": 1.5115248226950353,
2393
+ "grad_norm": 2.5675365914148904,
2394
+ "learning_rate": 2.7206461780929867e-06,
2395
+ "loss": 0.0014,
2396
+ "step": 17050
2397
+ },
2398
+ {
2399
+ "epoch": 1.5159574468085106,
2400
+ "grad_norm": 0.0036317115050709474,
2401
+ "learning_rate": 2.69602048857368e-06,
2402
+ "loss": 0.0055,
2403
+ "step": 17100
2404
+ },
2405
+ {
2406
+ "epoch": 1.520390070921986,
2407
+ "grad_norm": 0.0009379009024431418,
2408
+ "learning_rate": 2.671394799054374e-06,
2409
+ "loss": 0.0169,
2410
+ "step": 17150
2411
+ },
2412
+ {
2413
+ "epoch": 1.524822695035461,
2414
+ "grad_norm": 0.0022970151160776,
2415
+ "learning_rate": 2.646769109535067e-06,
2416
+ "loss": 0.0053,
2417
+ "step": 17200
2418
+ },
2419
+ {
2420
+ "epoch": 1.5292553191489362,
2421
+ "grad_norm": 0.0019246470303829883,
2422
+ "learning_rate": 2.622143420015761e-06,
2423
+ "loss": 0.0025,
2424
+ "step": 17250
2425
+ },
2426
+ {
2427
+ "epoch": 1.5336879432624113,
2428
+ "grad_norm": 0.007413966936557272,
2429
+ "learning_rate": 2.5985027580772264e-06,
2430
+ "loss": 0.0113,
2431
+ "step": 17300
2432
+ },
2433
+ {
2434
+ "epoch": 1.5381205673758864,
2435
+ "grad_norm": 0.00024108723854638272,
2436
+ "learning_rate": 2.57387706855792e-06,
2437
+ "loss": 0.0056,
2438
+ "step": 17350
2439
+ },
2440
+ {
2441
+ "epoch": 1.5425531914893615,
2442
+ "grad_norm": 0.08798384755826176,
2443
+ "learning_rate": 2.549251379038613e-06,
2444
+ "loss": 0.0181,
2445
+ "step": 17400
2446
+ },
2447
+ {
2448
+ "epoch": 1.5469858156028369,
2449
+ "grad_norm": 0.17481763887643156,
2450
+ "learning_rate": 2.524625689519307e-06,
2451
+ "loss": 0.0056,
2452
+ "step": 17450
2453
+ },
2454
+ {
2455
+ "epoch": 1.5514184397163122,
2456
+ "grad_norm": 0.00012363471801080713,
2457
+ "learning_rate": 2.5e-06,
2458
+ "loss": 0.0048,
2459
+ "step": 17500
2460
+ },
2461
+ {
2462
+ "epoch": 1.5558510638297873,
2463
+ "grad_norm": 0.000281235141606495,
2464
+ "learning_rate": 2.4753743104806936e-06,
2465
+ "loss": 0.0041,
2466
+ "step": 17550
2467
+ },
2468
+ {
2469
+ "epoch": 1.5602836879432624,
2470
+ "grad_norm": 0.002074284717507087,
2471
+ "learning_rate": 2.450748620961387e-06,
2472
+ "loss": 0.0078,
2473
+ "step": 17600
2474
+ },
2475
+ {
2476
+ "epoch": 1.5647163120567376,
2477
+ "grad_norm": 0.011639759081317758,
2478
+ "learning_rate": 2.4261229314420807e-06,
2479
+ "loss": 0.001,
2480
+ "step": 17650
2481
+ },
2482
+ {
2483
+ "epoch": 1.5691489361702127,
2484
+ "grad_norm": 0.0002970237662211247,
2485
+ "learning_rate": 2.401497241922774e-06,
2486
+ "loss": 0.0107,
2487
+ "step": 17700
2488
+ },
2489
+ {
2490
+ "epoch": 1.5735815602836878,
2491
+ "grad_norm": 0.0007912621277405812,
