Upload folder using huggingface_hub
Browse files- added_tokens.json +5 -0
- gliner_config.json +141 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- special_tokens_map.json +15 -0
- spm.model +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +75 -0
- trainer_state.json +1161 -0
added_tokens.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<<ENT>>": 128001,
|
| 3 |
+
"<<SEP>>": 128002,
|
| 4 |
+
"[MASK]": 128000
|
| 5 |
+
}
|
gliner_config.json
ADDED
|
@@ -0,0 +1,141 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"class_token_index": 128001,
|
| 3 |
+
"dropout": 0.3,
|
| 4 |
+
"embed_ent_token": true,
|
| 5 |
+
"encoder_config": {
|
| 6 |
+
"_attn_implementation_autoset": false,
|
| 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 |
+
"legacy": true,
|
| 43 |
+
"length_penalty": 1.0,
|
| 44 |
+
"max_length": 20,
|
| 45 |
+
"max_position_embeddings": 512,
|
| 46 |
+
"max_relative_positions": -1,
|
| 47 |
+
"min_length": 0,
|
| 48 |
+
"model_type": "deberta-v2",
|
| 49 |
+
"no_repeat_ngram_size": 0,
|
| 50 |
+
"norm_rel_ebd": "layer_norm",
|
| 51 |
+
"num_attention_heads": 12,
|
| 52 |
+
"num_beam_groups": 1,
|
| 53 |
+
"num_beams": 1,
|
| 54 |
+
"num_hidden_layers": 12,
|
| 55 |
+
"num_return_sequences": 1,
|
| 56 |
+
"output_attentions": false,
|
| 57 |
+
"output_hidden_states": false,
|
| 58 |
+
"output_scores": false,
|
| 59 |
+
"pad_token_id": 0,
|
| 60 |
+
"pooler_dropout": 0,
|
| 61 |
+
"pooler_hidden_act": "gelu",
|
| 62 |
+
"pooler_hidden_size": 768,
|
| 63 |
+
"pos_att_type": [
|
| 64 |
+
"p2c",
|
| 65 |
+
"c2p"
|
| 66 |
+
],
|
| 67 |
+
"position_biased_input": false,
|
| 68 |
+
"position_buckets": 256,
|
| 69 |
+
"prefix": null,
|
| 70 |
+
"problem_type": null,
|
| 71 |
+
"pruned_heads": {},
|
| 72 |
+
"relative_attention": true,
|
| 73 |
+
"remove_invalid_values": false,
|
| 74 |
+
"repetition_penalty": 1.0,
|
| 75 |
+
"return_dict": true,
|
| 76 |
+
"return_dict_in_generate": false,
|
| 77 |
+
"sep_token_id": null,
|
| 78 |
+
"share_att_key": true,
|
| 79 |
+
"suppress_tokens": null,
|
| 80 |
+
"task_specific_params": null,
|
| 81 |
+
"temperature": 1.0,
|
| 82 |
+
"tf_legacy_loss": false,
|
| 83 |
+
"tie_encoder_decoder": false,
|
| 84 |
+
"tie_word_embeddings": true,
|
| 85 |
+
"tokenizer_class": null,
|
| 86 |
+
"top_k": 50,
|
| 87 |
+
"top_p": 1.0,
|
| 88 |
+
"torch_dtype": null,
|
| 89 |
+
"torchscript": false,
|
| 90 |
+
"type_vocab_size": 0,
|
| 91 |
+
"typical_p": 1.0,
|
| 92 |
+
"use_bfloat16": false,
|
| 93 |
+
"vocab_size": 128003
|
| 94 |
+
},
|
| 95 |
+
"ent_token": "<<ENT>>",
|
| 96 |
+
"eval_every": 20000,
|
| 97 |
+
"fine_tune": true,
|
| 98 |
+
"freeze_token_rep": false,
|
| 99 |
+
"fuse_layers": false,
|
| 100 |
+
"has_rnn": true,
|
| 101 |
+
"hidden_size": 512,
|
| 102 |
+
"label_smoothing": 0,
|
| 103 |
+
"labels_encoder": null,
|
| 104 |
+
"labels_encoder_config": null,
|
| 105 |
+
"log_dir": "models/",
|
| 106 |
+
"loss_alpha": 0.75,
|
| 107 |
+
"loss_gamma": 2,
|
| 108 |
+
"loss_reduction": "sum",
|
| 109 |
+
"lr_encoder": "1e-5",
|
| 110 |
+
"lr_others": "5e-5",
|
| 111 |
+
"max_grad_norm": 1.0,
|
| 112 |
+
"max_len": 2048,
|
| 113 |
+
"max_neg_type_ratio": 1,
|
| 114 |
+
"max_types": 50,
|
| 115 |
+
"max_width": 12,
|
| 116 |
+
"model_name": "microsoft/deberta-v3-base",
|
| 117 |
+
"model_type": "gliner",
|
| 118 |
+
"name": "span level gliner",
|
| 119 |
+
"num_post_fusion_layers": 1,
|
| 120 |
+
"num_steps": 80000,
|
| 121 |
+
"post_fusion_schema": "",
|
| 122 |
+
"prev_path": null,
|
| 123 |
+
"random_drop": true,
|
| 124 |
+
"root_dir": "gliner_logs",
|
| 125 |
+
"save_total_limit": 3,
|
| 126 |
+
"scheduler_type": "cosine",
|
| 127 |
+
"sep_token": "<<SEP>>",
|
| 128 |
+
"shuffle_types": true,
|
| 129 |
+
"size_sup": -1,
|
| 130 |
+
"span_mode": "markerV0",
|
| 131 |
+
"subtoken_pooling": "first",
|
| 132 |
+
"train_batch_size": 4,
|
| 133 |
+
"train_data": "data/ie_batches_8b_tuned_100k.json",
|
| 134 |
+
"transformers_version": "4.48.1",
|
| 135 |
+
"val_data_dir": "none",
|
| 136 |
+
"vocab_size": 128003,
|
| 137 |
+
"warmup_ratio": 0.1,
|
| 138 |
+
"weight_decay_encoder": 0.01,
|
| 139 |
+
"weight_decay_other": 0.01,
|
| 140 |
+
"words_splitter_type": "whitespace"
|
| 141 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:410ff1697596cbde00d028ce8d62729ed890d23777c581e80267199b72a83a27
|
| 3 |
+
size 780781742
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:89dcd75d920ff072428a791c2583b85d087c6404d507d67693acb5112c77fbf0
|
| 3 |
+
size 14244
|
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 +1,75 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"128001": {
|
| 44 |
+
"content": "<<ENT>>",
|
| 45 |
+
"lstrip": false,
|
| 46 |
+
"normalized": false,
|
| 47 |
+
"rstrip": false,
|
| 48 |
+
"single_word": false,
|
| 49 |
+
"special": true
|
| 50 |
+
},
|
| 51 |
+
"128002": {
|
| 52 |
+
"content": "<<SEP>>",
|
| 53 |
+
"lstrip": false,
|
| 54 |
+
"normalized": false,
|
| 55 |
+
"rstrip": false,
|
| 56 |
+
"single_word": false,
|
| 57 |
+
"special": true
|
| 58 |
+
}
|
| 59 |
+
},
|
| 60 |
+
"bos_token": "[CLS]",
|
| 61 |
+
"clean_up_tokenization_spaces": false,
|
| 62 |
+
"cls_token": "[CLS]",
|
| 63 |
+
"do_lower_case": false,
|
| 64 |
+
"eos_token": "[SEP]",
|
| 65 |
+
"extra_special_tokens": {},
