feat: donut model
Browse files- config.json +195 -0
- pytorch_model.bin +3 -0
- trainer_state.json +1822 -0
- training_args.bin +3 -0
config.json
ADDED
@@ -0,0 +1,195 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_commit_hash": "3bb6803e5ea104ff51596e7066792ef0db0317f3",
|
3 |
+
"_name_or_path": "laverdes/donut-commoncrawl-mid",
|
4 |
+
"architectures": [
|
5 |
+
"VisionEncoderDecoderModel"
|
6 |
+
],
|
7 |
+
"decoder": {
|
8 |
+
"_name_or_path": "",
|
9 |
+
"activation_dropout": 0.0,
|
10 |
+
"activation_function": "gelu",
|
11 |
+
"add_cross_attention": true,
|
12 |
+
"add_final_layer_norm": true,
|
13 |
+
"architectures": null,
|
14 |
+
"attention_dropout": 0.0,
|
15 |
+
"bad_words_ids": null,
|
16 |
+
"begin_suppress_tokens": null,
|
17 |
+
"bos_token_id": 0,
|
18 |
+
"chunk_size_feed_forward": 0,
|
19 |
+
"classifier_dropout": 0.0,
|
20 |
+
"cross_attention_hidden_size": null,
|
21 |
+
"d_model": 1024,
|
22 |
+
"decoder_attention_heads": 16,
|
23 |
+
"decoder_ffn_dim": 4096,
|
24 |
+
"decoder_layerdrop": 0.0,
|
25 |
+
"decoder_layers": 4,
|
26 |
+
"decoder_start_token_id": null,
|
27 |
+
"diversity_penalty": 0.0,
|
28 |
+
"do_sample": false,
|
29 |
+
"dropout": 0.1,
|
30 |
+
"early_stopping": false,
|
31 |
+
"encoder_attention_heads": 16,
|
32 |
+
"encoder_ffn_dim": 4096,
|
33 |
+
"encoder_layerdrop": 0.0,
|
34 |
+
"encoder_layers": 12,
|
35 |
+
"encoder_no_repeat_ngram_size": 0,
|
36 |
+
"eos_token_id": 2,
|
37 |
+
"exponential_decay_length_penalty": null,
|
38 |
+
"finetuning_task": null,
|
39 |
+
"forced_bos_token_id": null,
|
40 |
+
"forced_eos_token_id": 2,
|
41 |
+
"id2label": {
|
42 |
+
"0": "LABEL_0",
|
43 |
+
"1": "LABEL_1"
|
44 |
+
},
|
45 |
+
"init_std": 0.02,
|
46 |
+
"is_decoder": true,
|
47 |
+
"is_encoder_decoder": false,
|
48 |
+
"label2id": {
|
49 |
+
"LABEL_0": 0,
|
50 |
+
"LABEL_1": 1
|
51 |
+
},
|
52 |
+
"length_penalty": 1.0,
|
53 |
+
"max_length": 1024,
|
54 |
+
"max_position_embeddings": 1536,
|
55 |
+
"min_length": 0,
|
56 |
+
"model_type": "mbart",
|
57 |
+
"no_repeat_ngram_size": 0,
|
58 |
+
"num_beam_groups": 1,
|
59 |
+
"num_beams": 1,
|
60 |
+
"num_hidden_layers": 12,
|
61 |
+
"num_return_sequences": 1,
|
62 |
+
"output_attentions": false,
|
63 |
+
"output_hidden_states": false,
|
64 |
+
"output_scores": false,
|
65 |
+
"pad_token_id": 1,
|
66 |
+
"prefix": null,
|
67 |
+
"problem_type": null,
|
68 |
+
"pruned_heads": {},
|
69 |
+
"remove_invalid_values": false,
|
70 |
+
"repetition_penalty": 1.0,
|
71 |
+
"return_dict": true,
|
72 |
+
"return_dict_in_generate": false,
|
73 |
+
"scale_embedding": true,
|
74 |
+
"sep_token_id": null,
|
75 |
+
"suppress_tokens": null,
|
76 |
+
"task_specific_params": null,
|
77 |
+
"temperature": 1.0,
|
78 |
+
"tf_legacy_loss": false,
|
79 |
+
"tie_encoder_decoder": false,
|
80 |
+
"tie_word_embeddings": true,
|
81 |
+
"tokenizer_class": null,
|
82 |
+
"top_k": 50,
|
83 |
+
"top_p": 1.0,
|
84 |
+
"torch_dtype": null,
|
85 |
+
"torchscript": false,
|
86 |