2492
+ "learning_rate": 2.3768715524034674e-06,
2493
+ "loss": 0.0114,
2494
+ "step": 17750
2495
+ },
2496
+ {
2497
+ "epoch": 1.5780141843971631,
2498
+ "grad_norm": 2.5015768800092233,
2499
+ "learning_rate": 2.352245862884161e-06,
2500
+ "loss": 0.0131,
2501
+ "step": 17800
2502
+ },
2503
+ {
2504
+ "epoch": 1.5824468085106385,
2505
+ "grad_norm": 1.1611881483542175,
2506
+ "learning_rate": 2.327620173364854e-06,
2507
+ "loss": 0.0048,
2508
+ "step": 17850
2509
+ },
2510
+ {
2511
+ "epoch": 1.5868794326241136,
2512
+ "grad_norm": 8.345867902582887,
2513
+ "learning_rate": 2.302994483845548e-06,
2514
+ "loss": 0.0062,
2515
+ "step": 17900
2516
+ },
2517
+ {
2518
+ "epoch": 1.5913120567375887,
2519
+ "grad_norm": 2.226488828276802,
2520
+ "learning_rate": 2.2783687943262413e-06,
2521
+ "loss": 0.0033,
2522
+ "step": 17950
2523
+ },
2524
+ {
2525
+ "epoch": 1.5957446808510638,
2526
+ "grad_norm": 0.2250974162797621,
2527
+ "learning_rate": 2.2537431048069346e-06,
2528
+ "loss": 0.0039,
2529
+ "step": 18000
2530
+ },
2531
+ {
2532
+ "epoch": 1.600177304964539,
2533
+ "grad_norm": 3.541017240933167,
2534
+ "learning_rate": 2.2291174152876284e-06,
2535
+ "loss": 0.0025,
2536
+ "step": 18050
2537
+ },
2538
+ {
2539
+ "epoch": 1.604609929078014,
2540
+ "grad_norm": 0.0010147231973171634,
2541
+ "learning_rate": 2.2044917257683217e-06,
2542
+ "loss": 0.0014,
2543
+ "step": 18100
2544
+ },
2545
+ {
2546
+ "epoch": 1.6090425531914894,
2547
+ "grad_norm": 0.005109813507854219,
2548
+ "learning_rate": 2.179866036249015e-06,
2549
+ "loss": 0.0077,
2550
+ "step": 18150
2551
+ },
2552
+ {
2553
+ "epoch": 1.6134751773049647,
2554
+ "grad_norm": 0.5285122394100901,
2555
+ "learning_rate": 2.1552403467297085e-06,
2556
+ "loss": 0.003,
2557
+ "step": 18200
2558
+ },
2559
+ {
2560
+ "epoch": 1.6179078014184398,
2561
+ "grad_norm": 0.007852944630078929,
2562
+ "learning_rate": 2.130614657210402e-06,
2563
+ "loss": 0.0063,
2564
+ "step": 18250
2565
+ },
2566
+ {
2567
+ "epoch": 1.622340425531915,
2568
+ "grad_norm": 5.6094508415245195e-05,
2569
+ "learning_rate": 2.1059889676910956e-06,
2570
+ "loss": 0.004,
2571
+ "step": 18300
2572
+ },
2573
+ {
2574
+ "epoch": 1.62677304964539,
2575
+ "grad_norm": 0.5491008148980681,
2576
+ "learning_rate": 2.081363278171789e-06,
2577
+ "loss": 0.0055,
2578
+ "step": 18350
2579
+ },
2580
+ {
2581
+ "epoch": 1.6312056737588652,
2582
+ "grad_norm": 1.5525025008668276,
2583
+ "learning_rate": 2.0567375886524823e-06,
2584
+ "loss": 0.0021,
2585
+ "step": 18400
2586
+ },
2587
+ {
2588
+ "epoch": 1.6356382978723403,
2589
+ "grad_norm": 0.019464790999853782,
2590
+ "learning_rate": 2.032111899133176e-06,
2591
