Upload folder using huggingface_hub
Browse files- added_tokens.json +3 -0
- gliner_config.json +216 -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 +1433 -0
added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"[MASK]": 128000
|
| 3 |
+
}
|
gliner_config.json
ADDED
|
@@ -0,0 +1,216 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"class_token_index": -1,
|
| 3 |
+
"dropout": 0.3,
|
| 4 |
+
"embed_ent_token": true,
|
| 5 |
+
"encoder_config": {
|
| 6 |
+
"_name_or_path": "microsoft/deberta-v3-base",
|
| 7 |
+
"add_cross_attention": false,
|
| 8 |
+
"architectures": null,
|
| 9 |
+
"attention_probs_dropout_prob": 0.1,
|
| 10 |
+
"bad_words_ids": null,
|
| 11 |
+
"begin_suppress_tokens": null,
|
| 12 |
+
"bos_token_id": null,
|
| 13 |
+
"chunk_size_feed_forward": 0,
|
| 14 |
+
"cross_attention_hidden_size": null,
|
| 15 |
+
"decoder_start_token_id": null,
|
| 16 |
+
"diversity_penalty": 0.0,
|
| 17 |
+
"do_sample": false,
|
| 18 |
+
"early_stopping": false,
|
| 19 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 20 |
+
"eos_token_id": null,
|
| 21 |
+
"exponential_decay_length_penalty": null,
|
| 22 |
+
"finetuning_task": null,
|
| 23 |
+
"forced_bos_token_id": null,
|
| 24 |
+
"forced_eos_token_id": null,
|
| 25 |
+
"hidden_act": "gelu",
|
| 26 |
+
"hidden_dropout_prob": 0.1,
|
| 27 |
+
"hidden_size": 768,
|
| 28 |
+
"id2label": {
|
| 29 |
+
"0": "LABEL_0",
|
| 30 |
+
"1": "LABEL_1"
|
| 31 |
+
},
|
| 32 |
+
"initializer_range": 0.02,
|
| 33 |
+
"intermediate_size": 3072,
|
| 34 |
+
"is_decoder": false,
|
| 35 |
+
"is_encoder_decoder": false,
|
| 36 |
+
"label2id": {
|
| 37 |
+
"LABEL_0": 0,
|
| 38 |
+
"LABEL_1": 1
|
| 39 |
+
},
|
| 40 |
+
"layer_norm_eps": 1e-07,
|
| 41 |
+
"length_penalty": 1.0,
|
| 42 |
+
"max_length": 20,
|
| 43 |
+
"max_position_embeddings": 512,
|
| 44 |
+
"max_relative_positions": -1,
|
| 45 |
+
"min_length": 0,
|
| 46 |
+
"model_type": "deberta-v2",
|
| 47 |
+
"no_repeat_ngram_size": 0,
|
| 48 |
+
"norm_rel_ebd": "layer_norm",
|
| 49 |
+
"num_attention_heads": 12,
|
| 50 |
+
"num_beam_groups": 1,
|
| 51 |
+
"num_beams": 1,
|
| 52 |
+
"num_hidden_layers": 12,
|
| 53 |
+
"num_return_sequences": 1,
|
| 54 |
+
"output_attentions": false,
|
| 55 |
+
"output_hidden_states": false,
|
| 56 |
+
"output_scores": false,
|
| 57 |
+
"pad_token_id": 0,
|
| 58 |
+
"pooler_dropout": 0,
|
| 59 |
+
"pooler_hidden_act": "gelu",
|
| 60 |
+
"pooler_hidden_size": 768,
|
| 61 |
+
"pos_att_type": [
|
| 62 |
+
"p2c",
|
| 63 |
+
"c2p"
|
| 64 |
+
],
|
| 65 |
+
"position_biased_input": false,
|
| 66 |
+
"position_buckets": 256,
|
| 67 |
+
"prefix": null,
|
| 68 |
+
"problem_type": null,
|
| 69 |
+
"pruned_heads": {},
|
| 70 |
+
"relative_attention": true,
|
| 71 |
+
"remove_invalid_values": false,
|
| 72 |
+
"repetition_penalty": 1.0,
|
| 73 |
+
"return_dict": true,
|
| 74 |
+
"return_dict_in_generate": false,
|
| 75 |
+
"sep_token_id": null,
|
| 76 |
+
"share_att_key": true,
|
| 77 |
+
"suppress_tokens": null,
|
| 78 |
+
"task_specific_params": null,
|
| 79 |
+
"temperature": 1.0,
|
| 80 |
+
"tf_legacy_loss": false,
|
| 81 |
+
"tie_encoder_decoder": false,
|
| 82 |
+
"tie_word_embeddings": true,
|
| 83 |
+
"tokenizer_class": null,
|
| 84 |
+
"top_k": 50,
|
| 85 |
+
"top_p": 1.0,
|
| 86 |
+
"torch_dtype": null,
|
| 87 |
+
"torchscript": false,
|
| 88 |
+
"type_vocab_size": 0,
|
| 89 |
+
"typical_p": 1.0,
|
| 90 |
+
"use_bfloat16": false,
|
| 91 |
+
"vocab_size": 128100
|
| 92 |
+
},
|
| 93 |
+
"ent_token": "<<ENT>>",
|
| 94 |
+
"eval_every": 1000,
|
| 95 |
+
"fine_tune": true,
|
| 96 |
+
"freeze_token_rep": false,
|
| 97 |
+
"fuse_layers": false,
|
| 98 |
+
"has_rnn": true,
|
| 99 |
+
"hidden_size": 768,
|
| 100 |
+
"label_smoothing": 0,
|
| 101 |
+
"labels_encoder": "BAAI/bge-small-en-v1.5",
|
| 102 |
+
"labels_encoder_config": {
|
| 103 |
+
"_name_or_path": "BAAI/bge-small-en-v1.5",
|
| 104 |
+
"add_cross_attention": false,
|
| 105 |
+
"architectures": [
|
| 106 |
+
"BertModel"
|
| 107 |
+
],
|
| 108 |
+
"attention_probs_dropout_prob": 0.1,
|
| 109 |
+
"bad_words_ids": null,
|
| 110 |
+
"begin_suppress_tokens": null,
|
| 111 |
+
"bos_token_id": null,
|
| 112 |
+
"chunk_size_feed_forward": 0,
|
| 113 |
+
"classifier_dropout": null,
|
| 114 |
+
"cross_attention_hidden_size": null,
|
| 115 |
+
"decoder_start_token_id": null,
|
| 116 |
+
"diversity_penalty": 0.0,
|
| 117 |
+
"do_sample": false,
|
| 118 |
+
"early_stopping": false,
|
| 119 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 120 |
+
"eos_token_id": null,
|
| 121 |
+
"exponential_decay_length_penalty": null,
|
| 122 |
+
"finetuning_task": null,
|
| 123 |
+
"forced_bos_token_id": null,
|
| 124 |
+
"forced_eos_token_id": null,
|
| 125 |
+
"hidden_act": "gelu",
|
| 126 |
+
"hidden_dropout_prob": 0.1,
|
| 127 |
+
"hidden_size": 384,
|
| 128 |
+
"id2label": {
|
| 129 |
+
"0": "LABEL_0"
|
| 130 |
+
},
|
| 131 |
+
"initializer_range": 0.02,
|
| 132 |
+
"intermediate_size": 1536,
|
| 133 |
+
"is_decoder": false,
|
| 134 |
+
"is_encoder_decoder": false,
|
| 135 |
+
"label2id": {
|
| 136 |
+
"LABEL_0": 0
|
| 137 |
+
},
|
| 138 |
+
"layer_norm_eps": 1e-12,
|
| 139 |
+
"length_penalty": 1.0,
|
| 140 |
+
"max_length": 20,
|
| 141 |
+
"max_position_embeddings": 512,
|
| 142 |
+
"min_length": 0,
|
| 143 |
+
"model_type": "bert",
|
| 144 |
+
"no_repeat_ngram_size": 0,
|
| 145 |
+
"num_attention_heads": 12,
|
| 146 |
+
"num_beam_groups": 1,
|
| 147 |
+
"num_beams": 1,
|
| 148 |
+
"num_hidden_layers": 12,
|
| 149 |
+
"num_return_sequences": 1,
|
| 150 |
+
"output_attentions": false,
|
| 151 |
+
"output_hidden_states": false,
|
| 152 |
+
"output_scores": false,
|
| 153 |
+
"pad_token_id": 0,
|
| 154 |
+
"position_embedding_type": "absolute",
|
| 155 |
+
"prefix": null,
|
| 156 |
+
"problem_type": null,
|
| 157 |
+
"pruned_heads": {},
|
| 158 |
+
"remove_invalid_values": false,
|
| 159 |
+
"repetition_penalty": 1.0,
|
| 160 |
+
"return_dict": true,
|
| 161 |
+
"return_dict_in_generate": false,
|
| 162 |
+
"sep_token_id": null,
|
| 163 |
+
"suppress_tokens": null,
|
| 164 |
+
"task_specific_params": null,
|
| 165 |
+
"temperature": 1.0,
|
| 166 |
+
"tf_legacy_loss": false,
|
| 167 |
+
"tie_encoder_decoder": false,
|
| 168 |
+
"tie_word_embeddings": true,
|
| 169 |
+
"tokenizer_class": null,
|
| 170 |
+
"top_k": 50,
|
| 171 |
+
"top_p": 1.0,
|
| 172 |
+
"torch_dtype": "float32",
|
| 173 |
+
"torchscript": false,
|
| 174 |
+
"type_vocab_size": 2,
|
| 175 |
+
"typical_p": 1.0,
|
| 176 |
+
"use_bfloat16": false,
|
| 177 |
+
"use_cache": true,
|
| 178 |
+
"vocab_size": 30522
|
| 179 |
+
},
|
| 180 |
+
"log_dir": "models-base_2/",
|
| 181 |
+