+
"transformers_version": "4.28.1",
|
87 |
+
"typical_p": 1.0,
|
88 |
+
"use_bfloat16": false,
|
89 |
+
"use_cache": true,
|
90 |
+
"vocab_size": 57569
|
91 |
+
},
|
92 |
+
"decoder_start_token_id": 0,
|
93 |
+
"encoder": {
|
94 |
+
"_name_or_path": "",
|
95 |
+
"add_cross_attention": false,
|
96 |
+
"architectures": null,
|
97 |
+
"attention_probs_dropout_prob": 0.0,
|
98 |
+
"bad_words_ids": null,
|
99 |
+
"begin_suppress_tokens": null,
|
100 |
+
"bos_token_id": null,
|
101 |
+
"chunk_size_feed_forward": 0,
|
102 |
+
"cross_attention_hidden_size": null,
|
103 |
+
"decoder_start_token_id": null,
|
104 |
+
"depths": [
|
105 |
+
2,
|
106 |
+
2,
|
107 |
+
14,
|
108 |
+
2
|
109 |
+
],
|
110 |
+
"diversity_penalty": 0.0,
|
111 |
+
"do_sample": false,
|
112 |
+
"drop_path_rate": 0.1,
|
113 |
+
"early_stopping": false,
|
114 |
+
"embed_dim": 128,
|
115 |
+
"encoder_no_repeat_ngram_size": 0,
|
116 |
+
"eos_token_id": null,
|
117 |
+
"exponential_decay_length_penalty": null,
|
118 |
+
"finetuning_task": null,
|
119 |
+
"forced_bos_token_id": null,
|
120 |
+
"forced_eos_token_id": null,
|
121 |
+
"hidden_act": "gelu",
|
122 |
+
"hidden_dropout_prob": 0.0,
|
123 |
+
"hidden_size": 1024,
|
124 |
+
"id2label": {
|
125 |
+
"0": "LABEL_0",
|
126 |
+
"1": "LABEL_1"
|
127 |
+
},
|
128 |
+
"image_size": [
|
129 |
+
960,
|
130 |
+
720
|
131 |
+
],
|
132 |
+
"initializer_range": 0.02,
|
133 |
+
"is_decoder": false,
|
134 |
+
"is_encoder_decoder": false,
|
135 |
+
"label2id": {
|
136 |
+
"LABEL_0": 0,
|
137 |
+
"LABEL_1": 1
|
138 |
+
},
|
139 |
+
"layer_norm_eps": 1e-05,
|
140 |
+
"length_penalty": 1.0,
|
141 |
+
"max_length": 20,
|
142 |
+
"min_length": 0,
|
143 |
+
"mlp_ratio": 4.0,
|
144 |
+
"model_type": "donut-swin",
|
145 |
+
"no_repeat_ngram_size": 0,
|
146 |
+
"num_beam_groups": 1,
|
147 |
+
"num_beams": 1,
|
148 |
+
"num_channels": 3,
|
149 |
+
"num_heads": [
|
150 |
+
4,
|
151 |
+
8,
|
152 |
+
16,
|
153 |
+
32
|
154 |
+
],
|
155 |
+
"num_layers": 4,
|
156 |
+
"num_return_sequences": 1,
|
157 |
+
"output_attentions": false,
|
158 |
+
"output_hidden_states": false,
|
159 |
+
"output_scores": false,
|
160 |
+
"pad_token_id": null,
|
161 |
+
"patch_size": 4,
|
162 |
+
"path_norm": true,
|
163 |
+
"prefix": null,
|
164 |
+
"problem_type": null,
|
165 |
+
"pruned_heads": {},
|
166 |
+
"qkv_bias": true,
|
167 |
+
"remove_invalid_values": false,
|
168 |
+
"repetition_penalty": 1.0,
|
169 |
+
"return_dict": true,
|
170 |
+
"return_dict_in_generate": false,
|
171 |
+
"sep_token_id": null,
|
172 |
+
"suppress_tokens": null,
|
173 |
+
"task_specific_params": null,
|
174 |
+
"temperature": 1.0,
|
175 |
+
"tf_legacy_loss": false,
|
176 |
+
"tie_encoder_decoder": false,
|
177 |
+
"tie_word_embeddings": true,
|
178 |
+
"tokenizer_class": null,
|
179 |
+
"top_k": 50,
|
180 |
+
"top_p": 1.0,
|
181 |
+
"torch_dtype": null,
|
182 |
+
"torchscript": false,
|
183 |
+
"transformers_version": "4.28.1",
|
184 |
+
"typical_p": 1.0,
|
185 |
+
"use_absolute_embeddings": false,