+ "loss": 0.0015,
2592
+ "step": 18450
2593
+ },
2594
+ {
2595
+ "epoch": 1.6400709219858156,
2596
+ "grad_norm": 0.03354862267681748,
2597
+ "learning_rate": 2.0074862096138694e-06,
2598
+ "loss": 0.0009,
2599
+ "step": 18500
2600
+ },
2601
+ {
2602
+ "epoch": 1.6445035460992907,
2603
+ "grad_norm": 3.661779589498576e-05,
2604
+ "learning_rate": 1.982860520094563e-06,
2605
+ "loss": 0.005,
2606
+ "step": 18550
2607
+ },
2608
+ {
2609
+ "epoch": 1.648936170212766,
2610
+ "grad_norm": 29.226511335493843,
2611
+ "learning_rate": 1.958234830575256e-06,
2612
+ "loss": 0.018,
2613
+ "step": 18600
2614
+ },
2615
+ {
2616
+ "epoch": 1.6533687943262412,
2617
+ "grad_norm": 2.6959490480503963,
2618
+ "learning_rate": 1.9336091410559495e-06,
2619
+ "loss": 0.0057,
2620
+ "step": 18650
2621
+ },
2622
+ {
2623
+ "epoch": 1.6578014184397163,
2624
+ "grad_norm": 0.3429448060734279,
2625
+ "learning_rate": 1.9089834515366433e-06,
2626
+ "loss": 0.0019,
2627
+ "step": 18700
2628
+ },
2629
+ {
2630
+ "epoch": 1.6622340425531914,
2631
+ "grad_norm": 1.3403285429948613e-05,
2632
+ "learning_rate": 1.8843577620173366e-06,
2633
+ "loss": 0.0025,
2634
+ "step": 18750
2635
+ },
2636
+ {
2637
+ "epoch": 1.6666666666666665,
2638
+ "grad_norm": 0.0006843325116012807,
2639
+ "learning_rate": 1.85973207249803e-06,
2640
+ "loss": 0.0083,
2641
+ "step": 18800
2642
+ },
2643
+ {
2644
+ "epoch": 1.6710992907801419,
2645
+ "grad_norm": 2.896950341714159e-05,
2646
+ "learning_rate": 1.8351063829787236e-06,
2647
+ "loss": 0.0047,
2648
+ "step": 18850
2649
+ },
2650
+ {
2651
+ "epoch": 1.675531914893617,
2652
+ "grad_norm": 0.03579736142023673,
2653
+ "learning_rate": 1.8104806934594171e-06,
2654
+ "loss": 0.0044,
2655
+ "step": 18900
2656
+ },
2657
+ {
2658
+ "epoch": 1.6799645390070923,
2659
+ "grad_norm": 0.0007634411592739593,
2660
+ "learning_rate": 1.7858550039401105e-06,
2661
+ "loss": 0.0074,
2662
+ "step": 18950
2663
+ },
2664
+ {
2665
+ "epoch": 1.6843971631205674,
2666
+ "grad_norm": 0.00025338095037660184,
2667
+ "learning_rate": 1.761229314420804e-06,
2668
+ "loss": 0.0028,
2669
+ "step": 19000
2670
+ },
2671
+ {
2672
+ "epoch": 1.6888297872340425,
2673
+ "grad_norm": 0.8367439016912624,
2674
+ "learning_rate": 1.7366036249014972e-06,
2675
+ "loss": 0.0123,
2676
+ "step": 19050
2677
+ },
2678
+ {
2679
+ "epoch": 1.6932624113475176,
2680
+ "grad_norm": 0.0009791356139938734,
2681
+ "learning_rate": 1.7119779353821908e-06,
2682
+ "loss": 0.0043,
2683
+ "step": 19100
2684
+ },
2685
+ {
2686
+ "epoch": 1.6976950354609928,
2687
+ "grad_norm": 1.1288708215981906,
2688
+ "learning_rate": 1.6873522458628843e-06,
2689
+ "loss": 0.0014,
2690
+ "step": 19150
2691
+ },
2692