"loss_alpha": -1,
|
| 182 |
+
"loss_gamma": 0,
|
| 183 |
+
"loss_reduction": "sum",
|
| 184 |
+
"lr_encoder": "1e-5",
|
| 185 |
+
"lr_others": "5e-5",
|
| 186 |
+
"max_grad_norm": 10.0,
|
| 187 |
+
"max_len": 512,
|
| 188 |
+
"max_neg_type_ratio": 1,
|
| 189 |
+
"max_types": 128,
|
| 190 |
+
"max_width": 12,
|
| 191 |
+
"model_name": "microsoft/deberta-v3-base",
|
| 192 |
+
"model_type": "gliner",
|
| 193 |
+
"name": "span level gliner for entity linking",
|
| 194 |
+
"num_post_fusion_layers": 1,
|
| 195 |
+
"num_steps": 123750,
|
| 196 |
+
"post_fusion_schema": "",
|
| 197 |
+
"prev_path": null,
|
| 198 |
+
"random_drop": true,
|
| 199 |
+
"root_dir": "gliner_logs",
|
| 200 |
+
"save_total_limit": 3,
|
| 201 |
+
"scheduler_type": "cosine",
|
| 202 |
+
"sep_token": "<<SEP>>",
|
| 203 |
+
"shuffle_types": true,
|
| 204 |
+
"size_sup": -1,
|
| 205 |
+
"span_mode": "markerV0",
|
| 206 |
+
"subtoken_pooling": "first",
|
| 207 |
+
"train_batch_size": 8,
|
| 208 |
+
"train_data": "dataset_with_negatives.json",
|
| 209 |
+
"transformers_version": "4.45.2",
|
| 210 |
+
"val_data_dir": "none",
|
| 211 |
+
"vocab_size": -1,
|
| 212 |
+
"warmup_ratio": 0.1,
|
| 213 |
+
"weight_decay_encoder": 0.01,
|
| 214 |
+
"weight_decay_other": 0.01,
|
| 215 |
+
"words_splitter_type": "whitespace"
|
| 216 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d3b519a6bb1f4b649d1159f7eeb7dd370db30902da368687e8ab51981195b71f
|
| 3 |
+
size 1909053417
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:77c8331eca9d673588c51e6517e0bc71a24da3e050b6985bca487ea4c72d8e18
|
| 3 |
+
size 969281034
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fbc9e1f1f70526b118ed4ee14260d0cbc1d7dd38754828e022aee68be3ff5248
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f45166d22f482fc3bf0154a2dfde8eb06f603e94b9c8dbca4161b7386c3b7e1b
|
| 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 +1,1433 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.016160440856826573,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 2000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 8.080220428413287e-05,
|
| 13 |
+
"grad_norm": 4661457.0,
|
| 14 |
+
"learning_rate": 4.040404040404041e-08,
|
| 15 |
+
"loss": 678012.1,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.00016160440856826573,
|
| 20 |
+
"grad_norm": 9865970.0,
|
| 21 |
+
"learning_rate": 8.080808080808082e-08,
|
| 22 |
+
"loss": 841980.1,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.00024240661285239863,
|
| 27 |
+
"grad_norm": 10027139.0,
|
| 28 |
+
"learning_rate": 1.2121212121212122e-07,
|
| 29 |
+
"loss": 816764.25,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.00032320881713653147,
|
| 34 |
+
"grad_norm": 16910428.0,
|
| 35 |
+
"learning_rate": 1.6161616161616163e-07,
|
| 36 |
+
"loss": 961656.7,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.00040401102142066436,
|
| 41 |
+
"grad_norm": 5068422.0,
|
| 42 |
+
"learning_rate": 2.0202020202020202e-07,
|
| 43 |
+
"loss": 679086.95,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.00048481322570479725,
|
| 48 |
+
"grad_norm": 2293057.5,
|
| 49 |
+
"learning_rate": 2.4242424242424244e-07,
|
| 50 |
+
"loss": 949613.2,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.0005656154299889301,
|
| 55 |
+
"grad_norm": 7849363.5,
|
| 56 |
+
"learning_rate": 2.8282828282828283e-07,
|
| 57 |
+
"loss": 859798.0,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.0006464176342730629,
|
| 62 |
+
"grad_norm": 2669786.75,
|
| 63 |
+
"learning_rate": 3.2323232323232327e-07,
|
| 64 |
+
"loss": 863443.9,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.0007272198385571959,
|
| 69 |
+
"grad_norm": 6021406.5,
|
| 70 |
+
"learning_rate": 3.6363636363636366e-07,
|
| 71 |
+
"loss": 734401.3,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.0008080220428413287,
|
| 76 |
+
"grad_norm": 5000674.5,
|
| 77 |
+
"learning_rate": 4.0404040404040405e-07,
|
| 78 |
+
"loss": 526539.8,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.0008888242471254616,
|
| 83 |
+
"grad_norm": 4362635.5,
|
| 84 |
+
"learning_rate": 4.444444444444445e-07,
|
| 85 |
+
"loss": 422810.45,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.0009696264514095945,
|
| 90 |
+
"grad_norm": 5343355.0,
|
| 91 |
+
"learning_rate": 4.848484848484849e-07,
|
| 92 |
+
"loss": 490438.55,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.0010504286556937273,
|
| 97 |
+
"grad_norm": 11366241.0,
|
| 98 |
+
"learning_rate": 5.252525252525253e-07,
|
| 99 |
+
"loss": 704809.3,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.0011312308599778602,
|
| 104 |
+
"grad_norm": 5520147.0,
|
| 105 |
+
"learning_rate": 5.656565656565657e-07,
|
| 106 |
+
"loss": 526687.5,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.001212033064261993,
|
| 111 |
+
"grad_norm": 3778741.0,
|
| 112 |
+
"learning_rate": 6.060606060606061e-07,
|
| 113 |
+
"loss": 322092.525,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.0012928352685461259,
|
| 118 |
+
"grad_norm": 3378473.0,
|
| 119 |
+
"learning_rate": 6.464646464646465e-07,
|
| 120 |
+
"loss": 516098.85,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.0013736374728302587,
|
| 125 |
+
"grad_norm": 4408531.0,
|
| 126 |
+
"learning_rate": 6.868686868686869e-07,
|
| 127 |
+
"loss": 410817.35,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.0014544396771143918,
|
| 132 |
+
"grad_norm": 766378.375,
|
| 133 |
+
"learning_rate": 7.272727272727273e-07,
|
| 134 |
+
"loss": 197623.1625,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.0015352418813985246,
|
| 139 |
+
"grad_norm": 5468394.5,
|
| 140 |
+
"learning_rate": 7.676767676767678e-07,
|
| 141 |
+
"loss": 238133.6,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.0016160440856826574,
|
| 146 |
+
"grad_norm": 7980194.5,
|
| 147 |
+
"learning_rate": 8.080808080808081e-07,
|
| 148 |
+
"loss": 182498.725,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.0016968462899667903,
|
| 153 |
+
"grad_norm": 737060.125,
|
| 154 |
+
"learning_rate": 8.484848484848486e-07,
|
| 155 |
+
"loss": 122880.075,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.0017776484942509231,
|
| 160 |
+
"grad_norm": 476725.8125,
|
| 161 |
+
"learning_rate": 8.88888888888889e-07,
|
| 162 |
+
"loss": 100649.7563,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.001858450698535056,
|
| 167 |
+
"grad_norm": 353791.875,
|
| 168 |
+
"learning_rate": 9.292929292929294e-07,
|
| 169 |
+
"loss": 58150.4563,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.001939252902819189,
|
| 174 |
+
"grad_norm": 684820.5625,
|
| 175 |
+
"learning_rate": 9.696969696969698e-07,
|
| 176 |
+
