first commit
Browse files- config.json +185 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +1 -0
- tokenizer.json +0 -0
- tokenizer_config.json +1 -0
- trainer_state.json +370 -0
- vocab.txt +0 -0
config.json
ADDED
|
@@ -0,0 +1,185 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "metamong1/bigbird-tapt-ep3",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"EncoderDecoderModel"
|
| 5 |
+
],
|
| 6 |
+
"decoder": {
|
| 7 |
+
"_name_or_path": "gogamza/kobart-base-v1",
|
| 8 |
+
"activation_dropout": 0.0,
|
| 9 |
+
"activation_function": "gelu",
|
| 10 |
+
"add_bias_logits": false,
|
| 11 |
+
"add_cross_attention": true,
|
| 12 |
+
"add_final_layer_norm": false,
|
| 13 |
+
"architectures": [
|
| 14 |
+
"BartModel"
|
| 15 |
+
],
|
| 16 |
+
"attention_dropout": 0.0,
|
| 17 |
+
"author": "Heewon Jeon(madjakarta@gmail.com)",
|
| 18 |
+
"bad_words_ids": null,
|
| 19 |
+
"bos_token_id": 0,
|
| 20 |
+
"chunk_size_feed_forward": 0,
|
| 21 |
+
"classif_dropout": 0.1,
|
| 22 |
+
"classifier_dropout": 0.1,
|
| 23 |
+
"d_model": 768,
|
| 24 |
+
"decoder_attention_heads": 16,
|
| 25 |
+
"decoder_ffn_dim": 3072,
|
| 26 |
+
"decoder_layerdrop": 0.0,
|
| 27 |
+
"decoder_layers": 6,
|
| 28 |
+
"decoder_start_token_id": 1,
|
| 29 |
+
"diversity_penalty": 0.0,
|
| 30 |
+
"do_blenderbot_90_layernorm": false,
|
| 31 |
+
"do_sample": false,
|
| 32 |
+
"doc_type_size": 4,
|
| 33 |
+
"dropout": 0.1,
|
| 34 |
+
"early_stopping": false,
|
| 35 |
+
"encoder_attention_heads": 16,
|
| 36 |
+
"encoder_ffn_dim": 3072,
|
| 37 |
+
"encoder_layerdrop": 0.0,
|
| 38 |
+
"encoder_layers": 6,
|
| 39 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 40 |
+
"eos_token_id": 1,
|
| 41 |
+
"extra_pos_embeddings": 2,
|
| 42 |
+
"finetuning_task": null,
|
| 43 |
+
"force_bos_token_to_be_generated": false,
|
| 44 |
+
"forced_bos_token_id": null,
|
| 45 |
+
"forced_eos_token_id": 1,
|
| 46 |
+
"gradient_checkpointing": false,
|
| 47 |
+
"id2label": {
|
| 48 |
+
"0": "NEGATIVE",
|
| 49 |
+
"1": "POSITIVE"
|
| 50 |
+
},
|
| 51 |
+
"init_std": 0.02,
|
| 52 |
+
"is_decoder": true,
|
| 53 |
+
"is_encoder_decoder": true,
|
| 54 |
+
"kobart_version": 1.0,
|
| 55 |
+
"label2id": {
|
| 56 |
+
"NEGATIVE": 0,
|
| 57 |
+
"POSITIVE": 1
|
| 58 |
+
},
|
| 59 |
+
"length_penalty": 1.0,
|
| 60 |
+
"max_length": 20,
|
| 61 |
+
"max_position_embeddings": 2048,
|
| 62 |
+
"min_length": 0,
|
| 63 |
+
"model_type": "bart",
|
| 64 |
+
"no_repeat_ngram_size": 0,
|
| 65 |
+
"normalize_before": false,
|
| 66 |
+
"normalize_embedding": true,
|
| 67 |
+
"num_beam_groups": 1,
|
| 68 |
+
"num_beams": 1,
|
| 69 |
+
"num_hidden_layers": 6,
|
| 70 |
+
"num_return_sequences": 1,
|
| 71 |
+
"output_attentions": false,
|
| 72 |
+
"output_hidden_states": false,
|
| 73 |
+
"output_scores": false,
|
| 74 |
+
"pad_token_id": 0,
|
| 75 |
+
"prefix": null,
|
| 76 |
+
"problem_type": null,
|
| 77 |
+
"pruned_heads": {},
|
| 78 |
+
"remove_invalid_values": false,
|
| 79 |
+
"repetition_penalty": 1.0,
|
| 80 |
+
