Upload folder using huggingface_hub
Browse files- added_tokens.json +3 -0
- gliner_config.json +218 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +15 -0
- spm.model +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +58 -0
- trainer_state.json +2833 -0
added_tokens.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"[MASK]": 128000
|
3 |
+
}
|
gliner_config.json
ADDED
@@ -0,0 +1,218 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"blank_prompt_rep_layer": true,
|
3 |
+
"class_token_index": -1,
|
4 |
+
"dropout": 0.35,
|
5 |
+
"embed_ent_token": true,
|
6 |
+
"encoder_config": {
|
7 |
+
"_name_or_path": "microsoft/deberta-v3-base",
|
8 |
+
"add_cross_attention": false,
|
9 |
+
"architectures": null,
|
10 |
+
"attention_probs_dropout_prob": 0.1,
|
11 |
+
"bad_words_ids": null,
|
12 |
+
"begin_suppress_tokens": null,
|
13 |
+
"bos_token_id": null,
|
14 |
+
"chunk_size_feed_forward": 0,
|
15 |
+
"cross_attention_hidden_size": null,
|
16 |
+
"decoder_start_token_id": null,
|
17 |
+
"diversity_penalty": 0.0,
|
18 |
+
"do_sample": false,
|
19 |
+
"early_stopping": false,
|
20 |
+
"encoder_no_repeat_ngram_size": 0,
|
21 |
+
"eos_token_id": null,
|
22 |
+
"exponential_decay_length_penalty": null,
|
23 |
+
"finetuning_task": null,
|
24 |
+
"forced_bos_token_id": null,
|
25 |
+
"forced_eos_token_id": null,
|
26 |
+
"hidden_act": "gelu",
|
27 |
+
"hidden_dropout_prob": 0.1,
|
28 |
+
"hidden_size": 768,
|
29 |
+
"id2label": {
|
30 |
+
"0": "LABEL_0",
|
31 |
+
"1": "LABEL_1"
|
32 |
+
},
|
33 |
+
"initializer_range": 0.02,
|
34 |
+
"intermediate_size": 3072,
|
35 |
+
"is_decoder": false,
|
36 |
+
"is_encoder_decoder": false,
|
37 |
+
"label2id": {
|
38 |
+
"LABEL_0": 0,
|
39 |
+
"LABEL_1": 1
|
40 |
+
},
|
41 |
+
"layer_norm_eps": 1e-07,
|
42 |
+
"length_penalty": 1.0,
|
43 |
+
"max_length": 20,
|
44 |
+
"max_position_embeddings": 512,
|
45 |
+
"max_relative_positions": -1,
|
46 |
+
"min_length": 0,
|
47 |
+
"model_type": "deberta-v2",
|
48 |
+
"no_repeat_ngram_size": 0,
|
49 |
+
"norm_rel_ebd": "layer_norm",
|
50 |
+
"num_attention_heads": 12,
|
51 |
+
"num_beam_groups": 1,
|
52 |
+
"num_beams": 1,
|
53 |
+
"num_hidden_layers": 12,
|
54 |
+
"num_return_sequences": 1,
|
55 |
+
"output_attentions": false,
|
56 |
+
"output_hidden_states": false,
|
57 |
+
"output_scores": false,
|
58 |
+
"pad_token_id": 0,
|
59 |
+
"pooler_dropout": 0,
|
60 |
+
"pooler_hidden_act": "gelu",
|
61 |
+
"pooler_hidden_size": 768,
|
62 |
+
"pos_att_type": [
|
63 |
+
"p2c",
|
64 |
+
"c2p"
|
65 |
+
],
|
66 |
+
"position_biased_input": false,
|
67 |
+
"position_buckets": 256,
|
68 |
+
"prefix": null,
|
69 |
+
"problem_type": null,
|
70 |
+
"pruned_heads": {},
|
71 |
+
"relative_attention": true,
|
72 |
+
"remove_invalid_values": false,
|
73 |
+
"repetition_penalty": 1.0,
|
74 |
+
"return_dict": true,
|
75 |
+
"return_dict_in_generate": false,
|
76 |
+
"sep_token_id": null,
|
77 |
+
"share_att_key": true,
|
78 |
+
"suppress_tokens": null,
|
79 |
+
"task_specific_params": null,
|
80 |
+
"temperature": 1.0,
|
81 |
+
"tf_legacy_loss": false,
|
82 |
+
"tie_encoder_decoder": false,
|
83 |
+
"tie_word_embeddings": true,
|
84 |
+
"tokenizer_class": null,
|
85 |
+
"top_k": 50,
|
86 |
+
"top_p": 1.0,
|
87 |
+
"torch_dtype": null,
|
88 |
+
"torchscript": false,
|
89 |
+
"type_vocab_size": 0,
|
90 |
+
"typical_p": 1.0,
|
91 |
+
"use_bfloat16": false,
|
92 |
+
"vocab_size": 128100
|
93 |
+
},
|
94 |
+
"ent_token": "<<ENT>>",
|
95 |
+
"entity_linking": null,
|
96 |
+
"eval_every": 4000,
|
97 |
+
"fine_tune": true,
|
98 |
+
"freeze_token_rep": false,
|
99 |
+
"fuse_layers": false,
|
100 |
+
"has_rnn": true,
|
101 |
+
"hidden_size": 768,
|
102 |
+
"label_smoothing": 0,
|
103 |
+
"labels_encoder": "BAAI/bge-small-en-v1.5",
|
104 |
+
"labels_encoder_config": {
|
105 |
+
"_name_or_path": "BAAI/bge-small-en-v1.5",
|
106 |
+
"add_cross_attention": false,
|
107 |
+
"architectures": [
|
108 |
+
"BertModel"
|
109 |
+
],
|
110 |
+
"attention_probs_dropout_prob": 0.1,
|
111 |
+
"bad_words_ids": null,
|
112 |
+
"begin_suppress_tokens": null,
|
113 |
+
"bos_token_id": null,
|
114 |
+
"chunk_size_feed_forward": 0,
|
115 |
+
"classifier_dropout": null,
|
116 |
+
"cross_attention_hidden_size": null,
|
117 |
+
"decoder_start_token_id": null,
|
118 |
+
"diversity_penalty": 0.0,
|
119 |
+
"do_sample": false,
|
120 |
+
"early_stopping": false,
|
121 |
+
"encoder_no_repeat_ngram_size": 0,
|
122 |
+
"eos_token_id": null,
|
123 |
+
"exponential_decay_length_penalty": null,
|
124 |
+
"finetuning_task": null,
|
125 |
+
"forced_bos_token_id": null,
|
126 |
+
"forced_eos_token_id": null,
|
127 |
+
"hidden_act": "gelu",
|
128 |
+
"hidden_dropout_prob": 0.1,
|
129 |
+
"hidden_size": 384,
|
130 |
+
"id2label": {
|
131 |
+
"0": "LABEL_0"
|
132 |
+
},
|
133 |
+
"initializer_range": 0.02,
|
134 |
+
"intermediate_size": 1536,
|
135 |
+
"is_decoder": false,
|
136 |
+
"is_encoder_decoder": false,
|
137 |
+
"label2id": {
|
138 |
+
"LABEL_0": 0
|
139 |
+
},
|
140 |
+
"layer_norm_eps": 1e-12,
|
141 |
+
"length_penalty": 1.0,
|
142 |
+
"max_length": 20,
|
143 |
+
"max_position_embeddings": 512,
|
144 |
+
"min_length": 0,
|
145 |
+
"model_type": "bert",
|
146 |
+
"no_repeat_ngram_size": 0,
|
147 |
+
"num_attention_heads": 12,
|
148 |
+
"num_beam_groups": 1,
|
149 |
+
"num_beams": 1,
|
150 |
+
"num_hidden_layers": 12,
|
151 |
+
"num_return_sequences": 1,
|
152 |
+
"output_attentions": false,
|
153 |
+
"output_hidden_states": false,
|
154 |
+
"output_scores": false,
|
155 |
+
"pad_token_id": 0,
|
156 |
+
"position_embedding_type": "absolute",
|
157 |
+
"prefix": null,
|
158 |
+
"problem_type": null,
|
159 |
+
"pruned_heads": {},
|
160 |
+
"remove_invalid_values": false,
|
161 |
+
"repetition_penalty": 1.0,
|
162 |
+
"return_dict": true,
|
163 |
+
"return_dict_in_generate": false,
|
164 |
+
"sep_token_id": null,
|
165 |
+
"suppress_tokens": null,
|
166 |
+
"task_specific_params": null,
|
167 |
+
"temperature": 1.0,
|
168 |
+
"tf_legacy_loss": false,
|
169 |
+
"tie_encoder_decoder": false,
|
170 |
+
"tie_word_embeddings": true,
|
171 |
+
"tokenizer_class": null,
|
172 |
+
"top_k": 50,
|
173 |
+
"top_p": 1.0,
|
174 |
+
"torch_dtype": "float32",
|
175 |
+
"torchscript": false,
|
176 |
+
"type_vocab_size": 2,
|
177 |
+
"typical_p": 1.0,
|
178 |
+
"use_bfloat16": false,
|
179 |
+
"use_cache": true,
|
180 |
+
"vocab_size": 30522
|
181 |
+
},
|
182 |
+
"log_dir": "models-base-2/",
|
183 |
+
"loss_alpha": 0.8,
|
184 |
+
"loss_gamma": 2,
|
185 |
+
"loss_reduction": "sum",
|
186 |
+
"lr_encoder": "1e-5",
|
187 |
+
"lr_others": "4e-5",
|
188 |
+
"max_grad_norm": 10.0,
|
189 |
+
"max_len": 768,
|
190 |
+
"max_neg_type_ratio": 5,
|
191 |
+
"max_types": 256,
|
192 |
+
"max_width": 12,
|
193 |
+
"model_name": "microsoft/deberta-v3-base",
|
194 |
+
"model_type": "gliner",
|
195 |
+
"name": "span level gliner for entity linking",
|
196 |
+
"num_post_fusion_layers": 1,
|
197 |
+
"num_steps": 250000,
|
198 |
+
"post_fusion_schema": "",
|
199 |
+
"prev_path": null,
|
200 |
+
"random_drop": true,
|
201 |
+
"root_dir": "gliner_logs",
|
202 |
+
"save_total_limit": 20,
|
203 |
+
"scheduler_type": "cosine",
|
204 |
+
"sep_token": "<<SEP>>",
|
205 |
+
"shuffle_types": true,
|
206 |
+
"size_sup": -1,
|
207 |
+
"span_mode": "markerV0",
|
208 |
+
"subtoken_pooling": "first",
|
209 |
+
"train_batch_size": 4,
|
210 |
+
"train_data": "final_dataset_1m.json",
|
211 |
+
"transformers_version": "4.45.2",
|
212 |
+
"val_data_dir": null,
|
213 |
+
"vocab_size": -1,
|
214 |
+
"warmup_ratio": 0.1,
|
215 |
+
"weight_decay_encoder": 0.01,
|
216 |
+
"weight_decay_other": 0.02,
|
217 |
+
"words_splitter_type": "whitespace"
|
218 |
+
}
|
optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d2e4187d893f68525c247a52723c2022fbc4ef6d889834dd9b19895213df0b81
|
3 |
+
size 1871270621
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:546033b12df9b34299187cd7df70f84197ed5cb18ec12e7d0521595115a85429
|
3 |
+
size 950390010
|
rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0fde4226575f68849adb4d7a5b74752c1734bd95e78eb8e0a10cb7b13642a66f
|
3 |
+
size 14244
|
scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4b5ae7b72b4dc79eecb0822bc5cb0128951fc297710abc018e5a6297ef6e2514
|
3 |
+
size 1064
|
special_tokens_map.json
ADDED
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "[CLS]",
|
3 |
+
"cls_token": "[CLS]",
|
4 |
+
"eos_token": "[SEP]",
|
5 |
+
"mask_token": "[MASK]",
|
6 |
+
"pad_token": "[PAD]",
|
7 |
+
"sep_token": "[SEP]",
|
8 |
+
"unk_token": {
|
9 |
+
"content": "[UNK]",
|
10 |
+
"lstrip": false,
|
11 |
+
"normalized": true,
|
12 |
+
"rstrip": false,
|
13 |
+
"single_word": false
|
14 |
+
}
|
15 |
+
}
|
spm.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c679fbf93643d19aab7ee10c0b99e460bdbc02fedf34b92b05af343b4af586fd
|
3 |
+
size 2464616
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"added_tokens_decoder": {
|
3 |
+
"0": {
|
4 |
+
"content": "[PAD]",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false,
|
9 |
+
"special": true
|
10 |
+
},
|
11 |
+
"1": {
|
12 |
+
"content": "[CLS]",
|
13 |
+
"lstrip": false,
|
14 |
+
"normalized": false,
|
15 |
+
"rstrip": false,
|
16 |
+
"single_word": false,
|
17 |
+
"special": true
|
18 |
+
},
|
19 |
+
"2": {
|
20 |
+
"content": "[SEP]",
|
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": true,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false,
|
33 |
+
"special": true
|
34 |
+
},
|
35 |
+
"128000": {
|
36 |
+
"content": "[MASK]",
|
37 |
+
"lstrip": false,
|
38 |
+
"normalized": false,
|
39 |
+
"rstrip": false,
|
40 |
+
"single_word": false,
|
41 |
+
"special": true
|
42 |
+
}
|
43 |
+
},
|
44 |
+
"bos_token": "[CLS]",
|
45 |
+
"clean_up_tokenization_spaces": false,
|
46 |
+
"cls_token": "[CLS]",
|
47 |
+
"do_lower_case": false,
|
48 |
+
"eos_token": "[SEP]",
|
49 |
+
"mask_token": "[MASK]",
|
50 |
+
"model_max_length": 1000000000000000019884624838656,
|
51 |
+
"pad_token": "[PAD]",
|
52 |
+
"sep_token": "[SEP]",
|
53 |
+
"sp_model_kwargs": {},
|
54 |
+
"split_by_punct": false,
|
55 |
+
"tokenizer_class": "DebertaV2Tokenizer",
|
56 |
+
"unk_token": "[UNK]",
|
57 |
+
"vocab_type": "spm"
|
58 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,2833 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.016160506147052525,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 4000,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 4.040126536763131e-05,
|
13 |
+
"grad_norm": 34776792.0,
|
14 |
+
"learning_rate": 1.6e-08,
|
15 |
+
"loss": 1112513.0,
|
16 |
+
"step": 10
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 8.080253073526263e-05,
|
20 |
+
"grad_norm": 8441178.0,
|
21 |
+
"learning_rate": 3.2e-08,
|
22 |
+
"loss": 1301709.2,
|
23 |
+
"step": 20
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.00012120379610289395,
|
27 |
+
"grad_norm": 12542169.0,
|
28 |
+
"learning_rate": 4.8e-08,
|
29 |
+
"loss": 1175913.9,
|
30 |
+
"step": 30
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.00016160506147052525,
|
34 |
+
"grad_norm": 42386256.0,
|
35 |
+
"learning_rate": 6.4e-08,
|
36 |
+
"loss": 1313873.9,
|
37 |
+
"step": 40
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.00020200632683815657,
|
41 |
+
"grad_norm": 33162978.0,
|
42 |
+
"learning_rate": 8e-08,
|
43 |
+
"loss": 1142017.2,
|
44 |
+
"step": 50
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.0002424075922057879,
|
48 |
+
"grad_norm": 52537692.0,
|
49 |
+
"learning_rate": 9.6e-08,
|
50 |
+
"loss": 1040597.1,
|
51 |
+
"step": 60
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.0002828088575734192,
|
55 |
+
"grad_norm": 17375308.0,
|
56 |
+
"learning_rate": 1.1200000000000001e-07,
|
57 |
+
"loss": 1494345.4,
|
58 |
+
"step": 70
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.0003232101229410505,
|
62 |
