model+tokenizer: add initial version
Browse files- added_tokens.json +4 -0
- config.json +206 -0
- special_tokens_map.json +7 -0
- tokenizer.json +0 -0
- tokenizer_config.json +19 -0
- trainer_state.json +634 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
added_tokens.json
ADDED
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"<end>": 31103,
|
3 |
+
"<start>": 31102
|
4 |
+
}
|
config.json
ADDED
@@ -0,0 +1,206 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"architectures": [
|
3 |
+
"SpanMarkerModel"
|
4 |
+
],
|
5 |
+
"encoder": {
|
6 |
+
"_name_or_path": "deepset/gelectra-large",
|
7 |
+
"add_cross_attention": false,
|
8 |
+
"architectures": [
|
9 |
+
"ElectraForPreTraining"
|
10 |
+
],
|
11 |
+
"attention_probs_dropout_prob": 0.1,
|
12 |
+
"bad_words_ids": null,
|
13 |
+
"begin_suppress_tokens": null,
|
14 |
+
"bos_token_id": null,
|
15 |
+
"chunk_size_feed_forward": 0,
|
16 |
+
"classifier_dropout": null,
|
17 |
+
"cross_attention_hidden_size": null,
|
18 |
+
"decoder_start_token_id": null,
|
19 |
+
"diversity_penalty": 0.0,
|
20 |
+
"do_sample": false,
|
21 |
+
"early_stopping": false,
|
22 |
+
"embedding_size": 1024,
|
23 |
+
"encoder_no_repeat_ngram_size": 0,
|
24 |
+
"eos_token_id": null,
|
25 |
+
"exponential_decay_length_penalty": null,
|
26 |
+
"finetuning_task": null,
|
27 |
+
"forced_bos_token_id": null,
|
28 |
+
"forced_eos_token_id": null,
|
29 |
+
"hidden_act": "gelu",
|
30 |
+
"hidden_dropout_prob": 0.1,
|
31 |
+
"hidden_size": 1024,
|
32 |
+
"id2label": {
|
33 |
+
"0": "O",
|
34 |
+
"1": "B-LOC",
|
35 |
+
"2": "I-LOC",
|
36 |
+
"3": "B-LOCderiv",
|
37 |
+
"4": "I-LOCderiv",
|
38 |
+
"5": "B-LOCpart",
|
39 |
+
"6": "I-LOCpart",
|
40 |
+
"7": "B-ORG",
|
41 |
+
"8": "I-ORG",
|
42 |
+
"9": "B-ORGderiv",
|
43 |
+
"10": "I-ORGderiv",
|
44 |
+
"11": "B-ORGpart",
|
45 |
+
"12": "I-ORGpart",
|
46 |
+
"13": "B-OTH",
|
47 |
+
"14": "I-OTH",
|
48 |
+
"15": "B-OTHderiv",
|
49 |
+
"16": "I-OTHderiv",
|
50 |
+
"17": "B-OTHpart",
|
51 |
+
"18": "I-OTHpart",
|
52 |
+
"19": "B-PER",
|
53 |
+
"20": "I-PER",
|
54 |
+
"21": "B-PERderiv",
|
55 |
+
"22": "I-PERderiv",
|
56 |
+
"23": "B-PERpart",
|
57 |
+
"24": "I-PERpart"
|
58 |
+
},
|
59 |
+
"initializer_range": 0.02,
|
60 |
+
"intermediate_size": 4096,
|
61 |
+
"is_decoder": false,
|
62 |
+
"is_encoder_decoder": false,
|
63 |
+
"label2id": {
|
64 |
+
"B-LOC": 1,
|
65 |
+
"B-LOCderiv": 3,
|
66 |
+
"B-LOCpart": 5,
|
67 |
+
"B-ORG": 7,
|
68 |
+
"B-ORGderiv": 9,
|
69 |
+
"B-ORGpart": 11,
|
70 |
+
"B-OTH": 13,
|
71 |
+
"B-OTHderiv": 15,
|
72 |
+
"B-OTHpart": 17,
|
73 |
+
"B-PER": 19,
|
74 |
+
"B-PERderiv": 21,
|
75 |
+
"B-PERpart": 23,
|
76 |
+
"I-LOC": 2,
|
77 |
+
"I-LOCderiv": 4,
|
78 |
+
"I-LOCpart": 6,
|
79 |
+
"I-ORG": 8,
|
80 |
+
"I-ORGderiv": 10,
|
81 |
+
"I-ORGpart": 12,
|
82 |
+
"I-OTH": 14,
|
83 |
+
"I-OTHderiv": 16,
|
84 |
+
"I-OTHpart": 18,
|
85 |
+
"I-PER": 20,
|
86 |
+
"I-PERderiv": 22,
|
87 |
+
"I-PERpart": 24,
|
88 |
+
"O": 0
|
89 |
+
},
|
90 |
+
"layer_norm_eps": 1e-12,
|
91 |
+
"length_penalty": 1.0,