"return_dict": true,
|
| 81 |
+
"return_dict_in_generate": false,
|
| 82 |
+
"scale_embedding": false,
|
| 83 |
+
"sep_token_id": null,
|
| 84 |
+
"static_position_embeddings": false,
|
| 85 |
+
"task_specific_params": null,
|
| 86 |
+
"temperature": 1.0,
|
| 87 |
+
"tie_encoder_decoder": false,
|
| 88 |
+
"tie_word_embeddings": true,
|
| 89 |
+
"tokenizer_class": "PreTrainedTokenizerFast",
|
| 90 |
+
"top_k": 50,
|
| 91 |
+
"top_p": 1.0,
|
| 92 |
+
"torch_dtype": null,
|
| 93 |
+
"torchscript": false,
|
| 94 |
+
"transformers_version": "4.11.0",
|
| 95 |
+
"use_bfloat16": false,
|
| 96 |
+
"use_cache": true,
|
| 97 |
+
"vocab_size": 32500
|
| 98 |
+
},
|
| 99 |
+
"encoder": {
|
| 100 |
+
"_name_or_path": "monologg/kobigbird-bert-base",
|
| 101 |
+
"add_cross_attention": false,
|
| 102 |
+
"architectures": [
|
| 103 |
+
"BigBirdForMaskedLM"
|
| 104 |
+
],
|
| 105 |
+
"attention_probs_dropout_prob": 0.1,
|
| 106 |
+
"attention_type": "block_sparse",
|
| 107 |
+
"bad_words_ids": null,
|
| 108 |
+
"block_size": 64,
|
| 109 |
+
"bos_token_id": 5,
|
| 110 |
+
"chunk_size_feed_forward": 0,
|
| 111 |
+
"classifier_dropout": null,
|
| 112 |
+
"decoder_start_token_id": null,
|
| 113 |
+
"diversity_penalty": 0.0,
|
| 114 |
+
"do_sample": false,
|
| 115 |
+
"doc_type_size": 4,
|
| 116 |
+
"early_stopping": false,
|
| 117 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 118 |
+
"eos_token_id": 6,
|
| 119 |
+
"finetuning_task": null,
|
| 120 |
+
"forced_bos_token_id": null,
|
| 121 |
+
"forced_eos_token_id": null,
|
| 122 |
+
"gradient_checkpointing": false,
|
| 123 |
+
"hidden_act": "gelu_new",
|
| 124 |
+
"hidden_dropout_prob": 0.1,
|
| 125 |
+
"hidden_size": 768,
|
| 126 |
+
"id2label": {
|
| 127 |
+
"0": "LABEL_0",
|
| 128 |
+
"1": "LABEL_1"
|
| 129 |
+
},
|
| 130 |
+
"initializer_range": 0.02,
|
| 131 |
+
"intermediate_size": 3072,
|
| 132 |
+
"is_decoder": false,
|
| 133 |
+
"is_encoder_decoder": false,
|
| 134 |
+
"label2id": {
|
| 135 |
+
"LABEL_0": 0,
|
| 136 |
+
"LABEL_1": 1
|
| 137 |
+
},
|
| 138 |
+
"layer_norm_eps": 1e-12,
|
| 139 |
+
"length_penalty": 1.0,
|
| 140 |
+
"max_length": 20,
|
| 141 |
+
"max_position_embeddings": 4096,
|
| 142 |
+
"min_length": 0,
|
| 143 |
+
"model_type": "big_bird",
|
| 144 |
+
"no_repeat_ngram_size": 0,
|
| 145 |
+
"num_attention_heads": 12,
|
| 146 |
+
"num_beam_groups": 1,
|
| 147 |
+
"num_beams": 1,
|
| 148 |
+
"num_hidden_layers": 6,
|
| 149 |
+
"num_random_blocks": 3,
|
| 150 |
+
"num_return_sequences": 1,
|
| 151 |
+
"output_attentions": false,
|
| 152 |
+
"output_hidden_states": false,
|
| 153 |
+
"output_scores": false,
|
| 154 |
+
"pad_token_id": 0,
|
| 155 |
+
"position_embedding_type": "absolute",
|
| 156 |
+
"prefix": null,
|
| 157 |
+
"problem_type": null,
|
| 158 |
+
"pruned_heads": {},
|
| 159 |
+
"remove_invalid_values": false,
|
| 160 |
+
"repetition_penalty": 1.0,
|
| 161 |
+
"rescale_embeddings": false,
|
| 162 |
+
"return_dict": true,
|
| 163 |
+
"return_dict_in_generate": false,
|