"loss": 44040.125,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.0020200551071033216,
|
| 181 |
+
"grad_norm": 228699.796875,
|
| 182 |
+
"learning_rate": 1.0101010101010103e-06,
|
| 183 |
+
"loss": 21505.6156,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.0021008573113874547,
|
| 188 |
+
"grad_norm": 91799.625,
|
| 189 |
+
"learning_rate": 1.0505050505050506e-06,
|
| 190 |
+
"loss": 11209.5648,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.0021816595156715873,
|
| 195 |
+
"grad_norm": 64077.56640625,
|
| 196 |
+
"learning_rate": 1.090909090909091e-06,
|
| 197 |
+
"loss": 5405.107,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.0022624617199557204,
|
| 202 |
+
"grad_norm": 68140.5078125,
|
| 203 |
+
"learning_rate": 1.1313131313131313e-06,
|
| 204 |
+
"loss": 2509.3971,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.0023432639242398534,
|
| 209 |
+
"grad_norm": 38560.15625,
|
| 210 |
+
"learning_rate": 1.1717171717171719e-06,
|
| 211 |
+
"loss": 1295.3724,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.002424066128523986,
|
| 216 |
+
"grad_norm": 2864.86474609375,
|
| 217 |
+
"learning_rate": 1.2121212121212122e-06,
|
| 218 |
+
"loss": 923.2405,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.002504868332808119,
|
| 223 |
+
"grad_norm": 2044.932861328125,
|
| 224 |
+
"learning_rate": 1.2525252525252527e-06,
|
| 225 |
+
"loss": 950.31,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.0025856705370922517,
|
| 230 |
+
"grad_norm": 2294.186279296875,
|
| 231 |
+
"learning_rate": 1.292929292929293e-06,
|
| 232 |
+
"loss": 830.1033,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.002666472741376385,
|
| 237 |
+
"grad_norm": 941.1571044921875,
|
| 238 |
+
"learning_rate": 1.3333333333333334e-06,
|
| 239 |
+
"loss": 711.3893,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.0027472749456605174,
|
| 244 |
+
"grad_norm": 6470.8056640625,
|
| 245 |
+
"learning_rate": 1.3737373737373738e-06,
|
| 246 |
+
"loss": 746.1014,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.0028280771499446505,
|
| 251 |
+
"grad_norm": 2412.6904296875,
|
| 252 |
+
"learning_rate": 1.4141414141414143e-06,
|
| 253 |
+
"loss": 774.8737,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.0029088793542287835,
|
| 258 |
+
"grad_norm": 2536.60791015625,
|
| 259 |
+
"learning_rate": 1.4545454545454546e-06,
|
| 260 |
+
"loss": 832.4265,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.002989681558512916,
|
| 265 |
+
"grad_norm": 5404.72265625,
|
| 266 |
+
"learning_rate": 1.4949494949494952e-06,
|
| 267 |
+
"loss": 676.5023,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.003070483762797049,
|
| 272 |
+
"grad_norm": 1271.8477783203125,
|
| 273 |
+
"learning_rate": 1.5353535353535355e-06,
|
| 274 |
+
"loss": 518.433,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.003151285967081182,
|
| 279 |
+
"grad_norm": 736.135009765625,
|
| 280 |
+
"learning_rate": 1.5757575757575759e-06,
|
| 281 |
+
"loss": 768.4572,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.003232088171365315,
|
| 286 |
+
"grad_norm": 1238.1878662109375,
|
| 287 |
+
"learning_rate": 1.6161616161616162e-06,
|
| 288 |
+
"loss": 696.9558,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.0033128903756494475,
|
| 293 |
+
"grad_norm": 1593.6436767578125,
|
| 294 |
+
"learning_rate": 1.6565656565656565e-06,
|
| 295 |
+
"loss": 670.7958,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.0033936925799335806,
|
| 300 |
+
"grad_norm": 819.6708374023438,
|
| 301 |
+
"learning_rate": 1.6969696969696973e-06,
|
| 302 |
+
"loss": 581.9092,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.0034744947842177136,
|
| 307 |
+
"grad_norm": 775.354248046875,
|
| 308 |
+
"learning_rate": 1.7373737373737376e-06,
|
| 309 |
+
"loss": 744.2987,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.0035552969885018462,
|
| 314 |
+
"grad_norm": 744.1176147460938,
|
| 315 |
+
"learning_rate": 1.777777777777778e-06,
|
| 316 |
+
"loss": 665.0154,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.0036360991927859793,
|
| 321 |
+
"grad_norm": 1123.35986328125,
|
| 322 |
+
"learning_rate": 1.818181818181818e-06,
|
| 323 |
+
"loss": 681.7526,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.003716901397070112,
|
| 328 |
+
"grad_norm": 2373.274658203125,
|
| 329 |
+
"learning_rate": 1.8585858585858588e-06,
|
| 330 |
+
"loss": 654.0196,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.003797703601354245,
|
| 335 |
+
"grad_norm": 1098.4403076171875,
|
| 336 |
+
"learning_rate": 1.8989898989898992e-06,
|
| 337 |
+
"loss": 657.3389,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.003878505805638378,
|
| 342 |
+
"grad_norm": 1008.40966796875,
|
| 343 |
+
"learning_rate": 1.9393939393939395e-06,
|
| 344 |
+
"loss": 564.041,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.003959308009922511,
|
| 349 |
+
"grad_norm": 1227.1414794921875,
|
| 350 |
+
"learning_rate": 1.9797979797979796e-06,
|
| 351 |
+
"loss": 614.5699,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.004040110214206643,
|
| 356 |
+
"grad_norm": 57328.9609375,
|
| 357 |
+
"learning_rate": 2.0202020202020206e-06,
|
| 358 |
+
"loss": 596.2117,
|
| 359 |
+
"step": 500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.004120912418490777,
|
| 363 |
+
"grad_norm": 2744.44873046875,
|
| 364 |
+
"learning_rate": 2.0606060606060607e-06,
|
| 365 |
+
"loss": 676.9374,
|
| 366 |
+
"step": 510
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.004201714622774909,
|
| 370 |
+
"grad_norm": 667.0958862304688,
|
| 371 |
+
"learning_rate": 2.1010101010101013e-06,
|
| 372 |
+
"loss": 812.1784,
|
| 373 |
+
"step": 520
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.004282516827059042,
|
| 377 |
+
"grad_norm": 1447.203857421875,
|
| 378 |
+
"learning_rate": 2.1414141414141414e-06,
|
| 379 |
+
"loss": 640.1749,
|
| 380 |
+
"step": 530
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.004363319031343175,
|
| 384 |
+
"grad_norm": 1186.56982421875,
|
| 385 |
+
"learning_rate": 2.181818181818182e-06,
|
| 386 |
+
"loss": 665.659,
|
| 387 |
+
"step": 540
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.004444121235627308,
|
| 391 |
+
"grad_norm": 1263.9395751953125,
|
| 392 |
+
"learning_rate": 2.2222222222222225e-06,
|
| 393 |
+
"loss": 696.9437,
|
| 394 |
+
"step": 550
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.004524923439911441,
|
| 398 |
+
"grad_norm": 1353.597412109375,
|
| 399 |
+
"learning_rate": 2.2626262626262626e-06,
|
| 400 |
+
"loss": 654.0818,
|
| 401 |
+
"step": 560
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.004605725644195573,
|
| 405 |
+