+
"grad_norm": 95491040.0,
|
63 |
+
"learning_rate": 1.28e-07,
|
64 |
+
"loss": 1404674.8,
|
65 |
+
"step": 80
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.0003636113883086818,
|
69 |
+
"grad_norm": 13118159.0,
|
70 |
+
"learning_rate": 1.4400000000000002e-07,
|
71 |
+
"loss": 1204737.6,
|
72 |
+
"step": 90
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.00040401265367631315,
|
76 |
+
"grad_norm": 18581922.0,
|
77 |
+
"learning_rate": 1.6e-07,
|
78 |
+
"loss": 783122.2,
|
79 |
+
"step": 100
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.00044441391904394446,
|
83 |
+
"grad_norm": 15600268.0,
|
84 |
+
"learning_rate": 1.7600000000000001e-07,
|
85 |
+
"loss": 1342865.9,
|
86 |
+
"step": 110
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.0004848151844115758,
|
90 |
+
"grad_norm": 5757326.5,
|
91 |
+
"learning_rate": 1.92e-07,
|
92 |
+
"loss": 998477.5,
|
93 |
+
"step": 120
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.000525216449779207,
|
97 |
+
"grad_norm": 37603152.0,
|
98 |
+
"learning_rate": 2.08e-07,
|
99 |
+
"loss": 1113327.2,
|
100 |
+
"step": 130
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.0005656177151468384,
|
104 |
+
"grad_norm": 3987340.5,
|
105 |
+
"learning_rate": 2.2400000000000002e-07,
|
106 |
+
"loss": 815339.55,
|
107 |
+
"step": 140
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.0006060189805144697,
|
111 |
+
"grad_norm": 17012036.0,
|
112 |
+
"learning_rate": 2.4000000000000003e-07,
|
113 |
+
"loss": 799237.55,
|
114 |
+
"step": 150
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.000646420245882101,
|
118 |
+
"grad_norm": 15123844.0,
|
119 |
+
"learning_rate": 2.56e-07,
|
120 |
+
"loss": 795830.9,
|
121 |
+
"step": 160
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.0006868215112497323,
|
125 |
+
"grad_norm": 13699358.0,
|
126 |
+
"learning_rate": 2.72e-07,
|
127 |
+
"loss": 675074.05,
|
128 |
+
"step": 170
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.0007272227766173637,
|
132 |
+
"grad_norm": 5762779.5,
|
133 |
+
"learning_rate": 2.8800000000000004e-07,
|
134 |
+
"loss": 397814.875,
|
135 |
+
"step": 180
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.000767624041984995,
|
139 |
+
"grad_norm": 3545711.25,
|
140 |
+
"learning_rate": 3.04e-07,
|
141 |
+
"loss": 278445.225,
|
142 |
+
"step": 190
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.0008080253073526263,
|
146 |
+
"grad_norm": 22483670.0,
|
147 |
+
"learning_rate": 3.2e-07,
|
148 |
+
"loss": 350395.65,
|
149 |
+
"step": 200
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.0008484265727202576,
|
153 |
+
"grad_norm": 3111300.75,
|
154 |
+
"learning_rate": 3.36e-07,
|
155 |
+
"loss": 215067.95,
|
156 |
+
"step": 210
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.0008888278380878889,
|
160 |
+
"grad_norm": 4985113.0,
|
161 |
+
"learning_rate": 3.5200000000000003e-07,
|
162 |
+
"loss": 160062.15,
|
163 |
+
"step": 220
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.0009292291034555202,
|
167 |
+
"grad_norm": 10360142.0,
|
168 |
+
"learning_rate": 3.68e-07,
|
169 |
+
"loss": 177085.5,
|
170 |
+
"step": 230
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.0009696303688231516,
|
174 |
+
"grad_norm": 4212897.0,
|
175 |
+
"learning_rate": 3.84e-07,
|
176 |
+
"loss": 156687.3375,
|
177 |
+
"step": 240
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.0010100316341907828,
|
181 |
+
"grad_norm": 2492394.75,
|
182 |
+
"learning_rate": 4.0000000000000003e-07,
|
183 |
+
"loss": 218777.15,
|
184 |
+
"step": 250
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.001050432899558414,
|
188 |
+
"grad_norm": 1271087.375,
|
189 |
+
"learning_rate": 4.16e-07,
|
190 |
+
"loss": 74811.1875,
|
191 |
+
"step": 260
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.0010908341649260454,
|
195 |
+
"grad_norm": 11904261.0,
|
196 |
+
"learning_rate": 4.3200000000000006e-07,
|
197 |
+
"loss": 78403.6125,
|
198 |
+
"step": 270
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.0011312354302936767,
|
202 |
+
"grad_norm": 1911185.875,
|
203 |
+
"learning_rate": 4.4800000000000004e-07,
|
204 |
+
"loss": 112190.75,
|
205 |
+
"step": 280
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.001171636695661308,
|
209 |
+
"grad_norm": 1619918.75,
|
210 |
+
"learning_rate": 4.64e-07,
|
211 |
+
"loss": 33613.9688,
|
212 |
+
"step": 290
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.0012120379610289394,
|
216 |
+
"grad_norm": 1365019.625,
|
217 |
+
"learning_rate": 4.800000000000001e-07,
|
218 |
+
"loss": 32836.45,
|
219 |
+
"step": 300
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.0012524392263965707,
|
223 |
+
"grad_norm": 1140846.75,
|
224 |
+
"learning_rate": 4.96e-07,
|
225 |
+
"loss": 38892.625,
|
226 |
+
"step": 310
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.001292840491764202,
|
230 |
+
"grad_norm": 950318.8125,
|
231 |
+
"learning_rate": 5.12e-07,
|
232 |
+
"loss": 43860.9844,
|
233 |
+
"step": 320
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.0013332417571318333,
|
237 |
+
"grad_norm": 191924.40625,
|
238 |
+
"learning_rate": 5.280000000000001e-07,
|
239 |
+
"loss": 19675.0312,
|
240 |
+
"step": 330
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.0013736430224994647,
|
244 |
+
"grad_norm": 354602.96875,
|
245 |
+
"learning_rate": 5.44e-07,
|
246 |
+
"loss": 17043.5047,
|
247 |
+
"step": 340
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.001414044287867096,
|
251 |
+
"grad_norm": 64334.99609375,
|
252 |
+
"learning_rate": 5.6e-07,
|
253 |
+
"loss": 6989.507,
|
254 |
+
"step": 350
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.0014544455532347273,
|
258 |
+
"grad_norm": 29138.234375,
|
259 |
+
"learning_rate": 5.760000000000001e-07,
|
260 |
+
"loss": 2483.216,
|
261 |
+
"step": 360
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.0014948468186023586,
|
265 |
+
"grad_norm": 4881.9619140625,
|
266 |
+
"learning_rate": 5.920000000000001e-07,
|
267 |
+
"loss": 4704.4551,
|
268 |
+
"step": 370
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.00153524808396999,
|
272 |
+
"grad_norm": 73014.8671875,
|
273 |
+
"learning_rate": 6.08e-07,
|
274 |
+
"loss": 1492.4164,
|
275 |
+
"step": 380
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.0015756493493376213,
|
279 |
+
"grad_norm": 86664.8203125,
|
280 |
+
"learning_rate": 6.24e-07,
|
281 |
+
"loss": 1407.5383,
|
282 |
+
"step": 390
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.0016160506147052526,
|
286 |
+
"grad_norm": 91074.265625,
|
287 |
+
"learning_rate": 6.4e-07,
|
288 |
+
"loss": 932.89,
|
289 |
+
"step": 400
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.001656451880072884,
|
293 |
+
"grad_norm": 5759.06396484375,
|
294 |
+
"learning_rate": 6.560000000000002e-07,
|
295 |
+
"loss": 856.515,
|
296 |
+
"step": 410
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.0016968531454405152,
|
300 |
+
"grad_norm": 1454.9195556640625,
|
301 |
+
"learning_rate": 6.72e-07,
|
302 |
+
"loss": 419.0969,
|
303 |
+
"step": 420
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.0017372544108081465,
|
307 |
+
"grad_norm": 673.9178466796875,
|
308 |
+
"learning_rate": 6.88e-07,
|
309 |
+
"loss": 621.3434,
|
310 |
+
"step": 430
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.0017776556761757779,
|
314 |
+
"grad_norm": 1015.685546875,
|
315 |
+
"learning_rate": 7.040000000000001e-07,
|
316 |
+
"loss": 498.2878,
|
317 |
+
"step": 440
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.0018180569415434092,
|
321 |
+
"grad_norm": 6716.2080078125,
|
322 |
+
"learning_rate": 7.2e-07,
|
323 |
+
"loss": 556.3153,
|
324 |
+
"step": 450
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.0018584582069110405,
|
328 |
+
"grad_norm": 1468.1954345703125,
|
329 |
+
"learning_rate": 7.36e-07,
|
330 |
+
"loss": 488.196,
|
331 |
+
"step": 460
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.0018988594722786718,
|
335 |
+
"grad_norm": 5355.9794921875,
|
336 |
+
"learning_rate": 7.520000000000001e-07,
|
337 |
+
"loss": 553.4161,
|
338 |
+
"step": 470
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.0019392607376463031,
|
342 |
+
"grad_norm": 1276.7708740234375,
|
343 |
+
"learning_rate": 7.68e-07,
|
344 |
+
"loss": 575.7385,
|
345 |
+
"step": 480
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.0019796620030139342,
|
349 |
+
"grad_norm": 1997.7783203125,
|
350 |
+
"learning_rate": 7.84e-07,
|
351 |
+
"loss": 449.5601,
|
352 |
+
"step": 490
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.0020200632683815656,
|
356 |
+
"grad_norm": 1234.7044677734375,
|
357 |
+
"learning_rate": 8.000000000000001e-07,
|
358 |
+
"loss": 454.7207,
|
359 |
+
"step": 500
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.002060464533749197,
|
363 |
+
"grad_norm": 1192.5169677734375,
|
364 |
+
"learning_rate": 8.160000000000001e-07,
|
365 |
+
"loss": 459.7037,
|
366 |
+
"step": 510
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.002100865799116828,
|
370 |
+
"grad_norm": 1192.2779541015625,
|
371 |
+
"learning_rate": 8.32e-07,
|
372 |
+
"loss": 406.1772,
|
373 |
+
"step": 520
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.0021412670644844595,
|
377 |
+
"grad_norm": 1168.502685546875,
|
378 |
+
"learning_rate": 8.480000000000001e-07,
|
379 |
+
"loss": 520.4339,
|
380 |
+
"step": 530
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.002181668329852091,
|
384 |
+
"grad_norm": 1580.9521484375,
|
385 |
+
"learning_rate": 8.640000000000001e-07,
|
386 |
+
"loss": 282.541,
|
387 |
+
"step": 540
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.002222069595219722,
|
391 |
+
"grad_norm": 12831.9287109375,
|
392 |
+
"learning_rate": 8.8e-07,
|
393 |
+
"loss": 531.3956,
|
394 |
+
"step": 550
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.0022624708605873535,
|
398 |
+
"grad_norm": 6975.4140625,
|
399 |
+
"learning_rate": 8.960000000000001e-07,
|
400 |
+
"loss": 377.4974,
|
401 |
+
"step": 560
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.002302872125954985,
|
405 |
+
"grad_norm": 3642.0546875,
|
406 |
+
"learning_rate": 9.120000000000001e-07,
|
407 |
+
"loss": 325.7552,
|
408 |
+
"step": 570
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.002343273391322616,
|
412 |
+
"grad_norm": 1394.17822265625,
|
413 |
+
"learning_rate": 9.28e-07,
|
414 |
+
"loss": 450.4489,
|
415 |
+
"step": 580
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.0023836746566902474,
|
419 |
+
"grad_norm": 6102.92626953125,
|
420 |
+
"learning_rate": 9.440000000000001e-07,
|
421 |
+
"loss": 581.4113,
|
422 |
+
"step": 590
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.0024240759220578788,
|
426 |
+
"grad_norm": 942.1673583984375,
|
427 |
+
"learning_rate": 9.600000000000001e-07,
|
428 |
+
"loss": 327.3541,
|
429 |
+
"step": 600
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.00246447718742551,
|
433 |
+
"grad_norm": 7506.9111328125,
|
434 |
+
"learning_rate": 9.76e-07,
|
435 |
+
"loss": 562.9349,
|
436 |
+
"step": 610
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.0025048784527931414,
|
440 |
+
"grad_norm": 828.693603515625,
|
441 |
+
"learning_rate": 9.92e-07,
|
442 |
+
"loss": 480.7408,
|
443 |
+
"step": 620
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.0025452797181607727,
|
447 |
+
"grad_norm": 1101.0645751953125,
|
448 |
+
"learning_rate": 1.0080000000000001e-06,
|
449 |
+
"loss": 510.9642,
|
450 |
+
"step": 630
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 0.002585680983528404,
|
454 |
+
"grad_norm": 4109.21728515625,
|
455 |
+
"learning_rate": 1.024e-06,
|
456 |
+
"loss": 380.4708,
|
457 |
+
"step": 640
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.0026260822488960354,
|
461 |
+
"grad_norm": 1035.9979248046875,
|
462 |
+