|
186 |
+
"use_bfloat16": false,
|
187 |
+
"window_size": 10
|
188 |
+
},
|
189 |
+
"is_encoder_decoder": true,
|
190 |
+
"model_type": "vision-encoder-decoder",
|
191 |
+
"pad_token_id": 1,
|
192 |
+
"tie_word_embeddings": false,
|
193 |
+
"torch_dtype": "float32",
|
194 |
+
"transformers_version": null
|
195 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8621032501532d3c68ad696ceceebafbcd91d56f56d2778dad83e463eef9d490
|
3 |
+
size 809352409
|
trainer_state.json
ADDED
@@ -0,0 +1,1822 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 11.0,
|
5 |
+
"global_step": 12914,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.09,
|
12 |
+
"learning_rate": 1.9914821124361162e-05,
|
13 |
+
"loss": 4.8491,
|
14 |
+
"step": 100
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.09,
|
18 |
+
"eval_loss": 4.696099281311035,
|
19 |
+
"eval_runtime": 907.3534,
|
20 |
+
"eval_samples_per_second": 0.323,
|
21 |
+
"eval_steps_per_second": 0.041,
|
22 |
+
"step": 100
|
23 |
+
},
|
24 |
+
{
|
25 |
+
"epoch": 0.17,
|
26 |
+
"learning_rate": 1.982964224872232e-05,
|
27 |
+
"loss": 4.88,
|
28 |
+
"step": 200
|
29 |
+
},
|
30 |
+
{
|
31 |
+
"epoch": 0.17,
|
32 |
+
"eval_loss": 4.71148681640625,
|
33 |
+
"eval_runtime": 901.0601,
|
34 |
+
"eval_samples_per_second": 0.325,
|
35 |
+
"eval_steps_per_second": 0.041,
|
36 |
+
"step": 200
|
37 |
+
},
|
38 |
+
{
|
39 |
+
"epoch": 0.26,
|
40 |
+
"learning_rate": 1.9744463373083477e-05,
|
41 |
+
"loss": 4.6409,
|
42 |
+
"step": 300
|
43 |
+
},
|
44 |
+
{
|
45 |
+
"epoch": 0.26,
|
46 |
+
"eval_loss": 4.671314239501953,
|
47 |
+
"eval_runtime": 906.18,
|
48 |
+
"eval_samples_per_second": 0.323,
|
49 |
+
"eval_steps_per_second": 0.041,
|
50 |
+
"step": 300
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.34,
|
54 |
+
"learning_rate": 1.9659284497444635e-05,
|
55 |
+
"loss": 4.7222,
|
56 |
+
"step": 400
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.34,
|
60 |
+
"eval_loss": 4.637036323547363,
|
61 |
+
"eval_runtime": 903.4161,
|
62 |
+
"eval_samples_per_second": 0.324,
|
63 |
+
"eval_steps_per_second": 0.041,
|
64 |
+
"step": 400
|
65 |
+
},
|
66 |
+
{
|
67 |
+
"epoch": 0.43,
|
68 |
+
"learning_rate": 1.9574105621805795e-05,
|
69 |
+
"loss": 4.6357,
|
70 |
+
"step": 500
|
71 |
+
},
|
72 |
+
{
|
73 |
+
"epoch": 0.43,
|
74 |
+
"eval_loss": 4.6376471519470215,
|
75 |
+
"eval_runtime": 912.9746,
|
76 |
+
"eval_samples_per_second": 0.321,
|
77 |
+
"eval_steps_per_second": 0.041,
|
78 |
+
"step": 500
|
79 |
+
},
|
80 |
+
{
|
81 |
+
"epoch": 0.51,
|
82 |
+
"learning_rate": 1.9488926746166953e-05,
|
83 |
+
"loss": 4.6993,
|
84 |
+
"step": 600
|
85 |
+
},
|
86 |
+
{
|
87 |
+
"epoch": 0.51,
|
88 |
+
"eval_loss": 4.662974834442139,
|
89 |
+
"eval_runtime": 908.9893,
|
90 |
+
"eval_samples_per_second": 0.322,
|
91 |
+
"eval_steps_per_second": 0.041,
|
92 |
+
"step": 600
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 0.6,
|
96 |
+
"learning_rate": 1.940374787052811e-05,
|
97 |
+