+ {
2693
+ "epoch": 1.702127659574468,
2694
+ "grad_norm": 6.775118398159698,
2695
+ "learning_rate": 1.6627265563435777e-06,
2696
+ "loss": 0.0027,
2697
+ "step": 19200
2698
+ },
2699
+ {
2700
+ "epoch": 1.7065602836879432,
2701
+ "grad_norm": 0.07111263325664992,
2702
+ "learning_rate": 1.6381008668242713e-06,
2703
+ "loss": 0.0054,
2704
+ "step": 19250
2705
+ },
2706
+ {
2707
+ "epoch": 1.7109929078014185,
2708
+ "grad_norm": 5.746382572617145e-05,
2709
+ "learning_rate": 1.6134751773049648e-06,
2710
+ "loss": 0.0023,
2711
+ "step": 19300
2712
+ },
2713
+ {
2714
+ "epoch": 1.7154255319148937,
2715
+ "grad_norm": 0.0005896955942330902,
2716
+ "learning_rate": 1.588849487785658e-06,
2717
+ "loss": 0.0028,
2718
+ "step": 19350
2719
+ },
2720
+ {
2721
+ "epoch": 1.7198581560283688,
2722
+ "grad_norm": 0.7760810372694008,
2723
+ "learning_rate": 1.5642237982663515e-06,
2724
+ "loss": 0.0017,
2725
+ "step": 19400
2726
+ },
2727
+ {
2728
+ "epoch": 1.724290780141844,
2729
+ "grad_norm": 0.006697039214016149,
2730
+ "learning_rate": 1.5400906225374313e-06,
2731
+ "loss": 0.0187,
2732
+ "step": 19450
2733
+ },
2734
+ {
2735
+ "epoch": 1.728723404255319,
2736
+ "grad_norm": 0.00022903477785452336,
2737
+ "learning_rate": 1.5154649330181245e-06,
2738
+ "loss": 0.0004,
2739
+ "step": 19500
2740
+ },
2741
+ {
2742
+ "epoch": 1.7331560283687943,
2743
+ "grad_norm": 2.7535469066175384e-05,
2744
+ "learning_rate": 1.490839243498818e-06,
2745
+ "loss": 0.0049,
2746
+ "step": 19550
2747
+ },
2748
+ {
2749
+ "epoch": 1.7375886524822695,
2750
+ "grad_norm": 0.02985438719947455,
2751
+ "learning_rate": 1.4662135539795116e-06,
2752
+ "loss": 0.0044,
2753
+ "step": 19600
2754
+ },
2755
+ {
2756
+ "epoch": 1.7420212765957448,
2757
+ "grad_norm": 0.00019760588466390203,
2758
+ "learning_rate": 1.441587864460205e-06,
2759
+ "loss": 0.0142,
2760
+ "step": 19650
2761
+ },
2762
+ {
2763
+ "epoch": 1.74645390070922,
2764
+ "grad_norm": 3.047480283427747,
2765
+ "learning_rate": 1.4169621749408985e-06,
2766
+ "loss": 0.0071,
2767
+ "step": 19700
2768
+ },
2769
+ {
2770
+ "epoch": 1.750886524822695,
2771
+ "grad_norm": 0.0037448846335741332,
2772
+ "learning_rate": 1.3923364854215921e-06,
2773
+ "loss": 0.0031,
2774
+ "step": 19750
2775
+ },
2776
+ {
2777
+ "epoch": 1.7553191489361701,
2778
+ "grad_norm": 0.00020748441056905848,
2779
+ "learning_rate": 1.3677107959022853e-06,
2780
+ "loss": 0.0057,
2781
+ "step": 19800
2782
+ },
2783
+ {
2784
+ "epoch": 1.7597517730496453,
2785
+ "grad_norm": 0.00016023525645660029,
2786
+ "learning_rate": 1.3430851063829788e-06,
2787
+ "loss": 0.0126,
2788
+ "step": 19850
2789
+ },
2790
+ {
2791
+ "epoch": 1.7641843971631206,