"grad_norm": 876.721923828125,
|
| 406 |
+
"learning_rate": 2.303030303030303e-06,
|
| 407 |
+
"loss": 683.512,
|
| 408 |
+
"step": 570
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.004686527848479707,
|
| 412 |
+
"grad_norm": 1866.14697265625,
|
| 413 |
+
"learning_rate": 2.3434343434343437e-06,
|
| 414 |
+
"loss": 727.845,
|
| 415 |
+
"step": 580
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.0047673300527638395,
|
| 419 |
+
"grad_norm": 667.1629028320312,
|
| 420 |
+
"learning_rate": 2.383838383838384e-06,
|
| 421 |
+
"loss": 625.2918,
|
| 422 |
+
"step": 590
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.004848132257047972,
|
| 426 |
+
"grad_norm": 614.8545532226562,
|
| 427 |
+
"learning_rate": 2.4242424242424244e-06,
|
| 428 |
+
"loss": 600.5201,
|
| 429 |
+
"step": 600
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.004928934461332105,
|
| 433 |
+
"grad_norm": 879.8363647460938,
|
| 434 |
+
"learning_rate": 2.4646464646464645e-06,
|
| 435 |
+
"loss": 509.5772,
|
| 436 |
+
"step": 610
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.005009736665616238,
|
| 440 |
+
"grad_norm": 1344.7303466796875,
|
| 441 |
+
"learning_rate": 2.5050505050505055e-06,
|
| 442 |
+
"loss": 639.9966,
|
| 443 |
+
"step": 620
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.005090538869900371,
|
| 447 |
+
"grad_norm": 1223.520751953125,
|
| 448 |
+
"learning_rate": 2.5454545454545456e-06,
|
| 449 |
+
"loss": 706.7377,
|
| 450 |
+
"step": 630
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.0051713410741845035,
|
| 454 |
+
"grad_norm": 1871.1324462890625,
|
| 455 |
+
"learning_rate": 2.585858585858586e-06,
|
| 456 |
+
"loss": 712.2218,
|
| 457 |
+
"step": 640
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.005252143278468637,
|
| 461 |
+
"grad_norm": 804.082763671875,
|
| 462 |
+
"learning_rate": 2.6262626262626263e-06,
|
| 463 |
+
"loss": 666.623,
|
| 464 |
+
"step": 650
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.00533294548275277,
|
| 468 |
+
"grad_norm": 10994.2314453125,
|
| 469 |
+
"learning_rate": 2.666666666666667e-06,
|
| 470 |
+
"loss": 726.5861,
|
| 471 |
+
"step": 660
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.005413747687036902,
|
| 475 |
+
"grad_norm": 892.67919921875,
|
| 476 |
+
"learning_rate": 2.7070707070707074e-06,
|
| 477 |
+
"loss": 596.617,
|
| 478 |
+
"step": 670
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.005494549891321035,
|
| 482 |
+
"grad_norm": 681.8887939453125,
|
| 483 |
+
"learning_rate": 2.7474747474747475e-06,
|
| 484 |
+
"loss": 496.8398,
|
| 485 |
+
"step": 680
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.005575352095605168,
|
| 489 |
+
"grad_norm": 1190.68310546875,
|
| 490 |
+
"learning_rate": 2.787878787878788e-06,
|
| 491 |
+
"loss": 684.7522,
|
| 492 |
+
"step": 690
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.005656154299889301,
|
| 496 |
+
"grad_norm": 1649.9376220703125,
|
| 497 |
+
"learning_rate": 2.8282828282828286e-06,
|
| 498 |
+
"loss": 717.8954,
|
| 499 |
+
"step": 700
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.0057369565041734336,
|
| 503 |
+
"grad_norm": 2140.240234375,
|
| 504 |
+
"learning_rate": 2.8686868686868687e-06,
|
| 505 |
+
"loss": 633.3918,
|
| 506 |
+
"step": 710
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.005817758708457567,
|
| 510 |
+
"grad_norm": 1765.3192138671875,
|
| 511 |
+
"learning_rate": 2.9090909090909093e-06,
|
| 512 |
+
"loss": 656.2145,
|
| 513 |
+
"step": 720
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.0058985609127417,
|
| 517 |
+
"grad_norm": 2076.052001953125,
|
| 518 |
+
"learning_rate": 2.9494949494949494e-06,
|
| 519 |
+
"loss": 620.9487,
|
| 520 |
+
"step": 730
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.005979363117025832,
|
| 524 |
+
"grad_norm": 817.6283569335938,
|
| 525 |
+
"learning_rate": 2.9898989898989904e-06,
|
| 526 |
+
"loss": 427.1178,
|
| 527 |
+
"step": 740
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.006060165321309965,
|
| 531 |
+
"grad_norm": 1084.86767578125,
|
| 532 |
+
"learning_rate": 3.0303030303030305e-06,
|
| 533 |
+
"loss": 560.9343,
|
| 534 |
+
"step": 750
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.006140967525594098,
|
| 538 |
+
"grad_norm": 786.8311767578125,
|
| 539 |
+
"learning_rate": 3.070707070707071e-06,
|
| 540 |
+
"loss": 520.7236,
|
| 541 |
+
"step": 760
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.006221769729878231,
|
| 545 |
+
"grad_norm": 4558.533203125,
|
| 546 |
+
"learning_rate": 3.111111111111111e-06,
|
| 547 |
+
"loss": 555.48,
|
| 548 |
+
"step": 770
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.006302571934162364,
|
| 552 |
+
"grad_norm": 785.0099487304688,
|
| 553 |
+
"learning_rate": 3.1515151515151517e-06,
|
| 554 |
+
"loss": 514.0267,
|
| 555 |
+
"step": 780
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.006383374138446497,
|
| 559 |
+
"grad_norm": 1063.466796875,
|
| 560 |
+
"learning_rate": 3.191919191919192e-06,
|
| 561 |
+
"loss": 592.0211,
|
| 562 |
+
"step": 790
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.00646417634273063,
|
| 566 |
+
"grad_norm": 2228.42041015625,
|
| 567 |
+
"learning_rate": 3.2323232323232324e-06,
|
| 568 |
+
"loss": 613.0331,
|
| 569 |
+
"step": 800
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.006544978547014762,
|
| 573 |
+
"grad_norm": 1026.750732421875,
|
| 574 |
+
"learning_rate": 3.2727272727272733e-06,
|
| 575 |
+
"loss": 639.7108,
|
| 576 |
+
"step": 810
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.006625780751298895,
|
| 580 |
+
"grad_norm": 2451.453369140625,
|
| 581 |
+
"learning_rate": 3.313131313131313e-06,
|
| 582 |
+
"loss": 659.9323,
|
| 583 |
+
"step": 820
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.0067065829555830285,
|
| 587 |
+
"grad_norm": 1183.8045654296875,
|
| 588 |
+
"learning_rate": 3.3535353535353536e-06,
|
| 589 |
+
"loss": 601.1871,
|
| 590 |
+
"step": 830
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.006787385159867161,
|
| 594 |
+
"grad_norm": 1861.18701171875,
|
| 595 |
+
"learning_rate": 3.3939393939393946e-06,
|
| 596 |
+
"loss": 589.3404,
|
| 597 |
+
"step": 840
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.006868187364151294,
|
| 601 |
+
"grad_norm": 1340.5020751953125,
|
| 602 |
+
"learning_rate": 3.4343434343434343e-06,
|
| 603 |
+
"loss": 588.7653,
|
| 604 |
+
"step": 850
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.006948989568435427,
|
| 608 |
+
"grad_norm": 8932.84375,
|
| 609 |
+
"learning_rate": 3.4747474747474752e-06,
|
| 610 |
+
"loss": 667.0873,
|
| 611 |
+
"step": 860