"learning_rate": 1.04e-06,
|
463 |
+
"loss": 438.6869,
|
464 |
+
"step": 650
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.0026664835142636667,
|
468 |
+
"grad_norm": 857.4368896484375,
|
469 |
+
"learning_rate": 1.0560000000000001e-06,
|
470 |
+
"loss": 333.5929,
|
471 |
+
"step": 660
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.002706884779631298,
|
475 |
+
"grad_norm": 861.3447875976562,
|
476 |
+
"learning_rate": 1.072e-06,
|
477 |
+
"loss": 369.1007,
|
478 |
+
"step": 670
|
479 |
+
},
|
480 |
+
{
|
481 |
+
"epoch": 0.0027472860449989293,
|
482 |
+
"grad_norm": 1937.781005859375,
|
483 |
+
"learning_rate": 1.088e-06,
|
484 |
+
"loss": 467.2207,
|
485 |
+
"step": 680
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 0.0027876873103665606,
|
489 |
+
"grad_norm": 11980.4130859375,
|
490 |
+
"learning_rate": 1.1040000000000001e-06,
|
491 |
+
"loss": 461.9273,
|
492 |
+
"step": 690
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 0.002828088575734192,
|
496 |
+
"grad_norm": 1039.8837890625,
|
497 |
+
"learning_rate": 1.12e-06,
|
498 |
+
"loss": 376.5904,
|
499 |
+
"step": 700
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 0.0028684898411018233,
|
503 |
+
"grad_norm": 2001.8681640625,
|
504 |
+
"learning_rate": 1.1360000000000002e-06,
|
505 |
+
"loss": 397.0597,
|
506 |
+
"step": 710
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 0.0029088911064694546,
|
510 |
+
"grad_norm": 4143.15087890625,
|
511 |
+
"learning_rate": 1.1520000000000002e-06,
|
512 |
+
"loss": 524.1538,
|
513 |
+
"step": 720
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 0.002949292371837086,
|
517 |
+
"grad_norm": 1036.8919677734375,
|
518 |
+
"learning_rate": 1.168e-06,
|
519 |
+
"loss": 367.9916,
|
520 |
+
"step": 730
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 0.0029896936372047172,
|
524 |
+
"grad_norm": 8330.5068359375,
|
525 |
+
"learning_rate": 1.1840000000000002e-06,
|
526 |
+
"loss": 394.5002,
|
527 |
+
"step": 740
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.0030300949025723486,
|
531 |
+
"grad_norm": 2307.091064453125,
|
532 |
+
"learning_rate": 1.2000000000000002e-06,
|
533 |
+
"loss": 295.3556,
|
534 |
+
"step": 750
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 0.00307049616793998,
|
538 |
+
"grad_norm": 1893.0789794921875,
|
539 |
+
"learning_rate": 1.216e-06,
|
540 |
+
"loss": 265.5479,
|
541 |
+
"step": 760
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 0.003110897433307611,
|
545 |
+
"grad_norm": 2477.446533203125,
|
546 |
+
"learning_rate": 1.2320000000000002e-06,
|
547 |
+
"loss": 534.005,
|
548 |
+
"step": 770
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 0.0031512986986752425,
|
552 |
+
"grad_norm": 916.7561645507812,
|
553 |
+
"learning_rate": 1.248e-06,
|
554 |
+
"loss": 334.4991,
|
555 |
+
"step": 780
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.003191699964042874,
|
559 |
+
"grad_norm": 13106.6064453125,
|
560 |
+
"learning_rate": 1.2640000000000003e-06,
|
561 |
+
"loss": 464.4099,
|
562 |
+
"step": 790
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 0.003232101229410505,
|
566 |
+
"grad_norm": 1100.711669921875,
|
567 |
+
"learning_rate": 1.28e-06,
|
568 |
+
"loss": 302.2139,
|
569 |
+
"step": 800
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.0032725024947781365,
|
573 |
+
"grad_norm": 2172.443603515625,
|
574 |
+
"learning_rate": 1.296e-06,
|
575 |
+
"loss": 389.6274,
|
576 |
+
"step": 810
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.003312903760145768,
|
580 |
+
"grad_norm": 1690.6312255859375,
|
581 |
+
"learning_rate": 1.3120000000000003e-06,
|
582 |
+
"loss": 361.3152,
|
583 |
+
"step": 820
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 0.003353305025513399,
|
587 |
+
"grad_norm": 924.0753173828125,
|
588 |
+
"learning_rate": 1.328e-06,
|
589 |
+
"loss": 247.9785,
|
590 |
+
"step": 830
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 0.0033937062908810304,
|
594 |
+
"grad_norm": 1367.6988525390625,
|
595 |
+
"learning_rate": 1.344e-06,
|
596 |
+
"loss": 320.1396,
|
597 |
+
"step": 840
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.0034341075562486618,
|
601 |
+
"grad_norm": 731.0768432617188,
|
602 |
+
"learning_rate": 1.3600000000000001e-06,
|
603 |
+
"loss": 394.588,
|
604 |
+
"step": 850
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 0.003474508821616293,
|
608 |
+
"grad_norm": 788.498046875,
|
609 |
+
"learning_rate": 1.376e-06,
|
610 |
+
"loss": 430.4721,
|
611 |
+
"step": 860
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.0035149100869839244,
|
615 |
+
"grad_norm": 989.5845336914062,
|
616 |
+
"learning_rate": 1.392e-06,
|
617 |
+
"loss": 343.7414,
|
618 |
+
"step": 870
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 0.0035553113523515557,
|
622 |
+
"grad_norm": 661.3421020507812,
|
623 |
+
"learning_rate": 1.4080000000000001e-06,
|
624 |
+
"loss": 382.3532,
|
625 |
+
"step": 880
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 0.003595712617719187,
|
629 |
+
"grad_norm": 831.7238159179688,
|
630 |
+
"learning_rate": 1.424e-06,
|
631 |
+
"loss": 442.5967,
|
632 |
+
"step": 890
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 0.0036361138830868184,
|
636 |
+
"grad_norm": 1258.17236328125,
|
637 |
+
"learning_rate": 1.44e-06,
|
638 |
+
"loss": 329.9442,
|
639 |
+
"step": 900
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 0.0036765151484544497,
|
643 |
+
"grad_norm": 1166.2576904296875,
|
644 |
+
"learning_rate": 1.4560000000000001e-06,
|
645 |
+
"loss": 332.621,
|
646 |
+
"step": 910
|
647 |
+
},
|
648 |
+
{
|
649 |
+
"epoch": 0.003716916413822081,
|
650 |
+
"grad_norm": 877.9427490234375,
|
651 |
+
"learning_rate": 1.472e-06,
|
652 |
+
"loss": 464.8878,
|
653 |
+
"step": 920
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 0.0037573176791897123,
|
657 |
+
"grad_norm": 1044.3133544921875,
|
658 |
+
"learning_rate": 1.488e-06,
|
659 |
+
"loss": 292.7577,
|
660 |
+
"step": 930
|
661 |
+
},
|
662 |
+
{
|
663 |
+
"epoch": 0.0037977189445573436,
|
664 |
+
"grad_norm": 1089.46728515625,
|
665 |
+
"learning_rate": 1.5040000000000001e-06,
|
666 |
+
"loss": 460.0909,
|
667 |
+
"step": 940
|
668 |
+
},
|
669 |
+
{
|
670 |
+
"epoch": 0.003838120209924975,
|
671 |
+
"grad_norm": 797.8550415039062,
|
672 |
+
"learning_rate": 1.52e-06,
|
673 |
+
"loss": 346.0366,
|
674 |
+
"step": 950
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 0.0038785214752926063,
|
678 |
+
"grad_norm": 798.5408325195312,
|
679 |
+
"learning_rate": 1.536e-06,
|
680 |
+
"loss": 310.8797,
|
681 |
+
"step": 960
|
682 |
+
},
|
683 |
+
{
|
684 |
+
"epoch": 0.003918922740660237,
|
685 |
+
"grad_norm": 2262.073974609375,
|
686 |
+
"learning_rate": 1.5520000000000001e-06,
|
687 |
+
"loss": 370.4966,
|
688 |
+
"step": 970
|
689 |
+
},
|
690 |
+
{
|
691 |
+
"epoch": 0.0039593240060278685,
|
692 |
+
"grad_norm": 2735.94482421875,
|
693 |
+
"learning_rate": 1.568e-06,
|
694 |
+
"loss": 323.6035,
|
695 |
+
"step": 980
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 0.0039997252713955,
|
699 |
+
"grad_norm": 646.0745849609375,
|
700 |
+
"learning_rate": 1.5840000000000002e-06,
|
701 |
+
"loss": 339.8437,
|
702 |
+
"step": 990
|
703 |
+
},
|
704 |
+
{
|
705 |
+
"epoch": 0.004040126536763131,
|
706 |
+
"grad_norm": 1748.4075927734375,
|
707 |
+
"learning_rate": 1.6000000000000001e-06,
|
708 |
+
"loss": 317.4156,
|
709 |
+
"step": 1000
|
710 |
+
},
|
711 |
+
{
|
712 |
+
"epoch": 0.0040805278021307624,
|
713 |
+
"grad_norm": 780.6202392578125,
|
714 |
+
"learning_rate": 1.616e-06,
|
715 |
+
"loss": 296.7254,
|
716 |
+
"step": 1010
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 0.004120929067498394,
|
720 |
+
"grad_norm": 1253.2996826171875,
|
721 |
+
"learning_rate": 1.6320000000000002e-06,
|
722 |
+
"loss": 384.1156,
|
723 |
+
"step": 1020
|
724 |
+
},
|
725 |
+
{
|
726 |
+
"epoch": 0.004161330332866025,
|
727 |
+
"grad_norm": 1055.1898193359375,
|
728 |
+
"learning_rate": 1.6480000000000001e-06,
|
729 |
+
"loss": 397.7781,
|
730 |
+
"step": 1030
|
731 |
+
},
|
732 |
+
{
|
733 |
+
"epoch": 0.004201731598233656,
|
734 |
+
"grad_norm": 803.1446533203125,
|
735 |
+
"learning_rate": 1.664e-06,
|
736 |
+
"loss": 359.3899,
|
737 |
+
"step": 1040
|
738 |
+
},
|
739 |
+
{
|
740 |
+
"epoch": 0.004242132863601288,
|
741 |
+
"grad_norm": 798.8219604492188,
|
742 |
+
"learning_rate": 1.6800000000000002e-06,
|
743 |
+
"loss": 381.0167,
|
744 |
+
"step": 1050
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 0.004282534128968919,
|
748 |
+
"grad_norm": 3304.813232421875,
|
749 |
+
"learning_rate": 1.6960000000000002e-06,
|
750 |
+
"loss": 391.1282,
|
751 |
+
"step": 1060
|
752 |
+
},
|
753 |
+
{
|
754 |
+
"epoch": 0.00432293539433655,
|
755 |
+
"grad_norm": 2308.608154296875,
|
756 |
+
"learning_rate": 1.712e-06,
|
757 |
+
"loss": 453.3599,
|
758 |
+
"step": 1070
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 0.004363336659704182,
|
762 |
+
"grad_norm": 744.8621215820312,
|
763 |
+
"learning_rate": 1.7280000000000002e-06,
|
764 |
+
"loss": 306.6811,
|
765 |
+
"step": 1080
|
766 |
+
},
|
767 |
+
{
|
768 |
+
"epoch": 0.004403737925071813,
|
769 |
+
"grad_norm": 5821.24267578125,
|
770 |
+
"learning_rate": 1.7440000000000002e-06,
|
771 |
+
"loss": 608.5389,
|
772 |
+
"step": 1090
|
773 |
+
},
|
774 |
+
{
|
775 |
+
"epoch": 0.004444139190439444,
|
776 |
+
"grad_norm": 865.3998413085938,
|
777 |
+
"learning_rate": 1.76e-06,
|
778 |
+
"loss": 242.1831,
|
779 |
+
"step": 1100
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 0.004484540455807076,
|
783 |
+
"grad_norm": 955.5283203125,
|
784 |
+
"learning_rate": 1.7760000000000002e-06,
|
785 |
+
"loss": 406.3821,
|
786 |
+
"step": 1110
|
787 |
+
},
|
788 |
+
{
|
789 |
+
"epoch": 0.004524941721174707,
|
790 |
+
"grad_norm": 1257.974853515625,
|
791 |
+
"learning_rate": 1.7920000000000002e-06,
|
792 |
+
"loss": 337.8827,
|
793 |
+
"step": 1120
|
794 |
+
},
|
795 |
+
{
|
796 |
+
"epoch": 0.004565342986542338,
|
797 |
+
"grad_norm": 848.3350830078125,
|
798 |
+
"learning_rate": 1.808e-06,
|
799 |
+
"loss": 383.7443,
|
800 |
+
"step": 1130
|
801 |
+
},
|
802 |
+
{
|
803 |
+
"epoch": 0.00460574425190997,
|
804 |
+
"grad_norm": 2609.029541015625,
|
805 |
+
"learning_rate": 1.8240000000000002e-06,
|
806 |
+
"loss": 394.2932,
|
807 |
+
"step": 1140
|
808 |
+
},
|
809 |
+
{
|
810 |
+
"epoch": 0.004646145517277601,
|
811 |
+
"grad_norm": 1440.8753662109375,
|
812 |
+
"learning_rate": 1.8400000000000002e-06,
|
813 |
+
"loss": 433.8044,
|
814 |
+
"step": 1150
|
815 |
+
},
|
816 |
+
{
|
817 |
+
"epoch": 0.004686546782645232,
|
818 |
+
"grad_norm": 1531.273681640625,
|
819 |
+
"learning_rate": 1.856e-06,
|
820 |
+
"loss": 424.2002,
|
821 |
+
"step": 1160
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 0.0047269480480128636,
|
825 |
+
"grad_norm": 2845.63623046875,
|
826 |
+
"learning_rate": 1.8720000000000002e-06,
|
827 |
+
"loss": 343.2191,
|
828 |
+
"step": 1170
|
829 |
+
},
|
830 |
+
{
|
831 |
+
"epoch": 0.004767349313380495,
|
832 |
+
"grad_norm": 679.157470703125,
|
833 |
+
"learning_rate": 1.8880000000000002e-06,
|
834 |
+
"loss": 327.4853,
|
835 |
+
"step": 1180
|
836 |
+
},
|
837 |
+
{
|
838 |
+
"epoch": 0.004807750578748126,
|
839 |
+
"grad_norm": 1685.8907470703125,
|
840 |
+
"learning_rate": 1.9040000000000003e-06,
|
841 |
+
"loss": 343.0018,
|
842 |
+
"step": 1190
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 0.0048481518441157575,
|
846 |
+
"grad_norm": 961.8662719726562,
|
847 |
+
"learning_rate": 1.9200000000000003e-06,
|
848 |
+
"loss": 345.5244,
|
849 |
+
"step": 1200
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 0.004888553109483389,
|
853 |
+
"grad_norm": 1039.9063720703125,
|
854 |
+
"learning_rate": 1.936e-06,
|
855 |
+
"loss": 286.9379,
|
856 |
+