|
92 |
+
"max_length": 20,
|
93 |
+
"max_position_embeddings": 512,
|
94 |
+
"min_length": 0,
|
95 |
+
"model_type": "electra",
|
96 |
+
"no_repeat_ngram_size": 0,
|
97 |
+
"num_attention_heads": 16,
|
98 |
+
"num_beam_groups": 1,
|
99 |
+
"num_beams": 1,
|
100 |
+
"num_hidden_layers": 24,
|
101 |
+
"num_return_sequences": 1,
|
102 |
+
"output_attentions": false,
|
103 |
+
"output_hidden_states": false,
|
104 |
+
"output_scores": false,
|
105 |
+
"pad_token_id": 0,
|
106 |
+
"position_embedding_type": "absolute",
|
107 |
+
"prefix": null,
|
108 |
+
"problem_type": null,
|
109 |
+
"pruned_heads": {},
|
110 |
+
"remove_invalid_values": false,
|
111 |
+
"repetition_penalty": 1.0,
|
112 |
+
"return_dict": true,
|
113 |
+
"return_dict_in_generate": false,
|
114 |
+
"sep_token_id": null,
|
115 |
+
"summary_activation": "gelu",
|
116 |
+
"summary_last_dropout": 0.1,
|
117 |
+
"summary_type": "first",
|
118 |
+
"summary_use_proj": true,
|
119 |
+
"suppress_tokens": null,
|
120 |
+
"task_specific_params": null,
|
121 |
+
"temperature": 1.0,
|
122 |
+
"tf_legacy_loss": false,
|
123 |
+
"tie_encoder_decoder": false,
|
124 |
+
"tie_word_embeddings": true,
|
125 |
+
"tokenizer_class": null,
|
126 |
+
"top_k": 50,
|
127 |
+
"top_p": 1.0,
|
128 |
+
"torch_dtype": null,
|
129 |
+
"torchscript": false,
|
130 |
+
"transformers_version": "4.31.0",
|
131 |
+
"type_vocab_size": 2,
|
132 |
+
"typical_p": 1.0,
|
133 |
+
"use_bfloat16": false,
|
134 |
+
"use_cache": true,
|
135 |
+
"vocab_size": 31104
|
136 |
+
},
|
137 |
+
"entity_max_length": 8,
|
138 |
+
"id2label": {
|
139 |
+
"0": "O",
|
140 |
+
"1": "LOC",
|
141 |
+
"2": "LOCderiv",
|
142 |
+
"3": "LOCpart",
|
143 |
+
"4": "ORG",
|
144 |
+
"5": "ORGderiv",
|
145 |
+
"6": "ORGpart",
|
146 |
+
"7": "OTH",
|
147 |
+
"8": "OTHderiv",
|
148 |
+
"9": "OTHpart",
|
149 |
+
"10": "PER",
|
150 |
+
"11": "PERderiv",
|
151 |
+
"12": "PERpart"
|
152 |
+
},
|
153 |
+
"id2reduced_id": {
|
154 |
+
"0": 0,
|
155 |
+
"1": 1,
|
156 |
+
"2": 1,
|
157 |
+
"3": 2,
|
158 |
+
"4": 2,
|
159 |
+
"5": 3,
|
160 |
+
"6": 3,
|
161 |
+
"7": 4,
|
162 |
+
"8": 4,
|
163 |
+
"9": 5,
|
164 |
+
"10": 5,
|
165 |
+
"11": 6,
|
166 |
+
"12": 6,
|
167 |
+
"13": 7,
|
168 |
+
"14": 7,
|
169 |
+
"15": 8,
|
170 |
+
"16": 8,
|
171 |
+
"17": 9,
|
172 |
+
"18": 9,
|
173 |
+
"19": 10,
|
174 |
+
"20": 10,
|
175 |
+
"21": 11,
|
176 |
+
"22": 11,
|
177 |
+
"23": 12,
|
178 |
+
"24": 12
|
179 |
+
},
|
180 |
+
"label2id": {
|
181 |
+
"LOC": 1,
|
182 |
+
"LOCderiv": 2,
|
183 |
+
"LOCpart": 3,
|
184 |
+
"O": 0,
|
185 |
+
"ORG": 4,
|
186 |
+
"ORGderiv": 5,
|
187 |
+
"ORGpart": 6,
|
188 |
+
"OTH": 7,
|
189 |
+
"OTHderiv": 8,
|
190 |
+
"OTHpart": 9,
|
191 |
+
"PER": 10,
|
192 |
+
"PERderiv": 11,
|
193 |
+
"PERpart": 12
|
194 |
+
},
|
195 |
+
"marker_max_length": 128,
|
196 |
+
"max_next_context": null,
|
197 |
+
"max_prev_context": null,
|
198 |
+
"model_max_length": 256,
|
199 |
+
"model_max_length_default": 512,
|
200 |
+
"model_type": "span-marker",
|
201 |
+
"span_marker_version": "1.2.4",