| 164 |
+
"sep_token_id": 3,
|
| 165 |
+
"task_specific_params": null,
|
| 166 |
+
"temperature": 1.0,
|
| 167 |
+
"tie_encoder_decoder": false,
|
| 168 |
+
"tie_word_embeddings": true,
|
| 169 |
+
"tokenizer_class": "BertTokenizer",
|
| 170 |
+
"top_k": 50,
|
| 171 |
+
"top_p": 1.0,
|
| 172 |
+
"torch_dtype": "float32",
|
| 173 |
+
"torchscript": false,
|
| 174 |
+
"transformers_version": "4.11.0",
|
| 175 |
+
"type_vocab_size": 2,
|
| 176 |
+
"use_bfloat16": false,
|
| 177 |
+
"use_bias": true,
|
| 178 |
+
"use_cache": true,
|
| 179 |
+
"vocab_size": 32500
|
| 180 |
+
},
|
| 181 |
+
"is_encoder_decoder": true,
|
| 182 |
+
"model_type": "encoder-decoder",
|
| 183 |
+
"torch_dtype": "float32",
|
| 184 |
+
"transformers_version": null
|
| 185 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:66a5fc7cfec99b85e7d781abc6cb38ebcab0e3ccd22a90b10adf082fcacb8354
|
| 3 |
+
size 718021289
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"do_lower_case": false, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "do_basic_tokenize": true, "never_split": null, "bos_token": "<s>", "eos_token": "</s>", "model_max_length": 4096, "special_tokens_map_file": "/opt/ml/.cache/huggingface/transformers/9bea998b48658e35dd618115a266f6c173183a9a4261fc6e40730d74c4b67899.e3640e465e51ce85d94923a0b396029ecc2e3e4c7764031eee57ab272637652d", "name_or_path": "metamong1/bigbird-tapt-ep3", "tokenizer_class": "BertTokenizer"}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,370 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 41.2618,
|
| 3 |
+
"best_model_checkpoint": "checkpoint/kobigbirdbart_tapt_ep3_bs16_pre_noam_LB/checkpoint-34340",
|
| 4 |
+
"epoch": 2.9999708802888674,
|
| 5 |
+
"global_step": 51510,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.06,
|
| 12 |
+
"learning_rate": 0.0001,
|
| 13 |
+
"loss": 4.0417,
|
| 14 |
+
"step": 1000
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.12,
|
| 18 |
+
"learning_rate": 7.072834474681765e-05,
|
| 19 |
+
"loss": 3.6688,
|
| 20 |
+
"step": 2000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.17,
|
| 24 |
+
"learning_rate": 5.775426231077362e-05,
|
| 25 |
+
"loss": 3.5243,
|
| 26 |
+
"step": 3000
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.23,
|
| 30 |
+
"learning_rate": 5.001874180112048e-05,
|
| 31 |
+
"loss": 3.4178,
|
| 32 |
+
"step": 4000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.29,
|
| 36 |
+
"learning_rate": 4.473924094197383e-05,
|
| 37 |
+
"loss": 3.367,
|
| 38 |
+
"step": 5000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.35,
|
| 42 |
+
"learning_rate": 4.084183301607125e-05,
|
| 43 |
+
"loss": 3.3013,
|
| 44 |
+
"step": 6000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.41,
|
| 48 |
+
"learning_rate": 3.7812639995975144e-05,
|
| 49 |
+
"loss": 3.2593,
|
| 50 |
+
"step": 7000
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.47,
|
| 54 |
+
"learning_rate": 3.537080170562066e-05,
|
| 55 |
+
"loss": 3.212,
|
| 56 |
+