|
| 66 |
+
"mask_token": "[MASK]",
|
| 67 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 68 |
+
"pad_token": "[PAD]",
|
| 69 |
+
"sep_token": "[SEP]",
|
| 70 |
+
"sp_model_kwargs": {},
|
| 71 |
+
"split_by_punct": false,
|
| 72 |
+
"tokenizer_class": "DebertaV2Tokenizer",
|
| 73 |
+
"unk_token": "[UNK]",
|
| 74 |
+
"vocab_type": "spm"
|
| 75 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1161 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.802654408616688,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 80000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.011266590053854301,
|
| 13 |
+
"grad_norm": 3268.47607421875,
|
| 14 |
+
"learning_rate": 3.125e-06,
|
| 15 |
+
"loss": 80243.872,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.022533180107708602,
|
| 20 |
+
"grad_norm": 1101.09423828125,
|
| 21 |
+
"learning_rate": 6.25e-06,
|
| 22 |
+
"loss": 455.9426,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0337997701615629,
|
| 27 |
+
"grad_norm": 1345.1741943359375,
|
| 28 |
+
"learning_rate": 9.375000000000001e-06,
|
| 29 |
+
"loss": 398.2565,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.045066360215417205,
|
| 34 |
+
"grad_norm": 957.8522338867188,
|
| 35 |
+
"learning_rate": 1.25e-05,
|
| 36 |
+
"loss": 361.662,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.0563329502692715,
|
| 41 |
+
"grad_norm": 1153.3565673828125,
|
| 42 |
+
"learning_rate": 1.5625e-05,
|
| 43 |
+
"loss": 351.6057,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.0675995403231258,
|
| 48 |
+
"grad_norm": 3220.947021484375,
|
| 49 |
+
"learning_rate": 1.8750000000000002e-05,
|
| 50 |
+
"loss": 324.4043,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.0788661303769801,
|
| 55 |
+
"grad_norm": 1318.7249755859375,
|
| 56 |
+
"learning_rate": 2.1875e-05,
|
| 57 |
+
"loss": 303.6339,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.09013272043083441,
|
| 62 |
+
"grad_norm": 1085.81982421875,
|
| 63 |
+
"learning_rate": 2.5e-05,
|
| 64 |
+
"loss": 290.8328,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.1013993104846887,
|
| 69 |
+
"grad_norm": 1732.97412109375,
|
| 70 |
+
"learning_rate": 2.8125000000000003e-05,
|
| 71 |
+
"loss": 286.8244,
|
| 72 |
+
"step": 4500
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.112665900538543,
|
| 76 |
+
"grad_norm": 1018.1327514648438,
|
| 77 |
+
"learning_rate": 3.125e-05,
|
| 78 |
+
"loss": 274.4126,
|
| 79 |
+
"step": 5000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.1239324905923973,
|
| 83 |
+
"grad_norm": 1115.255859375,
|
| 84 |
+
"learning_rate": 3.4375e-05,
|
| 85 |
+
"loss": 258.1825,
|
| 86 |
+
"step": 5500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.1351990806462516,
|
| 90 |
+
"grad_norm": 2264.675537109375,
|
| 91 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 92 |
+
"loss": 249.4162,
|
| 93 |
+
"step": 6000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.1464656707001059,
|
| 97 |
+
"grad_norm": 1202.668212890625,
|
| 98 |
+
"learning_rate": 4.0625000000000005e-05,
|
| 99 |
+
"loss": 238.756,
|
| 100 |
+
"step": 6500
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.1577322607539602,
|
| 104 |
+
"grad_norm": 1271.8997802734375,
|
| 105 |
+
"learning_rate": 4.375e-05,
|
| 106 |
+
"loss": 234.6415,
|
| 107 |
+
"step": 7000
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.16899885080781452,
|
| 111 |
+
"grad_norm": 880.48291015625,
|
| 112 |
+
"learning_rate": 4.6875e-05,
|
| 113 |
+
"loss": 220.2277,
|
| 114 |
+
"step": 7500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.18026544086166882,
|
| 118 |
+
"grad_norm": 1199.511962890625,
|
| 119 |
+
"learning_rate": 5e-05,
|
| 120 |
+
"loss": 212.284,
|
| 121 |
+
"step": 8000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.19153203091552312,
|
| 125 |
+
"grad_norm": 885.7015991210938,
|
| 126 |
+
"learning_rate": 4.999405067699773e-05,
|
| 127 |
+
"loss": 197.9255,
|
| 128 |
+
"step": 8500
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.2027986209693774,
|
| 132 |
+
"grad_norm": 1130.3941650390625,
|
| 133 |
+
"learning_rate": 4.997620553954645e-05,
|
| 134 |
+
"loss": 201.8366,
|
| 135 |
+
"step": 9000
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.2140652110232317,
|
| 139 |
+
"grad_norm": 1032.0145263671875,
|
| 140 |
+
"learning_rate": 4.994647308096509e-05,
|
| 141 |
+
"loss": 195.4686,
|
| 142 |
+
"step": 9500
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.225331801077086,
|
| 146 |
+
"grad_norm": 1835.482666015625,
|
| 147 |
+
"learning_rate": 4.990486745229364e-05,
|
| 148 |
+
"loss": 186.7062,
|
| 149 |
+
"step": 10000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.2365983911309403,
|
| 153 |
+
"grad_norm": 1477.0096435546875,
|
| 154 |
+
"learning_rate": 4.985140845555799e-05,
|
| 155 |
+
"loss": 188.2111,
|
| 156 |
+
"step": 10500
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.2478649811847946,
|
| 160 |
+
"grad_norm": 688.68310546875,
|
| 161 |
+
"learning_rate": 4.9786121534345265e-05,
|
| 162 |
+
"loss": 180.3121,
|
| 163 |
+
"step": 11000
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.2591315712386489,
|
| 167 |
+
"grad_norm": 1981.5882568359375,
|
| 168 |
+
"learning_rate": 4.970903776169402e-05,
|
| 169 |
+
"loss": 171.624,
|
| 170 |
+