|
202 |
+
"torch_dtype": "float32",
|
203 |
+
"trained_with_document_context": false,
|
204 |
+
"transformers_version": "4.31.0",
|
205 |
+
"vocab_size": 31104
|
206 |
+
}
|
special_tokens_map.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"cls_token": "[CLS]",
|
3 |
+
"mask_token": "[MASK]",
|
4 |
+
"pad_token": "[PAD]",
|
5 |
+
"sep_token": "[SEP]",
|
6 |
+
"unk_token": "[UNK]"
|
7 |
+
}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_prefix_space": true,
|
3 |
+
"clean_up_tokenization_spaces": true,
|
4 |
+
"cls_token": "[CLS]",
|
5 |
+
"do_basic_tokenize": true,
|
6 |
+
"do_lower_case": false,
|
7 |
+
"entity_max_length": 8,
|
8 |
+
"marker_max_length": 128,
|
9 |
+
"mask_token": "[MASK]",
|
10 |
+
"max_len": 512,
|
11 |
+
"model_max_length": 256,
|
12 |
+
"never_split": null,
|
13 |
+
"pad_token": "[PAD]",
|
14 |
+
"sep_token": "[SEP]",
|
15 |
+
"strip_accents": false,
|
16 |
+
"tokenize_chinese_chars": true,
|
17 |
+
"tokenizer_class": "ElectraTokenizer",
|
18 |
+
"unk_token": "[UNK]"
|
19 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,634 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.006780553143471479,
|
3 |
+
"best_model_checkpoint": "./span_marker-gelectra-large-bs16-lr5e-05-1/checkpoint-3208",
|
4 |
+
"epoch": 3.0,
|
5 |
+
"global_step": 4812,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.0,
|
12 |
+
"learning_rate": 1.037344398340249e-07,
|
13 |
+
"loss": 3.4599,
|
14 |
+
"step": 1
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.03,
|
18 |
+
"learning_rate": 4.564315352697096e-06,
|
19 |
+
"loss": 2.5081,
|
20 |
+
"step": 50
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.06,
|
24 |
+
"learning_rate": 9.647302904564317e-06,
|
25 |
+
"loss": 0.1807,
|
26 |
+
"step": 100
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.09,
|
30 |
+
"learning_rate": 1.4834024896265561e-05,
|
31 |
+
"loss": 0.056,
|
32 |
+
"step": 150
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.12,
|
36 |
+
"learning_rate": 2.0020746887966806e-05,
|
37 |
+
"loss": 0.0411,
|
38 |
+
"step": 200
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.16,
|
42 |
+
"learning_rate": 2.520746887966805e-05,
|
43 |
+
"loss": 0.0309,
|
44 |
+
"step": 250
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.19,
|
48 |
+
"learning_rate": 3.0394190871369292e-05,
|
49 |
+
"loss": 0.0239,
|
50 |
+
"step": 300
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.22,
|
54 |
+
"learning_rate": 3.558091286307054e-05,
|
55 |
+
"loss": 0.0192,
|
56 |
+
"step": 350
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.25,
|
60 |
+
"learning_rate": 4.0767634854771785e-05,
|
61 |
+
"loss": 0.0165,
|
62 |
+
"step": 400
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.28,
|
66 |
+
"learning_rate": 4.5954356846473026e-05,
|
67 |
+
"loss": 0.0158,
|
68 |
+
"step": 450
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.31,
|
72 |
+
"learning_rate": 4.9872979214780604e-05,
|
73 |
+
"loss": 0.0147,
|
74 |
+
"step": 500
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.34,
