Upload 10 files
Browse files- config.json +172 -0
- merges.txt +0 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +1 -0
- tokenizer_config.json +1 -0
- trainer_state.json +472 -0
- training_args.bin +3 -0
- vocab.json +0 -0
config.json
ADDED
|
@@ -0,0 +1,172 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"EncoderDecoderModel"
|
| 4 |
+
],
|
| 5 |
+
"decoder": {
|
| 6 |
+
"_name_or_path": "cahya/gpt2-small-indonesian-522M",
|
| 7 |
+
"activation_function": "gelu_new",
|
| 8 |
+
"add_cross_attention": true,
|
| 9 |
+
"architectures": [
|
| 10 |
+
"GPT2LMHeadModel"
|
| 11 |
+
],
|
| 12 |
+
"attn_pdrop": 0.1,
|
| 13 |
+
"bad_words_ids": null,
|
| 14 |
+
"bos_token_id": 50256,
|
| 15 |
+
"chunk_size_feed_forward": 0,
|
| 16 |
+
"decoder_start_token_id": null,
|
| 17 |
+
"diversity_penalty": 0.0,
|
| 18 |
+
"do_sample": false,
|
| 19 |
+
"early_stopping": false,
|
| 20 |
+
"embd_pdrop": 0.1,
|
| 21 |
+
"eos_token_id": 50256,
|
| 22 |
+
"finetuning_task": null,
|
| 23 |
+
"gradient_checkpointing": false,
|
| 24 |
+
"id2label": {
|
| 25 |
+
"0": "LABEL_0",
|
| 26 |
+
"1": "LABEL_1"
|
| 27 |
+
},
|
| 28 |
+
"initializer_range": 0.02,
|
| 29 |
+
"is_decoder": true,
|
| 30 |
+
"is_encoder_decoder": false,
|
| 31 |
+
"label2id": {
|
| 32 |
+
"LABEL_0": 0,
|
| 33 |
+
"LABEL_1": 1
|
| 34 |
+
},
|
| 35 |
+
"layer_norm_epsilon": 1e-05,
|
| 36 |
+
"length_penalty": 1.0,
|
| 37 |
+
"max_length": 20,
|
| 38 |
+
"min_length": 0,
|
| 39 |
+
"model_type": "gpt2",
|
| 40 |
+
"n_ctx": 1024,
|
| 41 |
+
"n_embd": 768,
|
| 42 |
+
"n_head": 12,
|
| 43 |
+
"n_inner": null,
|
| 44 |
+
"n_layer": 12,
|
| 45 |
+
"n_positions": 1024,
|
| 46 |
+
"no_repeat_ngram_size": 0,
|
| 47 |
+
"num_beam_groups": 1,
|
| 48 |
+
"num_beams": 1,
|
| 49 |
+
"num_return_sequences": 1,
|
| 50 |
+
"output_attentions": false,
|
| 51 |
+
"output_hidden_states": false,
|
| 52 |
+
"output_scores": false,
|
| 53 |
+
"pad_token_id": null,
|
| 54 |
+
"prefix": null,
|
| 55 |
+
"pruned_heads": {},
|
| 56 |
+
"repetition_penalty": 1.0,
|
| 57 |
+
"resid_pdrop": 0.1,
|
| 58 |
+
"return_dict": true,
|
| 59 |
+
"return_dict_in_generate": false,
|
| 60 |
+
"sep_token_id": null,
|
| 61 |
+
"summary_activation": null,
|
| 62 |
+
"summary_first_dropout": 0.1,
|
| 63 |
+
"summary_proj_to_labels": true,
|
| 64 |
+
"summary_type": "cls_index",
|
| 65 |
+
"summary_use_proj": true,
|
| 66 |
+
"task_specific_params": null,
|
| 67 |
+
"temperature": 1.0,
|
| 68 |
+
"tie_encoder_decoder": false,
|
| 69 |
+
"tie_word_embeddings": true,
|
| 70 |
+
"tokenizer_class": null,
|
| 71 |
+
"top_k": 50,
|
| 72 |
+
"top_p": 1.0,
|
| 73 |
+
"torchscript": false,
|
| 74 |
+
"transformers_version": "4.2.1",
|
| 75 |
+
"use_bfloat16": false,
|
| 76 |
+
"use_cache": true,
|
| 77 |
+
"vocab_size": 50257,
|
| 78 |
+
"xla_device": null
|
| 79 |
+
},
|
| 80 |
+
"decoder_start_token_id": 0,
|
| 81 |
+
"encoder": {
|
| 82 |
+
"_name_or_path": "indobenchmark/indobert-base-p1",
|
| 83 |
+
"_num_labels": 5,
|
| 84 |
+
"add_cross_attention": false,
|
| 85 |
+
"architectures": [
|
| 86 |
+
"BertModel"
|
| 87 |
+
],
|
| 88 |
+
"attention_probs_dropout_prob": 0.1,
|
| 89 |
+
"bad_words_ids": null,
|
| 90 |
+
"bos_token_id": null,
|
| 91 |
+
"chunk_size_feed_forward": 0,
|
| 92 |
+
"decoder_start_token_id": null,
|
| 93 |
+
"directionality": "bidi",
|
| 94 |
+
"diversity_penalty": 0.0,
|
| 95 |
+
"do_sample": false,
|
| 96 |
+
"early_stopping": false,
|
| 97 |