"step": 11500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.2703981612925032,
|
| 174 |
+
"grad_norm": 1986.2034912109375,
|
| 175 |
+
"learning_rate": 4.962019382530521e-05,
|
| 176 |
+
"loss": 178.433,
|
| 177 |
+
"step": 12000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.2816647513463575,
|
| 181 |
+
"grad_norm": 1576.7353515625,
|
| 182 |
+
"learning_rate": 4.951963201008076e-05,
|
| 183 |
+
"loss": 164.5357,
|
| 184 |
+
"step": 12500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.2929313414002118,
|
| 188 |
+
"grad_norm": 1520.091796875,
|
| 189 |
+
"learning_rate": 4.940740017799833e-05,
|
| 190 |
+
"loss": 162.9244,
|
| 191 |
+
"step": 13000
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.3041979314540661,
|
| 195 |
+
"grad_norm": 1361.819091796875,
|
| 196 |
+
"learning_rate": 4.9283551745331534e-05,
|
| 197 |
+
"loss": 164.2202,
|
| 198 |
+
"step": 13500
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.3154645215079204,
|
| 202 |
+
"grad_norm": 1085.12548828125,
|
| 203 |
+
"learning_rate": 4.914814565722671e-05,
|
| 204 |
+
"loss": 164.809,
|
| 205 |
+
"step": 14000
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.3267311115617747,
|
| 209 |
+
"grad_norm": 1452.0218505859375,
|
| 210 |
+
"learning_rate": 4.9001246359648224e-05,
|
| 211 |
+
"loss": 162.6041,
|
| 212 |
+
"step": 14500
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.33799770161562903,
|
| 216 |
+
"grad_norm": 1728.5352783203125,
|
| 217 |
+
"learning_rate": 4.884292376870567e-05,
|
| 218 |
+
"loss": 156.4712,
|
| 219 |
+
"step": 15000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.3492642916694833,
|
| 223 |
+
"grad_norm": 1888.999267578125,
|
| 224 |
+
"learning_rate": 4.867325323737765e-05,
|
| 225 |
+
"loss": 151.6581,
|
| 226 |
+
"step": 15500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.36053088172333764,
|
| 230 |
+
"grad_norm": 1127.0830078125,
|
| 231 |
+
"learning_rate": 4.849231551964771e-05,
|
| 232 |
+
"loss": 154.7625,
|
| 233 |
+
"step": 16000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.3717974717771919,
|
| 237 |
+
"grad_norm": 1579.301513671875,
|
| 238 |
+
"learning_rate": 4.830019673206997e-05,
|
| 239 |
+
"loss": 158.3963,
|
| 240 |
+
"step": 16500
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.38306406183104624,
|
| 244 |
+
"grad_norm": 1466.5936279296875,
|
| 245 |
+
"learning_rate": 4.8096988312782174e-05,
|
| 246 |
+
"loss": 151.9632,
|
| 247 |
+
"step": 17000
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.3943306518849005,
|
| 251 |
+
"grad_norm": 1501.413330078125,
|
| 252 |
+
"learning_rate": 4.788278697798618e-05,
|
| 253 |
+
"loss": 152.0479,
|
| 254 |
+
"step": 17500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.4055972419387548,
|
| 258 |
+
"grad_norm": 938.9967651367188,
|
| 259 |
+
"learning_rate": 4.765769467591625e-05,
|
| 260 |
+
"loss": 147.7795,
|
| 261 |
+
"step": 18000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.4168638319926091,
|
| 265 |
+
"grad_norm": 1407.6708984375,
|
| 266 |
+
"learning_rate": 4.742181853831721e-05,
|
| 267 |
+
"loss": 145.714,
|
| 268 |
+
"step": 18500
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.4281304220464634,
|
| 272 |
+
"grad_norm": 1046.4781494140625,
|
| 273 |
+
"learning_rate": 4.717527082945554e-05,
|
| 274 |
+
"loss": 147.87,
|
| 275 |
+
"step": 19000
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.43939701210031773,
|
| 279 |
+
"grad_norm": 1437.9764404296875,
|
| 280 |
+
"learning_rate": 4.69181688926877e-05,
|
| 281 |
+
"loss": 139.7023,
|
| 282 |
+
"step": 19500
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.450663602154172,
|
| 286 |
+
"grad_norm": 1155.10595703125,
|
| 287 |
+
"learning_rate": 4.665063509461097e-05,
|
| 288 |
+
"loss": 146.1232,
|
| 289 |
+
"step": 20000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.46193019220802634,
|
| 293 |
+
"grad_norm": 1317.321044921875,
|
| 294 |
+
"learning_rate": 4.637279676682367e-05,
|
| 295 |
+
"loss": 139.9448,
|
| 296 |
+
"step": 20500
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.4731967822618806,
|
| 300 |
+
"grad_norm": 1005.6251831054688,
|
| 301 |
+
"learning_rate": 4.608478614532215e-05,
|
| 302 |
+
"loss": 142.7613,
|
| 303 |
+
"step": 21000
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.48446337231573494,
|
| 307 |
+
"grad_norm": 2557.992919921875,
|
| 308 |
+
"learning_rate": 4.5786740307563636e-05,
|
| 309 |
+
"loss": 139.5429,
|
| 310 |
+
"step": 21500
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.4957299623695892,
|
| 314 |
+
"grad_norm": 1912.8707275390625,
|
| 315 |
+
"learning_rate": 4.54788011072248e-05,
|
| 316 |
+
"loss": 138.4188,
|
| 317 |
+
"step": 22000
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.5069965524234435,
|
| 321 |
+
"grad_norm": 1349.5655517578125,
|
| 322 |
+
"learning_rate": 4.516111510668707e-05,
|
| 323 |
+
"loss": 135.984,
|
| 324 |
+
"step": 22500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.5182631424772978,
|
| 328 |
+
"grad_norm": 2255.250732421875,
|
| 329 |
+
"learning_rate": 4.4833833507280884e-05,
|
| 330 |
+
"loss": 136.2522,
|
| 331 |
+
"step": 23000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.5295297325311521,
|
| 335 |
+
"grad_norm": 947.2132568359375,
|
| 336 |