|
78 |
+
"learning_rate": 4.929561200923788e-05,
|
79 |
+
"loss": 0.0134,
|
80 |
+
"step": 550
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.37,
|
84 |
+
"learning_rate": 4.871824480369515e-05,
|
85 |
+
"loss": 0.0128,
|
86 |
+
"step": 600
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.41,
|
90 |
+
"learning_rate": 4.814087759815243e-05,
|
91 |
+
"loss": 0.0112,
|
92 |
+
"step": 650
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 0.44,
|
96 |
+
"learning_rate": 4.75635103926097e-05,
|
97 |
+
"loss": 0.0115,
|
98 |
+
"step": 700
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 0.47,
|
102 |
+
"learning_rate": 4.698614318706698e-05,
|
103 |
+
"loss": 0.0123,
|
104 |
+
"step": 750
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 0.5,
|
108 |
+
"learning_rate": 4.640877598152425e-05,
|
109 |
+
"loss": 0.0102,
|
110 |
+
"step": 800
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 0.53,
|
114 |
+
"learning_rate": 4.5831408775981525e-05,
|
115 |
+
"loss": 0.0089,
|
116 |
+
"step": 850
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 0.56,
|
120 |
+
"learning_rate": 4.52540415704388e-05,
|
121 |
+
"loss": 0.0104,
|
122 |
+
"step": 900
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 0.59,
|
126 |
+
"learning_rate": 4.4676674364896075e-05,
|
127 |
+
"loss": 0.0107,
|
128 |
+
"step": 950
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.62,
|
132 |
+
"learning_rate": 4.409930715935335e-05,
|
133 |
+
"loss": 0.01,
|
134 |
+
"step": 1000
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 0.65,
|
138 |
+
"learning_rate": 4.3521939953810626e-05,
|
139 |
+
"loss": 0.0096,
|
140 |
+
"step": 1050
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 0.69,
|
144 |
+
"learning_rate": 4.29445727482679e-05,
|
145 |
+
"loss": 0.0108,
|
146 |
+
"step": 1100
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 0.72,
|
150 |
+
"learning_rate": 4.236720554272518e-05,
|
151 |
+
"loss": 0.0093,
|
152 |
+
"step": 1150
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 0.75,
|
156 |
+
"learning_rate": 4.178983833718245e-05,
|
157 |
+
"loss": 0.0096,
|
158 |
+
"step": 1200
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 0.78,
|
162 |
+
"learning_rate": 4.121247113163972e-05,
|
163 |
+
"loss": 0.0098,
|
164 |
+
"step": 1250
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 0.81,
|
168 |
+
"learning_rate": 4.0635103926097e-05,
|
169 |
+
"loss": 0.0085,
|
170 |
+
"step": 1300
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.84,
|
174 |
+
"learning_rate": 4.005773672055427e-05,
|
175 |
+
"loss": 0.0079,
|
176 |
+
"step": 1350
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 0.87,
|
180 |
+
"learning_rate": 3.9480369515011554e-05,
|
181 |
+
"loss": 0.0098,
|
182 |
+
"step": 1400
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 0.9,
|
186 |
+
"learning_rate": 3.890300230946882e-05,
|
187 |
+
"loss": 0.0089,
|
188 |
+
"step": 1450
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 0.94,
|
192 |