"step": 1210
|
857 |
+
},
|
858 |
+
{
|
859 |
+
"epoch": 0.00492895437485102,
|
860 |
+
"grad_norm": 915.8055419921875,
|
861 |
+
"learning_rate": 1.952e-06,
|
862 |
+
"loss": 291.1856,
|
863 |
+
"step": 1220
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 0.0049693556402186515,
|
867 |
+
"grad_norm": 2453.09716796875,
|
868 |
+
"learning_rate": 1.968e-06,
|
869 |
+
"loss": 357.5581,
|
870 |
+
"step": 1230
|
871 |
+
},
|
872 |
+
{
|
873 |
+
"epoch": 0.005009756905586283,
|
874 |
+
"grad_norm": 1281.601318359375,
|
875 |
+
"learning_rate": 1.984e-06,
|
876 |
+
"loss": 370.0242,
|
877 |
+
"step": 1240
|
878 |
+
},
|
879 |
+
{
|
880 |
+
"epoch": 0.005050158170953914,
|
881 |
+
"grad_norm": 1927.247802734375,
|
882 |
+
"learning_rate": 2.0000000000000003e-06,
|
883 |
+
"loss": 420.42,
|
884 |
+
"step": 1250
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 0.0050905594363215454,
|
888 |
+
"grad_norm": 1159.7850341796875,
|
889 |
+
"learning_rate": 2.0160000000000003e-06,
|
890 |
+
"loss": 381.0742,
|
891 |
+
"step": 1260
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 0.005130960701689177,
|
895 |
+
"grad_norm": 1102.625244140625,
|
896 |
+
"learning_rate": 2.032e-06,
|
897 |
+
"loss": 441.1049,
|
898 |
+
"step": 1270
|
899 |
+
},
|
900 |
+
{
|
901 |
+
"epoch": 0.005171361967056808,
|
902 |
+
"grad_norm": 1439.484130859375,
|
903 |
+
"learning_rate": 2.048e-06,
|
904 |
+
"loss": 277.274,
|
905 |
+
"step": 1280
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 0.005211763232424439,
|
909 |
+
"grad_norm": 5610.380859375,
|
910 |
+
"learning_rate": 2.064e-06,
|
911 |
+
"loss": 421.7488,
|
912 |
+
"step": 1290
|
913 |
+
},
|
914 |
+
{
|
915 |
+
"epoch": 0.005252164497792071,
|
916 |
+
"grad_norm": 982.2139892578125,
|
917 |
+
"learning_rate": 2.08e-06,
|
918 |
+
"loss": 340.8586,
|
919 |
+
"step": 1300
|
920 |
+
},
|
921 |
+
{
|
922 |
+
"epoch": 0.005292565763159702,
|
923 |
+
"grad_norm": 791.9103393554688,
|
924 |
+
"learning_rate": 2.0960000000000003e-06,
|
925 |
+
"loss": 363.6415,
|
926 |
+
"step": 1310
|
927 |
+
},
|
928 |
+
{
|
929 |
+
"epoch": 0.005332967028527333,
|
930 |
+
"grad_norm": 0.0,
|
931 |
+
"learning_rate": 2.1120000000000003e-06,
|
932 |
+
"loss": 443.9375,
|
933 |
+
"step": 1320
|
934 |
+
},
|
935 |
+
{
|
936 |
+
"epoch": 0.005373368293894965,
|
937 |
+
"grad_norm": 1269.4005126953125,
|
938 |
+
"learning_rate": 2.128e-06,
|
939 |
+
"loss": 352.3543,
|
940 |
+
"step": 1330
|
941 |
+
},
|
942 |
+
{
|
943 |
+
"epoch": 0.005413769559262596,
|
944 |
+
"grad_norm": 1225.1302490234375,
|
945 |
+
"learning_rate": 2.144e-06,
|
946 |
+
"loss": 357.3038,
|
947 |
+
"step": 1340
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 0.005454170824630227,
|
951 |
+
"grad_norm": 827.63720703125,
|
952 |
+
"learning_rate": 2.16e-06,
|
953 |
+
"loss": 303.4535,
|
954 |
+
"step": 1350
|
955 |
+
},
|
956 |
+
{
|
957 |
+
"epoch": 0.005494572089997859,
|
958 |
+
"grad_norm": 848.7166137695312,
|
959 |
+
"learning_rate": 2.176e-06,
|
960 |
+
"loss": 268.09,
|
961 |
+
"step": 1360
|
962 |
+
},
|
963 |
+
{
|
964 |
+
"epoch": 0.00553497335536549,
|
965 |
+
"grad_norm": 1410.2564697265625,
|
966 |
+
"learning_rate": 2.1920000000000004e-06,
|
967 |
+
"loss": 389.759,
|
968 |
+
"step": 1370
|
969 |
+
},
|
970 |
+
{
|
971 |
+
"epoch": 0.005575374620733121,
|
972 |
+
"grad_norm": 1602.010009765625,
|
973 |
+
"learning_rate": 2.2080000000000003e-06,
|
974 |
+
"loss": 412.8606,
|
975 |
+
"step": 1380
|
976 |
+
},
|
977 |
+
{
|
978 |
+
"epoch": 0.005615775886100753,
|
979 |
+
"grad_norm": 1337.7384033203125,
|
980 |
+
"learning_rate": 2.2240000000000002e-06,
|
981 |
+
"loss": 428.547,
|
982 |
+
"step": 1390
|
983 |
+
},
|
984 |
+
{
|
985 |
+
"epoch": 0.005656177151468384,
|
986 |
+
"grad_norm": 867.4197387695312,
|
987 |
+
"learning_rate": 2.24e-06,
|
988 |
+
"loss": 378.6157,
|
989 |
+
"step": 1400
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 0.005696578416836015,
|
993 |
+
"grad_norm": 731.271728515625,
|
994 |
+
"learning_rate": 2.256e-06,
|
995 |
+
"loss": 374.0744,
|
996 |
+
"step": 1410
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 0.0057369796822036466,
|
1000 |
+
"grad_norm": 2343.665771484375,
|
1001 |
+
"learning_rate": 2.2720000000000004e-06,
|
1002 |
+
"loss": 379.0267,
|
1003 |
+
"step": 1420
|
1004 |
+
},
|
1005 |
+
{
|
1006 |
+
"epoch": 0.005777380947571278,
|
1007 |
+
"grad_norm": 1540.48388671875,
|
1008 |
+
"learning_rate": 2.2880000000000004e-06,
|
1009 |
+
"loss": 382.6617,
|
1010 |
+
"step": 1430
|
1011 |
+
},
|
1012 |
+
{
|
1013 |
+
"epoch": 0.005817782212938909,
|
1014 |
+
"grad_norm": 1533.5958251953125,
|
1015 |
+
"learning_rate": 2.3040000000000003e-06,
|
1016 |
+
"loss": 396.0985,
|
1017 |
+
"step": 1440
|
1018 |
+
},
|
1019 |
+
{
|
1020 |
+
"epoch": 0.0058581834783065405,
|
1021 |
+
"grad_norm": 723.0481567382812,
|
1022 |
+
"learning_rate": 2.3200000000000002e-06,
|
1023 |
+
"loss": 308.0563,
|
1024 |
+
"step": 1450
|
1025 |
+
},
|
1026 |
+
{
|
1027 |
+
"epoch": 0.005898584743674172,
|
1028 |
+
"grad_norm": 2209.6083984375,
|
1029 |
+
"learning_rate": 2.336e-06,
|
1030 |
+
"loss": 424.7519,
|
1031 |
+
"step": 1460
|
1032 |
+
},
|
1033 |
+
{
|
1034 |
+
"epoch": 0.005938986009041803,
|
1035 |
+
"grad_norm": 566.4544067382812,
|
1036 |
+
"learning_rate": 2.352e-06,
|
1037 |
+
"loss": 248.9047,
|
1038 |
+
"step": 1470
|
1039 |
+
},
|
1040 |
+
{
|
1041 |
+
"epoch": 0.0059793872744094345,
|
1042 |
+
"grad_norm": 868.7781982421875,
|
1043 |
+
"learning_rate": 2.3680000000000005e-06,
|
1044 |
+
"loss": 257.8537,
|
1045 |
+
"step": 1480
|
1046 |
+
},
|
1047 |
+
{
|
1048 |
+
"epoch": 0.006019788539777066,
|
1049 |
+
"grad_norm": 732.210205078125,
|
1050 |
+
"learning_rate": 2.3840000000000004e-06,
|
1051 |
+
"loss": 304.146,
|
1052 |
+
"step": 1490
|
1053 |
+
},
|
1054 |
+
{
|
1055 |
+
"epoch": 0.006060189805144697,
|
1056 |
+
"grad_norm": 946.9075317382812,
|
1057 |
+
"learning_rate": 2.4000000000000003e-06,
|
1058 |
+
"loss": 239.7318,
|
1059 |
+
"step": 1500
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 0.0061005910705123284,
|
1063 |
+
"grad_norm": 13990.4541015625,
|
1064 |
+
"learning_rate": 2.4160000000000002e-06,
|
1065 |
+
"loss": 317.3043,
|
1066 |
+
"step": 1510
|
1067 |
+
},
|
1068 |
+
{
|
1069 |
+
"epoch": 0.00614099233587996,
|
1070 |
+
"grad_norm": 1053.9326171875,
|
1071 |
+
"learning_rate": 2.432e-06,
|
1072 |
+
"loss": 306.3782,
|
1073 |
+
"step": 1520
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 0.006181393601247591,
|
1077 |
+
"grad_norm": 1479.0665283203125,
|
1078 |
+
"learning_rate": 2.448e-06,
|
1079 |
+
"loss": 321.1809,
|
1080 |
+
"step": 1530
|
1081 |
+
},
|
1082 |
+
{
|
1083 |
+
"epoch": 0.006221794866615222,
|
1084 |
+
"grad_norm": 0.0,
|
1085 |
+
"learning_rate": 2.4640000000000005e-06,
|
1086 |
+
"loss": 272.3122,
|
1087 |
+
"step": 1540
|
1088 |
+
},
|
1089 |
+
{
|
1090 |
+
"epoch": 0.006262196131982854,
|
1091 |
+
"grad_norm": 836.43994140625,
|
1092 |
+
"learning_rate": 2.4800000000000004e-06,
|
1093 |
+
"loss": 276.743,
|
1094 |
+
"step": 1550
|
1095 |
+
},
|
1096 |
+
{
|
1097 |
+
"epoch": 0.006302597397350485,
|
1098 |
+
"grad_norm": 1576.191650390625,
|
1099 |
+
"learning_rate": 2.496e-06,
|
1100 |
+
"loss": 293.9756,
|
1101 |
+
"step": 1560
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 0.006342998662718116,
|
1105 |
+
"grad_norm": 1241.993408203125,
|
1106 |
+
"learning_rate": 2.512e-06,
|
1107 |
+
"loss": 324.3042,
|
1108 |
+
"step": 1570
|
1109 |
+
},
|
1110 |
+
{
|
1111 |
+
"epoch": 0.006383399928085748,
|
1112 |
+
"grad_norm": 3129.779052734375,
|
1113 |
+
"learning_rate": 2.5280000000000006e-06,
|
1114 |
+
"loss": 282.3287,
|
1115 |
+
"step": 1580
|
1116 |
+
},
|
1117 |
+
{
|
1118 |
+
"epoch": 0.006423801193453379,
|
1119 |
+
"grad_norm": 824.8917846679688,
|
1120 |
+
"learning_rate": 2.5440000000000005e-06,
|
1121 |
+
"loss": 332.7758,
|
1122 |
+
"step": 1590
|
1123 |
+
},
|
1124 |
+
{
|
1125 |
+
"epoch": 0.00646420245882101,
|
1126 |
+
"grad_norm": 1429.83154296875,
|
1127 |
+
"learning_rate": 2.56e-06,
|
1128 |
+
"loss": 341.1543,
|
1129 |
+
"step": 1600
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 0.006504603724188642,
|
1133 |
+
"grad_norm": 1096.0162353515625,
|
1134 |
+
"learning_rate": 2.576e-06,
|
1135 |
+
"loss": 370.956,
|
1136 |
+
"step": 1610
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 0.006545004989556273,
|
1140 |
+
"grad_norm": 2190.672607421875,
|
1141 |
+
"learning_rate": 2.592e-06,
|
1142 |
+
"loss": 393.7615,
|
1143 |
+
"step": 1620
|
1144 |
+
},
|
1145 |
+
{
|
1146 |
+
"epoch": 0.006585406254923904,
|
1147 |
+
"grad_norm": 659.6456909179688,
|
1148 |
+
"learning_rate": 2.608e-06,
|
1149 |
+
"loss": 371.8131,
|
1150 |
+
"step": 1630
|
1151 |
+
},
|
1152 |
+
{
|
1153 |
+
"epoch": 0.006625807520291536,
|
1154 |
+
"grad_norm": 1114.4139404296875,
|
1155 |
+
"learning_rate": 2.6240000000000006e-06,
|
1156 |
+
"loss": 387.3432,
|
1157 |
+
"step": 1640
|
1158 |
+
},
|
1159 |
+
{
|
1160 |
+
"epoch": 0.006666208785659167,
|
1161 |
+
"grad_norm": 1945.1435546875,
|
1162 |
+
"learning_rate": 2.64e-06,
|
1163 |
+
"loss": 419.9613,
|
1164 |
+
"step": 1650
|
1165 |
+
},
|
1166 |
+
{
|
1167 |
+
"epoch": 0.006706610051026798,
|
1168 |
+
"grad_norm": 1134.198486328125,
|
1169 |
+
"learning_rate": 2.656e-06,
|
1170 |
+
"loss": 261.7552,
|
1171 |
+
"step": 1660
|
1172 |
+
},
|
1173 |
+
{
|
1174 |
+
"epoch": 0.0067470113163944296,
|
1175 |
+
"grad_norm": 625.1832885742188,
|
1176 |
+
"learning_rate": 2.672e-06,
|
1177 |
+
"loss": 348.9073,
|
1178 |
+
"step": 1670
|
1179 |
+
},
|
1180 |
+
{
|
1181 |
+
"epoch": 0.006787412581762061,
|
1182 |
+
"grad_norm": 1558.600830078125,
|
1183 |
+
"learning_rate": 2.688e-06,
|
1184 |
+
"loss": 329.5783,
|
1185 |
+
"step": 1680
|
1186 |
+
},
|
1187 |
+
{
|
1188 |
+
"epoch": 0.006827813847129692,
|
1189 |
+
"grad_norm": 903.2876586914062,
|
1190 |
+
"learning_rate": 2.704e-06,
|
1191 |
+
"loss": 288.3402,
|
1192 |
+
"step": 1690
|
1193 |
+
},
|
1194 |
+
{
|
1195 |
+
"epoch": 0.0068682151124973235,
|
1196 |
+
"grad_norm": 4166.87158203125,
|
1197 |
+
"learning_rate": 2.7200000000000002e-06,
|
1198 |
+
"loss": 357.7317,
|
1199 |
+
"step": 1700
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 0.006908616377864955,
|
1203 |
+
"grad_norm": 784.2251586914062,
|
1204 |
+
"learning_rate": 2.736e-06,
|
1205 |
+
"loss": 318.5063,
|
1206 |
+
"step": 1710
|
1207 |
+
},
|
1208 |
+
{
|
1209 |
+
"epoch": 0.006949017643232586,
|
1210 |
+
"grad_norm": 5029.16552734375,
|
1211 |
+
"learning_rate": 2.752e-06,
|
1212 |
+
"loss": 442.4195,
|
1213 |
+
"step": 1720
|
1214 |
+
},
|
1215 |
+
{
|
1216 |
+
"epoch": 0.0069894189086002175,
|
1217 |
+
"grad_norm": 781.5169677734375,
|
1218 |
+
"learning_rate": 2.768e-06,
|
1219 |
+
"loss": 306.0392,
|
1220 |
+
"step": 1730
|
1221 |
+
},
|
1222 |
+
{
|
1223 |
+
"epoch": 0.007029820173967849,
|
1224 |
+
"grad_norm": 1048.294921875,
|
1225 |
+
"learning_rate": 2.784e-06,
|
1226 |
+
"loss": 359.3845,
|
1227 |
+
"step": 1740
|
1228 |
+
},
|
1229 |
+
{
|
1230 |
+
"epoch": 0.00707022143933548,
|
1231 |
+
"grad_norm": 1155.3671875,
|
1232 |
+
"learning_rate": 2.8000000000000003e-06,
|
1233 |
+
"loss": 359.3311,
|
1234 |
+
"step": 1750
|
1235 |
+
},
|
1236 |
+
{
|
1237 |
+
"epoch": 0.0071106227047031114,
|
1238 |
+
"grad_norm": 857.3798217773438,
|
1239 |
+
"learning_rate": 2.8160000000000002e-06,
|
1240 |
+
"loss": 391.5449,
|
1241 |
+
"step": 1760
|
1242 |
+
},
|
1243 |
+
{
|
1244 |
+
"epoch": 0.007151023970070743,
|
1245 |
+
"grad_norm": 933.0081787109375,
|
1246 |
+
"learning_rate": 2.832e-06,
|
1247 |
+
"loss": 297.32,
|
1248 |
+
"step": 1770
|
1249 |
+
},
|
1250 |
+
{
|
1251 |
+
"epoch": 0.007191425235438374,
|
1252 |
+
"grad_norm": 1060.47314453125,
|
1253 |
+