2792
+ "grad_norm": 7.791162563353672e-06,
2793
+ "learning_rate": 1.3184594168636722e-06,
2794
+ "loss": 0.012,
2795
+ "step": 19900
2796
+ },
2797
+ {
2798
+ "epoch": 1.7686170212765957,
2799
+ "grad_norm": 0.9465404763088664,
2800
+ "learning_rate": 1.2938337273443658e-06,
2801
+ "loss": 0.0052,
2802
+ "step": 19950
2803
+ },
2804
+ {
2805
+ "epoch": 1.773049645390071,
2806
+ "grad_norm": 0.07414027193365252,
2807
+ "learning_rate": 1.2692080378250593e-06,
2808
+ "loss": 0.0019,
2809
+ "step": 20000
2810
+ },
2811
+ {
2812
+ "epoch": 1.7774822695035462,
2813
+ "grad_norm": 0.0055180504122557894,
2814
+ "learning_rate": 1.2445823483057527e-06,
2815
+ "loss": 0.0006,
2816
+ "step": 20050
2817
+ },
2818
+ {
2819
+ "epoch": 1.7819148936170213,
2820
+ "grad_norm": 0.0006138651547325419,
2821
+ "learning_rate": 1.2199566587864462e-06,
2822
+ "loss": 0.0031,
2823
+ "step": 20100
2824
+ },
2825
+ {
2826
+ "epoch": 1.7863475177304964,
2827
+ "grad_norm": 1.2899272619970438e-05,
2828
+ "learning_rate": 1.1953309692671396e-06,
2829
+ "loss": 0.0032,
2830
+ "step": 20150
2831
+ },
2832
+ {
2833
+ "epoch": 1.7907801418439715,
2834
+ "grad_norm": 15.895355390743765,
2835
+ "learning_rate": 1.170705279747833e-06,
2836
+ "loss": 0.0064,
2837
+ "step": 20200
2838
+ },
2839
+ {
2840
+ "epoch": 1.7952127659574468,
2841
+ "grad_norm": 0.0004655677253189525,
2842
+ "learning_rate": 1.1460795902285265e-06,
2843
+ "loss": 0.0045,
2844
+ "step": 20250
2845
+ },
2846
+ {
2847
+ "epoch": 1.799645390070922,
2848
+ "grad_norm": 4.462022318771812,
2849
+ "learning_rate": 1.1214539007092199e-06,
2850
+ "loss": 0.0051,
2851
+ "step": 20300
2852
+ },
2853
+ {
2854
+ "epoch": 1.8040780141843973,
2855
+ "grad_norm": 7.791263127884315,
2856
+ "learning_rate": 1.0968282111899134e-06,
2857
+ "loss": 0.0024,
2858
+ "step": 20350
2859
+ },
2860
+ {
2861
+ "epoch": 1.8085106382978724,
2862
+ "grad_norm": 0.001981428634821069,
2863
+ "learning_rate": 1.0722025216706068e-06,
2864
+ "loss": 0.0047,
2865
+ "step": 20400
2866
+ },
2867
+ {
2868
+ "epoch": 1.8129432624113475,
2869
+ "grad_norm": 0.9791286474652084,
2870
+ "learning_rate": 1.0480693459416864e-06,
2871
+ "loss": 0.0067,
2872
+ "step": 20450
2873
+ },
2874
+ {
2875
+ "epoch": 1.8173758865248226,
2876
+ "grad_norm": 4.49890759411509e-05,
2877
+ "learning_rate": 1.02344365642238e-06,
2878
+ "loss": 0.003,
2879
+ "step": 20500
2880
+ },
2881
+ {
2882
+ "epoch": 1.8218085106382977,
2883
+ "grad_norm": 0.0008872317773727661,
2884
+ "learning_rate": 9.988179669030735e-07,
2885
+ "loss": 0.0198,
2886
+ "step": 20550
2887
+ },
2888
+ {
2889
+ "epoch": 1.826241134751773,
2890
+ "grad_norm": 0.02413008880272575,
2891