"loss": 4.7104,
|
98 |
+
"step": 700
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 0.6,
|
102 |
+
"eval_loss": 4.587018966674805,
|
103 |
+
"eval_runtime": 913.6781,
|
104 |
+
"eval_samples_per_second": 0.321,
|
105 |
+
"eval_steps_per_second": 0.04,
|
106 |
+
"step": 700
|
107 |
+
},
|
108 |
+
{
|
109 |
+
"epoch": 0.68,
|
110 |
+
"learning_rate": 1.9318568994889268e-05,
|
111 |
+
"loss": 4.5864,
|
112 |
+
"step": 800
|
113 |
+
},
|
114 |
+
{
|
115 |
+
"epoch": 0.68,
|
116 |
+
"eval_loss": 4.565457344055176,
|
117 |
+
"eval_runtime": 900.1349,
|
118 |
+
"eval_samples_per_second": 0.326,
|
119 |
+
"eval_steps_per_second": 0.041,
|
120 |
+
"step": 800
|
121 |
+
},
|
122 |
+
{
|
123 |
+
"epoch": 0.77,
|
124 |
+
"learning_rate": 1.923339011925043e-05,
|
125 |
+
"loss": 4.6003,
|
126 |
+
"step": 900
|
127 |
+
},
|
128 |
+
{
|
129 |
+
"epoch": 0.77,
|
130 |
+
"eval_loss": 4.560632705688477,
|
131 |
+
"eval_runtime": 913.1928,
|
132 |
+
"eval_samples_per_second": 0.321,
|
133 |
+
"eval_steps_per_second": 0.041,
|
134 |
+
"step": 900
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 0.85,
|
138 |
+
"learning_rate": 1.9148211243611586e-05,
|
139 |
+
"loss": 4.6166,
|
140 |
+
"step": 1000
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 0.85,
|
144 |
+
"eval_loss": 4.5305914878845215,
|
145 |
+
"eval_runtime": 898.0505,
|
146 |
+
"eval_samples_per_second": 0.326,
|
147 |
+
"eval_steps_per_second": 0.041,
|
148 |
+
"step": 1000
|
149 |
+
},
|
150 |
+
{
|
151 |
+
"epoch": 0.94,
|
152 |
+
"learning_rate": 1.9063032367972743e-05,
|
153 |
+
"loss": 4.6019,
|
154 |
+
"step": 1100
|
155 |
+
},
|
156 |
+
{
|
157 |
+
"epoch": 0.94,
|
158 |
+
"eval_loss": 4.547730922698975,
|
159 |
+
"eval_runtime": 916.6743,
|
160 |
+
"eval_samples_per_second": 0.32,
|
161 |
+
"eval_steps_per_second": 0.04,
|
162 |
+
"step": 1100
|
163 |
+
},
|
164 |
+
{
|
165 |
+
"epoch": 1.02,
|
166 |
+
"learning_rate": 1.89778534923339e-05,
|
167 |
+
"loss": 4.4833,
|
168 |
+
"step": 1200
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"epoch": 1.02,
|
172 |
+
"eval_loss": 4.527022838592529,
|
173 |
+
"eval_runtime": 905.1447,
|
174 |
+
"eval_samples_per_second": 0.324,
|
175 |
+
"eval_steps_per_second": 0.041,
|
176 |
+
"step": 1200
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 1.11,
|
180 |
+
"learning_rate": 1.889267461669506e-05,
|
181 |
+
"loss": 4.4664,
|
182 |
+
"step": 1300
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 1.11,
|
186 |
+
"eval_loss": 4.533600807189941,
|
187 |
+
"eval_runtime": 900.1154,
|
188 |
+
"eval_samples_per_second": 0.326,
|
189 |
+
"eval_steps_per_second": 0.041,
|
190 |
+
"step": 1300
|
191 |
+
},
|
192 |
+
{
|
193 |
+
"epoch": 1.19,
|
194 |
+
"learning_rate": 1.880749574105622e-05,
|
195 |
+
"loss": 4.4533,
|
196 |
+
"step": 1400
|
197 |
+
},
|
198 |
+
{
|
199 |
+
"epoch": 1.19,
|
200 |
+
"eval_loss": 4.512392044067383,
|
201 |
+
"eval_runtime": 897.9913,
|
202 |
+
"eval_samples_per_second": 0.326,
|
203 |
+