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.00702979177271956,
|
| 615 |
+
"grad_norm": 1353.702392578125,
|
| 616 |
+
"learning_rate": 3.515151515151515e-06,
|
| 617 |
+
"loss": 574.4604,
|
| 618 |
+
"step": 870
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.0071105939770036925,
|
| 622 |
+
"grad_norm": 1281.541748046875,
|
| 623 |
+
"learning_rate": 3.555555555555556e-06,
|
| 624 |
+
"loss": 683.6188,
|
| 625 |
+
"step": 880
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.007191396181287825,
|
| 629 |
+
"grad_norm": 3347.4111328125,
|
| 630 |
+
"learning_rate": 3.5959595959595965e-06,
|
| 631 |
+
"loss": 591.1179,
|
| 632 |
+
"step": 890
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.007272198385571959,
|
| 636 |
+
"grad_norm": 832.118896484375,
|
| 637 |
+
"learning_rate": 3.636363636363636e-06,
|
| 638 |
+
"loss": 509.1054,
|
| 639 |
+
"step": 900
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.007353000589856091,
|
| 643 |
+
"grad_norm": 3215.1875,
|
| 644 |
+
"learning_rate": 3.676767676767677e-06,
|
| 645 |
+
"loss": 634.5958,
|
| 646 |
+
"step": 910
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.007433802794140224,
|
| 650 |
+
"grad_norm": 1072.3865966796875,
|
| 651 |
+
"learning_rate": 3.7171717171717177e-06,
|
| 652 |
+
"loss": 525.3159,
|
| 653 |
+
"step": 920
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.007514604998424357,
|
| 657 |
+
"grad_norm": 1002.2363891601562,
|
| 658 |
+
"learning_rate": 3.757575757575758e-06,
|
| 659 |
+
"loss": 659.7474,
|
| 660 |
+
"step": 930
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.00759540720270849,
|
| 664 |
+
"grad_norm": 6355.693359375,
|
| 665 |
+
"learning_rate": 3.7979797979797984e-06,
|
| 666 |
+
"loss": 636.0396,
|
| 667 |
+
"step": 940
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.007676209406992623,
|
| 671 |
+
"grad_norm": 1635.6080322265625,
|
| 672 |
+
"learning_rate": 3.8383838383838385e-06,
|
| 673 |
+
"loss": 624.7128,
|
| 674 |
+
"step": 950
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.007757011611276756,
|
| 678 |
+
"grad_norm": 1284.3531494140625,
|
| 679 |
+
"learning_rate": 3.878787878787879e-06,
|
| 680 |
+
"loss": 637.0934,
|
| 681 |
+
"step": 960
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.007837813815560889,
|
| 685 |
+
"grad_norm": 2921.176025390625,
|
| 686 |
+
"learning_rate": 3.9191919191919196e-06,
|
| 687 |
+
"loss": 648.8942,
|
| 688 |
+
"step": 970
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.007918616019845021,
|
| 692 |
+
"grad_norm": 845.0542602539062,
|
| 693 |
+
"learning_rate": 3.959595959595959e-06,
|
| 694 |
+
"loss": 536.0546,
|
| 695 |
+
"step": 980
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.007999418224129154,
|
| 699 |
+
"grad_norm": 986.8812866210938,
|
| 700 |
+
"learning_rate": 4.000000000000001e-06,
|
| 701 |
+
"loss": 643.056,
|
| 702 |
+
"step": 990
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.008080220428413287,
|
| 706 |
+
"grad_norm": 748.7238159179688,
|
| 707 |
+
"learning_rate": 4.040404040404041e-06,
|
| 708 |
+
"loss": 587.7603,
|
| 709 |
+
"step": 1000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.00816102263269742,
|
| 713 |
+
"grad_norm": 4340.18017578125,
|
| 714 |
+
"learning_rate": 4.080808080808081e-06,
|
| 715 |
+
"loss": 545.2501,
|
| 716 |
+
"step": 1010
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.008241824836981554,
|
| 720 |
+
"grad_norm": 1614.0966796875,
|
| 721 |
+
"learning_rate": 4.1212121212121215e-06,
|
| 722 |
+
"loss": 652.8924,
|
| 723 |
+
"step": 1020
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.008322627041265686,
|
| 727 |
+
"grad_norm": 10709.900390625,
|
| 728 |
+
"learning_rate": 4.161616161616161e-06,
|
| 729 |
+
"loss": 598.342,
|
| 730 |
+
"step": 1030
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.008403429245549819,
|
| 734 |
+
"grad_norm": 870.67578125,
|
| 735 |
+
"learning_rate": 4.2020202020202026e-06,
|
| 736 |
+
"loss": 612.1897,
|
| 737 |
+
"step": 1040
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.008484231449833951,
|
| 741 |
+
"grad_norm": 686.4441528320312,
|
| 742 |
+
"learning_rate": 4.242424242424243e-06,
|
| 743 |
+
"loss": 564.3604,
|
| 744 |
+
"step": 1050
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.008565033654118084,
|
| 748 |
+
"grad_norm": 800.3753051757812,
|
| 749 |
+
"learning_rate": 4.282828282828283e-06,
|
| 750 |
+
"loss": 515.465,
|
| 751 |
+
"step": 1060
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.008645835858402217,
|
| 755 |
+
"grad_norm": 1612.2467041015625,
|
| 756 |
+
"learning_rate": 4.323232323232323e-06,
|
| 757 |
+
"loss": 615.4043,
|
| 758 |
+
"step": 1070
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.00872663806268635,
|
| 762 |
+
"grad_norm": 1488.6268310546875,
|
| 763 |
+
"learning_rate": 4.363636363636364e-06,
|
| 764 |
+
"loss": 584.7924,
|
| 765 |
+
"step": 1080
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.008807440266970484,
|
| 769 |
+
"grad_norm": 1085.609619140625,
|
| 770 |
+
"learning_rate": 4.4040404040404044e-06,
|
| 771 |
+
"loss": 568.4938,
|
| 772 |
+
"step": 1090
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.008888242471254616,
|
| 776 |
+
"grad_norm": 1548.931884765625,
|
| 777 |
+
"learning_rate": 4.444444444444445e-06,
|
| 778 |
+
"loss": 515.5535,
|
| 779 |
+
"step": 1100
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.008969044675538749,
|
| 783 |
+
"grad_norm": 1010.5916137695312,
|
| 784 |
+
"learning_rate": 4.484848484848485e-06,
|
| 785 |
+
"loss": 496.3722,
|
| 786 |
+
"step": 1110
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.009049846879822881,
|
| 790 |
+
"grad_norm": 1959.3551025390625,
|
| 791 |
+
"learning_rate": 4.525252525252525e-06,
|
| 792 |
+
"loss": 697.1435,
|
| 793 |
+
"step": 1120
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.009130649084107014,
|
| 797 |
+
"grad_norm": 2370.032470703125,
|
| 798 |
+
"learning_rate": 4.565656565656566e-06,
|
| 799 |
+
"loss": 541.7345,
|
| 800 |
+
"step": 1130
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.009211451288391147,
|
| 804 |
+
"grad_norm": 1848.2596435546875,
|
| 805 |
+
"learning_rate": 4.606060606060606e-06,
|
| 806 |
+
"loss": 539.6271,
|
| 807 |
+
"step": 1140
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.00929225349267528,
|
| 811 |
+
"grad_norm": 2245.12548828125,
|
| 812 |
+
"learning_rate": 4.646464646464647e-06,
|
| 813 |
+
"loss": 485.2028,
|
| 814 |
+
"step": 1150
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.009373055696959414,
|
| 818 |
+
"grad_norm": 1303.244140625,
|
| 819 |
+