+
"learning_rate": 3.83256351039261e-05,
|
193 |
+
"loss": 0.0095,
|
194 |
+
"step": 1500
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 0.97,
|
198 |
+
"learning_rate": 3.774826789838337e-05,
|
199 |
+
"loss": 0.008,
|
200 |
+
"step": 1550
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 1.0,
|
204 |
+
"learning_rate": 3.717090069284065e-05,
|
205 |
+
"loss": 0.0076,
|
206 |
+
"step": 1600
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 1.0,
|
210 |
+
"eval_loss": 0.0075011407025158405,
|
211 |
+
"eval_overall_accuracy": 0.9820421098120183,
|
212 |
+
"eval_overall_f1": 0.880326046683957,
|
213 |
+
"eval_overall_precision": 0.8693743139407245,
|
214 |
+
"eval_overall_recall": 0.8915572232645403,
|
215 |
+
"eval_runtime": 25.2918,
|
216 |
+
"eval_samples_per_second": 93.153,
|
217 |
+
"eval_steps_per_second": 5.852,
|
218 |
+
"step": 1604
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 1.03,
|
222 |
+
"learning_rate": 3.6593533487297924e-05,
|
223 |
+
"loss": 0.0068,
|
224 |
+
"step": 1650
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 1.06,
|
228 |
+
"learning_rate": 3.60161662817552e-05,
|
229 |
+
"loss": 0.0062,
|
230 |
+
"step": 1700
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 1.09,
|
234 |
+
"learning_rate": 3.5438799076212475e-05,
|
235 |
+
"loss": 0.0054,
|
236 |
+
"step": 1750
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 1.12,
|
240 |
+
"learning_rate": 3.486143187066975e-05,
|
241 |
+
"loss": 0.0062,
|
242 |
+
"step": 1800
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 1.15,
|
246 |
+
"learning_rate": 3.4284064665127025e-05,
|
247 |
+
"loss": 0.0063,
|
248 |
+
"step": 1850
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 1.18,
|
252 |
+
"learning_rate": 3.3706697459584294e-05,
|
253 |
+
"loss": 0.0053,
|
254 |
+
"step": 1900
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 1.22,
|
258 |
+
"learning_rate": 3.3129330254041576e-05,
|
259 |
+
"loss": 0.0056,
|
260 |
+
"step": 1950
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 1.25,
|
264 |
+
"learning_rate": 3.2551963048498845e-05,
|
265 |
+
"loss": 0.0066,
|
266 |
+
"step": 2000
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 1.28,
|
270 |
+
"learning_rate": 3.197459584295612e-05,
|
271 |
+
"loss": 0.0054,
|
272 |
+
"step": 2050
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 1.31,
|
276 |
+
"learning_rate": 3.1397228637413396e-05,
|
277 |
+
"loss": 0.0061,
|
278 |
+
"step": 2100
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 1.34,
|
282 |
+
"learning_rate": 3.081986143187067e-05,
|
283 |
+
"loss": 0.0063,
|
284 |
+
"step": 2150
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 1.37,
|
288 |
+
"learning_rate": 3.0242494226327943e-05,
|
289 |
+
"loss": 0.0062,
|
290 |
+
"step": 2200
|
291 |
+
},
|
292 |
+
{
|
293 |
+
"epoch": 1.4,
|
294 |
+
"learning_rate": 2.9665127020785222e-05,
|
295 |
+
"loss": 0.0063,
|
296 |
+
"step": 2250
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 1.43,
|
300 |
+