+
"eos_token_id": null,
|
| 98 |
+
"finetuning_task": null,
|
| 99 |
+
"gradient_checkpointing": false,
|
| 100 |
+
"hidden_act": "gelu",
|
| 101 |
+
"hidden_dropout_prob": 0.1,
|
| 102 |
+
"hidden_size": 768,
|
| 103 |
+
"id2label": {
|
| 104 |
+
"0": "LABEL_0",
|
| 105 |
+
"1": "LABEL_1",
|
| 106 |
+
"2": "LABEL_2",
|
| 107 |
+
"3": "LABEL_3",
|
| 108 |
+
"4": "LABEL_4"
|
| 109 |
+
},
|
| 110 |
+
"initializer_range": 0.02,
|
| 111 |
+
"intermediate_size": 3072,
|
| 112 |
+
"is_decoder": false,
|
| 113 |
+
"is_encoder_decoder": false,
|
| 114 |
+
"label2id": {
|
| 115 |
+
"LABEL_0": 0,
|
| 116 |
+
"LABEL_1": 1,
|
| 117 |
+
"LABEL_2": 2,
|
| 118 |
+
"LABEL_3": 3,
|
| 119 |
+
"LABEL_4": 4
|
| 120 |
+
},
|
| 121 |
+
"layer_norm_eps": 1e-12,
|
| 122 |
+
"length_penalty": 1.0,
|
| 123 |
+
"max_length": 20,
|
| 124 |
+
"max_position_embeddings": 512,
|
| 125 |
+
"min_length": 0,
|
| 126 |
+
"model_type": "bert",
|
| 127 |
+
"no_repeat_ngram_size": 0,
|
| 128 |
+
"num_attention_heads": 12,
|
| 129 |
+
"num_beam_groups": 1,
|
| 130 |
+
"num_beams": 1,
|
| 131 |
+
"num_hidden_layers": 12,
|
| 132 |
+
"num_return_sequences": 1,
|
| 133 |
+
"output_attentions": false,
|
| 134 |
+
"output_hidden_states": false,
|
| 135 |
+
"output_past": true,
|
| 136 |
+
"output_scores": false,
|
| 137 |
+
"pad_token_id": 0,
|
| 138 |
+
"pooler_fc_size": 768,
|
| 139 |
+
"pooler_num_attention_heads": 12,
|
| 140 |
+
"pooler_num_fc_layers": 3,
|
| 141 |
+
"pooler_size_per_head": 128,
|
| 142 |
+
"pooler_type": "first_token_transform",
|
| 143 |
+
"position_embedding_type": "absolute",
|
| 144 |
+
"prefix": null,
|
| 145 |
+
"pruned_heads": {},
|
| 146 |
+
"repetition_penalty": 1.0,
|
| 147 |
+
"return_dict": true,
|
| 148 |
+
"return_dict_in_generate": false,
|
| 149 |
+
"sep_token_id": null,
|
| 150 |
+
"task_specific_params": null,
|
| 151 |
+
"temperature": 1.0,
|
| 152 |
+
"tie_encoder_decoder": false,
|
| 153 |
+
"tie_word_embeddings": true,
|
| 154 |
+
"tokenizer_class": null,
|
| 155 |
+
"top_k": 50,
|
| 156 |
+
"top_p": 1.0,
|
| 157 |
+
"torchscript": false,
|
| 158 |
+
"transformers_version": "4.2.1",
|
| 159 |
+
"type_vocab_size": 2,
|
| 160 |
+
"use_bfloat16": false,
|
| 161 |
+
"use_cache": true,
|
| 162 |
+
"vocab_size": 50000,
|
| 163 |
+
"xla_device": null
|
| 164 |
+
},
|
| 165 |
+
"eos_token_id": 0,
|
| 166 |
+
"is_encoder_decoder": true,
|
| 167 |
+
"max_length": 80,
|
| 168 |
+
"min_length": 3,
|
| 169 |
+
"model_type": "encoder-decoder",
|
| 170 |
+
"no_repeat_ngram_size": 3,
|
| 171 |
+
"pad_token_id": 0
|
| 172 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8d5c4cb1e3c67d1f94154a70261331682ca80d6b6c6f1bba0ce52add539f73b0
|
| 3 |
+
size 2213518341
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a22d3bd8a9c8508279ece362585da8bfd67d4cc8401a865361aa19d6bec19e0e
|
| 3 |
+
size 1134357885
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0cf3d146428b3c6d5d3ccc9aa4d544a648a7758efb66452d7f7270c48c06db6d
|
| 3 |
+
size 627