+ "learning_rate": 9.741922773837669e-07,
2892
+ "loss": 0.0027,
2893
+ "step": 20600
2894
+ },
2895
+ {
2896
+ "epoch": 1.8306737588652482,
2897
+ "grad_norm": 0.0064929573762096335,
2898
+ "learning_rate": 9.495665878644602e-07,
2899
+ "loss": 0.0051,
2900
+ "step": 20650
2901
+ },
2902
+ {
2903
+ "epoch": 1.8351063829787235,
2904
+ "grad_norm": 2.0483772647244787,
2905
+ "learning_rate": 9.249408983451537e-07,
2906
+ "loss": 0.0024,
2907
+ "step": 20700
2908
+ },
2909
+ {
2910
+ "epoch": 1.8395390070921986,
2911
+ "grad_norm": 0.008048228817919003,
2912
+ "learning_rate": 9.003152088258473e-07,
2913
+ "loss": 0.0062,
2914
+ "step": 20750
2915
+ },
2916
+ {
2917
+ "epoch": 1.8439716312056738,
2918
+ "grad_norm": 0.0011709925053889282,
2919
+ "learning_rate": 8.756895193065406e-07,
2920
+ "loss": 0.0008,
2921
+ "step": 20800
2922
+ },
2923
+ {
2924
+ "epoch": 1.8484042553191489,
2925
+ "grad_norm": 2.6180834531581985e-05,
2926
+ "learning_rate": 8.510638297872341e-07,
2927
+ "loss": 0.0087,
2928
+ "step": 20850
2929
+ },
2930
+ {
2931
+ "epoch": 1.852836879432624,
2932
+ "grad_norm": 0.00019269222943037962,
2933
+ "learning_rate": 8.264381402679275e-07,
2934
+ "loss": 0.0045,
2935
+ "step": 20900
2936
+ },
2937
+ {
2938
+ "epoch": 1.8572695035460993,
2939
+ "grad_norm": 3.881137417498048,
2940
+ "learning_rate": 8.018124507486211e-07,
2941
+ "loss": 0.0055,
2942
+ "step": 20950
2943
+ },
2944
+ {
2945
+ "epoch": 1.8617021276595744,
2946
+ "grad_norm": 0.005418221882606532,
2947
+ "learning_rate": 7.771867612293145e-07,
2948
+ "loss": 0.0018,
2949
+ "step": 21000
2950
+ },
2951
+ {
2952
+ "epoch": 1.8661347517730498,
2953
+ "grad_norm": 1.07656255093645,
2954
+ "learning_rate": 7.525610717100079e-07,
2955
+ "loss": 0.0002,
2956
+ "step": 21050
2957
+ },
2958
+ {
2959
+ "epoch": 1.8705673758865249,
2960
+ "grad_norm": 9.912206549708396,
2961
+ "learning_rate": 7.279353821907014e-07,
2962
+ "loss": 0.0055,
2963
+ "step": 21100
2964
+ },
2965
+ {
2966
+ "epoch": 1.875,
2967
+ "grad_norm": 2.1879933703908734e-05,
2968
+ "learning_rate": 7.033096926713949e-07,
2969
+ "loss": 0.0033,
2970
+ "step": 21150
2971
+ },
2972
+ {
2973
+ "epoch": 1.8794326241134751,
2974
+ "grad_norm": 0.00016572161736391663,
2975
+ "learning_rate": 6.786840031520883e-07,
2976
+ "loss": 0.0177,
2977
+ "step": 21200
2978
+ },
2979
+ {
2980
+ "epoch": 1.8838652482269502,
2981
+ "grad_norm": 0.27069341578831,
2982
+ "learning_rate": 6.540583136327818e-07,
2983
+ "loss": 0.0121,
2984
+ "step": 21250
2985
+ },
2986
+ {
2987
+ "epoch": 1.8882978723404256,
2988
+ "grad_norm": 0.004855813506598364,
2989
+ "learning_rate": 6.294326241134751e-07,
2990
+ "loss": 0.0028,