"step": 8000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.52,
|
| 60 |
+
"learning_rate": 3.3348143212252787e-05,
|
| 61 |
+
"loss": 3.1631,
|
| 62 |
+
"step": 9000
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.58,
|
| 66 |
+
"learning_rate": 3.16370022285456e-05,
|
| 67 |
+
"loss": 3.1368,
|
| 68 |
+
"step": 10000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.64,
|
| 72 |
+
"learning_rate": 3.0164835160286395e-05,
|
| 73 |
+
"loss": 3.106,
|
| 74 |
+
"step": 11000
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.7,
|
| 78 |
+
"learning_rate": 2.8880740270466042e-05,
|
| 79 |
+
"loss": 3.0851,
|
| 80 |
+
"step": 12000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.76,
|
| 84 |
+
"learning_rate": 2.774780664821333e-05,
|
| 85 |
+
"loss": 3.0639,
|
| 86 |
+
"step": 13000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.82,
|
| 90 |
+
"learning_rate": 2.6738528983830193e-05,
|
| 91 |
+
"loss": 3.0412,
|
| 92 |
+
"step": 14000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.87,
|
| 96 |
+
"learning_rate": 2.583193464319257e-05,
|
| 97 |
+
"loss": 3.0153,
|
| 98 |
+
"step": 15000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.93,
|
| 102 |
+
"learning_rate": 2.5011715272671623e-05,
|
| 103 |
+
"loss": 2.9905,
|
| 104 |
+
"step": 16000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.99,
|
| 108 |
+
"learning_rate": 2.426497258953522e-05,
|
| 109 |
+
"loss": 2.9887,
|
| 110 |
+
"step": 17000
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 1.0,
|
| 114 |
+
"eval_gen_len": 20.0,
|
| 115 |
+
"eval_loss": 3.2666966915130615,
|
| 116 |
+
"eval_rouge1": 43.6697,
|
| 117 |
+
"eval_rouge2": 28.3814,
|
| 118 |
+
"eval_rougeL": 40.2632,
|
| 119 |
+
"eval_rougeLsum": 40.2565,
|
| 120 |
+
"eval_runtime": 2924.633,
|
| 121 |
+
"eval_samples_per_second": 31.314,
|
| 122 |
+
"eval_steps_per_second": 1.957,
|
| 123 |
+
"step": 17170
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.05,
|
| 127 |
+
"learning_rate": 2.3581353179274502e-05,
|
| 128 |
+
"loss": 2.7893,
|
| 129 |
+
"step": 18000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.11,
|
| 133 |
+
"learning_rate": 2.2952437303925296e-05,
|
| 134 |
+
"loss": 2.724,
|
| 135 |
+
"step": 19000
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 1.16,
|
| 139 |
+
"learning_rate": 2.2371298045736806e-05,
|
| 140 |
+
"loss": 2.722,
|
| 141 |
+
"step": 20000
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 1.22,
|
| 145 |
+
"learning_rate": 2.183217738419129e-05,
|
| 146 |
+
"loss": 2.7157,
|
| 147 |
+
"step": 21000
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 1.28,
|
| 151 |
+
"learning_rate": 2.1330244234932196e-05,
|
| 152 |
+
"loss": 2.7097,
|
| 153 |
+
"step": 22000
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 1.34,
|
| 157 |
+
"learning_rate": 2.0861411017276734e-05,
|
| 158 |
+
"loss": 2.6939,
|
| 159 |
+
"step": 23000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.4,
|
| 163 |
+
"learning_rate": 2.0422192722261335e-05,
|
| 164 |
+