"learning_rate": 2.9087759815242494e-05,
|
301 |
+
"loss": 0.005,
|
302 |
+
"step": 2300
|
303 |
+
},
|
304 |
+
{
|
305 |
+
"epoch": 1.47,
|
306 |
+
"learning_rate": 2.8510392609699772e-05,
|
307 |
+
"loss": 0.0049,
|
308 |
+
"step": 2350
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 1.5,
|
312 |
+
"learning_rate": 2.793302540415704e-05,
|
313 |
+
"loss": 0.006,
|
314 |
+
"step": 2400
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 1.53,
|
318 |
+
"learning_rate": 2.735565819861432e-05,
|
319 |
+
"loss": 0.0059,
|
320 |
+
"step": 2450
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 1.56,
|
324 |
+
"learning_rate": 2.67782909930716e-05,
|
325 |
+
"loss": 0.006,
|
326 |
+
"step": 2500
|
327 |
+
},
|
328 |
+
{
|
329 |
+
"epoch": 1.59,
|
330 |
+
"learning_rate": 2.620092378752887e-05,
|
331 |
+
"loss": 0.0057,
|
332 |
+
"step": 2550
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 1.62,
|
336 |
+
"learning_rate": 2.5623556581986146e-05,
|
337 |
+
"loss": 0.0051,
|
338 |
+
"step": 2600
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 1.65,
|
342 |
+
"learning_rate": 2.5046189376443418e-05,
|
343 |
+
"loss": 0.005,
|
344 |
+
"step": 2650
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 1.68,
|
348 |
+
"learning_rate": 2.4468822170900693e-05,
|
349 |
+
"loss": 0.0056,
|
350 |
+
"step": 2700
|
351 |
+
},
|
352 |
+
{
|
353 |
+
"epoch": 1.71,
|
354 |
+
"learning_rate": 2.389145496535797e-05,
|
355 |
+
"loss": 0.0054,
|
356 |
+
"step": 2750
|
357 |
+
},
|
358 |
+
{
|
359 |
+
"epoch": 1.75,
|
360 |
+
"learning_rate": 2.3314087759815244e-05,
|
361 |
+
"loss": 0.0053,
|
362 |
+
"step": 2800
|
363 |
+
},
|
364 |
+
{
|
365 |
+
"epoch": 1.78,
|
366 |
+
"learning_rate": 2.273672055427252e-05,
|
367 |
+
"loss": 0.0056,
|
368 |
+
"step": 2850
|
369 |
+
},
|
370 |
+
{
|
371 |
+
"epoch": 1.81,
|
372 |
+
"learning_rate": 2.2159353348729795e-05,
|
373 |
+
"loss": 0.0053,
|
374 |
+
"step": 2900
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 1.84,
|
378 |
+
"learning_rate": 2.158198614318707e-05,
|
379 |
+
"loss": 0.0044,
|
380 |
+
"step": 2950
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 1.87,
|
384 |
+
"learning_rate": 2.1004618937644342e-05,
|
385 |
+
"loss": 0.0044,
|
386 |
+
"step": 3000
|
387 |
+
},
|
388 |
+
{
|
389 |
+
"epoch": 1.9,
|
390 |
+
"learning_rate": 2.0427251732101618e-05,
|
391 |
+
"loss": 0.0057,
|
392 |
+
"step": 3050
|
393 |
+
},
|
394 |
+
{
|
395 |
+
"epoch": 1.93,
|
396 |
+
"learning_rate": 1.9849884526558893e-05,
|
397 |
+
"loss": 0.0058,
|
398 |
+
"step": 3100
|
399 |
+
},
|
400 |
+
{
|
401 |
+
"epoch": 1.96,
|
402 |
+
"learning_rate": 1.927251732101617e-05,
|
403 |
+
"loss": 0.0056,
|
404 |
+
"step": 3150
|
405 |
+
},
|
406 |
+
{
|
407 |
+
"epoch": 2.0,
|
408 |
+
"learning_rate": 1.8695150115473444e-05,
|
409 |
+
"loss": 0.0059,
|
410 |
+
"step": 3200
|
411 |
+
},
|
412 |
+
{
|
413 |
+
"epoch": 2.0,
|
414 |
+
"eval_loss": 0.006780553143471479,
|
415 |
+
"eval_overall_accuracy": 0.983866708280316,
|
416 |
+
"eval_overall_f1": 0.8873682263732198,