"learning_rate": 2.848e-06,
|
1254 |
+
"loss": 411.9166,
|
1255 |
+
"step": 1780
|
1256 |
+
},
|
1257 |
+
{
|
1258 |
+
"epoch": 0.007231826500806005,
|
1259 |
+
"grad_norm": 707.4050903320312,
|
1260 |
+
"learning_rate": 2.864e-06,
|
1261 |
+
"loss": 238.9362,
|
1262 |
+
"step": 1790
|
1263 |
+
},
|
1264 |
+
{
|
1265 |
+
"epoch": 0.007272227766173637,
|
1266 |
+
"grad_norm": 526.6619873046875,
|
1267 |
+
"learning_rate": 2.88e-06,
|
1268 |
+
"loss": 290.1179,
|
1269 |
+
"step": 1800
|
1270 |
+
},
|
1271 |
+
{
|
1272 |
+
"epoch": 0.007312629031541268,
|
1273 |
+
"grad_norm": 829.2198486328125,
|
1274 |
+
"learning_rate": 2.8960000000000003e-06,
|
1275 |
+
"loss": 347.0842,
|
1276 |
+
"step": 1810
|
1277 |
+
},
|
1278 |
+
{
|
1279 |
+
"epoch": 0.007353030296908899,
|
1280 |
+
"grad_norm": 716.2144165039062,
|
1281 |
+
"learning_rate": 2.9120000000000002e-06,
|
1282 |
+
"loss": 295.0978,
|
1283 |
+
"step": 1820
|
1284 |
+
},
|
1285 |
+
{
|
1286 |
+
"epoch": 0.007393431562276531,
|
1287 |
+
"grad_norm": 1365.7164306640625,
|
1288 |
+
"learning_rate": 2.928e-06,
|
1289 |
+
"loss": 399.1327,
|
1290 |
+
"step": 1830
|
1291 |
+
},
|
1292 |
+
{
|
1293 |
+
"epoch": 0.007433832827644162,
|
1294 |
+
"grad_norm": 1474.3541259765625,
|
1295 |
+
"learning_rate": 2.944e-06,
|
1296 |
+
"loss": 249.1963,
|
1297 |
+
"step": 1840
|
1298 |
+
},
|
1299 |
+
{
|
1300 |
+
"epoch": 0.007474234093011793,
|
1301 |
+
"grad_norm": 958.6668090820312,
|
1302 |
+
"learning_rate": 2.96e-06,
|
1303 |
+
"loss": 405.7022,
|
1304 |
+
"step": 1850
|
1305 |
+
},
|
1306 |
+
{
|
1307 |
+
"epoch": 0.007514635358379425,
|
1308 |
+
"grad_norm": 928.7461547851562,
|
1309 |
+
"learning_rate": 2.976e-06,
|
1310 |
+
"loss": 392.5513,
|
1311 |
+
"step": 1860
|
1312 |
+
},
|
1313 |
+
{
|
1314 |
+
"epoch": 0.007555036623747056,
|
1315 |
+
"grad_norm": 0.0,
|
1316 |
+
"learning_rate": 2.9920000000000003e-06,
|
1317 |
+
"loss": 235.2757,
|
1318 |
+
"step": 1870
|
1319 |
+
},
|
1320 |
+
{
|
1321 |
+
"epoch": 0.007595437889114687,
|
1322 |
+
"grad_norm": 13806.5771484375,
|
1323 |
+
"learning_rate": 3.0080000000000003e-06,
|
1324 |
+
"loss": 338.4041,
|
1325 |
+
"step": 1880
|
1326 |
+
},
|
1327 |
+
{
|
1328 |
+
"epoch": 0.007635839154482319,
|
1329 |
+
"grad_norm": 2829.76513671875,
|
1330 |
+
"learning_rate": 3.024e-06,
|
1331 |
+
"loss": 365.0503,
|
1332 |
+
"step": 1890
|
1333 |
+
},
|
1334 |
+
{
|
1335 |
+
"epoch": 0.00767624041984995,
|
1336 |
+
"grad_norm": 1206.556640625,
|
1337 |
+
"learning_rate": 3.04e-06,
|
1338 |
+
"loss": 372.7524,
|
1339 |
+
"step": 1900
|
1340 |
+
},
|
1341 |
+
{
|
1342 |
+
"epoch": 0.007716641685217581,
|
1343 |
+
"grad_norm": 1795.8287353515625,
|
1344 |
+
"learning_rate": 3.056e-06,
|
1345 |
+
"loss": 355.855,
|
1346 |
+
"step": 1910
|
1347 |
+
},
|
1348 |
+
{
|
1349 |
+
"epoch": 0.0077570429505852126,
|
1350 |
+
"grad_norm": 733.5438232421875,
|
1351 |
+
"learning_rate": 3.072e-06,
|
1352 |
+
"loss": 369.4895,
|
1353 |
+
"step": 1920
|
1354 |
+
},
|
1355 |
+
{
|
1356 |
+
"epoch": 0.007797444215952844,
|
1357 |
+
"grad_norm": 1936.3282470703125,
|
1358 |
+
"learning_rate": 3.0880000000000003e-06,
|
1359 |
+
"loss": 331.1103,
|
1360 |
+
"step": 1930
|
1361 |
+
},
|
1362 |
+
{
|
1363 |
+
"epoch": 0.007837845481320474,
|
1364 |
+
"grad_norm": 2228.966552734375,
|
1365 |
+
"learning_rate": 3.1040000000000003e-06,
|
1366 |
+
"loss": 441.473,
|
1367 |
+
"step": 1940
|
1368 |
+
},
|
1369 |
+
{
|
1370 |
+
"epoch": 0.007878246746688106,
|
1371 |
+
"grad_norm": 1527.3521728515625,
|
1372 |
+
"learning_rate": 3.12e-06,
|
1373 |
+
"loss": 286.5484,
|
1374 |
+
"step": 1950
|
1375 |
+
},
|
1376 |
+
{
|
1377 |
+
"epoch": 0.007918648012055737,
|
1378 |
+
"grad_norm": 1155.413330078125,
|
1379 |
+
"learning_rate": 3.136e-06,
|
1380 |
+
"loss": 325.0518,
|
1381 |
+
"step": 1960
|
1382 |
+
},
|
1383 |
+
{
|
1384 |
+
"epoch": 0.007959049277423368,
|
1385 |
+
"grad_norm": 5305.16845703125,
|
1386 |
+
"learning_rate": 3.152e-06,
|
1387 |
+
"loss": 426.8875,
|
1388 |
+
"step": 1970
|
1389 |
+
},
|
1390 |
+
{
|
1391 |
+
"epoch": 0.007999450542791,
|
1392 |
+
"grad_norm": 497.835693359375,
|
1393 |
+
"learning_rate": 3.1680000000000004e-06,
|
1394 |
+
"loss": 305.8856,
|
1395 |
+
"step": 1980
|
1396 |
+
},
|
1397 |
+
{
|
1398 |
+
"epoch": 0.008039851808158631,
|
1399 |
+
"grad_norm": 950.1622924804688,
|
1400 |
+
"learning_rate": 3.1840000000000003e-06,
|
1401 |
+
"loss": 298.3484,
|
1402 |
+
"step": 1990
|
1403 |
+
},
|
1404 |
+
{
|
1405 |
+
"epoch": 0.008080253073526262,
|
1406 |
+
"grad_norm": 977.10546875,
|
1407 |
+
"learning_rate": 3.2000000000000003e-06,
|
1408 |
+
"loss": 385.7309,
|
1409 |
+
"step": 2000
|
1410 |
+
},
|
1411 |
+
{
|
1412 |
+
"epoch": 0.008120654338893894,
|
1413 |
+
"grad_norm": 1347.1170654296875,
|
1414 |
+
"learning_rate": 3.216e-06,
|
1415 |
+
"loss": 311.8849,
|
1416 |
+
"step": 2010
|
1417 |
+
},
|
1418 |
+
{
|
1419 |
+
"epoch": 0.008161055604261525,
|
1420 |
+
"grad_norm": 7115.32470703125,
|
1421 |
+
"learning_rate": 3.232e-06,
|
1422 |
+
"loss": 310.0182,
|
1423 |
+
"step": 2020
|
1424 |
+
},
|
1425 |
+
{
|
1426 |
+
"epoch": 0.008201456869629156,
|
1427 |
+
"grad_norm": 979.557861328125,
|
1428 |
+
"learning_rate": 3.248e-06,
|
1429 |
+
"loss": 378.8968,
|
1430 |
+
"step": 2030
|
1431 |
+
},
|
1432 |
+
{
|
1433 |
+
"epoch": 0.008241858134996788,
|
1434 |
+
"grad_norm": 713.5515747070312,
|
1435 |
+
"learning_rate": 3.2640000000000004e-06,
|
1436 |
+
"loss": 333.8391,
|
1437 |
+
"step": 2040
|
1438 |
+
},
|
1439 |
+
{
|
1440 |
+
"epoch": 0.008282259400364419,
|
1441 |
+
"grad_norm": 1746.4747314453125,
|
1442 |
+
"learning_rate": 3.2800000000000004e-06,
|
1443 |
+
"loss": 211.6272,
|
1444 |
+
"step": 2050
|
1445 |
+
},
|
1446 |
+
{
|
1447 |
+
"epoch": 0.00832266066573205,
|
1448 |
+
"grad_norm": 3959.4423828125,
|
1449 |
+
"learning_rate": 3.2960000000000003e-06,
|
1450 |
+
"loss": 418.0941,
|
1451 |
+
"step": 2060
|
1452 |
+
},
|
1453 |
+
{
|
1454 |
+
"epoch": 0.008363061931099681,
|
1455 |
+
"grad_norm": 1449.259765625,
|
1456 |
+
"learning_rate": 3.3120000000000002e-06,
|
1457 |
+
"loss": 366.1021,
|
1458 |
+
"step": 2070
|
1459 |
+
},
|
1460 |
+
{
|
1461 |
+
"epoch": 0.008403463196467313,
|
1462 |
+
"grad_norm": 686.3818359375,
|
1463 |
+
"learning_rate": 3.328e-06,
|
1464 |
+
"loss": 317.5829,
|
1465 |
+
"step": 2080
|
1466 |
+
},
|
1467 |
+
{
|
1468 |
+
"epoch": 0.008443864461834944,
|
1469 |
+
"grad_norm": 1431.3048095703125,
|
1470 |
+
"learning_rate": 3.344e-06,
|
1471 |
+
"loss": 341.1351,
|
1472 |
+
"step": 2090
|
1473 |
+
},
|
1474 |
+
{
|
1475 |
+
"epoch": 0.008484265727202575,
|
1476 |
+
"grad_norm": 1141.1195068359375,
|
1477 |
+
"learning_rate": 3.3600000000000004e-06,
|
1478 |
+
"loss": 284.9502,
|
1479 |
+
"step": 2100
|
1480 |
+
},
|
1481 |
+
{
|
1482 |
+
"epoch": 0.008524666992570207,
|
1483 |
+
"grad_norm": 583.6392211914062,
|
1484 |
+
"learning_rate": 3.3760000000000004e-06,
|
1485 |
+
"loss": 246.5204,
|
1486 |
+
"step": 2110
|
1487 |
+
},
|
1488 |
+
{
|
1489 |
+
"epoch": 0.008565068257937838,
|
1490 |
+
"grad_norm": 1383.4288330078125,
|
1491 |
+
"learning_rate": 3.3920000000000003e-06,
|
1492 |
+
"loss": 335.3397,
|
1493 |
+
"step": 2120
|
1494 |
+
},
|
1495 |
+
{
|
1496 |
+
"epoch": 0.00860546952330547,
|
1497 |
+
"grad_norm": 720.5582275390625,
|
1498 |
+
"learning_rate": 3.4080000000000002e-06,
|
1499 |
+
"loss": 300.3475,
|
1500 |
+
"step": 2130
|
1501 |
+
},
|
1502 |
+
{
|
1503 |
+
"epoch": 0.0086458707886731,
|
1504 |
+
"grad_norm": 1018.2730712890625,
|
1505 |
+
"learning_rate": 3.424e-06,
|
1506 |
+
"loss": 364.7046,
|
1507 |
+
"step": 2140
|
1508 |
+
},
|
1509 |
+
{
|
1510 |
+
"epoch": 0.008686272054040732,
|
1511 |
+
"grad_norm": 836.5111694335938,
|
1512 |
+
"learning_rate": 3.44e-06,
|
1513 |
+
"loss": 215.2347,
|
1514 |
+
"step": 2150
|
1515 |
+
},
|
1516 |
+
{
|
1517 |
+
"epoch": 0.008726673319408363,
|
1518 |
+
"grad_norm": 1106.7752685546875,
|
1519 |
+
"learning_rate": 3.4560000000000005e-06,
|
1520 |
+
"loss": 392.8156,
|
1521 |
+
"step": 2160
|
1522 |
+
},
|
1523 |
+
{
|
1524 |
+
"epoch": 0.008767074584775995,
|
1525 |
+
"grad_norm": 838.0346069335938,
|
1526 |
+
"learning_rate": 3.4720000000000004e-06,
|
1527 |
+
"loss": 328.1276,
|
1528 |
+
"step": 2170
|
1529 |
+
},
|
1530 |
+
{
|
1531 |
+
"epoch": 0.008807475850143626,
|
1532 |
+
"grad_norm": 870.4414672851562,
|
1533 |
+
"learning_rate": 3.4880000000000003e-06,
|
1534 |
+
"loss": 313.8927,
|
1535 |
+
"step": 2180
|
1536 |
+
},
|
1537 |
+
{
|
1538 |
+
"epoch": 0.008847877115511257,
|
1539 |
+
"grad_norm": 2859.623046875,
|
1540 |
+
"learning_rate": 3.5040000000000002e-06,
|
1541 |
+
"loss": 241.9949,
|
1542 |
+
"step": 2190
|
1543 |
+
},
|
1544 |
+
{
|
1545 |
+
"epoch": 0.008888278380878889,
|
1546 |
+
"grad_norm": 971.9884033203125,
|
1547 |
+
"learning_rate": 3.52e-06,
|
1548 |
+
"loss": 349.5033,
|
1549 |
+
"step": 2200
|
1550 |
+
},
|
1551 |
+
{
|
1552 |
+
"epoch": 0.00892867964624652,
|
1553 |
+
"grad_norm": 670.4691772460938,
|
1554 |
+
"learning_rate": 3.5360000000000005e-06,
|
1555 |
+
"loss": 239.4784,
|
1556 |
+
"step": 2210
|
1557 |
+
},
|
1558 |
+
{
|
1559 |
+
"epoch": 0.008969080911614151,
|
1560 |
+
"grad_norm": 784.5670166015625,
|
1561 |
+
"learning_rate": 3.5520000000000005e-06,
|
1562 |
+
"loss": 310.9449,
|
1563 |
+
"step": 2220
|
1564 |
+
},
|
1565 |
+
{
|
1566 |
+
"epoch": 0.009009482176981783,
|
1567 |
+
"grad_norm": 915.0491333007812,
|
1568 |
+
"learning_rate": 3.5680000000000004e-06,
|
1569 |
+
"loss": 413.7835,
|
1570 |
+
"step": 2230
|
1571 |
+
},
|
1572 |
+
{
|
1573 |
+
"epoch": 0.009049883442349414,
|
1574 |
+
"grad_norm": 893.2843627929688,
|
1575 |
+
"learning_rate": 3.5840000000000003e-06,
|
1576 |
+
"loss": 365.2546,
|
1577 |
+
"step": 2240
|
1578 |
+
},
|
1579 |
+
{
|
1580 |
+
"epoch": 0.009090284707717045,
|
1581 |
+
"grad_norm": 1068.404296875,
|
1582 |
+
"learning_rate": 3.6000000000000003e-06,
|
1583 |
+
"loss": 302.3378,
|
1584 |
+
"step": 2250
|
1585 |
+
},
|
1586 |
+
{
|
1587 |
+
"epoch": 0.009130685973084677,
|
1588 |
+
"grad_norm": 1454.05859375,
|
1589 |
+
"learning_rate": 3.616e-06,
|
1590 |
+
"loss": 285.7942,
|
1591 |
+
"step": 2260
|
1592 |
+
},
|
1593 |
+
{
|
1594 |
+
"epoch": 0.009171087238452308,
|
1595 |
+
"grad_norm": 3212.861083984375,
|
1596 |
+
"learning_rate": 3.6320000000000005e-06,
|
1597 |
+
"loss": 296.5446,
|
1598 |
+
"step": 2270
|
1599 |
+
},
|
1600 |
+
{
|
1601 |
+
"epoch": 0.00921148850381994,
|
1602 |
+
"grad_norm": 700.9654541015625,
|
1603 |
+
"learning_rate": 3.6480000000000005e-06,
|
1604 |
+
"loss": 277.7706,
|
1605 |
+
"step": 2280
|
1606 |
+
},
|
1607 |
+
{
|
1608 |
+
"epoch": 0.00925188976918757,
|
1609 |
+
"grad_norm": 601.0996704101562,
|
1610 |
+
"learning_rate": 3.6640000000000004e-06,
|
1611 |
+
"loss": 299.2936,
|
1612 |
+
"step": 2290
|
1613 |
+
},
|
1614 |
+
{
|
1615 |
+
"epoch": 0.009292291034555202,
|
1616 |
+
"grad_norm": 1073.6248779296875,
|
1617 |
+
"learning_rate": 3.6800000000000003e-06,
|
1618 |
+
"loss": 227.1634,
|
1619 |
+
"step": 2300
|
1620 |
+
},
|
1621 |
+
{
|
1622 |
+
"epoch": 0.009332692299922833,
|
1623 |
+
"grad_norm": 803.8406372070312,
|
1624 |
+
"learning_rate": 3.6960000000000003e-06,
|
1625 |
+
"loss": 344.3006,
|
1626 |
+
"step": 2310
|
1627 |
+
},
|
1628 |
+
{
|
1629 |
+
"epoch": 0.009373093565290464,
|
1630 |
+
"grad_norm": 940.9983520507812,
|
1631 |
+
"learning_rate": 3.712e-06,
|
1632 |
+
"loss": 266.4591,
|
1633 |
+
"step": 2320
|
1634 |
+
},
|
1635 |
+
{
|
1636 |
+
"epoch": 0.009413494830658096,
|
1637 |
+
"grad_norm": 13408.8291015625,
|
1638 |
+
"learning_rate": 3.7280000000000006e-06,
|
1639 |
+
"loss": 295.9835,
|
1640 |
+
"step": 2330
|
1641 |
+
},
|
1642 |
+
{
|
1643 |
+
"epoch": 0.009453896096025727,
|
1644 |
+
"grad_norm": 1009.2761840820312,
|