"eval_steps_per_second": 0.041,
|
204 |
+
"step": 1400
|
205 |
+
},
|
206 |
+
{
|
207 |
+
"epoch": 1.28,
|
208 |
+
"learning_rate": 1.872231686541738e-05,
|
209 |
+
"loss": 4.2969,
|
210 |
+
"step": 1500
|
211 |
+
},
|
212 |
+
{
|
213 |
+
"epoch": 1.28,
|
214 |
+
"eval_loss": 4.5288238525390625,
|
215 |
+
"eval_runtime": 902.8396,
|
216 |
+
"eval_samples_per_second": 0.325,
|
217 |
+
"eval_steps_per_second": 0.041,
|
218 |
+
"step": 1500
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 1.36,
|
222 |
+
"learning_rate": 1.8637137989778534e-05,
|
223 |
+
"loss": 4.4203,
|
224 |
+
"step": 1600
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 1.36,
|
228 |
+
"eval_loss": 4.510149002075195,
|
229 |
+
"eval_runtime": 896.5776,
|
230 |
+
"eval_samples_per_second": 0.327,
|
231 |
+
"eval_steps_per_second": 0.041,
|
232 |
+
"step": 1600
|
233 |
+
},
|
234 |
+
{
|
235 |
+
"epoch": 1.45,
|
236 |
+
"learning_rate": 1.8551959114139694e-05,
|
237 |
+
"loss": 4.4828,
|
238 |
+
"step": 1700
|
239 |
+
},
|
240 |
+
{
|
241 |
+
"epoch": 1.45,
|
242 |
+
"eval_loss": 4.4747633934021,
|
243 |
+
"eval_runtime": 900.0595,
|
244 |
+
"eval_samples_per_second": 0.326,
|
245 |
+
"eval_steps_per_second": 0.041,
|
246 |
+
"step": 1700
|
247 |
+
},
|
248 |
+
{
|
249 |
+
"epoch": 1.53,
|
250 |
+
"learning_rate": 1.8466780238500855e-05,
|
251 |
+
"loss": 4.5045,
|
252 |
+
"step": 1800
|
253 |
+
},
|
254 |
+
{
|
255 |
+
"epoch": 1.53,
|
256 |
+
"eval_loss": 4.477636337280273,
|
257 |
+
"eval_runtime": 930.3322,
|
258 |
+
"eval_samples_per_second": 0.315,
|
259 |
+
"eval_steps_per_second": 0.04,
|
260 |
+
"step": 1800
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 1.62,
|
264 |
+
"learning_rate": 1.8381601362862013e-05,
|
265 |
+
"loss": 4.2977,
|
266 |
+
"step": 1900
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 1.62,
|
270 |
+
"eval_loss": 4.475375175476074,
|
271 |
+
"eval_runtime": 897.0451,
|
272 |
+
"eval_samples_per_second": 0.327,
|
273 |
+
"eval_steps_per_second": 0.041,
|
274 |
+
"step": 1900
|
275 |
+
},
|
276 |
+
{
|
277 |
+
"epoch": 1.7,
|
278 |
+
"learning_rate": 1.829642248722317e-05,
|
279 |
+
"loss": 4.4022,
|
280 |
+
"step": 2000
|
281 |
+
},
|
282 |
+
{
|
283 |
+
"epoch": 1.7,
|
284 |
+
"eval_loss": 4.451622486114502,
|
285 |
+
"eval_runtime": 924.228,
|
286 |
+
"eval_samples_per_second": 0.317,
|
287 |
+
"eval_steps_per_second": 0.04,
|
288 |
+
"step": 2000
|
289 |
+
},
|
290 |
+
{
|
291 |
+
"epoch": 1.79,
|
292 |
+
"learning_rate": 1.8211243611584328e-05,
|
293 |
+
"loss": 4.3429,
|
294 |
+
"step": 2100
|
295 |
+
},
|
296 |
+
{
|
297 |
+
"epoch": 1.79,
|
298 |
+
"eval_loss": 4.43123722076416,
|
299 |
+
"eval_runtime": 903.4516,
|
300 |
+
"eval_samples_per_second": 0.324,
|
301 |
+
"eval_steps_per_second": 0.041,
|
302 |
+
"step": 2100
|
303 |
+
},
|
304 |
+
{
|
305 |
+
"epoch": 1.87,
|
306 |
+
"learning_rate": 1.812606473594549e-05,
|
307 |
+
"loss": 4.3748,
|
308 |
+
"step": 2200
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 1.87,