|
417 |
+
"eval_overall_precision": 0.87490882567469,
|
418 |
+
"eval_overall_recall": 0.900187617260788,
|
419 |
+
"eval_runtime": 25.1096,
|
420 |
+
"eval_samples_per_second": 93.829,
|
421 |
+
"eval_steps_per_second": 5.894,
|
422 |
+
"step": 3208
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 2.03,
|
426 |
+
"learning_rate": 1.8117782909930716e-05,
|
427 |
+
"loss": 0.0037,
|
428 |
+
"step": 3250
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 2.06,
|
432 |
+
"learning_rate": 1.754041570438799e-05,
|
433 |
+
"loss": 0.0027,
|
434 |
+
"step": 3300
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 2.09,
|
438 |
+
"learning_rate": 1.6963048498845266e-05,
|
439 |
+
"loss": 0.003,
|
440 |
+
"step": 3350
|
441 |
+
},
|
442 |
+
{
|
443 |
+
"epoch": 2.12,
|
444 |
+
"learning_rate": 1.6385681293302542e-05,
|
445 |
+
"loss": 0.003,
|
446 |
+
"step": 3400
|
447 |
+
},
|
448 |
+
{
|
449 |
+
"epoch": 2.15,
|
450 |
+
"learning_rate": 1.5808314087759814e-05,
|
451 |
+
"loss": 0.003,
|
452 |
+
"step": 3450
|
453 |
+
},
|
454 |
+
{
|
455 |
+
"epoch": 2.18,
|
456 |
+
"learning_rate": 1.523094688221709e-05,
|
457 |
+
"loss": 0.0036,
|
458 |
+
"step": 3500
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 2.21,
|
462 |
+
"learning_rate": 1.4653579676674365e-05,
|
463 |
+
"loss": 0.0024,
|
464 |
+
"step": 3550
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 2.24,
|
468 |
+
"learning_rate": 1.4076212471131642e-05,
|
469 |
+
"loss": 0.0031,
|
470 |
+
"step": 3600
|
471 |
+
},
|
472 |
+
{
|
473 |
+
"epoch": 2.28,
|
474 |
+
"learning_rate": 1.3498845265588917e-05,
|
475 |
+
"loss": 0.003,
|
476 |
+
"step": 3650
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 2.31,
|
480 |
+
"learning_rate": 1.292147806004619e-05,
|
481 |
+
"loss": 0.003,
|
482 |
+
"step": 3700
|
483 |
+
},
|
484 |
+
{
|
485 |
+
"epoch": 2.34,
|
486 |
+
"learning_rate": 1.2344110854503464e-05,
|
487 |
+
"loss": 0.0027,
|
488 |
+
"step": 3750
|
489 |
+
},
|
490 |
+
{
|
491 |
+
"epoch": 2.37,
|
492 |
+
"learning_rate": 1.176674364896074e-05,
|
493 |
+
"loss": 0.0028,
|
494 |
+
"step": 3800
|
495 |
+
},
|
496 |
+
{
|
497 |
+
"epoch": 2.4,
|
498 |
+
"learning_rate": 1.1189376443418015e-05,
|
499 |
+
"loss": 0.0025,
|
500 |
+
"step": 3850
|
501 |
+
},
|
502 |
+
{
|
503 |
+
"epoch": 2.43,
|
504 |
+
"learning_rate": 1.0612009237875289e-05,
|
505 |
+
"loss": 0.0026,
|
506 |
+
"step": 3900
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 2.46,
|
510 |
+
"learning_rate": 1.0034642032332564e-05,
|
511 |
+
"loss": 0.0029,
|
512 |
+
"step": 3950
|
513 |
+
},
|
514 |
+
{
|
515 |
+
"epoch": 2.49,
|
516 |
+
"learning_rate": 9.457274826789838e-06,
|
517 |
+
"loss": 0.0025,
|
518 |
+
"step": 4000
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 2.52,
|
522 |
+
"learning_rate": 8.879907621247113e-06,
|
523 |
+
"loss": 0.0022,
|
524 |
+
"step": 4050
|
525 |
+
},
|
526 |
+
{
|
527 |
+
"epoch": 2.56,
|
528 |
+
"learning_rate": 8.302540415704389e-06,
|
529 |
+
"loss": 0.0025,
|
530 |
+
"step": 4100