1645 |
+
"learning_rate": 3.7440000000000005e-06,
|
1646 |
+
"loss": 292.3968,
|
1647 |
+
"step": 2340
|
1648 |
+
},
|
1649 |
+
{
|
1650 |
+
"epoch": 0.009494297361393358,
|
1651 |
+
"grad_norm": 556.4901733398438,
|
1652 |
+
"learning_rate": 3.7600000000000004e-06,
|
1653 |
+
"loss": 262.7075,
|
1654 |
+
"step": 2350
|
1655 |
+
},
|
1656 |
+
{
|
1657 |
+
"epoch": 0.00953469862676099,
|
1658 |
+
"grad_norm": 913.5816040039062,
|
1659 |
+
"learning_rate": 3.7760000000000004e-06,
|
1660 |
+
"loss": 508.4715,
|
1661 |
+
"step": 2360
|
1662 |
+
},
|
1663 |
+
{
|
1664 |
+
"epoch": 0.009575099892128621,
|
1665 |
+
"grad_norm": 1582.836181640625,
|
1666 |
+
"learning_rate": 3.7920000000000003e-06,
|
1667 |
+
"loss": 321.3247,
|
1668 |
+
"step": 2370
|
1669 |
+
},
|
1670 |
+
{
|
1671 |
+
"epoch": 0.009615501157496252,
|
1672 |
+
"grad_norm": 1292.945068359375,
|
1673 |
+
"learning_rate": 3.8080000000000006e-06,
|
1674 |
+
"loss": 266.3162,
|
1675 |
+
"step": 2380
|
1676 |
+
},
|
1677 |
+
{
|
1678 |
+
"epoch": 0.009655902422863884,
|
1679 |
+
"grad_norm": 741.0372314453125,
|
1680 |
+
"learning_rate": 3.824e-06,
|
1681 |
+
"loss": 333.5866,
|
1682 |
+
"step": 2390
|
1683 |
+
},
|
1684 |
+
{
|
1685 |
+
"epoch": 0.009696303688231515,
|
1686 |
+
"grad_norm": 594.6494750976562,
|
1687 |
+
"learning_rate": 3.8400000000000005e-06,
|
1688 |
+
"loss": 370.6466,
|
1689 |
+
"step": 2400
|
1690 |
+
},
|
1691 |
+
{
|
1692 |
+
"epoch": 0.009736704953599146,
|
1693 |
+
"grad_norm": 671.8253173828125,
|
1694 |
+
"learning_rate": 3.856e-06,
|
1695 |
+
"loss": 176.7593,
|
1696 |
+
"step": 2410
|
1697 |
+
},
|
1698 |
+
{
|
1699 |
+
"epoch": 0.009777106218966778,
|
1700 |
+
"grad_norm": 1318.59912109375,
|
1701 |
+
"learning_rate": 3.872e-06,
|
1702 |
+
"loss": 296.8816,
|
1703 |
+
"step": 2420
|
1704 |
+
},
|
1705 |
+
{
|
1706 |
+
"epoch": 0.009817507484334409,
|
1707 |
+
"grad_norm": 4490.33837890625,
|
1708 |
+
"learning_rate": 3.888e-06,
|
1709 |
+
"loss": 353.2209,
|
1710 |
+
"step": 2430
|
1711 |
+
},
|
1712 |
+
{
|
1713 |
+
"epoch": 0.00985790874970204,
|
1714 |
+
"grad_norm": 1145.6590576171875,
|
1715 |
+
"learning_rate": 3.904e-06,
|
1716 |
+
"loss": 232.9775,
|
1717 |
+
"step": 2440
|
1718 |
+
},
|
1719 |
+
{
|
1720 |
+
"epoch": 0.009898310015069672,
|
1721 |
+
"grad_norm": 669.175537109375,
|
1722 |
+
"learning_rate": 3.920000000000001e-06,
|
1723 |
+
"loss": 245.2131,
|
1724 |
+
"step": 2450
|
1725 |
+
},
|
1726 |
+
{
|
1727 |
+
"epoch": 0.009938711280437303,
|
1728 |
+
"grad_norm": 945.0543212890625,
|
1729 |
+
"learning_rate": 3.936e-06,
|
1730 |
+
"loss": 189.2121,
|
1731 |
+
"step": 2460
|
1732 |
+
},
|
1733 |
+
{
|
1734 |
+
"epoch": 0.009979112545804934,
|
1735 |
+
"grad_norm": 871.2230224609375,
|
1736 |
+
"learning_rate": 3.9520000000000004e-06,
|
1737 |
+
"loss": 354.6668,
|
1738 |
+
"step": 2470
|
1739 |
+
},
|
1740 |
+
{
|
1741 |
+
"epoch": 0.010019513811172566,
|
1742 |
+
"grad_norm": 1143.5745849609375,
|
1743 |
+
"learning_rate": 3.968e-06,
|
1744 |
+
"loss": 309.6017,
|
1745 |
+
"step": 2480
|
1746 |
+
},
|
1747 |
+
{
|
1748 |
+
"epoch": 0.010059915076540197,
|
1749 |
+
"grad_norm": 921.1688842773438,
|
1750 |
+
"learning_rate": 3.984e-06,
|
1751 |
+
"loss": 393.8875,
|
1752 |
+
"step": 2490
|
1753 |
+
},
|
1754 |
+
{
|
1755 |
+
"epoch": 0.010100316341907828,
|
1756 |
+
"grad_norm": 670.0267333984375,
|
1757 |
+
"learning_rate": 4.000000000000001e-06,
|
1758 |
+
"loss": 187.9601,
|
1759 |
+
"step": 2500
|
1760 |
+
},
|
1761 |
+
{
|
1762 |
+
"epoch": 0.01014071760727546,
|
1763 |
+
"grad_norm": 906.67626953125,
|
1764 |
+
"learning_rate": 4.016e-06,
|
1765 |
+
"loss": 263.1041,
|
1766 |
+
"step": 2510
|
1767 |
+
},
|
1768 |
+
{
|
1769 |
+
"epoch": 0.010181118872643091,
|
1770 |
+
"grad_norm": 501.8714599609375,
|
1771 |
+
"learning_rate": 4.0320000000000005e-06,
|
1772 |
+
"loss": 249.8009,
|
1773 |
+
"step": 2520
|
1774 |
+
},
|
1775 |
+
{
|
1776 |
+
"epoch": 0.010221520138010722,
|
1777 |
+
"grad_norm": 1012.7957153320312,
|
1778 |
+
"learning_rate": 4.048e-06,
|
1779 |
+
"loss": 215.6589,
|
1780 |
+
"step": 2530
|
1781 |
+
},
|
1782 |
+
{
|
1783 |
+
"epoch": 0.010261921403378354,
|
1784 |
+
"grad_norm": 1067.8087158203125,
|
1785 |
+
"learning_rate": 4.064e-06,
|
1786 |
+
"loss": 277.4538,
|
1787 |
+
"step": 2540
|
1788 |
+
},
|
1789 |
+
{
|
1790 |
+
"epoch": 0.010302322668745985,
|
1791 |
+
"grad_norm": 821.8876342773438,
|
1792 |
+
"learning_rate": 4.08e-06,
|
1793 |
+
"loss": 357.288,
|
1794 |
+
"step": 2550
|
1795 |
+
},
|
1796 |
+
{
|
1797 |
+
"epoch": 0.010342723934113616,
|
1798 |
+
"grad_norm": 695.7543334960938,
|
1799 |
+
"learning_rate": 4.096e-06,
|
1800 |
+
"loss": 192.3841,
|
1801 |
+
"step": 2560
|
1802 |
+
},
|
1803 |
+
{
|
1804 |
+
"epoch": 0.010383125199481247,
|
1805 |
+
"grad_norm": 2217.32080078125,
|
1806 |
+
"learning_rate": 4.112000000000001e-06,
|
1807 |
+
"loss": 311.4946,
|
1808 |
+
"step": 2570
|
1809 |
+
},
|
1810 |
+
{
|
1811 |
+
"epoch": 0.010423526464848879,
|
1812 |
+
"grad_norm": 1252.262451171875,
|
1813 |
+
"learning_rate": 4.128e-06,
|
1814 |
+
"loss": 299.7565,
|
1815 |
+
"step": 2580
|
1816 |
+
},
|
1817 |
+
{
|
1818 |
+
"epoch": 0.01046392773021651,
|
1819 |
+
"grad_norm": 1227.412109375,
|
1820 |
+
"learning_rate": 4.1440000000000005e-06,
|
1821 |
+
"loss": 258.2722,
|
1822 |
+
"step": 2590
|
1823 |
+
},
|
1824 |
+
{
|
1825 |
+
"epoch": 0.010504328995584141,
|
1826 |
+
"grad_norm": 1460.778076171875,
|
1827 |
+
"learning_rate": 4.16e-06,
|
1828 |
+
"loss": 310.2179,
|
1829 |
+
"step": 2600
|
1830 |
+
},
|
1831 |
+
{
|
1832 |
+
"epoch": 0.010544730260951773,
|
1833 |
+
"grad_norm": 3191.095947265625,
|
1834 |
+
"learning_rate": 4.176e-06,
|
1835 |
+
"loss": 276.1908,
|
1836 |
+
"step": 2610
|
1837 |
+
},
|
1838 |
+
{
|
1839 |
+
"epoch": 0.010585131526319404,
|
1840 |
+
"grad_norm": 841.0155029296875,
|
1841 |
+
"learning_rate": 4.192000000000001e-06,
|
1842 |
+
"loss": 409.2921,
|
1843 |
+
"step": 2620
|
1844 |
+
},
|
1845 |
+
{
|
1846 |
+
"epoch": 0.010625532791687035,
|
1847 |
+
"grad_norm": 979.8448486328125,
|
1848 |
+
"learning_rate": 4.208e-06,
|
1849 |
+
"loss": 370.0784,
|
1850 |
+
"step": 2630
|
1851 |
+
},
|
1852 |
+
{
|
1853 |
+
"epoch": 0.010665934057054667,
|
1854 |
+
"grad_norm": 779.2421264648438,
|
1855 |
+
"learning_rate": 4.2240000000000006e-06,
|
1856 |
+
"loss": 327.6988,
|
1857 |
+
"step": 2640
|
1858 |
+
},
|
1859 |
+
{
|
1860 |
+
"epoch": 0.010706335322422298,
|
1861 |
+
"grad_norm": 1958.304443359375,
|
1862 |
+
"learning_rate": 4.24e-06,
|
1863 |
+
"loss": 270.3837,
|
1864 |
+
"step": 2650
|
1865 |
+
},
|
1866 |
+
{
|
1867 |
+
"epoch": 0.01074673658778993,
|
1868 |
+
"grad_norm": 612.7789916992188,
|
1869 |
+
"learning_rate": 4.256e-06,
|
1870 |
+
"loss": 208.5096,
|
1871 |
+
"step": 2660
|
1872 |
+
},
|
1873 |
+
{
|
1874 |
+
"epoch": 0.01078713785315756,
|
1875 |
+
"grad_norm": 913.646484375,
|
1876 |
+
"learning_rate": 4.272000000000001e-06,
|
1877 |
+
"loss": 326.1478,
|
1878 |
+
"step": 2670
|
1879 |
+
},
|
1880 |
+
{
|
1881 |
+
"epoch": 0.010827539118525192,
|
1882 |
+
"grad_norm": 609.9149780273438,
|
1883 |
+
"learning_rate": 4.288e-06,
|
1884 |
+
"loss": 323.9391,
|
1885 |
+
"step": 2680
|
1886 |
+
},
|
1887 |
+
{
|
1888 |
+
"epoch": 0.010867940383892823,
|
1889 |
+
"grad_norm": 4146.9677734375,
|
1890 |
+
"learning_rate": 4.304000000000001e-06,
|
1891 |
+
"loss": 331.8204,
|
1892 |
+
"step": 2690
|
1893 |
+
},
|
1894 |
+
{
|
1895 |
+
"epoch": 0.010908341649260455,
|
1896 |
+
"grad_norm": 1144.696044921875,
|
1897 |
+
"learning_rate": 4.32e-06,
|
1898 |
+
"loss": 344.9819,
|
1899 |
+
"step": 2700
|
1900 |
+
},
|
1901 |
+
{
|
1902 |
+
"epoch": 0.010948742914628086,
|
1903 |
+
"grad_norm": 1466.951904296875,
|
1904 |
+
"learning_rate": 4.3360000000000005e-06,
|
1905 |
+
"loss": 364.1662,
|
1906 |
+
"step": 2710
|
1907 |
+
},
|
1908 |
+
{
|
1909 |
+
"epoch": 0.010989144179995717,
|
1910 |
+
"grad_norm": 955.5458984375,
|
1911 |
+
"learning_rate": 4.352e-06,
|
1912 |
+
"loss": 284.9516,
|
1913 |
+
"step": 2720
|
1914 |
+
},
|
1915 |
+
{
|
1916 |
+
"epoch": 0.011029545445363349,
|
1917 |
+
"grad_norm": 591.5704345703125,
|
1918 |
+
"learning_rate": 4.368e-06,
|
1919 |
+
"loss": 279.7713,
|
1920 |
+
"step": 2730
|
1921 |
+
},
|
1922 |
+
{
|
1923 |
+
"epoch": 0.01106994671073098,
|
1924 |
+
"grad_norm": 1922.548095703125,
|
1925 |
+
"learning_rate": 4.384000000000001e-06,
|
1926 |
+
"loss": 269.3951,
|
1927 |
+
"step": 2740
|
1928 |
+
},
|
1929 |
+
{
|
1930 |
+
"epoch": 0.011110347976098611,
|
1931 |
+
"grad_norm": 705.4920654296875,
|
1932 |
+
"learning_rate": 4.4e-06,
|
1933 |
+
"loss": 202.7344,
|
1934 |
+
"step": 2750
|
1935 |
+
},
|
1936 |
+
{
|
1937 |
+
"epoch": 0.011150749241466243,
|
1938 |
+
"grad_norm": 6034.74951171875,
|
1939 |
+
"learning_rate": 4.416000000000001e-06,
|
1940 |
+
"loss": 319.9652,
|
1941 |
+
"step": 2760
|
1942 |
+
},
|
1943 |
+
{
|
1944 |
+
"epoch": 0.011191150506833874,
|
1945 |
+
"grad_norm": 1474.4591064453125,
|
1946 |
+
"learning_rate": 4.432e-06,
|
1947 |
+
"loss": 239.9669,
|
1948 |
+
"step": 2770
|
1949 |
+
},
|
1950 |
+
{
|
1951 |
+
"epoch": 0.011231551772201505,
|
1952 |
+
"grad_norm": 998.685302734375,
|
1953 |
+
"learning_rate": 4.4480000000000004e-06,
|
1954 |
+
"loss": 254.0082,
|
1955 |
+
"step": 2780
|
1956 |
+
},
|
1957 |
+
{
|
1958 |
+
"epoch": 0.011271953037569137,
|
1959 |
+
"grad_norm": 1730.3958740234375,
|
1960 |
+
"learning_rate": 4.464000000000001e-06,
|
1961 |
+
"loss": 285.7142,
|
1962 |
+
"step": 2790
|
1963 |
+
},
|
1964 |
+
{
|
1965 |
+
"epoch": 0.011312354302936768,
|
1966 |
+
"grad_norm": 666.0034790039062,
|
1967 |
+
"learning_rate": 4.48e-06,
|
1968 |
+
"loss": 243.7346,
|
1969 |
+
"step": 2800
|
1970 |
+
},
|
1971 |
+
{
|
1972 |
+
"epoch": 0.0113527555683044,
|
1973 |
+
"grad_norm": 1185.6009521484375,
|
1974 |
+
"learning_rate": 4.496000000000001e-06,
|
1975 |
+
"loss": 243.6461,
|
1976 |
+
"step": 2810
|
1977 |
+
},
|
1978 |
+
{
|
1979 |
+
"epoch": 0.01139315683367203,
|
1980 |
+
"grad_norm": 902.0358276367188,
|
1981 |
+
"learning_rate": 4.512e-06,
|
1982 |
+
"loss": 193.6116,
|
1983 |
+
"step": 2820
|
1984 |
+
},
|
1985 |
+
{
|
1986 |
+
"epoch": 0.011433558099039662,
|
1987 |
+
"grad_norm": 1310.3072509765625,
|
1988 |
+
"learning_rate": 4.5280000000000005e-06,
|
1989 |
+
"loss": 327.7516,
|
1990 |
+
"step": 2830
|
1991 |
+
},
|
1992 |
+
{
|
1993 |
+
"epoch": 0.011473959364407293,
|
1994 |
+
"grad_norm": 693.1176147460938,
|
1995 |
+
"learning_rate": 4.544000000000001e-06,
|
1996 |
+
"loss": 239.3185,
|
1997 |
+
"step": 2840
|
1998 |
+
},
|
1999 |
+
{
|
2000 |
+
"epoch": 0.011514360629774924,
|
2001 |
+
"grad_norm": 968.5886840820312,
|
2002 |
+
"learning_rate": 4.56e-06,
|
2003 |
+
"loss": 287.3573,
|
2004 |
+
"step": 2850
|
2005 |
+
},
|
2006 |
+
{
|
2007 |
+
"epoch": 0.011554761895142556,
|
2008 |
+
"grad_norm": 705.2077026367188,
|
2009 |
+
"learning_rate": 4.576000000000001e-06,
|
2010 |
+
"loss": 282.05,
|
2011 |
+
"step": 2860
|
2012 |
+
},
|
2013 |
+
{
|
2014 |
+
"epoch": 0.011595163160510187,
|
2015 |
+
"grad_norm": 1975.849365234375,
|
2016 |
+
"learning_rate": 4.592e-06,
|
2017 |
+
"loss": 296.7342,
|
2018 |
+
"step": 2870
|
2019 |
+
},
|
2020 |
+
{
|
2021 |
+
"epoch": 0.011635564425877818,
|
2022 |
+
"grad_norm": 746.0858764648438,
|
2023 |
+
"learning_rate": 4.608000000000001e-06,
|
2024 |
+
"loss": 204.9099,
|
2025 |
+
"step": 2880
|
2026 |
+
},
|
2027 |
+
{
|
2028 |
+
"epoch": 0.01167596569124545,
|
2029 |
+
"grad_norm": 1130.648681640625,
|
2030 |
+
"learning_rate": 4.624e-06,
|
2031 |
+
"loss": 342.3003,
|
2032 |
+
"step": 2890
|
2033 |
+
},
|
2034 |
+
{
|
2035 |
+
"epoch": 0.011716366956613081,
|
2036 |
+
"grad_norm": 1225.900634765625,
|
2037 |
+
"learning_rate": 4.6400000000000005e-06,
|
2038 |
+