|
312 |
+
"eval_loss": 4.475191116333008,
|
313 |
+
"eval_runtime": 901.3458,
|
314 |
+
"eval_samples_per_second": 0.325,
|
315 |
+
"eval_steps_per_second": 0.041,
|
316 |
+
"step": 2200
|
317 |
+
},
|
318 |
+
{
|
319 |
+
"epoch": 1.96,
|
320 |
+
"learning_rate": 1.8040885860306646e-05,
|
321 |
+
"loss": 4.343,
|
322 |
+
"step": 2300
|
323 |
+
},
|
324 |
+
{
|
325 |
+
"epoch": 1.96,
|
326 |
+
"eval_loss": 4.455069541931152,
|
327 |
+
"eval_runtime": 919.5227,
|
328 |
+
"eval_samples_per_second": 0.319,
|
329 |
+
"eval_steps_per_second": 0.04,
|
330 |
+
"step": 2300
|
331 |
+
},
|
332 |
+
{
|
333 |
+
"epoch": 2.04,
|
334 |
+
"learning_rate": 1.7955706984667803e-05,
|
335 |
+
"loss": 4.1796,
|
336 |
+
"step": 2400
|
337 |
+
},
|
338 |
+
{
|
339 |
+
"epoch": 2.04,
|
340 |
+
"eval_loss": 4.424516201019287,
|
341 |
+
"eval_runtime": 912.0297,
|
342 |
+
"eval_samples_per_second": 0.321,
|
343 |
+
"eval_steps_per_second": 0.041,
|
344 |
+
"step": 2400
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 2.13,
|
348 |
+
"learning_rate": 1.787052810902896e-05,
|
349 |
+
"loss": 4.2325,
|
350 |
+
"step": 2500
|
351 |
+
},
|
352 |
+
{
|
353 |
+
"epoch": 2.13,
|
354 |
+
"eval_loss": 4.425054550170898,
|
355 |
+
"eval_runtime": 900.4656,
|
356 |
+
"eval_samples_per_second": 0.325,
|
357 |
+
"eval_steps_per_second": 0.041,
|
358 |
+
"step": 2500
|
359 |
+
},
|
360 |
+
{
|
361 |
+
"epoch": 2.21,
|
362 |
+
"learning_rate": 1.778534923339012e-05,
|
363 |
+
"loss": 4.2419,
|
364 |
+
"step": 2600
|
365 |
+
},
|
366 |
+
{
|
367 |
+
"epoch": 2.21,
|
368 |
+
"eval_loss": 4.426344871520996,
|
369 |
+
"eval_runtime": 907.7136,
|
370 |
+
"eval_samples_per_second": 0.323,
|
371 |
+
"eval_steps_per_second": 0.041,
|
372 |
+
"step": 2600
|
373 |
+
},
|
374 |
+
{
|
375 |
+
"epoch": 2.3,
|
376 |
+
"learning_rate": 1.770017035775128e-05,
|
377 |
+
"loss": 4.2556,
|
378 |
+
"step": 2700
|
379 |
+
},
|
380 |
+
{
|
381 |
+
"epoch": 2.3,
|
382 |
+
"eval_loss": 4.392487525939941,
|
383 |
+
"eval_runtime": 904.6208,
|
384 |
+
"eval_samples_per_second": 0.324,
|
385 |
+
"eval_steps_per_second": 0.041,
|
386 |
+
"step": 2700
|
387 |
+
},
|
388 |
+
{
|
389 |
+
"epoch": 2.39,
|
390 |
+
"learning_rate": 1.761499148211244e-05,
|
391 |
+
"loss": 4.2501,
|
392 |
+
"step": 2800
|
393 |
+
},
|
394 |
+
{
|
395 |
+
"epoch": 2.39,
|
396 |
+
"eval_loss": 4.414584636688232,
|
397 |
+
"eval_runtime": 904.1877,
|
398 |
+
"eval_samples_per_second": 0.324,
|
399 |
+
"eval_steps_per_second": 0.041,
|
400 |
+
"step": 2800
|
401 |
+
},
|
402 |
+
{
|
403 |
+
"epoch": 2.47,
|
404 |
+
"learning_rate": 1.7529812606473594e-05,
|
405 |
+
"loss": 4.2199,
|
406 |
+
"step": 2900
|
407 |
+
},
|
408 |
+
{
|
409 |
+
"epoch": 2.47,
|
410 |
+
"eval_loss": 4.391025543212891,
|
411 |
+
"eval_runtime": 904.8782,
|
412 |
+
"eval_samples_per_second": 0.324,
|
413 |
+
"eval_steps_per_second": 0.041,
|
414 |
+
"step": 2900
|
415 |
+
},
|
416 |
+
{
|
417 |
+
"epoch": 2.56,
|
418 |
+