+
"learning_rate": 4.4497112077322044e-05,
|
| 337 |
+
"loss": 135.873,
|
| 338 |
+
"step": 23500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.5407963225850064,
|
| 342 |
+
"grad_norm": 1092.2021484375,
|
| 343 |
+
"learning_rate": 4.415111107797445e-05,
|
| 344 |
+
"loss": 136.0968,
|
| 345 |
+
"step": 24000
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.5520629126388608,
|
| 349 |
+
"grad_norm": 2551.856201171875,
|
| 350 |
+
"learning_rate": 4.379599518697444e-05,
|
| 351 |
+
"loss": 133.4127,
|
| 352 |
+
"step": 24500
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.563329502692715,
|
| 356 |
+
"grad_norm": 1087.750732421875,
|
| 357 |
+
"learning_rate": 4.34319334202531e-05,
|
| 358 |
+
"loss": 137.0111,
|
| 359 |
+
"step": 25000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.5745960927465693,
|
| 363 |
+
"grad_norm": 710.9840698242188,
|
| 364 |
+
"learning_rate": 4.305909905149389e-05,
|
| 365 |
+
"loss": 128.7961,
|
| 366 |
+
"step": 25500
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.5858626828004236,
|
| 370 |
+
"grad_norm": 1028.0732421875,
|
| 371 |
+
"learning_rate": 4.267766952966369e-05,
|
| 372 |
+
"loss": 131.5538,
|
| 373 |
+
"step": 26000
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.597129272854278,
|
| 377 |
+
"grad_norm": 1650.9874267578125,
|
| 378 |
+
"learning_rate": 4.228782639455674e-05,
|
| 379 |
+
"loss": 131.9968,
|
| 380 |
+
"step": 26500
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.6083958629081322,
|
| 384 |
+
"grad_norm": 3326.53564453125,
|
| 385 |
+
"learning_rate": 4.188975519039151e-05,
|
| 386 |
+
"loss": 129.6315,
|
| 387 |
+
"step": 27000
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.6196624529619865,
|
| 391 |
+
"grad_norm": 2218.9794921875,
|
| 392 |
+
"learning_rate": 4.148364537750172e-05,
|
| 393 |
+
"loss": 126.7289,
|
| 394 |
+
"step": 27500
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.6309290430158409,
|
| 398 |
+
"grad_norm": 1092.4063720703125,
|
| 399 |
+
"learning_rate": 4.1069690242163484e-05,
|
| 400 |
+
"loss": 131.7434,
|
| 401 |
+
"step": 28000
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.6421956330696951,
|
| 405 |
+
"grad_norm": 1642.626220703125,
|
| 406 |
+
"learning_rate": 4.064808680460148e-05,
|
| 407 |
+
"loss": 131.2289,
|
| 408 |
+
"step": 28500
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.6534622231235494,
|
| 412 |
+
"grad_norm": 2752.57470703125,
|
| 413 |
+
"learning_rate": 4.021903572521802e-05,
|
| 414 |
+
"loss": 125.4669,
|
| 415 |
+
"step": 29000
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.6647288131774037,
|
| 419 |
+
"grad_norm": 2134.545654296875,
|
| 420 |
+
"learning_rate": 3.978274120908956e-05,
|
| 421 |
+
"loss": 128.8268,
|
| 422 |
+
"step": 29500
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.6759954032312581,
|
| 426 |
+
"grad_norm": 1844.7005615234375,
|
| 427 |
+
"learning_rate": 3.933941090877615e-05,
|
| 428 |
+
"loss": 126.0543,
|
| 429 |
+
"step": 30000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.6872619932851123,
|
| 433 |
+
"grad_norm": 911.3765869140625,
|
| 434 |
+
"learning_rate": 3.888925582549006e-05,
|
| 435 |
+
"loss": 124.5692,
|
| 436 |
+
"step": 30500
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.6985285833389666,
|
| 440 |
+
"grad_norm": 1240.784423828125,
|
| 441 |
+
"learning_rate": 3.84324902086706e-05,
|
| 442 |
+
"loss": 129.9127,
|
| 443 |
+
"step": 31000
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.7097951733928209,
|
| 447 |
+
"grad_norm": 1387.1654052734375,
|
| 448 |
+
"learning_rate": 3.796933145401304e-05,
|
| 449 |
+
"loss": 128.229,
|
| 450 |
+
"step": 31500
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.7210617634466753,
|
| 454 |
+
"grad_norm": 5207.958984375,
|
| 455 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 456 |
+
"loss": 123.6102,
|
| 457 |
+
"step": 32000
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.7323283535005295,
|
| 461 |
+
"grad_norm": 1478.238525390625,
|
| 462 |
+
"learning_rate": 3.702471922298469e-05,
|
| 463 |
+
"loss": 122.5027,
|
| 464 |
+
"step": 32500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.7435949435543838,
|
| 468 |
+
"grad_norm": 1219.868408203125,
|
| 469 |
+
"learning_rate": 3.654371533087586e-05,
|
| 470 |
+
"loss": 121.6549,
|
| 471 |
+
"step": 33000
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.7548615336082382,
|
| 475 |
+
"grad_norm": 1724.080078125,
|
| 476 |
+
"learning_rate": 3.6057217255475034e-05,
|
| 477 |
+
"loss": 122.2066,
|
| 478 |
+
"step": 33500
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.7661281236620925,
|
| 482 |
+
"grad_norm": 2315.779052734375,
|
| 483 |
+
"learning_rate": 3.556545654351749e-05,
|
| 484 |
+
"loss": 124.5871,
|
| 485 |
+
"step": 34000
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.7773947137159467,
|
| 489 |
+
"grad_norm": 1415.927978515625,
|
| 490 |
+
"learning_rate": 3.5068667246468436e-05,
|
| 491 |
+
"loss": 119.864,
|
| 492 |
+
"step": 34500
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.788661303769801,
|
| 496 |
+
"grad_norm": 2292.79736328125,
|
| 497 |
+
"learning_rate": 3.456708580912725e-05,
|
| 498 |
+
"loss": 124.4397,
|
| 499 |
+
"step": 35000
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.7999278938236554,