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"bos_token": {"content": "<|endoftext|>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "eos_token": {"content": "<|endoftext|>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "unk_token": {"content": "<|endoftext|>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "pad_token": "<|endoftext|>"}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"errors": "replace", "unk_token": {"content": "<|endoftext|>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "bos_token": {"content": "<|endoftext|>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "eos_token": {"content": "<|endoftext|>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "add_prefix_space": false, "max_len": 512, "special_tokens_map_file": "/home/.cache/huggingface/transformers/57902062674d8ea91c522ec7c31f93f4b3947d525b682597edac4bc76b6d99c2.2b8bf81243d08385c806171bc7ced6d2a0dcc7f896ca637f4e777418f7f0cc3c", "tokenizer_file": null, "name_or_path": "cahya/gpt2-small-indonesian-522M"}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,472 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 126.66666666666667,
|
| 5 |
+
"global_step": 3800,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1.67,
|
| 12 |
+
"learning_rate": 4.9542483660130725e-05,
|
| 13 |
+
"loss": 4.085,
|
| 14 |
+
"step": 50
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 3.33,
|
| 18 |
+
"learning_rate": 4.888888888888889e-05,
|
| 19 |
+
"loss": 2.4616,
|
| 20 |
+
"step": 100
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 5.0,
|
| 24 |
+
"learning_rate": 4.823529411764706e-05,
|
| 25 |
+
"loss": 1.4793,
|
| 26 |
+
"step": 150
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 6.67,
|
| 30 |
+
"learning_rate": 4.7581699346405226e-05,
|
| 31 |
+
"loss": 0.7873,
|
| 32 |
+
"step": 200
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 8.33,
|
| 36 |
+
"learning_rate": 4.69281045751634e-05,
|
| 37 |
+
"loss": 0.4859,
|
| 38 |
+
"step": 250
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 10.0,
|
| 42 |
+
"learning_rate": 4.6274509803921576e-05,
|
| 43 |
+
"loss": 0.3261,
|
| 44 |
+
"step": 300
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 11.67,
|
| 48 |
+
"learning_rate": 4.562091503267974e-05,
|
| 49 |
+
"loss": 0.2257,
|
| 50 |
+
"step": 350
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 13.33,
|
| 54 |
+
"learning_rate": 4.496732026143791e-05,
|
| 55 |
+
"loss": 0.1711,
|
| 56 |
+
"step": 400
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 15.0,
|
| 60 |
+
"learning_rate": 4.431372549019608e-05,
|
| 61 |
+
"loss": 0.126,
|
| 62 |
+
"step": 450
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 16.67,
|
| 66 |
+
"learning_rate": 4.366013071895425e-05,
|
| 67 |
+
"loss": 0.0819,
|
| 68 |
+
"step": 500
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 18.33,
|
| 72 |
+
"learning_rate": 4.300653594771242e-05,
|
| 73 |
+
"loss": 0.0699,
|
| 74 |
+
"step": 550
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 20.0,
|
| 78 |
+
"learning_rate": 4.235294117647059e-05,
|
| 79 |
+
"loss": 0.0568,
|
| 80 |
+
"step": 600
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 21.67,
|
| 84 |
+
"learning_rate": 4.169934640522876e-05,
|
| 85 |
+
"loss": 0.0473,
|
| 86 |
+
"step": 650
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 23.33,
|
| 90 |
+
"learning_rate": 4.104575163398693e-05,
|
| 91 |
+
"loss": 0.0428,
|
| 92 |
+
"step": 700
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 25.0,
|
| 96 |
+
"learning_rate": 4.03921568627451e-05,
|
| 97 |
+
"loss": 0.0382,
|
| 98 |
+
"step": 750
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 26.67,
|
| 102 |
+
"learning_rate": 3.973856209150327e-05,
|
| 103 |
+
"loss": 0.033,
|
| 104 |
+
"step": 800
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 28.33,
|
| 108 |
+
"learning_rate": 3.908496732026144e-05,