"loss": 2.7026,
|
| 165 |
+
"step": 24000
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.46,
|
| 169 |
+
"learning_rate": 2.0009597313304793e-05,
|
| 170 |
+
"loss": 2.6871,
|
| 171 |
+
"step": 25000
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.51,
|
| 175 |
+
"learning_rate": 1.9621039546293084e-05,
|
| 176 |
+
"loss": 2.6776,
|
| 177 |
+
"step": 26000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.57,
|
| 181 |
+
"learning_rate": 1.925427251574924e-05,
|
| 182 |
+
"loss": 2.6799,
|
| 183 |
+
"step": 27000
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 1.63,
|
| 187 |
+
"learning_rate": 1.8907332773261164e-05,
|
| 188 |
+
"loss": 2.6613,
|
| 189 |
+
"step": 28000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.69,
|
| 193 |
+
"learning_rate": 1.8578495948447844e-05,
|
| 194 |
+
"loss": 2.6756,
|
| 195 |
+
"step": 29000
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 1.75,
|
| 199 |
+
"learning_rate": 1.8266240576954328e-05,
|
| 200 |
+
"loss": 2.6459,
|
| 201 |
+
"step": 30000
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 1.81,
|
| 205 |
+
"learning_rate": 1.796921840006843e-05,
|
| 206 |
+
"loss": 2.6352,
|
| 207 |
+
"step": 31000
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 1.86,
|
| 211 |
+
"learning_rate": 1.768622981064213e-05,
|
| 212 |
+
"loss": 2.6379,
|
| 213 |
+
"step": 32000
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.92,
|
| 217 |
+
"learning_rate": 1.7416203423613194e-05,
|
| 218 |
+
"loss": 2.6138,
|
| 219 |
+
"step": 33000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.98,
|
| 223 |
+
"learning_rate": 1.7158178976540783e-05,
|
| 224 |
+
"loss": 2.6193,
|
| 225 |
+
"step": 34000
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 2.0,
|
| 229 |
+
"eval_gen_len": 20.0,
|
| 230 |
+
"eval_loss": 3.2624921798706055,
|
| 231 |
+
"eval_rouge1": 44.7484,
|
| 232 |
+
"eval_rouge2": 29.1819,
|
| 233 |
+
"eval_rougeL": 41.2553,
|
| 234 |
+
"eval_rougeLsum": 41.2618,
|
| 235 |
+
"eval_runtime": 2925.4113,
|
| 236 |
+
"eval_samples_per_second": 31.306,
|
| 237 |
+
"eval_steps_per_second": 1.957,
|
| 238 |
+
"step": 34340
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 2.04,
|
| 242 |
+
"learning_rate": 1.6911292937114278e-05,
|
| 243 |
+
"loss": 2.5111,
|
| 244 |
+
"step": 35000
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 2.1,
|
| 248 |
+
"learning_rate": 1.6674766325339214e-05,
|
| 249 |
+
"loss": 2.4361,
|
| 250 |
+
"step": 36000
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 2.15,
|
| 254 |
+
"learning_rate": 1.644789435859962e-05,
|
| 255 |
+
"loss": 2.4523,
|
| 256 |
+
"step": 37000
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 2.21,
|
| 260 |
+
"learning_rate": 1.6230037605656824e-05,
|
| 261 |
+
"loss": 2.4485,
|
| 262 |
+
"step": 38000
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 2.27,
|
| 266 |
+
"learning_rate": 1.6020614396421562e-05,
|
| 267 |
+
"loss": 2.443,
|
| 268 |
+
"step": 39000
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 2.33,