|
| 503 |
+
"grad_norm": 2198.923583984375,
|
| 504 |
+
"learning_rate": 3.406095095709254e-05,
|
| 505 |
+
"loss": 119.8706,
|
| 506 |
+
"step": 35500
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.8111944838775096,
|
| 510 |
+
"grad_norm": 774.8341064453125,
|
| 511 |
+
"learning_rate": 3.355050358314172e-05,
|
| 512 |
+
"loss": 122.3335,
|
| 513 |
+
"step": 36000
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.8224610739313639,
|
| 517 |
+
"grad_norm": 1356.7291259765625,
|
| 518 |
+
"learning_rate": 3.303598663257904e-05,
|
| 519 |
+
"loss": 117.6119,
|
| 520 |
+
"step": 36500
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.8337276639852182,
|
| 524 |
+
"grad_norm": 0.0,
|
| 525 |
+
"learning_rate": 3.251764498760683e-05,
|
| 526 |
+
"loss": 122.5679,
|
| 527 |
+
"step": 37000
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.8449942540390726,
|
| 531 |
+
"grad_norm": 0.0,
|
| 532 |
+
"learning_rate": 3.1995725350774806e-05,
|
| 533 |
+
"loss": 116.4365,
|
| 534 |
+
"step": 37500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.8562608440929268,
|
| 538 |
+
"grad_norm": 1365.280517578125,
|
| 539 |
+
"learning_rate": 3.147047612756302e-05,
|
| 540 |
+
"loss": 119.871,
|
| 541 |
+
"step": 38000
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.8675274341467811,
|
| 545 |
+
"grad_norm": 1635.288330078125,
|
| 546 |
+
"learning_rate": 3.094214730815433e-05,
|
| 547 |
+
"loss": 116.4286,
|
| 548 |
+
"step": 38500
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.8787940242006355,
|
| 552 |
+
"grad_norm": 1430.511962890625,
|
| 553 |
+
"learning_rate": 3.0410990348452573e-05,
|
| 554 |
+
"loss": 113.9206,
|
| 555 |
+
"step": 39000
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.8900606142544898,
|
| 559 |
+
"grad_norm": 1990.450927734375,
|
| 560 |
+
"learning_rate": 2.9877258050403212e-05,
|
| 561 |
+
"loss": 118.977,
|
| 562 |
+
"step": 39500
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.901327204308344,
|
| 566 |
+
"grad_norm": 875.0726928710938,
|
| 567 |
+
"learning_rate": 2.9341204441673266e-05,
|
| 568 |
+
"loss": 114.4106,
|
| 569 |
+
"step": 40000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.9125937943621983,
|
| 573 |
+
"grad_norm": 1655.7935791015625,
|
| 574 |
+
"learning_rate": 2.8803084654747918e-05,
|
| 575 |
+
"loss": 115.3111,
|
| 576 |
+
"step": 40500
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.9238603844160527,
|
| 580 |
+
"grad_norm": 1956.72216796875,
|
| 581 |
+
"learning_rate": 2.8263154805501297e-05,
|
| 582 |
+
"loss": 117.3135,
|
| 583 |
+
"step": 41000
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.9351269744699069,
|
| 587 |
+
"grad_norm": 937.9488525390625,
|
| 588 |
+
"learning_rate": 2.7721671871299116e-05,
|
| 589 |
+
"loss": 116.4852,
|
| 590 |
+
"step": 41500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.9463935645237612,
|
| 594 |
+
"grad_norm": 1579.9736328125,
|
| 595 |
+
"learning_rate": 2.717889356869146e-05,
|
| 596 |
+
"loss": 118.1533,
|
| 597 |
+
"step": 42000
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.9576601545776156,
|
| 601 |
+
"grad_norm": 1284.749755859375,
|
| 602 |
+
"learning_rate": 2.663507823075358e-05,
|
| 603 |
+
"loss": 113.1541,
|
| 604 |
+
"step": 42500
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.9689267446314699,
|
| 608 |
+
"grad_norm": 1050.632080078125,
|
| 609 |
+
"learning_rate": 2.6090484684133404e-05,
|
| 610 |
+
"loss": 115.3209,
|
| 611 |
+
"step": 43000
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.9801933346853241,
|
| 615 |
+
"grad_norm": 1058.0616455078125,
|
| 616 |
+
"learning_rate": 2.5545372125864032e-05,
|
| 617 |
+
"loss": 119.323,
|
| 618 |
+
"step": 43500
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.9914599247391784,
|
| 622 |
+
"grad_norm": 1310.6236572265625,
|
| 623 |
+
"learning_rate": 2.5e-05,
|
| 624 |
+
"loss": 111.4601,
|
| 625 |
+
"step": 44000
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.0,
|
| 629 |
+
"eval_loss": 114.70393371582031,
|
| 630 |
+
"eval_runtime": 1385.5714,
|
| 631 |
+
"eval_samples_per_second": 14.238,
|
| 632 |
+
"eval_steps_per_second": 3.56,
|
| 633 |
+
"step": 44379
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 1.0027265147930327,
|
| 637 |
+
"grad_norm": 1699.930419921875,
|
| 638 |
+
"learning_rate": 2.4454627874135974e-05,
|
| 639 |
+
"loss": 111.6484,
|
| 640 |
+
"step": 44500
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 1.013993104846887,
|
| 644 |
+
"grad_norm": 0.0,
|
| 645 |
+
"learning_rate": 2.3909515315866605e-05,
|
| 646 |
+
"loss": 111.7967,
|
| 647 |
+
"step": 45000
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.0252596949007413,
|
| 651 |
+
"grad_norm": 1440.404052734375,
|
| 652 |
+
"learning_rate": 2.3364921769246423e-05,
|
| 653 |
+
"loss": 107.7449,
|
| 654 |
+
"step": 45500
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.0365262849545958,
|
| 658 |
+
"grad_norm": 1180.1439208984375,
|
| 659 |
+
"learning_rate": 2.2821106431308544e-05,
|
| 660 |
+
"loss": 109.0394,
|
| 661 |
+
"step": 46000
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.04779287500845,
|
| 665 |
+
"grad_norm": 1600.07568359375,
|
| 666 |
+
"learning_rate": 2.2278328128700893e-05,
|
| 667 |
+