|
| 109 |
+
"loss": 0.0306,
|
| 110 |
+
"step": 850
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 30.0,
|
| 114 |
+
"learning_rate": 3.8431372549019614e-05,
|
| 115 |
+
"loss": 0.0272,
|
| 116 |
+
"step": 900
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 31.67,
|
| 120 |
+
"learning_rate": 3.777777777777778e-05,
|
| 121 |
+
"loss": 0.0225,
|
| 122 |
+
"step": 950
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 33.33,
|
| 126 |
+
"learning_rate": 3.712418300653595e-05,
|
| 127 |
+
"loss": 0.0211,
|
| 128 |
+
"step": 1000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 35.0,
|
| 132 |
+
"learning_rate": 3.6470588235294114e-05,
|
| 133 |
+
"loss": 0.0196,
|
| 134 |
+
"step": 1050
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 36.67,
|
| 138 |
+
"learning_rate": 3.581699346405229e-05,
|
| 139 |
+
"loss": 0.0167,
|
| 140 |
+
"step": 1100
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 38.33,
|
| 144 |
+
"learning_rate": 3.5163398692810464e-05,
|
| 145 |
+
"loss": 0.0158,
|
| 146 |
+
"step": 1150
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 40.0,
|
| 150 |
+
"learning_rate": 3.450980392156863e-05,
|
| 151 |
+
"loss": 0.0168,
|
| 152 |
+
"step": 1200
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 41.67,
|
| 156 |
+
"learning_rate": 3.38562091503268e-05,
|
| 157 |
+
"loss": 0.0101,
|
| 158 |
+
"step": 1250
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 43.33,
|
| 162 |
+
"learning_rate": 3.3202614379084965e-05,
|
| 163 |
+
"loss": 0.0128,
|
| 164 |
+
"step": 1300
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 45.0,
|
| 168 |
+
"learning_rate": 3.254901960784314e-05,
|
| 169 |
+
"loss": 0.0077,
|
| 170 |
+
"step": 1350
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 46.67,
|
| 174 |
+
"learning_rate": 3.189542483660131e-05,
|
| 175 |
+
"loss": 0.0065,
|
| 176 |
+
"step": 1400
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 48.33,
|
| 180 |
+
"learning_rate": 3.124183006535948e-05,
|
| 181 |
+
"loss": 0.0053,
|
| 182 |
+
"step": 1450
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 50.0,
|
| 186 |
+
"learning_rate": 3.058823529411765e-05,
|
| 187 |
+
"loss": 0.0064,
|
| 188 |
+
"step": 1500
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 51.67,
|
| 192 |
+
"learning_rate": 2.9934640522875816e-05,
|
| 193 |
+
"loss": 0.0059,
|
| 194 |
+
"step": 1550
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 53.33,
|
| 198 |
+
"learning_rate": 2.928104575163399e-05,
|
| 199 |
+
"loss": 0.0044,
|
| 200 |
+
"step": 1600
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 55.0,
|
| 204 |
+
"learning_rate": 2.8627450980392155e-05,
|
| 205 |
+
"loss": 0.0057,
|
| 206 |
+
"step": 1650
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 56.67,
|
| 210 |
+
"learning_rate": 2.7973856209150327e-05,
|
| 211 |
+
"loss": 0.0043,
|
| 212 |
+
"step": 1700
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 58.33,
|
| 216 |
+
"learning_rate": 2.7320261437908495e-05,
|
| 217 |
+
"loss": 0.0042,
|
| 218 |
+
"step": 1750
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 60.0,
|
| 222 |
+
"learning_rate": 2.6666666666666667e-05,
|
| 223 |
+
"loss": 0.0024,
|
| 224 |
+
"step": 1800
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 61.67,
|
| 228 |
+
"learning_rate": 2.601307189542484e-05,
|
| 229 |
+
"loss": 0.003,
|
| 230 |
+
"step": 1850
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 63.33,