"loss": 302.3731,
|
2039 |
+
"step": 2900
|
2040 |
+
},
|
2041 |
+
{
|
2042 |
+
"epoch": 0.011756768221980712,
|
2043 |
+
"grad_norm": 965.9699096679688,
|
2044 |
+
"learning_rate": 4.656000000000001e-06,
|
2045 |
+
"loss": 403.5073,
|
2046 |
+
"step": 2910
|
2047 |
+
},
|
2048 |
+
{
|
2049 |
+
"epoch": 0.011797169487348344,
|
2050 |
+
"grad_norm": 700.615966796875,
|
2051 |
+
"learning_rate": 4.672e-06,
|
2052 |
+
"loss": 220.3576,
|
2053 |
+
"step": 2920
|
2054 |
+
},
|
2055 |
+
{
|
2056 |
+
"epoch": 0.011837570752715975,
|
2057 |
+
"grad_norm": 744.499267578125,
|
2058 |
+
"learning_rate": 4.688000000000001e-06,
|
2059 |
+
"loss": 227.1823,
|
2060 |
+
"step": 2930
|
2061 |
+
},
|
2062 |
+
{
|
2063 |
+
"epoch": 0.011877972018083606,
|
2064 |
+
"grad_norm": 588.2945556640625,
|
2065 |
+
"learning_rate": 4.704e-06,
|
2066 |
+
"loss": 294.2805,
|
2067 |
+
"step": 2940
|
2068 |
+
},
|
2069 |
+
{
|
2070 |
+
"epoch": 0.011918373283451238,
|
2071 |
+
"grad_norm": 8179.02490234375,
|
2072 |
+
"learning_rate": 4.7200000000000005e-06,
|
2073 |
+
"loss": 360.6105,
|
2074 |
+
"step": 2950
|
2075 |
+
},
|
2076 |
+
{
|
2077 |
+
"epoch": 0.011958774548818869,
|
2078 |
+
"grad_norm": 4027.3515625,
|
2079 |
+
"learning_rate": 4.736000000000001e-06,
|
2080 |
+
"loss": 286.0583,
|
2081 |
+
"step": 2960
|
2082 |
+
},
|
2083 |
+
{
|
2084 |
+
"epoch": 0.0119991758141865,
|
2085 |
+
"grad_norm": 8620.037109375,
|
2086 |
+
"learning_rate": 4.752e-06,
|
2087 |
+
"loss": 338.6114,
|
2088 |
+
"step": 2970
|
2089 |
+
},
|
2090 |
+
{
|
2091 |
+
"epoch": 0.012039577079554132,
|
2092 |
+
"grad_norm": 1030.6036376953125,
|
2093 |
+
"learning_rate": 4.768000000000001e-06,
|
2094 |
+
"loss": 293.7208,
|
2095 |
+
"step": 2980
|
2096 |
+
},
|
2097 |
+
{
|
2098 |
+
"epoch": 0.012079978344921763,
|
2099 |
+
"grad_norm": 2113.48095703125,
|
2100 |
+
"learning_rate": 4.784e-06,
|
2101 |
+
"loss": 338.7042,
|
2102 |
+
"step": 2990
|
2103 |
+
},
|
2104 |
+
{
|
2105 |
+
"epoch": 0.012120379610289394,
|
2106 |
+
"grad_norm": 664.2764892578125,
|
2107 |
+
"learning_rate": 4.800000000000001e-06,
|
2108 |
+
"loss": 180.6681,
|
2109 |
+
"step": 3000
|
2110 |
+
},
|
2111 |
+
{
|
2112 |
+
"epoch": 0.012160780875657026,
|
2113 |
+
"grad_norm": 2052.64111328125,
|
2114 |
+
"learning_rate": 4.816e-06,
|
2115 |
+
"loss": 269.5311,
|
2116 |
+
"step": 3010
|
2117 |
+
},
|
2118 |
+
{
|
2119 |
+
"epoch": 0.012201182141024657,
|
2120 |
+
"grad_norm": 646.8975219726562,
|
2121 |
+
"learning_rate": 4.8320000000000005e-06,
|
2122 |
+
"loss": 199.238,
|
2123 |
+
"step": 3020
|
2124 |
+
},
|
2125 |
+
{
|
2126 |
+
"epoch": 0.012241583406392288,
|
2127 |
+
"grad_norm": 1087.690673828125,
|
2128 |
+
"learning_rate": 4.848000000000001e-06,
|
2129 |
+
"loss": 300.9158,
|
2130 |
+
"step": 3030
|
2131 |
+
},
|
2132 |
+
{
|
2133 |
+
"epoch": 0.01228198467175992,
|
2134 |
+
"grad_norm": 1012.8238525390625,
|
2135 |
+
"learning_rate": 4.864e-06,
|
2136 |
+
"loss": 405.1533,
|
2137 |
+
"step": 3040
|
2138 |
+
},
|
2139 |
+
{
|
2140 |
+
"epoch": 0.01232238593712755,
|
2141 |
+
"grad_norm": 1519.6015625,
|
2142 |
+
"learning_rate": 4.880000000000001e-06,
|
2143 |
+
"loss": 340.4835,
|
2144 |
+
"step": 3050
|
2145 |
+
},
|
2146 |
+
{
|
2147 |
+
"epoch": 0.012362787202495182,
|
2148 |
+
"grad_norm": 787.8037109375,
|
2149 |
+
"learning_rate": 4.896e-06,
|
2150 |
+
"loss": 249.336,
|
2151 |
+
"step": 3060
|
2152 |
+
},
|
2153 |
+
{
|
2154 |
+
"epoch": 0.012403188467862813,
|
2155 |
+
"grad_norm": 1030.4306640625,
|
2156 |
+
"learning_rate": 4.9120000000000006e-06,
|
2157 |
+
"loss": 248.3195,
|
2158 |
+
"step": 3070
|
2159 |
+
},
|
2160 |
+
{
|
2161 |
+
"epoch": 0.012443589733230445,
|
2162 |
+
"grad_norm": 1620.253173828125,
|
2163 |
+
"learning_rate": 4.928000000000001e-06,
|
2164 |
+
"loss": 210.1467,
|
2165 |
+
"step": 3080
|
2166 |
+
},
|
2167 |
+
{
|
2168 |
+
"epoch": 0.012483990998598076,
|
2169 |
+
"grad_norm": 2037.361328125,
|
2170 |
+
"learning_rate": 4.9440000000000004e-06,
|
2171 |
+
"loss": 331.2714,
|
2172 |
+
"step": 3090
|
2173 |
+
},
|
2174 |
+
{
|
2175 |
+
"epoch": 0.012524392263965707,
|
2176 |
+
"grad_norm": 4152.18701171875,
|
2177 |
+
"learning_rate": 4.960000000000001e-06,
|
2178 |
+
"loss": 369.5478,
|
2179 |
+
"step": 3100
|
2180 |
+
},
|
2181 |
+
{
|
2182 |
+
"epoch": 0.012564793529333339,
|
2183 |
+
"grad_norm": 751.2320556640625,
|
2184 |
+
"learning_rate": 4.976e-06,
|
2185 |
+
"loss": 339.7245,
|
2186 |
+
"step": 3110
|
2187 |
+
},
|
2188 |
+
{
|
2189 |
+
"epoch": 0.01260519479470097,
|
2190 |
+
"grad_norm": 1463.79833984375,
|
2191 |
+
"learning_rate": 4.992e-06,
|
2192 |
+
"loss": 349.4508,
|
2193 |
+
"step": 3120
|
2194 |
+
},
|
2195 |
+
{
|
2196 |
+
"epoch": 0.012645596060068601,
|
2197 |
+
"grad_norm": 931.794189453125,
|
2198 |
+
"learning_rate": 5.008000000000001e-06,
|
2199 |
+
"loss": 219.2512,
|
2200 |
+
"step": 3130
|
2201 |
+
},
|
2202 |
+
{
|
2203 |
+
"epoch": 0.012685997325436233,
|
2204 |
+
"grad_norm": 587.0802001953125,
|
2205 |
+
"learning_rate": 5.024e-06,
|
2206 |
+
"loss": 249.6072,
|
2207 |
+
"step": 3140
|
2208 |
+
},
|
2209 |
+
{
|
2210 |
+
"epoch": 0.012726398590803864,
|
2211 |
+
"grad_norm": 4629.9833984375,
|
2212 |
+
"learning_rate": 5.04e-06,
|
2213 |
+
"loss": 274.248,
|
2214 |
+
"step": 3150
|
2215 |
+
},
|
2216 |
+
{
|
2217 |
+
"epoch": 0.012766799856171495,
|
2218 |
+
"grad_norm": 1912.8997802734375,
|
2219 |
+
"learning_rate": 5.056000000000001e-06,
|
2220 |
+
"loss": 265.1541,
|
2221 |
+
"step": 3160
|
2222 |
+
},
|
2223 |
+
{
|
2224 |
+
"epoch": 0.012807201121539127,
|
2225 |
+
"grad_norm": 714.0695190429688,
|
2226 |
+
"learning_rate": 5.072e-06,
|
2227 |
+
"loss": 189.4337,
|
2228 |
+
"step": 3170
|
2229 |
+
},
|
2230 |
+
{
|
2231 |
+
"epoch": 0.012847602386906758,
|
2232 |
+
"grad_norm": 1278.3006591796875,
|
2233 |
+
"learning_rate": 5.088000000000001e-06,
|
2234 |
+
"loss": 368.549,
|
2235 |
+
"step": 3180
|
2236 |
+
},
|
2237 |
+
{
|
2238 |
+
"epoch": 0.01288800365227439,
|
2239 |
+
"grad_norm": 749.0835571289062,
|
2240 |
+
"learning_rate": 5.104e-06,
|
2241 |
+
"loss": 228.1246,
|
2242 |
+
"step": 3190
|
2243 |
+
},
|
2244 |
+
{
|
2245 |
+
"epoch": 0.01292840491764202,
|
2246 |
+
"grad_norm": 0.0,
|
2247 |
+
"learning_rate": 5.12e-06,
|
2248 |
+
"loss": 227.3482,
|
2249 |
+
"step": 3200
|
2250 |
+
},
|
2251 |
+
{
|
2252 |
+
"epoch": 0.012968806183009652,
|
2253 |
+
"grad_norm": 888.6599731445312,
|
2254 |
+
"learning_rate": 5.136e-06,
|
2255 |
+
"loss": 310.4809,
|
2256 |
+
"step": 3210
|
2257 |
+
},
|
2258 |
+
{
|
2259 |
+
"epoch": 0.013009207448377283,
|
2260 |
+
"grad_norm": 760.9104614257812,
|
2261 |
+
"learning_rate": 5.152e-06,
|
2262 |
+
"loss": 229.1965,
|
2263 |
+
"step": 3220
|
2264 |
+
},
|
2265 |
+
{
|
2266 |
+
"epoch": 0.013049608713744915,
|
2267 |
+
"grad_norm": 1096.23291015625,
|
2268 |
+
"learning_rate": 5.168000000000001e-06,
|
2269 |
+
"loss": 325.4543,
|
2270 |
+
"step": 3230
|
2271 |
+
},
|
2272 |
+
{
|
2273 |
+
"epoch": 0.013090009979112546,
|
2274 |
+
"grad_norm": 2388.175537109375,
|
2275 |
+
"learning_rate": 5.184e-06,
|
2276 |
+
"loss": 270.4536,
|
2277 |
+
"step": 3240
|
2278 |
+
},
|
2279 |
+
{
|
2280 |
+
"epoch": 0.013130411244480177,
|
2281 |
+
"grad_norm": 7733.79296875,
|
2282 |
+
"learning_rate": 5.2e-06,
|
2283 |
+
"loss": 364.0487,
|
2284 |
+
"step": 3250
|
2285 |
+
},
|
2286 |
+
{
|
2287 |
+
"epoch": 0.013170812509847809,
|
2288 |
+
"grad_norm": 1177.9654541015625,
|
2289 |
+
"learning_rate": 5.216e-06,
|
2290 |
+
"loss": 272.8813,
|
2291 |
+
"step": 3260
|
2292 |
+
},
|
2293 |
+
{
|
2294 |
+
"epoch": 0.01321121377521544,
|
2295 |
+
"grad_norm": 1095.551025390625,
|
2296 |
+
"learning_rate": 5.232e-06,
|
2297 |
+
"loss": 356.0726,
|
2298 |
+
"step": 3270
|
2299 |
+
},
|
2300 |
+
{
|
2301 |
+
"epoch": 0.013251615040583071,
|
2302 |
+
"grad_norm": 1486.8251953125,
|
2303 |
+
"learning_rate": 5.248000000000001e-06,
|
2304 |
+
"loss": 269.6024,
|
2305 |
+
"step": 3280
|
2306 |
+
},
|
2307 |
+
{
|
2308 |
+
"epoch": 0.013292016305950703,
|
2309 |
+
"grad_norm": 1649.369384765625,
|
2310 |
+
"learning_rate": 5.264e-06,
|
2311 |
+
"loss": 406.2197,
|
2312 |
+
"step": 3290
|
2313 |
+
},
|
2314 |
+
{
|
2315 |
+
"epoch": 0.013332417571318334,
|
2316 |
+
"grad_norm": 901.133544921875,
|
2317 |
+
"learning_rate": 5.28e-06,
|
2318 |
+
"loss": 334.5719,
|
2319 |
+
"step": 3300
|
2320 |
+
},
|
2321 |
+
{
|
2322 |
+
"epoch": 0.013372818836685965,
|
2323 |
+
"grad_norm": 1080.5753173828125,
|
2324 |
+
"learning_rate": 5.296e-06,
|
2325 |
+
"loss": 256.0706,
|
2326 |
+
"step": 3310
|
2327 |
+
},
|
2328 |
+
{
|
2329 |
+
"epoch": 0.013413220102053596,
|
2330 |
+
"grad_norm": 903.6046752929688,
|
2331 |
+
"learning_rate": 5.312e-06,
|
2332 |
+
"loss": 229.897,
|
2333 |
+
"step": 3320
|
2334 |
+
},
|
2335 |
+
{
|
2336 |
+
"epoch": 0.013453621367421228,
|
2337 |
+
"grad_norm": 984.148193359375,
|
2338 |
+
"learning_rate": 5.328000000000001e-06,
|
2339 |
+
"loss": 268.5443,
|
2340 |
+
"step": 3330
|
2341 |
+
},
|
2342 |
+
{
|
2343 |
+
"epoch": 0.013494022632788859,
|
2344 |
+
"grad_norm": 1167.3309326171875,
|
2345 |
+
"learning_rate": 5.344e-06,
|
2346 |
+
"loss": 339.0956,
|
2347 |
+
"step": 3340
|
2348 |
+
},
|
2349 |
+
{
|
2350 |
+
"epoch": 0.01353442389815649,
|
2351 |
+
"grad_norm": 2445.7763671875,
|
2352 |
+
"learning_rate": 5.36e-06,
|
2353 |
+
"loss": 237.0434,
|
2354 |
+
"step": 3350
|
2355 |
+
},
|
2356 |
+
{
|
2357 |
+
"epoch": 0.013574825163524122,
|
2358 |
+
"grad_norm": 814.3775634765625,
|
2359 |
+
"learning_rate": 5.376e-06,
|
2360 |
+
"loss": 182.9125,
|
2361 |
+
"step": 3360
|
2362 |
+
},
|
2363 |
+
{
|
2364 |
+
"epoch": 0.013615226428891753,
|
2365 |
+
"grad_norm": 883.6320190429688,
|
2366 |
+
"learning_rate": 5.392e-06,
|
2367 |
+
"loss": 346.1443,
|
2368 |
+
"step": 3370
|
2369 |
+
},
|
2370 |
+
{
|
2371 |
+
"epoch": 0.013655627694259384,
|
2372 |
+
"grad_norm": 788.3290405273438,
|
2373 |
+
"learning_rate": 5.408e-06,
|
2374 |
+
"loss": 143.6525,
|
2375 |
+
"step": 3380
|
2376 |
+
},
|
2377 |
+
{
|
2378 |
+
"epoch": 0.013696028959627016,
|
2379 |
+
"grad_norm": 1109.12744140625,
|
2380 |
+
"learning_rate": 5.424e-06,
|
2381 |
+
"loss": 293.3582,
|
2382 |
+
"step": 3390
|
2383 |
+
},
|
2384 |
+
{
|
2385 |
+
"epoch": 0.013736430224994647,
|
2386 |
+
"grad_norm": 968.6015014648438,
|
2387 |
+
"learning_rate": 5.4400000000000004e-06,
|
2388 |
+
"loss": 228.7989,
|
2389 |
+
"step": 3400
|
2390 |
+
},
|
2391 |
+
{
|
2392 |
+
"epoch": 0.013776831490362278,
|
2393 |
+
"grad_norm": 1213.0040283203125,
|
2394 |
+
"learning_rate": 5.456e-06,
|
2395 |
+
"loss": 432.3414,
|
2396 |
+
"step": 3410
|
2397 |
+
},
|
2398 |
+
{
|
2399 |
+
"epoch": 0.01381723275572991,
|
2400 |
+
"grad_norm": 1326.287109375,
|
2401 |
+
"learning_rate": 5.472e-06,
|
2402 |
+
"loss": 332.6496,
|
2403 |
+
"step": 3420
|
2404 |
+
},
|
2405 |
+
{
|
2406 |
+
"epoch": 0.013857634021097541,
|
2407 |
+
"grad_norm": 6259.54248046875,
|
2408 |
+
"learning_rate": 5.488e-06,
|
2409 |
+
"loss": 271.7205,
|
2410 |
+
"step": 3430
|
2411 |
+
},
|
2412 |
+
{
|
2413 |
+
"epoch": 0.013898035286465172,
|
2414 |
+
"grad_norm": 708.2656860351562,
|
2415 |
+
"learning_rate": 5.504e-06,
|
2416 |
+
"loss": 180.486,
|
2417 |
+
"step": 3440
|
2418 |
+
},
|
2419 |
+
{
|
2420 |
+
"epoch": 0.013938436551832804,
|
2421 |
+
"grad_norm": 685.2205200195312,
|
2422 |
+
"learning_rate": 5.5200000000000005e-06,
|
2423 |
+
"loss": 268.5384,
|
2424 |
+
"step": 3450
|
2425 |
+
},
|
2426 |
+
{
|
2427 |
+
"epoch": 0.013978837817200435,
|
2428 |
+
"grad_norm": 1125.2498779296875,
|
2429 |
+
"learning_rate": 5.536e-06,
|
2430 |
+
"loss": 288.8181,
|
2431 |
+
"step": 3460
|
2432 |
+
},
|
2433 |
+
{
|
2434 |
+
"epoch": 0.014019239082568066,
|
2435 |
+
"grad_norm": 1228.8125,
|
2436 |
+