"learning_rate": 4.6868686868686874e-06,
|
| 820 |
+
"loss": 552.4869,
|
| 821 |
+
"step": 1160
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.009453857901243546,
|
| 825 |
+
"grad_norm": 813.8889770507812,
|
| 826 |
+
"learning_rate": 4.727272727272727e-06,
|
| 827 |
+
"loss": 566.426,
|
| 828 |
+
"step": 1170
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.009534660105527679,
|
| 832 |
+
"grad_norm": 2155.474609375,
|
| 833 |
+
"learning_rate": 4.767676767676768e-06,
|
| 834 |
+
"loss": 694.8829,
|
| 835 |
+
"step": 1180
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.009615462309811812,
|
| 839 |
+
"grad_norm": 3313.629150390625,
|
| 840 |
+
"learning_rate": 4.808080808080808e-06,
|
| 841 |
+
"loss": 541.8082,
|
| 842 |
+
"step": 1190
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.009696264514095944,
|
| 846 |
+
"grad_norm": 830.4761352539062,
|
| 847 |
+
"learning_rate": 4.848484848484849e-06,
|
| 848 |
+
"loss": 633.2839,
|
| 849 |
+
"step": 1200
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.009777066718380077,
|
| 853 |
+
"grad_norm": 2314.525634765625,
|
| 854 |
+
"learning_rate": 4.888888888888889e-06,
|
| 855 |
+
"loss": 465.8495,
|
| 856 |
+
"step": 1210
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.00985786892266421,
|
| 860 |
+
"grad_norm": 1821.2579345703125,
|
| 861 |
+
"learning_rate": 4.929292929292929e-06,
|
| 862 |
+
"loss": 543.8752,
|
| 863 |
+
"step": 1220
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.009938671126948344,
|
| 867 |
+
"grad_norm": 471.08380126953125,
|
| 868 |
+
"learning_rate": 4.96969696969697e-06,
|
| 869 |
+
"loss": 405.759,
|
| 870 |
+
"step": 1230
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.010019473331232476,
|
| 874 |
+
"grad_norm": 1476.8787841796875,
|
| 875 |
+
"learning_rate": 5.010101010101011e-06,
|
| 876 |
+
"loss": 617.1836,
|
| 877 |
+
"step": 1240
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.010100275535516609,
|
| 881 |
+
"grad_norm": 667.0234985351562,
|
| 882 |
+
"learning_rate": 5.050505050505051e-06,
|
| 883 |
+
"loss": 513.511,
|
| 884 |
+
"step": 1250
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.010181077739800742,
|
| 888 |
+
"grad_norm": 613.2576904296875,
|
| 889 |
+
"learning_rate": 5.090909090909091e-06,
|
| 890 |
+
"loss": 468.2697,
|
| 891 |
+
"step": 1260
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.010261879944084874,
|
| 895 |
+
"grad_norm": 1321.9991455078125,
|
| 896 |
+
"learning_rate": 5.131313131313131e-06,
|
| 897 |
+
"loss": 476.6512,
|
| 898 |
+
"step": 1270
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.010342682148369007,
|
| 902 |
+
"grad_norm": 840.0568237304688,
|
| 903 |
+
"learning_rate": 5.171717171717172e-06,
|
| 904 |
+
"loss": 607.1966,
|
| 905 |
+
"step": 1280
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.01042348435265314,
|
| 909 |
+
"grad_norm": 1686.292724609375,
|
| 910 |
+
"learning_rate": 5.212121212121213e-06,
|
| 911 |
+
"loss": 538.6932,
|
| 912 |
+
"step": 1290
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.010504286556937274,
|
| 916 |
+
"grad_norm": 2623.5302734375,
|
| 917 |
+
"learning_rate": 5.2525252525252526e-06,
|
| 918 |
+
"loss": 511.9114,
|
| 919 |
+
"step": 1300
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.010585088761221407,
|
| 923 |
+
"grad_norm": 1300.63671875,
|
| 924 |
+
"learning_rate": 5.292929292929293e-06,
|
| 925 |
+
"loss": 573.5399,
|
| 926 |
+
"step": 1310
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.01066589096550554,
|
| 930 |
+
"grad_norm": 828.8121948242188,
|
| 931 |
+
"learning_rate": 5.333333333333334e-06,
|
| 932 |
+
"loss": 617.9385,
|
| 933 |
+
"step": 1320
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.010746693169789672,
|
| 937 |
+
"grad_norm": 965.5888671875,
|
| 938 |
+
"learning_rate": 5.373737373737374e-06,
|
| 939 |
+
"loss": 428.5168,
|
| 940 |
+
"step": 1330
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.010827495374073804,
|
| 944 |
+
"grad_norm": 1123.3818359375,
|
| 945 |
+
"learning_rate": 5.414141414141415e-06,
|
| 946 |
+
"loss": 582.797,
|
| 947 |
+
"step": 1340
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.010908297578357937,
|
| 951 |
+
"grad_norm": 4098.775390625,
|
| 952 |
+
"learning_rate": 5.4545454545454545e-06,
|
| 953 |
+
"loss": 633.2868,
|
| 954 |
+
"step": 1350
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.01098909978264207,
|
| 958 |
+
"grad_norm": 792.64306640625,
|
| 959 |
+
"learning_rate": 5.494949494949495e-06,
|
| 960 |
+
"loss": 688.7412,
|
| 961 |
+
"step": 1360
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.011069901986926204,
|
| 965 |
+
"grad_norm": 1111.2113037109375,
|
| 966 |
+
"learning_rate": 5.5353535353535355e-06,
|
| 967 |
+
"loss": 473.5282,
|
| 968 |
+
"step": 1370
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.011150704191210337,
|
| 972 |
+
"grad_norm": 5420.9765625,
|
| 973 |
+
"learning_rate": 5.575757575757576e-06,
|
| 974 |
+
"loss": 516.3693,
|
| 975 |
+
"step": 1380
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.01123150639549447,
|
| 979 |
+
"grad_norm": 1263.883056640625,
|
| 980 |
+
"learning_rate": 5.616161616161617e-06,
|
| 981 |
+
"loss": 457.2747,
|
| 982 |
+
"step": 1390
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.011312308599778602,
|
| 986 |
+
"grad_norm": 944.7489624023438,
|
| 987 |
+
"learning_rate": 5.656565656565657e-06,
|
| 988 |
+
"loss": 539.8824,
|
| 989 |
+
"step": 1400
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.011393110804062734,
|
| 993 |
+
"grad_norm": 562.0636596679688,
|
| 994 |
+
"learning_rate": 5.696969696969697e-06,
|
| 995 |
+
"loss": 455.684,
|
| 996 |
+
"step": 1410
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.011473913008346867,
|
| 1000 |
+
"grad_norm": 1293.833984375,
|
| 1001 |
+
"learning_rate": 5.7373737373737374e-06,
|
| 1002 |
+
"loss": 527.1053,
|
| 1003 |
+
"step": 1420
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.011554715212631,
|
| 1007 |
+
"grad_norm": 1857.3572998046875,
|
| 1008 |
+
"learning_rate": 5.777777777777778e-06,
|
| 1009 |
+
"loss": 519.1355,
|
| 1010 |
+
"step": 1430
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.011635517416915134,
|
| 1014 |
+
"grad_norm": 1110.5677490234375,
|
| 1015 |
+
"learning_rate": 5.8181818181818185e-06,
|
| 1016 |
+
"loss": 470.4913,
|
| 1017 |
+
"step": 1440
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.011716319621199267,
|
| 1021 |
+
"grad_norm": 1656.9898681640625,
|
| 1022 |
+
"learning_rate": 5.858585858585859e-06,
|
| 1023 |
+
"loss": 570.9559,
|
| 1024 |
+
"step": 1450