"loss": 104.1056,
|
| 668 |
+
"step": 46500
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.0590594650623042,
|
| 672 |
+
"grad_norm": 1731.480224609375,
|
| 673 |
+
"learning_rate": 2.173684519449872e-05,
|
| 674 |
+
"loss": 106.6714,
|
| 675 |
+
"step": 47000
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 1.0703260551161586,
|
| 679 |
+
"grad_norm": 955.4271240234375,
|
| 680 |
+
"learning_rate": 2.1196915345252084e-05,
|
| 681 |
+
"loss": 105.6627,
|
| 682 |
+
"step": 47500
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 1.0815926451700129,
|
| 686 |
+
"grad_norm": 2818.5283203125,
|
| 687 |
+
"learning_rate": 2.0658795558326743e-05,
|
| 688 |
+
"loss": 106.0866,
|
| 689 |
+
"step": 48000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.092859235223867,
|
| 693 |
+
"grad_norm": 806.58447265625,
|
| 694 |
+
"learning_rate": 2.0122741949596797e-05,
|
| 695 |
+
"loss": 111.0945,
|
| 696 |
+
"step": 48500
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 1.1041258252777215,
|
| 700 |
+
"grad_norm": 1397.9573974609375,
|
| 701 |
+
"learning_rate": 1.958900965154743e-05,
|
| 702 |
+
"loss": 104.8343,
|
| 703 |
+
"step": 49000
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 1.1153924153315757,
|
| 707 |
+
"grad_norm": 827.4343872070312,
|
| 708 |
+
"learning_rate": 1.9057852691845677e-05,
|
| 709 |
+
"loss": 103.2378,
|
| 710 |
+
"step": 49500
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 1.12665900538543,
|
| 714 |
+
"grad_norm": 2430.602294921875,
|
| 715 |
+
"learning_rate": 1.852952387243698e-05,
|
| 716 |
+
"loss": 107.4349,
|
| 717 |
+
"step": 50000
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 1.1379255954392844,
|
| 721 |
+
"grad_norm": 1653.0751953125,
|
| 722 |
+
"learning_rate": 1.80042746492252e-05,
|
| 723 |
+
"loss": 111.0688,
|
| 724 |
+
"step": 50500
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 1.1491921854931386,
|
| 728 |
+
"grad_norm": 1318.4361572265625,
|
| 729 |
+
"learning_rate": 1.7482355012393177e-05,
|
| 730 |
+
"loss": 107.7423,
|
| 731 |
+
"step": 51000
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 1.160458775546993,
|
| 735 |
+
"grad_norm": 1956.568603515625,
|
| 736 |
+
"learning_rate": 1.6964013367420966e-05,
|
| 737 |
+
"loss": 103.8008,
|
| 738 |
+
"step": 51500
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 1.1717253656008473,
|
| 742 |
+
"grad_norm": 764.591552734375,
|
| 743 |
+
"learning_rate": 1.6449496416858284e-05,
|
| 744 |
+
"loss": 104.8202,
|
| 745 |
+
"step": 52000
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 1.1829919556547015,
|
| 749 |
+
"grad_norm": 1741.2701416015625,
|
| 750 |
+
"learning_rate": 1.5939049042907462e-05,
|
| 751 |
+
"loss": 107.5796,
|
| 752 |
+
"step": 52500
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.194258545708556,
|
| 756 |
+
"grad_norm": 2550.640625,
|
| 757 |
+
"learning_rate": 1.5432914190872757e-05,
|
| 758 |
+
"loss": 101.9078,
|
| 759 |
+
"step": 53000
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 1.2055251357624102,
|
| 763 |
+
"grad_norm": 698.9768676757812,
|
| 764 |
+
"learning_rate": 1.4931332753531574e-05,
|
| 765 |
+
"loss": 108.1872,
|
| 766 |
+
"step": 53500
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 1.2167917258162644,
|
| 770 |
+
"grad_norm": 3405.810546875,
|
| 771 |
+
"learning_rate": 1.443454345648252e-05,
|
| 772 |
+
"loss": 99.6448,
|
| 773 |
+
"step": 54000
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 1.2280583158701188,
|
| 777 |
+
"grad_norm": 3267.17236328125,
|
| 778 |
+
"learning_rate": 1.3942782744524973e-05,
|
| 779 |
+
"loss": 108.0154,
|
| 780 |
+
"step": 54500
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 1.239324905923973,
|
| 784 |
+
"grad_norm": 1127.849853515625,
|
| 785 |
+
"learning_rate": 1.3456284669124158e-05,
|
| 786 |
+
"loss": 100.0244,
|
| 787 |
+
"step": 55000
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 1.2505914959778273,
|
| 791 |
+
"grad_norm": 1082.1241455078125,
|
| 792 |
+
"learning_rate": 1.2975280777015314e-05,
|
| 793 |
+
"loss": 104.4581,
|
| 794 |
+
"step": 55500
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 1.2618580860316817,
|
| 798 |
+
"grad_norm": 1860.104248046875,
|
| 799 |
+
"learning_rate": 1.2500000000000006e-05,
|
| 800 |
+
"loss": 103.1496,
|
| 801 |
+
"step": 56000
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 1.273124676085536,
|
| 805 |
+
"grad_norm": 1682.26708984375,
|
| 806 |
+
"learning_rate": 1.2030668545986959e-05,
|
| 807 |
+
"loss": 107.3074,
|
| 808 |
+
"step": 56500
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 1.2843912661393904,
|
| 812 |
+
"grad_norm": 616.0757446289062,
|
| 813 |
+
"learning_rate": 1.1567509791329401e-05,
|
| 814 |
+
"loss": 102.4632,
|
| 815 |
+
"step": 57000
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 1.2956578561932446,
|
| 819 |
+
"grad_norm": 624.0526733398438,
|
| 820 |
+
"learning_rate": 1.1110744174509952e-05,
|
| 821 |
+
"loss": 100.0244,
|
| 822 |
+
"step": 57500
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 1.3069244462470988,
|
| 826 |
+
"grad_norm": 1387.1529541015625,
|
| 827 |
+
"learning_rate": 1.0660589091223855e-05,
|
| 828 |
+
"loss": 105.9446,
|
| 829 |
+
"step": 58000
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 1.318191036300953,