"learning_rate": 1.7444633730834754e-05,
|
419 |
+
"loss": 4.1842,
|
420 |
+
"step": 3000
|
421 |
+
},
|
422 |
+
{
|
423 |
+
"epoch": 2.56,
|
424 |
+
"eval_loss": 4.3928704261779785,
|
425 |
+
"eval_runtime": 911.7639,
|
426 |
+
"eval_samples_per_second": 0.321,
|
427 |
+
"eval_steps_per_second": 0.041,
|
428 |
+
"step": 3000
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 2.64,
|
432 |
+
"learning_rate": 1.7359454855195912e-05,
|
433 |
+
"loss": 4.1789,
|
434 |
+
"step": 3100
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 2.64,
|
438 |
+
"eval_loss": 4.373032569885254,
|
439 |
+
"eval_runtime": 903.6693,
|
440 |
+
"eval_samples_per_second": 0.324,
|
441 |
+
"eval_steps_per_second": 0.041,
|
442 |
+
"step": 3100
|
443 |
+
},
|
444 |
+
{
|
445 |
+
"epoch": 2.73,
|
446 |
+
"learning_rate": 1.7274275979557073e-05,
|
447 |
+
"loss": 4.1993,
|
448 |
+
"step": 3200
|
449 |
+
},
|
450 |
+
{
|
451 |
+
"epoch": 2.73,
|
452 |
+
"eval_loss": 4.401707172393799,
|
453 |
+
"eval_runtime": 906.6931,
|
454 |
+
"eval_samples_per_second": 0.323,
|
455 |
+
"eval_steps_per_second": 0.041,
|
456 |
+
"step": 3200
|
457 |
+
},
|
458 |
+
{
|
459 |
+
"epoch": 2.81,
|
460 |
+
"learning_rate": 1.718909710391823e-05,
|
461 |
+
"loss": 4.1348,
|
462 |
+
"step": 3300
|
463 |
+
},
|
464 |
+
{
|
465 |
+
"epoch": 2.81,
|
466 |
+
"eval_loss": 4.366961479187012,
|
467 |
+
"eval_runtime": 909.5878,
|
468 |
+
"eval_samples_per_second": 0.322,
|
469 |
+
"eval_steps_per_second": 0.041,
|
470 |
+
"step": 3300
|
471 |
+
},
|
472 |
+
{
|
473 |
+
"epoch": 2.9,
|
474 |
+
"learning_rate": 1.7103918228279387e-05,
|
475 |
+
"loss": 4.196,
|
476 |
+
"step": 3400
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 2.9,
|
480 |
+
"eval_loss": 4.354998588562012,
|
481 |
+
"eval_runtime": 914.5578,
|
482 |
+
"eval_samples_per_second": 0.32,
|
483 |
+
"eval_steps_per_second": 0.04,
|
484 |
+
"step": 3400
|
485 |
+
},
|
486 |
+
{
|
487 |
+
"epoch": 2.98,
|
488 |
+
"learning_rate": 1.7018739352640548e-05,
|
489 |
+
"loss": 4.1906,
|
490 |
+
"step": 3500
|
491 |
+
},
|
492 |
+
{
|
493 |
+
"epoch": 2.98,
|
494 |
+
"eval_loss": 4.349732875823975,
|
495 |
+
"eval_runtime": 898.3438,
|
496 |
+
"eval_samples_per_second": 0.326,
|
497 |
+
"eval_steps_per_second": 0.041,
|
498 |
+
"step": 3500
|
499 |
+
},
|
500 |
+
{
|
501 |
+
"epoch": 3.07,
|
502 |
+
"learning_rate": 1.6933560477001706e-05,
|
503 |
+
"loss": 4.0583,
|
504 |
+
"step": 3600
|
505 |
+
},
|
506 |
+
{
|
507 |
+
"epoch": 3.07,
|
508 |
+
"eval_loss": 4.3627471923828125,
|
509 |
+
"eval_runtime": 897.909,
|
510 |
+
"eval_samples_per_second": 0.326,
|
511 |
+
"eval_steps_per_second": 0.041,
|
512 |
+
"step": 3600
|
513 |
+
},
|
514 |
+
{
|
515 |
+
"epoch": 3.15,
|
516 |
+
"learning_rate": 1.6848381601362863e-05,
|
517 |
+
"loss": 4.1298,
|
518 |
+
"step": 3700
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 3.15,
|
522 |
+
"eval_loss": 4.362400531768799,
|
523 |
+
"eval_runtime": 912.2819,
|
524 |
+