"learning_rate": 5.552e-06,
|
2437 |
+
"loss": 244.6971,
|
2438 |
+
"step": 3470
|
2439 |
+
},
|
2440 |
+
{
|
2441 |
+
"epoch": 0.014059640347935698,
|
2442 |
+
"grad_norm": 2762.309814453125,
|
2443 |
+
"learning_rate": 5.568e-06,
|
2444 |
+
"loss": 273.8453,
|
2445 |
+
"step": 3480
|
2446 |
+
},
|
2447 |
+
{
|
2448 |
+
"epoch": 0.014100041613303329,
|
2449 |
+
"grad_norm": 945.8040161132812,
|
2450 |
+
"learning_rate": 5.584e-06,
|
2451 |
+
"loss": 240.8805,
|
2452 |
+
"step": 3490
|
2453 |
+
},
|
2454 |
+
{
|
2455 |
+
"epoch": 0.01414044287867096,
|
2456 |
+
"grad_norm": 2207.40185546875,
|
2457 |
+
"learning_rate": 5.600000000000001e-06,
|
2458 |
+
"loss": 165.5966,
|
2459 |
+
"step": 3500
|
2460 |
+
},
|
2461 |
+
{
|
2462 |
+
"epoch": 0.014180844144038592,
|
2463 |
+
"grad_norm": 622.099609375,
|
2464 |
+
"learning_rate": 5.616e-06,
|
2465 |
+
"loss": 202.4418,
|
2466 |
+
"step": 3510
|
2467 |
+
},
|
2468 |
+
{
|
2469 |
+
"epoch": 0.014221245409406223,
|
2470 |
+
"grad_norm": 1473.064208984375,
|
2471 |
+
"learning_rate": 5.6320000000000005e-06,
|
2472 |
+
"loss": 229.2896,
|
2473 |
+
"step": 3520
|
2474 |
+
},
|
2475 |
+
{
|
2476 |
+
"epoch": 0.014261646674773854,
|
2477 |
+
"grad_norm": 1057.5018310546875,
|
2478 |
+
"learning_rate": 5.648e-06,
|
2479 |
+
"loss": 211.9801,
|
2480 |
+
"step": 3530
|
2481 |
+
},
|
2482 |
+
{
|
2483 |
+
"epoch": 0.014302047940141486,
|
2484 |
+
"grad_norm": 4774.0947265625,
|
2485 |
+
"learning_rate": 5.664e-06,
|
2486 |
+
"loss": 204.8896,
|
2487 |
+
"step": 3540
|
2488 |
+
},
|
2489 |
+
{
|
2490 |
+
"epoch": 0.014342449205509117,
|
2491 |
+
"grad_norm": 676.964111328125,
|
2492 |
+
"learning_rate": 5.68e-06,
|
2493 |
+
"loss": 200.4312,
|
2494 |
+
"step": 3550
|
2495 |
+
},
|
2496 |
+
{
|
2497 |
+
"epoch": 0.014382850470876748,
|
2498 |
+
"grad_norm": 571.7015380859375,
|
2499 |
+
"learning_rate": 5.696e-06,
|
2500 |
+
"loss": 275.7348,
|
2501 |
+
"step": 3560
|
2502 |
+
},
|
2503 |
+
{
|
2504 |
+
"epoch": 0.01442325173624438,
|
2505 |
+
"grad_norm": 1923.3953857421875,
|
2506 |
+
"learning_rate": 5.7120000000000005e-06,
|
2507 |
+
"loss": 224.3793,
|
2508 |
+
"step": 3570
|
2509 |
+
},
|
2510 |
+
{
|
2511 |
+
"epoch": 0.01446365300161201,
|
2512 |
+
"grad_norm": 541.0182495117188,
|
2513 |
+
"learning_rate": 5.728e-06,
|
2514 |
+
"loss": 155.7405,
|
2515 |
+
"step": 3580
|
2516 |
+
},
|
2517 |
+
{
|
2518 |
+
"epoch": 0.014504054266979642,
|
2519 |
+
"grad_norm": 1030.9990234375,
|
2520 |
+
"learning_rate": 5.744e-06,
|
2521 |
+
"loss": 225.919,
|
2522 |
+
"step": 3590
|
2523 |
+
},
|
2524 |
+
{
|
2525 |
+
"epoch": 0.014544455532347273,
|
2526 |
+
"grad_norm": 1258.4619140625,
|
2527 |
+
"learning_rate": 5.76e-06,
|
2528 |
+
"loss": 242.7457,
|
2529 |
+
"step": 3600
|
2530 |
+
},
|
2531 |
+
{
|
2532 |
+
"epoch": 0.014584856797714905,
|
2533 |
+
"grad_norm": 1085.32177734375,
|
2534 |
+
"learning_rate": 5.776e-06,
|
2535 |
+
"loss": 251.3692,
|
2536 |
+
"step": 3610
|
2537 |
+
},
|
2538 |
+
{
|
2539 |
+
"epoch": 0.014625258063082536,
|
2540 |
+
"grad_norm": 880.9969482421875,
|
2541 |
+
"learning_rate": 5.792000000000001e-06,
|
2542 |
+
"loss": 228.1045,
|
2543 |
+
"step": 3620
|
2544 |
+
},
|
2545 |
+
{
|
2546 |
+
"epoch": 0.014665659328450167,
|
2547 |
+
"grad_norm": 616.6618041992188,
|
2548 |
+
"learning_rate": 5.808e-06,
|
2549 |
+
"loss": 216.2711,
|
2550 |
+
"step": 3630
|
2551 |
+
},
|
2552 |
+
{
|
2553 |
+
"epoch": 0.014706060593817799,
|
2554 |
+
"grad_norm": 2865.4169921875,
|
2555 |
+
"learning_rate": 5.8240000000000005e-06,
|
2556 |
+
"loss": 253.2795,
|
2557 |
+
"step": 3640
|
2558 |
+
},
|
2559 |
+
{
|
2560 |
+
"epoch": 0.01474646185918543,
|
2561 |
+
"grad_norm": 941.0968627929688,
|
2562 |
+
"learning_rate": 5.84e-06,
|
2563 |
+
"loss": 268.1058,
|
2564 |
+
"step": 3650
|
2565 |
+
},
|
2566 |
+
{
|
2567 |
+
"epoch": 0.014786863124553061,
|
2568 |
+
"grad_norm": 1495.3153076171875,
|
2569 |
+
"learning_rate": 5.856e-06,
|
2570 |
+
"loss": 262.6222,
|
2571 |
+
"step": 3660
|
2572 |
+
},
|
2573 |
+
{
|
2574 |
+
"epoch": 0.014827264389920693,
|
2575 |
+
"grad_norm": 1153.8773193359375,
|
2576 |
+
"learning_rate": 5.872000000000001e-06,
|
2577 |
+
"loss": 314.2995,
|
2578 |
+
"step": 3670
|
2579 |
+
},
|
2580 |
+
{
|
2581 |
+
"epoch": 0.014867665655288324,
|
2582 |
+
"grad_norm": 3272.336669921875,
|
2583 |
+
"learning_rate": 5.888e-06,
|
2584 |
+
"loss": 275.6237,
|
2585 |
+
"step": 3680
|
2586 |
+
},
|
2587 |
+
{
|
2588 |
+
"epoch": 0.014908066920655955,
|
2589 |
+
"grad_norm": 1363.574462890625,
|
2590 |
+
"learning_rate": 5.9040000000000006e-06,
|
2591 |
+
"loss": 408.7946,
|
2592 |
+
"step": 3690
|
2593 |
+
},
|
2594 |
+
{
|
2595 |
+
"epoch": 0.014948468186023587,
|
2596 |
+
"grad_norm": 1989.05810546875,
|
2597 |
+
"learning_rate": 5.92e-06,
|
2598 |
+
"loss": 347.1884,
|
2599 |
+
"step": 3700
|
2600 |
+
},
|
2601 |
+
{
|
2602 |
+
"epoch": 0.014988869451391218,
|
2603 |
+
"grad_norm": 893.3648071289062,
|
2604 |
+
"learning_rate": 5.9360000000000004e-06,
|
2605 |
+
"loss": 231.8197,
|
2606 |
+
"step": 3710
|
2607 |
+
},
|
2608 |
+
{
|
2609 |
+
"epoch": 0.01502927071675885,
|
2610 |
+
"grad_norm": 2015.3258056640625,
|
2611 |
+
"learning_rate": 5.952e-06,
|
2612 |
+
"loss": 241.9273,
|
2613 |
+
"step": 3720
|
2614 |
+
},
|
2615 |
+
{
|
2616 |
+
"epoch": 0.01506967198212648,
|
2617 |
+
"grad_norm": 1097.1138916015625,
|
2618 |
+
"learning_rate": 5.968e-06,
|
2619 |
+
"loss": 289.7548,
|
2620 |
+
"step": 3730
|
2621 |
+
},
|
2622 |
+
{
|
2623 |
+
"epoch": 0.015110073247494112,
|
2624 |
+
"grad_norm": 1198.3504638671875,
|
2625 |
+
"learning_rate": 5.984000000000001e-06,
|
2626 |
+
"loss": 257.8569,
|
2627 |
+
"step": 3740
|
2628 |
+
},
|
2629 |
+
{
|
2630 |
+
"epoch": 0.015150474512861743,
|
2631 |
+
"grad_norm": 2764.9599609375,
|
2632 |
+
"learning_rate": 6e-06,
|
2633 |
+
"loss": 394.4565,
|
2634 |
+
"step": 3750
|
2635 |
+
},
|
2636 |
+
{
|
2637 |
+
"epoch": 0.015190875778229375,
|
2638 |
+
"grad_norm": 758.348876953125,
|
2639 |
+
"learning_rate": 6.0160000000000005e-06,
|
2640 |
+
"loss": 249.2975,
|
2641 |
+
"step": 3760
|
2642 |
+
},
|
2643 |
+
{
|
2644 |
+
"epoch": 0.015231277043597006,
|
2645 |
+
"grad_norm": 2277.13818359375,
|
2646 |
+
"learning_rate": 6.032e-06,
|
2647 |
+
"loss": 294.9143,
|
2648 |
+
"step": 3770
|
2649 |
+
},
|
2650 |
+
{
|
2651 |
+
"epoch": 0.015271678308964637,
|
2652 |
+
"grad_norm": 2135.546142578125,
|
2653 |
+
"learning_rate": 6.048e-06,
|
2654 |
+
"loss": 235.1545,
|
2655 |
+
"step": 3780
|
2656 |
+
},
|
2657 |
+
{
|
2658 |
+
"epoch": 0.015312079574332269,
|
2659 |
+
"grad_norm": 1097.388916015625,
|
2660 |
+
"learning_rate": 6.064000000000001e-06,
|
2661 |
+
"loss": 241.006,
|
2662 |
+
"step": 3790
|
2663 |
+
},
|
2664 |
+
{
|
2665 |
+
"epoch": 0.0153524808396999,
|
2666 |
+
"grad_norm": 943.5238647460938,
|
2667 |
+
"learning_rate": 6.08e-06,
|
2668 |
+
"loss": 174.3143,
|
2669 |
+
"step": 3800
|
2670 |
+
},
|
2671 |
+
{
|
2672 |
+
"epoch": 0.015392882105067531,
|
2673 |
+
"grad_norm": 1239.0521240234375,
|
2674 |
+
"learning_rate": 6.096000000000001e-06,
|
2675 |
+
"loss": 307.3331,
|
2676 |
+
"step": 3810
|
2677 |
+
},
|
2678 |
+
{
|
2679 |
+
"epoch": 0.015433283370435162,
|
2680 |
+
"grad_norm": 643.6400146484375,
|
2681 |
+
"learning_rate": 6.112e-06,
|
2682 |
+
"loss": 217.4633,
|
2683 |
+
"step": 3820
|
2684 |
+
},
|
2685 |
+
{
|
2686 |
+
"epoch": 0.015473684635802794,
|
2687 |
+
"grad_norm": 1569.7869873046875,
|
2688 |
+
"learning_rate": 6.1280000000000005e-06,
|
2689 |
+
"loss": 322.4538,
|
2690 |
+
"step": 3830
|
2691 |
+
},
|
2692 |
+
{
|
2693 |
+
"epoch": 0.015514085901170425,
|
2694 |
+
"grad_norm": 3763.859130859375,
|
2695 |
+
"learning_rate": 6.144e-06,
|
2696 |
+
"loss": 182.1174,
|
2697 |
+
"step": 3840
|
2698 |
+
},
|
2699 |
+
{
|
2700 |
+
"epoch": 0.015554487166538056,
|
2701 |
+
"grad_norm": 717.2794799804688,
|
2702 |
+
"learning_rate": 6.16e-06,
|
2703 |
+
"loss": 207.8513,
|
2704 |
+
"step": 3850
|
2705 |
+
},
|
2706 |
+
{
|
2707 |
+
"epoch": 0.015594888431905688,
|
2708 |
+
"grad_norm": 1418.3277587890625,
|
2709 |
+
"learning_rate": 6.176000000000001e-06,
|
2710 |
+
"loss": 254.4121,
|
2711 |
+
"step": 3860
|
2712 |
+
},
|
2713 |
+
{
|
2714 |
+
"epoch": 0.01563528969727332,
|
2715 |
+
"grad_norm": 1528.9119873046875,
|
2716 |
+
"learning_rate": 6.192e-06,
|
2717 |
+
"loss": 223.6123,
|
2718 |
+
"step": 3870
|
2719 |
+
},
|
2720 |
+
{
|
2721 |
+
"epoch": 0.01567569096264095,
|
2722 |
+
"grad_norm": 544.6751708984375,
|
2723 |
+
"learning_rate": 6.2080000000000005e-06,
|
2724 |
+
"loss": 222.1293,
|
2725 |
+
"step": 3880
|
2726 |
+
},
|
2727 |
+
{
|
2728 |
+
"epoch": 0.01571609222800858,
|
2729 |
+
"grad_norm": 754.9981079101562,
|
2730 |
+
"learning_rate": 6.224e-06,
|
2731 |
+
"loss": 255.1399,
|
2732 |
+
"step": 3890
|
2733 |
+
},
|
2734 |
+
{
|
2735 |
+
"epoch": 0.01575649349337621,
|
2736 |
+
"grad_norm": 1966.5140380859375,
|
2737 |
+
"learning_rate": 6.24e-06,
|
2738 |
+
"loss": 275.0173,
|
2739 |
+
"step": 3900
|
2740 |
+
},
|
2741 |
+
{
|
2742 |
+
"epoch": 0.015796894758743844,
|
2743 |
+
"grad_norm": 810.1495361328125,
|
2744 |
+
"learning_rate": 6.256000000000001e-06,
|
2745 |
+
"loss": 219.6732,
|
2746 |
+
"step": 3910
|
2747 |
+
},
|
2748 |
+
{
|
2749 |
+
"epoch": 0.015837296024111474,
|
2750 |
+
"grad_norm": 3056.527099609375,
|
2751 |
+
"learning_rate": 6.272e-06,
|
2752 |
+
"loss": 184.322,
|
2753 |
+
"step": 3920
|
2754 |
+
},
|
2755 |
+
{
|
2756 |
+
"epoch": 0.015877697289479107,
|
2757 |
+
"grad_norm": 2820.2568359375,
|
2758 |
+
"learning_rate": 6.288000000000001e-06,
|
2759 |
+
"loss": 235.3666,
|
2760 |
+
"step": 3930
|
2761 |
+
},
|
2762 |
+
{
|
2763 |
+
"epoch": 0.015918098554846737,
|
2764 |
+
"grad_norm": 1320.5225830078125,
|
2765 |
+
"learning_rate": 6.304e-06,
|
2766 |
+
"loss": 220.3165,
|
2767 |
+
"step": 3940
|
2768 |
+
},
|
2769 |
+
{
|
2770 |
+
"epoch": 0.01595849982021437,
|
2771 |
+
"grad_norm": 933.7466430664062,
|
2772 |
+
"learning_rate": 6.3200000000000005e-06,
|
2773 |
+
"loss": 307.4496,
|
2774 |
+
"step": 3950
|
2775 |
+
},
|
2776 |
+
{
|
2777 |
+
"epoch": 0.015998901085582,
|
2778 |
+
"grad_norm": 867.0122680664062,
|
2779 |
+
"learning_rate": 6.336000000000001e-06,
|
2780 |
+
"loss": 263.1927,
|
2781 |
+
"step": 3960
|
2782 |
+
},
|
2783 |
+
{
|
2784 |
+
"epoch": 0.016039302350949632,
|
2785 |
+
"grad_norm": 747.3133544921875,
|
2786 |
+
"learning_rate": 6.352e-06,
|
2787 |
+
"loss": 306.4962,
|
2788 |
+
"step": 3970
|
2789 |
+
},
|
2790 |
+
{
|
2791 |
+
"epoch": 0.016079703616317262,
|
2792 |
+
"grad_norm": 711.7617797851562,
|
2793 |
+
"learning_rate": 6.368000000000001e-06,
|
2794 |
+
"loss": 243.1562,
|
2795 |
+
"step": 3980
|
2796 |
+
},
|
2797 |
+
{
|
2798 |
+
"epoch": 0.016120104881684895,
|
2799 |
+
"grad_norm": 732.2003173828125,
|
2800 |
+
"learning_rate": 6.384e-06,
|
2801 |
+
"loss": 231.337,
|
2802 |
+
"step": 3990
|
2803 |
+
},
|
2804 |
+
{
|
2805 |
+
"epoch": 0.016160506147052525,
|
2806 |
+
"grad_norm": 582.39990234375,
|
2807 |
+
"learning_rate": 6.4000000000000006e-06,
|
2808 |
+
"loss": 273.9559,
|
2809 |
+
"step": 4000
|
2810 |
+
}
|
2811 |
+
],
|
2812 |
+
"logging_steps": 10,
|
2813 |
+
"max_steps": 250000,
|
2814 |
+
"num_input_tokens_seen": 0,
|
2815 |
+
"num_train_epochs": 2,
|
2816 |
+
"save_steps": 4000,
|
2817 |
+
"stateful_callbacks": {
|
2818 |
+
"TrainerControl": {
|
2819 |
+
"args": {
|
2820 |
+
"should_epoch_stop": false,
|
2821 |
+
"should_evaluate": false,
|
2822 |
+
"should_log": false,
|
2823 |
+
"should_save": true,
|
2824 |
+
"should_training_stop": false
|
2825 |
+
},
|
2826 |
+
"attributes": {}
|
2827 |
+
}
|
2828 |
+
},
|
2829 |
+
"total_flos": 0.0,
|
2830 |
+
"train_batch_size": 4,
|
2831 |
+
"trial_name": null,
|
2832 |
+
"trial_params": null
|
2833 |
+
}
|