2991
+ "step": 21300
2992
+ },
2993
+ {
2994
+ "epoch": 1.8927304964539007,
2995
+ "grad_norm": 2.708412159481762,
2996
+ "learning_rate": 6.048069345941687e-07,
2997
+ "loss": 0.0041,
2998
+ "step": 21350
2999
+ },
3000
+ {
3001
+ "epoch": 1.897163120567376,
3002
+ "grad_norm": 0.0023739265966847512,
3003
+ "learning_rate": 5.801812450748622e-07,
3004
+ "loss": 0.0018,
3005
+ "step": 21400
3006
+ },
3007
+ {
3008
+ "epoch": 1.9015957446808511,
3009
+ "grad_norm": 6.753017191100636e-05,
3010
+ "learning_rate": 5.555555555555555e-07,
3011
+ "loss": 0.0022,
3012
+ "step": 21450
3013
+ },
3014
+ {
3015
+ "epoch": 1.9060283687943262,
3016
+ "grad_norm": 0.1679332551490689,
3017
+ "learning_rate": 5.309298660362491e-07,
3018
+ "loss": 0.0038,
3019
+ "step": 21500
3020
+ },
3021
+ {
3022
+ "epoch": 1.9104609929078014,
3023
+ "grad_norm": 0.0036223384079424226,
3024
+ "learning_rate": 5.063041765169424e-07,
3025
+ "loss": 0.0005,
3026
+ "step": 21550
3027
+ },
3028
+ {
3029
+ "epoch": 1.9148936170212765,
3030
+ "grad_norm": 0.005616534836631223,
3031
+ "learning_rate": 4.81678486997636e-07,
3032
+ "loss": 0.0015,
3033
+ "step": 21600
3034
+ },
3035
+ {
3036
+ "epoch": 1.9193262411347518,
3037
+ "grad_norm": 0.03037940885875586,
3038
+ "learning_rate": 4.570527974783294e-07,
3039
+ "loss": 0.0053,
3040
+ "step": 21650
3041
+ },
3042
+ {
3043
+ "epoch": 1.923758865248227,
3044
+ "grad_norm": 0.008440772200658765,
3045
+ "learning_rate": 4.32919621749409e-07,
3046
+ "loss": 0.0063,
3047
+ "step": 21700
3048
+ },
3049
+ {
3050
+ "epoch": 1.9281914893617023,
3051
+ "grad_norm": 0.16797682831433902,
3052
+ "learning_rate": 4.082939322301025e-07,
3053
+ "loss": 0.0161,
3054
+ "step": 21750
3055
+ },
3056
+ {
3057
+ "epoch": 1.9326241134751774,
3058
+ "grad_norm": 0.0003968882411133838,
3059
+ "learning_rate": 3.836682427107959e-07,
3060
+ "loss": 0.0016,
3061
+ "step": 21800
3062
+ },
3063
+ {
3064
+ "epoch": 1.9370567375886525,
3065
+ "grad_norm": 0.0025799262877277416,
3066
+ "learning_rate": 3.5904255319148944e-07,
3067
+ "loss": 0.0038,
3068
+ "step": 21850
3069
+ },
3070
+ {
3071
+ "epoch": 1.9414893617021276,
3072
+ "grad_norm": 2.369958646711284,
3073
+ "learning_rate": 3.3490937746256897e-07,
3074
+ "loss": 0.0163,
3075
+ "step": 21900
3076
+ },
3077
+ {
3078
+ "epoch": 1.9459219858156027,
3079
+ "grad_norm": 35.57388740134749,
3080
+ "learning_rate": 3.1028368794326243e-07,
3081
+ "loss": 0.0152,
3082
+ "step": 21950
3083
+ },
3084
+ {
3085
+ "epoch": 1.950354609929078,
3086
+ "grad_norm": 0.00020545397695902366,
3087
+ "learning_rate": 2.856579984239559e-07,
3088
+ "loss": 0.0081,
3089
+ "step": 22000
3090
+ },
3091
+ {
3092