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.0117971218254834,
|
| 1028 |
+
"grad_norm": 830.130126953125,
|
| 1029 |
+
"learning_rate": 5.898989898989899e-06,
|
| 1030 |
+
"loss": 551.6251,
|
| 1031 |
+
"step": 1460
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.011877924029767532,
|
| 1035 |
+
"grad_norm": 718.1753540039062,
|
| 1036 |
+
"learning_rate": 5.93939393939394e-06,
|
| 1037 |
+
"loss": 495.0048,
|
| 1038 |
+
"step": 1470
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.011958726234051665,
|
| 1042 |
+
"grad_norm": 2903.817138671875,
|
| 1043 |
+
"learning_rate": 5.979797979797981e-06,
|
| 1044 |
+
"loss": 742.8938,
|
| 1045 |
+
"step": 1480
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.012039528438335797,
|
| 1049 |
+
"grad_norm": 893.6466674804688,
|
| 1050 |
+
"learning_rate": 6.0202020202020204e-06,
|
| 1051 |
+
"loss": 542.4396,
|
| 1052 |
+
"step": 1490
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.01212033064261993,
|
| 1056 |
+
"grad_norm": 1576.75390625,
|
| 1057 |
+
"learning_rate": 6.060606060606061e-06,
|
| 1058 |
+
"loss": 513.3051,
|
| 1059 |
+
"step": 1500
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.012201132846904064,
|
| 1063 |
+
"grad_norm": 1925.69140625,
|
| 1064 |
+
"learning_rate": 6.101010101010101e-06,
|
| 1065 |
+
"loss": 450.3941,
|
| 1066 |
+
"step": 1510
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.012281935051188197,
|
| 1070 |
+
"grad_norm": 1998.0458984375,
|
| 1071 |
+
"learning_rate": 6.141414141414142e-06,
|
| 1072 |
+
"loss": 635.4773,
|
| 1073 |
+
"step": 1520
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.01236273725547233,
|
| 1077 |
+
"grad_norm": 1258.9659423828125,
|
| 1078 |
+
"learning_rate": 6.181818181818183e-06,
|
| 1079 |
+
"loss": 595.3451,
|
| 1080 |
+
"step": 1530
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.012443539459756462,
|
| 1084 |
+
"grad_norm": 1726.843017578125,
|
| 1085 |
+
"learning_rate": 6.222222222222222e-06,
|
| 1086 |
+
"loss": 556.3297,
|
| 1087 |
+
"step": 1540
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.012524341664040595,
|
| 1091 |
+
"grad_norm": 2486.63134765625,
|
| 1092 |
+
"learning_rate": 6.262626262626263e-06,
|
| 1093 |
+
"loss": 647.7714,
|
| 1094 |
+
"step": 1550
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.012605143868324727,
|
| 1098 |
+
"grad_norm": 1644.4110107421875,
|
| 1099 |
+
"learning_rate": 6.303030303030303e-06,
|
| 1100 |
+
"loss": 611.4104,
|
| 1101 |
+
"step": 1560
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.01268594607260886,
|
| 1105 |
+
"grad_norm": 2169.43408203125,
|
| 1106 |
+
"learning_rate": 6.343434343434344e-06,
|
| 1107 |
+
"loss": 445.9718,
|
| 1108 |
+
"step": 1570
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.012766748276892994,
|
| 1112 |
+
"grad_norm": 1554.9425048828125,
|
| 1113 |
+
"learning_rate": 6.383838383838384e-06,
|
| 1114 |
+
"loss": 561.5079,
|
| 1115 |
+
"step": 1580
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.012847550481177127,
|
| 1119 |
+
"grad_norm": 891.0722045898438,
|
| 1120 |
+
"learning_rate": 6.424242424242424e-06,
|
| 1121 |
+
"loss": 510.2931,
|
| 1122 |
+
"step": 1590
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.01292835268546126,
|
| 1126 |
+
"grad_norm": 22828.455078125,
|
| 1127 |
+
"learning_rate": 6.464646464646465e-06,
|
| 1128 |
+
"loss": 552.862,
|
| 1129 |
+
"step": 1600
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.013009154889745392,
|
| 1133 |
+
"grad_norm": 722.2283325195312,
|
| 1134 |
+
"learning_rate": 6.505050505050505e-06,
|
| 1135 |
+
"loss": 512.3786,
|
| 1136 |
+
"step": 1610
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.013089957094029525,
|
| 1140 |
+
"grad_norm": 1105.0716552734375,
|
| 1141 |
+
"learning_rate": 6.545454545454547e-06,
|
| 1142 |
+
"loss": 532.3522,
|
| 1143 |
+
"step": 1620
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.013170759298313657,
|
| 1147 |
+
"grad_norm": 1146.0550537109375,
|
| 1148 |
+
"learning_rate": 6.5858585858585856e-06,
|
| 1149 |
+
"loss": 647.0639,
|
| 1150 |
+
"step": 1630
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.01325156150259779,
|
| 1154 |
+
"grad_norm": 2632.366943359375,
|
| 1155 |
+
"learning_rate": 6.626262626262626e-06,
|
| 1156 |
+
"loss": 613.0408,
|
| 1157 |
+
"step": 1640
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.013332363706881924,
|
| 1161 |
+
"grad_norm": 1236.950439453125,
|
| 1162 |
+
"learning_rate": 6.666666666666667e-06,
|
| 1163 |
+
"loss": 614.0946,
|
| 1164 |
+
"step": 1650
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.013413165911166057,
|
| 1168 |
+
"grad_norm": 973.5294799804688,
|
| 1169 |
+
"learning_rate": 6.707070707070707e-06,
|
| 1170 |
+
"loss": 383.6397,
|
| 1171 |
+
"step": 1660
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.01349396811545019,
|
| 1175 |
+
"grad_norm": 1101.2989501953125,
|
| 1176 |
+
"learning_rate": 6.747474747474749e-06,
|
| 1177 |
+
"loss": 556.5678,
|
| 1178 |
+
"step": 1670
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.013574770319734322,
|
| 1182 |
+
"grad_norm": 694.1087646484375,
|
| 1183 |
+
"learning_rate": 6.787878787878789e-06,
|
| 1184 |
+
"loss": 415.0727,
|
| 1185 |
+
"step": 1680
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.013655572524018455,
|
| 1189 |
+
"grad_norm": 1031.21240234375,
|
| 1190 |
+
"learning_rate": 6.828282828282828e-06,
|
| 1191 |
+
"loss": 508.7032,
|
| 1192 |
+
"step": 1690
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.013736374728302587,
|
| 1196 |
+
"grad_norm": 734.2860717773438,
|
| 1197 |
+
"learning_rate": 6.8686868686868685e-06,
|
| 1198 |
+
"loss": 482.0532,
|
| 1199 |
+
"step": 1700
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.01381717693258672,
|
| 1203 |
+
"grad_norm": 879.1790771484375,
|
| 1204 |
+
"learning_rate": 6.909090909090909e-06,
|
| 1205 |
+
"loss": 655.4877,
|
| 1206 |
+
"step": 1710
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.013897979136870854,
|
| 1210 |
+
"grad_norm": 1699.49560546875,
|
| 1211 |
+
"learning_rate": 6.9494949494949505e-06,
|
| 1212 |
+
"loss": 400.7212,
|
| 1213 |
+
"step": 1720
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.013978781341154987,
|
| 1217 |
+
"grad_norm": 1901.2413330078125,
|
| 1218 |
+
"learning_rate": 6.989898989898991e-06,
|
| 1219 |
+
"loss": 496.4101,
|
| 1220 |
+
"step": 1730
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.01405958354543912,
|
| 1224 |
+
"grad_norm": 1368.836669921875,
|
| 1225 |
+
"learning_rate": 7.03030303030303e-06,
|
| 1226 |
+
"loss": 520.5295,
|
| 1227 |
+
"step": 1740