|
| 234 |
+
"learning_rate": 2.5359477124183006e-05,
|
| 235 |
+
"loss": 0.0037,
|
| 236 |
+
"step": 1900
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 65.0,
|
| 240 |
+
"learning_rate": 2.4705882352941178e-05,
|
| 241 |
+
"loss": 0.0035,
|
| 242 |
+
"step": 1950
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 66.67,
|
| 246 |
+
"learning_rate": 2.405228758169935e-05,
|
| 247 |
+
"loss": 0.0019,
|
| 248 |
+
"step": 2000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 68.33,
|
| 252 |
+
"learning_rate": 2.3398692810457517e-05,
|
| 253 |
+
"loss": 0.0027,
|
| 254 |
+
"step": 2050
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 70.0,
|
| 258 |
+
"learning_rate": 2.2745098039215685e-05,
|
| 259 |
+
"loss": 0.003,
|
| 260 |
+
"step": 2100
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 71.67,
|
| 264 |
+
"learning_rate": 2.2091503267973857e-05,
|
| 265 |
+
"loss": 0.0028,
|
| 266 |
+
"step": 2150
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 73.33,
|
| 270 |
+
"learning_rate": 2.1437908496732025e-05,
|
| 271 |
+
"loss": 0.0014,
|
| 272 |
+
"step": 2200
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 75.0,
|
| 276 |
+
"learning_rate": 2.0784313725490197e-05,
|
| 277 |
+
"loss": 0.0011,
|
| 278 |
+
"step": 2250
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 76.67,
|
| 282 |
+
"learning_rate": 2.0130718954248368e-05,
|
| 283 |
+
"loss": 0.0017,
|
| 284 |
+
"step": 2300
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 78.33,
|
| 288 |
+
"learning_rate": 1.9477124183006536e-05,
|
| 289 |
+
"loss": 0.0007,
|
| 290 |
+
"step": 2350
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 80.0,
|
| 294 |
+
"learning_rate": 1.8823529411764708e-05,
|
| 295 |
+
"loss": 0.0006,
|
| 296 |
+
"step": 2400
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 81.67,
|
| 300 |
+
"learning_rate": 1.8169934640522876e-05,
|
| 301 |
+
"loss": 0.0005,
|
| 302 |
+
"step": 2450
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 83.33,
|
| 306 |
+
"learning_rate": 1.7516339869281044e-05,
|
| 307 |
+
"loss": 0.0005,
|
| 308 |
+
"step": 2500
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 85.0,
|
| 312 |
+
"learning_rate": 1.686274509803922e-05,
|
| 313 |
+
"loss": 0.0005,
|
| 314 |
+
"step": 2550
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 86.67,
|
| 318 |
+
"learning_rate": 1.6209150326797387e-05,
|
| 319 |
+
"loss": 0.0005,
|
| 320 |
+
"step": 2600
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 88.33,
|
| 324 |
+
"learning_rate": 1.5555555555555555e-05,
|
| 325 |
+
"loss": 0.0005,
|
| 326 |
+
"step": 2650
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 90.0,
|
| 330 |
+
"learning_rate": 1.4901960784313726e-05,
|
| 331 |
+
"loss": 0.0006,
|
| 332 |
+
"step": 2700
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 91.67,
|
| 336 |
+
"learning_rate": 1.4248366013071896e-05,
|
| 337 |
+
"loss": 0.0017,
|
| 338 |
+
"step": 2750
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 93.33,
|
| 342 |
+
"learning_rate": 1.3594771241830064e-05,
|
| 343 |
+
"loss": 0.0009,
|
| 344 |
+
"step": 2800
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 95.0,
|
| 348 |
+
"learning_rate": 1.2941176470588238e-05,
|
| 349 |
+
"loss": 0.0008,
|
| 350 |
+
"step": 2850
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 96.67,
|
| 354 |
+
"learning_rate": 1.2287581699346406e-05,
|
| 355 |
+
"loss": 0.0005,
|
| 356 |