+ "epoch": 1.9547872340425532,
3093
+ "grad_norm": 0.10530696929365847,
3094
+ "learning_rate": 2.6103230890464936e-07,
3095
+ "loss": 0.001,
3096
+ "step": 22050
3097
+ },
3098
+ {
3099
+ "epoch": 1.9592198581560285,
3100
+ "grad_norm": 2.128160827831901e-05,
3101
+ "learning_rate": 2.364066193853428e-07,
3102
+ "loss": 0.0011,
3103
+ "step": 22100
3104
+ },
3105
+ {
3106
+ "epoch": 1.9636524822695036,
3107
+ "grad_norm": 22.61228357889454,
3108
+ "learning_rate": 2.1178092986603625e-07,
3109
+ "loss": 0.0128,
3110
+ "step": 22150
3111
+ },
3112
+ {
3113
+ "epoch": 1.9680851063829787,
3114
+ "grad_norm": 9.742101126804748e-05,
3115
+ "learning_rate": 1.8715524034672972e-07,
3116
+ "loss": 0.0011,
3117
+ "step": 22200
3118
+ },
3119
+ {
3120
+ "epoch": 1.9725177304964538,
3121
+ "grad_norm": 0.0001410965514704057,
3122
+ "learning_rate": 1.6252955082742318e-07,
3123
+ "loss": 0.0028,
3124
+ "step": 22250
3125
+ },
3126
+ {
3127
+ "epoch": 1.976950354609929,
3128
+ "grad_norm": 2.4346340548203285e-05,
3129
+ "learning_rate": 1.3790386130811664e-07,
3130
+ "loss": 0.006,
3131
+ "step": 22300
3132
+ },
3133
+ {
3134
+ "epoch": 1.9813829787234043,
3135
+ "grad_norm": 1.3746725472766317,
3136
+ "learning_rate": 1.132781717888101e-07,
3137
+ "loss": 0.0046,
3138
+ "step": 22350
3139
+ },
3140
+ {
3141
+ "epoch": 1.9858156028368794,
3142
+ "grad_norm": 0.0011067158800566529,
3143
+ "learning_rate": 8.865248226950356e-08,
3144
+ "loss": 0.0075,
3145
+ "step": 22400
3146
+ },
3147
+ {
3148
+ "epoch": 1.9902482269503547,
3149
+ "grad_norm": 0.003488971575134763,
3150
+ "learning_rate": 6.402679275019701e-08,
3151
+ "loss": 0.0053,
3152
+ "step": 22450
3153
+ },
3154
+ {
3155
+ "epoch": 1.9946808510638299,
3156
+ "grad_norm": 0.0002509487351378721,
3157
+ "learning_rate": 3.940110323089047e-08,
3158
+ "loss": 0.0056,
3159
+ "step": 22500
3160
+ },
3161
+ {
3162
+ "epoch": 1.999113475177305,
3163
+ "grad_norm": 0.7472621297480373,
3164
+ "learning_rate": 1.4775413711583924e-08,
3165
+ "loss": 0.0062,
3166
+ "step": 22550
3167
+ }
3168
+ ],
3169
+ "logging_steps": 50,
3170
+ "max_steps": 22560,
3171
+ "num_input_tokens_seen": 0,
3172
+ "num_train_epochs": 2,
3173
+ "save_steps": 4250,
3174
+ "stateful_callbacks": {
3175
+ "TrainerControl": {
3176
+ "args": {
3177
+ "should_epoch_stop": false,
3178
+ "should_evaluate": false,
3179
+ "should_log": false,
3180
+ "should_save": true,
3181
+ "should_training_stop": true
3182
+ },
3183
+ "attributes": {}
3184
+ }
3185
+ },
3186
+ "total_flos": 0.0,
3187
+ "train_batch_size": 1,
3188
+ "trial_name": null,
3189
+ "trial_params": null
3190
+ }
checkpoint-22560/training_args.bin ADDED
Binary file (7.42 kB). View file