|
| 833 |
+
"grad_norm": 1828.0877685546875,
|
| 834 |
+
"learning_rate": 1.0217258790910448e-05,
|
| 835 |
+
"loss": 104.6986,
|
| 836 |
+
"step": 58500
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 1.3294576263548075,
|
| 840 |
+
"grad_norm": 650.5953369140625,
|
| 841 |
+
"learning_rate": 9.780964274781984e-06,
|
| 842 |
+
"loss": 105.4167,
|
| 843 |
+
"step": 59000
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 1.3407242164086617,
|
| 847 |
+
"grad_norm": 996.328125,
|
| 848 |
+
"learning_rate": 9.351913195398524e-06,
|
| 849 |
+
"loss": 103.2462,
|
| 850 |
+
"step": 59500
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 1.3519908064625161,
|
| 854 |
+
"grad_norm": 1035.3780517578125,
|
| 855 |
+
"learning_rate": 8.930309757836517e-06,
|
| 856 |
+
"loss": 103.7156,
|
| 857 |
+
"step": 60000
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 1.3632573965163703,
|
| 861 |
+
"grad_norm": 1865.9571533203125,
|
| 862 |
+
"learning_rate": 8.51635462249828e-06,
|
| 863 |
+
"loss": 105.3619,
|
| 864 |
+
"step": 60500
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.3745239865702246,
|
| 868 |
+
"grad_norm": 2139.88671875,
|
| 869 |
+
"learning_rate": 8.110244809608495e-06,
|
| 870 |
+
"loss": 103.2506,
|
| 871 |
+
"step": 61000
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 1.385790576624079,
|
| 875 |
+
"grad_norm": 2942.61083984375,
|
| 876 |
+
"learning_rate": 7.712173605443269e-06,
|
| 877 |
+
"loss": 102.9886,
|
| 878 |
+
"step": 61500
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.3970571666779332,
|
| 882 |
+
"grad_norm": 415.38153076171875,
|
| 883 |
+
"learning_rate": 7.3223304703363135e-06,
|
| 884 |
+
"loss": 102.6045,
|
| 885 |
+
"step": 62000
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 1.4083237567317877,
|
| 889 |
+
"grad_norm": 1247.396728515625,
|
| 890 |
+
"learning_rate": 6.940900948506113e-06,
|
| 891 |
+
"loss": 102.5344,
|
| 892 |
+
"step": 62500
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 1.4195903467856419,
|
| 896 |
+
"grad_norm": 740.3623046875,
|
| 897 |
+
"learning_rate": 6.568066579746901e-06,
|
| 898 |
+
"loss": 104.4504,
|
| 899 |
+
"step": 63000
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 1.430856936839496,
|
| 903 |
+
"grad_norm": 1221.581298828125,
|
| 904 |
+
"learning_rate": 6.204004813025568e-06,
|
| 905 |
+
"loss": 101.7631,
|
| 906 |
+
"step": 63500
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 1.4421235268933503,
|
| 910 |
+
"grad_norm": 1514.0115966796875,
|
| 911 |
+
"learning_rate": 5.848888922025553e-06,
|
| 912 |
+
"loss": 99.2285,
|
| 913 |
+
"step": 64000
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.4533901169472048,
|
| 917 |
+
"grad_norm": 821.0801391601562,
|
| 918 |
+
"learning_rate": 5.50288792267796e-06,
|
| 919 |
+
"loss": 102.3846,
|
| 920 |
+
"step": 64500
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.464656707001059,
|
| 924 |
+
"grad_norm": 1313.3104248046875,
|
| 925 |
+
"learning_rate": 5.166166492719124e-06,
|
| 926 |
+
"loss": 103.6035,
|
| 927 |
+
"step": 65000
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 1.4759232970549134,
|
| 931 |
+
"grad_norm": 1329.42919921875,
|
| 932 |
+
"learning_rate": 4.8388848933129335e-06,
|
| 933 |
+
"loss": 104.2133,
|
| 934 |
+
"step": 65500
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.4871898871087676,
|
| 938 |
+
"grad_norm": 3106.43505859375,
|
| 939 |
+
"learning_rate": 4.521198892775203e-06,
|
| 940 |
+
"loss": 100.1638,
|
| 941 |
+
"step": 66000
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.4984564771626219,
|
| 945 |
+
"grad_norm": 996.626220703125,
|
| 946 |
+
"learning_rate": 4.213259692436367e-06,
|
| 947 |
+
"loss": 97.5407,
|
| 948 |
+
"step": 66500
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.5097230672164763,
|
| 952 |
+
"grad_norm": 1821.3323974609375,
|
| 953 |
+
"learning_rate": 3.9152138546778625e-06,
|
| 954 |
+
"loss": 102.4095,
|
| 955 |
+
"step": 67000
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.5209896572703305,
|
| 959 |
+
"grad_norm": 1764.0323486328125,
|
| 960 |
+
"learning_rate": 3.6272032331763408e-06,
|
| 961 |
+
"loss": 96.4079,
|
| 962 |
+
"step": 67500
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.532256247324185,
|
| 966 |
+
"grad_norm": 1377.978271484375,
|
| 967 |
+
"learning_rate": 3.3493649053890326e-06,
|
| 968 |
+
"loss": 104.5846,
|
| 969 |
+
"step": 68000
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.5435228373780392,
|
| 973 |
+
"grad_norm": 1291.5908203125,
|
| 974 |
+
"learning_rate": 3.081831107312308e-06,
|
| 975 |
+
"loss": 100.9904,
|
| 976 |
+
"step": 68500
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.5547894274318934,
|
| 980 |
+
"grad_norm": 2216.6796875,
|
| 981 |
+
"learning_rate": 2.8247291705444575e-06,
|
| 982 |
+
"loss": 99.7059,
|
| 983 |
+
"step": 69000
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.5660560174857476,
|
| 987 |
+
"grad_norm": 1271.5609130859375,
|
| 988 |
+
"learning_rate": 2.578181461682794e-06,
|
| 989 |
+
"loss": 99.0976,
|
| 990 |
+
"step": 69500
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.577322607539602,
|
| 994 |
+
"grad_norm": 2712.908203125,
|
| 995 |
+
"learning_rate": 2.3423053240837515e-06,
|
| 996 |