+
"step": 2900
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 98.33,
|
| 360 |
+
"learning_rate": 1.1633986928104575e-05,
|
| 361 |
+
"loss": 0.0004,
|
| 362 |
+
"step": 2950
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 100.0,
|
| 366 |
+
"learning_rate": 1.0980392156862745e-05,
|
| 367 |
+
"loss": 0.0005,
|
| 368 |
+
"step": 3000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 101.67,
|
| 372 |
+
"learning_rate": 1.0326797385620915e-05,
|
| 373 |
+
"loss": 0.0004,
|
| 374 |
+
"step": 3050
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 103.33,
|
| 378 |
+
"learning_rate": 9.673202614379087e-06,
|
| 379 |
+
"loss": 0.0004,
|
| 380 |
+
"step": 3100
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 105.0,
|
| 384 |
+
"learning_rate": 9.019607843137255e-06,
|
| 385 |
+
"loss": 0.0004,
|
| 386 |
+
"step": 3150
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 106.67,
|
| 390 |
+
"learning_rate": 8.366013071895424e-06,
|
| 391 |
+
"loss": 0.0004,
|
| 392 |
+
"step": 3200
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 108.33,
|
| 396 |
+
"learning_rate": 7.712418300653596e-06,
|
| 397 |
+
"loss": 0.0004,
|
| 398 |
+
"step": 3250
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 110.0,
|
| 402 |
+
"learning_rate": 7.058823529411765e-06,
|
| 403 |
+
"loss": 0.0003,
|
| 404 |
+
"step": 3300
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 111.67,
|
| 408 |
+
"learning_rate": 6.405228758169935e-06,
|
| 409 |
+
"loss": 0.0003,
|
| 410 |
+
"step": 3350
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 113.33,
|
| 414 |
+
"learning_rate": 5.7516339869281045e-06,
|
| 415 |
+
"loss": 0.0003,
|
| 416 |
+
"step": 3400
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 115.0,
|
| 420 |
+
"learning_rate": 5.098039215686274e-06,
|
| 421 |
+
"loss": 0.0004,
|
| 422 |
+
"step": 3450
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 116.67,
|
| 426 |
+
"learning_rate": 4.444444444444445e-06,
|
| 427 |
+
"loss": 0.0004,
|
| 428 |
+
"step": 3500
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 118.33,
|
| 432 |
+
"learning_rate": 3.7908496732026144e-06,
|
| 433 |
+
"loss": 0.0003,
|
| 434 |
+
"step": 3550
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 120.0,
|
| 438 |
+
"learning_rate": 3.1372549019607846e-06,
|
| 439 |
+
"loss": 0.0003,
|
| 440 |
+
"step": 3600
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 121.67,
|
| 444 |
+
"learning_rate": 2.4836601307189544e-06,
|
| 445 |
+
"loss": 0.0003,
|
| 446 |
+
"step": 3650
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 123.33,
|
| 450 |
+
"learning_rate": 1.8300653594771244e-06,
|
| 451 |
+
"loss": 0.0003,
|
| 452 |
+
"step": 3700
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 125.0,
|
| 456 |
+
"learning_rate": 1.1764705882352942e-06,
|
| 457 |
+
"loss": 0.0003,
|
| 458 |
+
"step": 3750
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 126.67,
|
| 462 |
+
"learning_rate": 5.22875816993464e-07,
|
| 463 |
+
"loss": 0.0003,
|
| 464 |
+
"step": 3800
|
| 465 |
+
}
|
| 466 |
+
],
|
| 467 |
+
"max_steps": 3840,
|
| 468 |
+
"num_train_epochs": 128,
|
| 469 |
+
"total_flos": 29937460838400000,
|
| 470 |
+
"trial_name": null,
|
| 471 |
+
"trial_params": null
|
| 472 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fef6dedf574fee1a84085c0e4ac02bee4e4596818ff62847ebfe1ad979dcefee
|
| 3 |
+
size 2107
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|