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.014140385749723252,
|
| 1231 |
+
"grad_norm": 1587.8297119140625,
|
| 1232 |
+
"learning_rate": 7.0707070707070704e-06,
|
| 1233 |
+
"loss": 397.7875,
|
| 1234 |
+
"step": 1750
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.014221187954007385,
|
| 1238 |
+
"grad_norm": 1054.773681640625,
|
| 1239 |
+
"learning_rate": 7.111111111111112e-06,
|
| 1240 |
+
"loss": 456.9818,
|
| 1241 |
+
"step": 1760
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.014301990158291518,
|
| 1245 |
+
"grad_norm": 3368.4013671875,
|
| 1246 |
+
"learning_rate": 7.151515151515152e-06,
|
| 1247 |
+
"loss": 397.1987,
|
| 1248 |
+
"step": 1770
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.01438279236257565,
|
| 1252 |
+
"grad_norm": 1371.4473876953125,
|
| 1253 |
+
"learning_rate": 7.191919191919193e-06,
|
| 1254 |
+
"loss": 433.1569,
|
| 1255 |
+
"step": 1780
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.014463594566859785,
|
| 1259 |
+
"grad_norm": 1193.6456298828125,
|
| 1260 |
+
"learning_rate": 7.232323232323232e-06,
|
| 1261 |
+
"loss": 368.5404,
|
| 1262 |
+
"step": 1790
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.014544396771143917,
|
| 1266 |
+
"grad_norm": 976.5448608398438,
|
| 1267 |
+
"learning_rate": 7.272727272727272e-06,
|
| 1268 |
+
"loss": 427.9085,
|
| 1269 |
+
"step": 1800
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.01462519897542805,
|
| 1273 |
+
"grad_norm": 665.8095092773438,
|
| 1274 |
+
"learning_rate": 7.313131313131314e-06,
|
| 1275 |
+
"loss": 461.1592,
|
| 1276 |
+
"step": 1810
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.014706001179712182,
|
| 1280 |
+
"grad_norm": 1352.773681640625,
|
| 1281 |
+
"learning_rate": 7.353535353535354e-06,
|
| 1282 |
+
"loss": 421.6953,
|
| 1283 |
+
"step": 1820
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.014786803383996315,
|
| 1287 |
+
"grad_norm": 1284.1273193359375,
|
| 1288 |
+
"learning_rate": 7.393939393939395e-06,
|
| 1289 |
+
"loss": 486.5889,
|
| 1290 |
+
"step": 1830
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.014867605588280448,
|
| 1294 |
+
"grad_norm": 3249.41650390625,
|
| 1295 |
+
"learning_rate": 7.434343434343435e-06,
|
| 1296 |
+
"loss": 534.9857,
|
| 1297 |
+
"step": 1840
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.01494840779256458,
|
| 1301 |
+
"grad_norm": 1162.9168701171875,
|
| 1302 |
+
"learning_rate": 7.474747474747475e-06,
|
| 1303 |
+
"loss": 684.8136,
|
| 1304 |
+
"step": 1850
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.015029209996848715,
|
| 1308 |
+
"grad_norm": 2700.0771484375,
|
| 1309 |
+
"learning_rate": 7.515151515151516e-06,
|
| 1310 |
+
"loss": 438.9466,
|
| 1311 |
+
"step": 1860
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.015110012201132847,
|
| 1315 |
+
"grad_norm": 1472.51171875,
|
| 1316 |
+
"learning_rate": 7.555555555555556e-06,
|
| 1317 |
+
"loss": 541.1812,
|
| 1318 |
+
"step": 1870
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.01519081440541698,
|
| 1322 |
+
"grad_norm": 763.5363159179688,
|
| 1323 |
+
"learning_rate": 7.595959595959597e-06,
|
| 1324 |
+
"loss": 574.8496,
|
| 1325 |
+
"step": 1880
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.015271616609701113,
|
| 1329 |
+
"grad_norm": 1097.047607421875,
|
| 1330 |
+
"learning_rate": 7.636363636363638e-06,
|
| 1331 |
+
"loss": 495.2874,
|
| 1332 |
+
"step": 1890
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.015352418813985245,
|
| 1336 |
+
"grad_norm": 1086.6981201171875,
|
| 1337 |
+
"learning_rate": 7.676767676767677e-06,
|
| 1338 |
+
"loss": 494.5146,
|
| 1339 |
+
"step": 1900
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.015433221018269378,
|
| 1343 |
+
"grad_norm": 1003.9575805664062,
|
| 1344 |
+
"learning_rate": 7.717171717171717e-06,
|
| 1345 |
+
"loss": 493.7404,
|
| 1346 |
+
"step": 1910
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.015514023222553512,
|
| 1350 |
+
"grad_norm": 1008.3009643554688,
|
| 1351 |
+
"learning_rate": 7.757575757575758e-06,
|
| 1352 |
+
"loss": 507.6875,
|
| 1353 |
+
"step": 1920
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.015594825426837645,
|
| 1357 |
+
"grad_norm": 2552.92626953125,
|
| 1358 |
+
"learning_rate": 7.797979797979799e-06,
|
| 1359 |
+
"loss": 420.8823,
|
| 1360 |
+
"step": 1930
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.015675627631121777,
|
| 1364 |
+
"grad_norm": 735.5524291992188,
|
| 1365 |
+
"learning_rate": 7.838383838383839e-06,
|
| 1366 |
+
"loss": 519.3292,
|
| 1367 |
+
"step": 1940
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.01575642983540591,
|
| 1371 |
+
"grad_norm": 1559.79296875,
|
| 1372 |
+
"learning_rate": 7.878787878787878e-06,
|
| 1373 |
+
"loss": 468.1747,
|
| 1374 |
+
"step": 1950
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.015837232039690043,
|
| 1378 |
+
"grad_norm": 1734.436767578125,
|
| 1379 |
+
"learning_rate": 7.919191919191919e-06,
|
| 1380 |
+
"loss": 417.3558,
|
| 1381 |
+
"step": 1960
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.015918034243974175,
|
| 1385 |
+
"grad_norm": 2985.763671875,
|
| 1386 |
+
"learning_rate": 7.959595959595959e-06,
|
| 1387 |
+
"loss": 427.185,
|
| 1388 |
+
"step": 1970
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.015998836448258308,
|
| 1392 |
+
"grad_norm": 1284.0130615234375,
|
| 1393 |
+
"learning_rate": 8.000000000000001e-06,
|
| 1394 |
+
"loss": 523.1269,
|
| 1395 |
+
"step": 1980
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.01607963865254244,
|
| 1399 |
+
"grad_norm": 1079.0145263671875,
|
| 1400 |
+
"learning_rate": 8.040404040404042e-06,
|
| 1401 |
+
"loss": 497.3142,
|
| 1402 |
+
"step": 1990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.016160440856826573,
|
| 1406 |
+
"grad_norm": 1260.9786376953125,
|
| 1407 |
+
"learning_rate": 8.080808080808082e-06,
|
| 1408 |
+
"loss": 495.9584,
|
| 1409 |
+
"step": 2000
|
| 1410 |
+
}
|
| 1411 |
+
],
|
| 1412 |
+
"logging_steps": 10,
|
| 1413 |
+
"max_steps": 123750,
|
| 1414 |
+
"num_input_tokens_seen": 0,
|
| 1415 |
+
"num_train_epochs": 1,
|
| 1416 |
+
"save_steps": 1000,
|
| 1417 |
+
"stateful_callbacks": {
|
| 1418 |
+
"TrainerControl": {
|
| 1419 |
+
"args": {
|
| 1420 |
+
"should_epoch_stop": false,
|
| 1421 |
+
"should_evaluate": false,
|
| 1422 |
+
"should_log": false,
|
| 1423 |
+
"should_save": true,
|
| 1424 |
+
"should_training_stop": false
|
| 1425 |
+
},
|
| 1426 |
+
"attributes": {}
|
| 1427 |
+
}
|
| 1428 |
+
},
|
| 1429 |
+
"total_flos": 0.0,
|
| 1430 |
+
"train_batch_size": 8,
|
| 1431 |
+
"trial_name": null,
|
| 1432 |
+
"trial_params": null
|
| 1433 |
+
}
|