+
"loss": 99.1392,
|
| 997 |
+
"step": 70000
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.5885891975934565,
|
| 1001 |
+
"grad_norm": 1653.12890625,
|
| 1002 |
+
"learning_rate": 2.1172130220138226e-06,
|
| 1003 |
+
"loss": 101.0942,
|
| 1004 |
+
"step": 70500
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.5998557876473107,
|
| 1008 |
+
"grad_norm": 1816.012451171875,
|
| 1009 |
+
"learning_rate": 1.9030116872178316e-06,
|
| 1010 |
+
"loss": 99.6448,
|
| 1011 |
+
"step": 71000
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 1.611122377701165,
|
| 1015 |
+
"grad_norm": 1233.8719482421875,
|
| 1016 |
+
"learning_rate": 1.6998032679300391e-06,
|
| 1017 |
+
"loss": 101.785,
|
| 1018 |
+
"step": 71500
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.6223889677550192,
|
| 1022 |
+
"grad_norm": 988.7132568359375,
|
| 1023 |
+
"learning_rate": 1.5076844803522922e-06,
|
| 1024 |
+
"loss": 95.7829,
|
| 1025 |
+
"step": 72000
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.6336555578088736,
|
| 1029 |
+
"grad_norm": 1261.22021484375,
|
| 1030 |
+
"learning_rate": 1.3267467626223606e-06,
|
| 1031 |
+
"loss": 105.201,
|
| 1032 |
+
"step": 72500
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.6449221478627278,
|
| 1036 |
+
"grad_norm": 934.3240966796875,
|
| 1037 |
+
"learning_rate": 1.1570762312943295e-06,
|
| 1038 |
+
"loss": 98.9473,
|
| 1039 |
+
"step": 73000
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.6561887379165823,
|
| 1043 |
+
"grad_norm": 664.5089111328125,
|
| 1044 |
+
"learning_rate": 9.98753640351785e-07,
|
| 1045 |
+
"loss": 95.5199,
|
| 1046 |
+
"step": 73500
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.6674553279704365,
|
| 1050 |
+
"grad_norm": 1963.7371826171875,
|
| 1051 |
+
"learning_rate": 8.51854342773295e-07,
|
| 1052 |
+
"loss": 101.1854,
|
| 1053 |
+
"step": 74000
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 1.6787219180242907,
|
| 1057 |
+
"grad_norm": 2443.130126953125,
|
| 1058 |
+
"learning_rate": 7.164482546684642e-07,
|
| 1059 |
+
"loss": 97.2825,
|
| 1060 |
+
"step": 74500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.689988508078145,
|
| 1064 |
+
"grad_norm": 1137.326416015625,
|
| 1065 |
+
"learning_rate": 5.925998220016659e-07,
|
| 1066 |
+
"loss": 100.3116,
|
| 1067 |
+
"step": 75000
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.7012550981319994,
|
| 1071 |
+
"grad_norm": 1520.0399169921875,
|
| 1072 |
+
"learning_rate": 4.803679899192392e-07,
|
| 1073 |
+
"loss": 100.4226,
|
| 1074 |
+
"step": 75500
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.7125216881858538,
|
| 1078 |
+
"grad_norm": 1065.4630126953125,
|
| 1079 |
+
"learning_rate": 3.7980617469479953e-07,
|
| 1080 |
+
"loss": 101.6046,
|
| 1081 |
+
"step": 76000
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.723788278239708,
|
| 1085 |
+
"grad_norm": 1042.942138671875,
|
| 1086 |
+
"learning_rate": 2.909622383059835e-07,
|
| 1087 |
+
"loss": 99.1135,
|
| 1088 |
+
"step": 76500
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.7350548682935623,
|
| 1092 |
+
"grad_norm": 841.5830078125,
|
| 1093 |
+
"learning_rate": 2.1387846565474045e-07,
|
| 1094 |
+
"loss": 101.7611,
|
| 1095 |
+
"step": 77000
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.7463214583474165,
|
| 1099 |
+
"grad_norm": 3369.8330078125,
|
| 1100 |
+
"learning_rate": 1.4859154444200884e-07,
|
| 1101 |
+
"loss": 99.4417,
|
| 1102 |
+
"step": 77500
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.757588048401271,
|
| 1106 |
+
"grad_norm": 2432.385498046875,
|
| 1107 |
+
"learning_rate": 9.513254770636137e-08,
|
| 1108 |
+
"loss": 100.2592,
|
| 1109 |
+
"step": 78000
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.7688546384551251,
|
| 1113 |
+
"grad_norm": 1680.1993408203125,
|
| 1114 |
+
"learning_rate": 5.352691903491303e-08,
|
| 1115 |
+
"loss": 98.8878,
|
| 1116 |
+
"step": 78500
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.7801212285089796,
|
| 1120 |
+
"grad_norm": 682.8377075195312,
|
| 1121 |
+
"learning_rate": 2.3794460453555047e-08,
|
| 1122 |
+
"loss": 97.455,
|
| 1123 |
+
"step": 79000
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.7913878185628338,
|
| 1127 |
+
"grad_norm": 0.0,
|
| 1128 |
+
"learning_rate": 5.94932300227169e-09,
|
| 1129 |
+
"loss": 102.4157,
|
| 1130 |
+
"step": 79500
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.802654408616688,
|
| 1134 |
+
"grad_norm": 3429.1298828125,
|
| 1135 |
+
"learning_rate": 0.0,
|
| 1136 |
+
"loss": 97.4711,
|
| 1137 |
+
"step": 80000
|
| 1138 |
+
}
|
| 1139 |
+
],
|
| 1140 |
+
"logging_steps": 500,
|
| 1141 |
+
"max_steps": 80000,
|
| 1142 |
+
"num_input_tokens_seen": 0,
|
| 1143 |
+
"num_train_epochs": 2,
|
| 1144 |
+
"save_steps": 20000,
|
| 1145 |
+
"stateful_callbacks": {
|
| 1146 |
+
"TrainerControl": {
|
| 1147 |
+
"args": {
|
| 1148 |
+
"should_epoch_stop": false,
|
| 1149 |
+
"should_evaluate": false,
|
| 1150 |
+
"should_log": false,
|
| 1151 |
+
"should_save": true,
|
| 1152 |
+
"should_training_stop": true
|
| 1153 |
+
},
|
| 1154 |
+
"attributes": {}
|
| 1155 |
+
}
|
| 1156 |
+
},
|
| 1157 |
+
"total_flos": 0.0,
|
| 1158 |
+
"train_batch_size": 4,
|
| 1159 |
+
"trial_name": null,
|
| 1160 |
+
"trial_params": null
|
| 1161 |
+
}
|