|
531 |
+
},
|
532 |
+
{
|
533 |
+
"epoch": 2.59,
|
534 |
+
"learning_rate": 7.725173210161662e-06,
|
535 |
+
"loss": 0.0026,
|
536 |
+
"step": 4150
|
537 |
+
},
|
538 |
+
{
|
539 |
+
"epoch": 2.62,
|
540 |
+
"learning_rate": 7.147806004618939e-06,
|
541 |
+
"loss": 0.0026,
|
542 |
+
"step": 4200
|
543 |
+
},
|
544 |
+
{
|
545 |
+
"epoch": 2.65,
|
546 |
+
"learning_rate": 6.570438799076213e-06,
|
547 |
+
"loss": 0.0019,
|
548 |
+
"step": 4250
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 2.68,
|
552 |
+
"learning_rate": 5.993071593533488e-06,
|
553 |
+
"loss": 0.0031,
|
554 |
+
"step": 4300
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 2.71,
|
558 |
+
"learning_rate": 5.415704387990762e-06,
|
559 |
+
"loss": 0.0027,
|
560 |
+
"step": 4350
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 2.74,
|
564 |
+
"learning_rate": 4.838337182448037e-06,
|
565 |
+
"loss": 0.003,
|
566 |
+
"step": 4400
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 2.77,
|
570 |
+
"learning_rate": 4.260969976905312e-06,
|
571 |
+
"loss": 0.0023,
|
572 |
+
"step": 4450
|
573 |
+
},
|
574 |
+
{
|
575 |
+
"epoch": 2.81,
|
576 |
+
"learning_rate": 3.683602771362587e-06,
|
577 |
+
"loss": 0.0031,
|
578 |
+
"step": 4500
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 2.84,
|
582 |
+
"learning_rate": 3.1062355658198616e-06,
|
583 |
+
"loss": 0.0033,
|
584 |
+
"step": 4550
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 2.87,
|
588 |
+
"learning_rate": 2.5288683602771365e-06,
|
589 |
+
"loss": 0.0022,
|
590 |
+
"step": 4600
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 2.9,
|
594 |
+
"learning_rate": 1.951501154734411e-06,
|
595 |
+
"loss": 0.0023,
|
596 |
+
"step": 4650
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 2.93,
|
600 |
+
"learning_rate": 1.374133949191686e-06,
|
601 |
+
"loss": 0.0022,
|
602 |
+
"step": 4700
|
603 |
+
},
|
604 |
+
{
|
605 |
+
"epoch": 2.96,
|
606 |
+
"learning_rate": 7.967667436489609e-07,
|
607 |
+
"loss": 0.0024,
|
608 |
+
"step": 4750
|
609 |
+
},
|
610 |
+
{
|
611 |
+
"epoch": 2.99,
|
612 |
+
"learning_rate": 2.1939953810623557e-07,
|
613 |
+
"loss": 0.003,
|
614 |
+
"step": 4800
|
615 |
+
},
|
616 |
+
{
|
617 |
+
"epoch": 3.0,
|
618 |
+
"eval_loss": 0.007023950573056936,
|
619 |
+
"eval_overall_accuracy": 0.9848990468873791,
|
620 |
+
"eval_overall_f1": 0.8998883513211762,
|
621 |
+
"eval_overall_precision": 0.8925802879291251,
|
622 |
+
"eval_overall_recall": 0.9073170731707317,
|
623 |
+
"eval_runtime": 25.1076,
|
624 |
+
"eval_samples_per_second": 93.836,
|
625 |
+
"eval_steps_per_second": 5.895,
|
626 |
+
"step": 4812
|
627 |
+
}
|
628 |
+
],
|
629 |
+
"max_steps": 4812,
|
630 |
+
"num_train_epochs": 3,
|
631 |
+
"total_flos": 7.147260152258458e+16,
|
632 |
+
"trial_name": null,
|
633 |
+
"trial_params": null
|
634 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d55d278499fcda606ccb98597a26de5e1f6319c722706453708b1a4b52f87fa7
|
3 |
+
size 4027
|
vocab.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|