|
| 272 |
+
"learning_rate": 1.58190942821137e-05,
|
| 273 |
+
"loss": 2.4408,
|
| 274 |
+
"step": 40000
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 2.39,
|
| 278 |
+
"learning_rate": 1.562499237823282e-05,
|
| 279 |
+
"loss": 2.4376,
|
| 280 |
+
"step": 41000
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 2.45,
|
| 284 |
+
"learning_rate": 1.5437864452872526e-05,
|
| 285 |
+
"loss": 2.4237,
|
| 286 |
+
"step": 42000
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 2.5,
|
| 290 |
+
"learning_rate": 1.5257302647033036e-05,
|
| 291 |
+
"loss": 2.4302,
|
| 292 |
+
"step": 43000
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 2.56,
|
| 296 |
+
"learning_rate": 1.508293173302068e-05,
|
| 297 |
+
"loss": 2.4272,
|
| 298 |
+
"step": 44000
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 2.62,
|
| 302 |
+
"learning_rate": 1.4914405832764997e-05,
|
| 303 |
+
"loss": 2.4191,
|
| 304 |
+
"step": 45000
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 2.68,
|
| 308 |
+
"learning_rate": 1.4751405530700888e-05,
|
| 309 |
+
"loss": 2.415,
|
| 310 |
+
"step": 46000
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 2.74,
|
| 314 |
+
"learning_rate": 1.4593635326349197e-05,
|
| 315 |
+
"loss": 2.4139,
|
| 316 |
+
"step": 47000
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 2.8,
|
| 320 |
+
"learning_rate": 1.4440821380348212e-05,
|
| 321 |
+
"loss": 2.409,
|
| 322 |
+
"step": 48000
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 2.85,
|
| 326 |
+
"learning_rate": 1.4292709514804612e-05,
|
| 327 |
+
"loss": 2.4043,
|
| 328 |
+
"step": 49000
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 2.91,
|
| 332 |
+
"learning_rate": 1.4149063434732295e-05,
|
| 333 |
+
"loss": 2.3984,
|
| 334 |
+
"step": 50000
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 2.97,
|
| 338 |
+
"learning_rate": 1.4009663142259388e-05,
|
| 339 |
+
"loss": 2.4011,
|
| 340 |
+
"step": 51000
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 3.0,
|
| 344 |
+
"eval_gen_len": 20.0,
|
| 345 |
+
"eval_loss": 3.302915573120117,
|
| 346 |
+
"eval_rouge1": 44.3235,
|
| 347 |
+
"eval_rouge2": 28.8361,
|
| 348 |
+
"eval_rougeL": 40.7694,
|
| 349 |
+
"eval_rougeLsum": 40.7674,
|
| 350 |
+
"eval_runtime": 2928.7847,
|
| 351 |
+
"eval_samples_per_second": 31.27,
|
| 352 |
+
"eval_steps_per_second": 1.954,
|
| 353 |
+
"step": 51510
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 3.0,
|
| 357 |
+
"step": 51510,
|
| 358 |
+
"total_flos": 3.55488261007809e+17,
|
| 359 |
+
"train_loss": 2.7859744007345357,
|
| 360 |
+
"train_runtime": 40305.1206,
|
| 361 |
+
"train_samples_per_second": 20.449,
|
| 362 |
+
"train_steps_per_second": 1.278
|
| 363 |
+
}
|
| 364 |
+
],
|
| 365 |
+
"max_steps": 51510,
|
| 366 |
+
"num_train_epochs": 3,
|
| 367 |
+
"total_flos": 3.55488261007809e+17,
|
| 368 |
+
"trial_name": null,
|
| 369 |
+
"trial_params": null
|
| 370 |
+
}
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|