"eval_samples_per_second": 0.321,
|
525 |
+
"eval_steps_per_second": 0.041,
|
526 |
+
"step": 3700
|
527 |
+
},
|
528 |
+
{
|
529 |
+
"epoch": 3.24,
|
530 |
+
"learning_rate": 1.676320272572402e-05,
|
531 |
+
"loss": 4.0244,
|
532 |
+
"step": 3800
|
533 |
+
},
|
534 |
+
{
|
535 |
+
"epoch": 3.24,
|
536 |
+
"eval_loss": 4.338736534118652,
|
537 |
+
"eval_runtime": 896.0396,
|
538 |
+
"eval_samples_per_second": 0.327,
|
539 |
+
"eval_steps_per_second": 0.041,
|
540 |
+
"step": 3800
|
541 |
+
},
|
542 |
+
{
|
543 |
+
"epoch": 3.32,
|
544 |
+
"learning_rate": 1.667802385008518e-05,
|
545 |
+
"loss": 4.0549,
|
546 |
+
"step": 3900
|
547 |
+
},
|
548 |
+
{
|
549 |
+
"epoch": 3.32,
|
550 |
+
"eval_loss": 4.341971397399902,
|
551 |
+
"eval_runtime": 901.737,
|
552 |
+
"eval_samples_per_second": 0.325,
|
553 |
+
"eval_steps_per_second": 0.041,
|
554 |
+
"step": 3900
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 3.41,
|
558 |
+
"learning_rate": 1.659284497444634e-05,
|
559 |
+
"loss": 4.0006,
|
560 |
+
"step": 4000
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 3.41,
|
564 |
+
"eval_loss": 4.326377868652344,
|
565 |
+
"eval_runtime": 898.1346,
|
566 |
+
"eval_samples_per_second": 0.326,
|
567 |
+
"eval_steps_per_second": 0.041,
|
568 |
+
"step": 4000
|
569 |
+
},
|
570 |
+
{
|
571 |
+
"epoch": 3.49,
|
572 |
+
"learning_rate": 1.65076660988075e-05,
|
573 |
+
"loss": 4.0435,
|
574 |
+
"step": 4100
|
575 |
+
},
|
576 |
+
{
|
577 |
+
"epoch": 3.49,
|
578 |
+
"eval_loss": 4.331975936889648,
|
579 |
+
"eval_runtime": 901.1995,
|
580 |
+
"eval_samples_per_second": 0.325,
|
581 |
+
"eval_steps_per_second": 0.041,
|
582 |
+
"step": 4100
|
583 |
+
},
|
584 |
+
{
|
585 |
+
"epoch": 3.58,
|
586 |
+
"learning_rate": 1.6422487223168653e-05,
|
587 |
+
"loss": 4.0948,
|
588 |
+
"step": 4200
|
589 |
+
},
|
590 |
+
{
|
591 |
+
"epoch": 3.58,
|
592 |
+
"eval_loss": 4.32388973236084,
|
593 |
+
"eval_runtime": 904.6488,
|
594 |
+
"eval_samples_per_second": 0.324,
|
595 |
+
"eval_steps_per_second": 0.041,
|
596 |
+
"step": 4200
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 3.66,
|
600 |
+
"learning_rate": 1.6337308347529814e-05,
|
601 |
+
"loss": 3.933,
|
602 |
+
"step": 4300
|
603 |
+
},
|
604 |
+
{
|
605 |
+
"epoch": 3.66,
|
606 |
+
"eval_loss": 4.314174175262451,
|
607 |
+
"eval_runtime": 902.3829,
|
608 |
+
"eval_samples_per_second": 0.325,
|
609 |
+
"eval_steps_per_second": 0.041,
|
610 |
+
"step": 4300
|
611 |
+
},
|
612 |
+
{
|
613 |
+
"epoch": 3.75,
|
614 |
+
"learning_rate": 1.625212947189097e-05,
|
615 |
+
"loss": 3.9773,
|
616 |
+
"step": 4400
|
617 |
+
},
|
618 |
+
{
|
619 |
+
"epoch": 3.75,
|
620 |
+
"eval_loss": 4.353682041168213,
|
621 |
+
"eval_runtime": 901.0754,
|
622 |
+
"eval_samples_per_second": 0.325,
|
623 |
+
"eval_steps_per_second": 0.041,
|
624 |
+
"step": 4400
|
625 |
+
},
|
626 |
+
{
|
627 |
+
"epoch": 3.83,
|
628 |
+
"learning_rate": 1.6166950596252132e-05,
|
629 |
+
"loss": 4.1143,
|
630 |
+
"step": 4500
|
631 |
+
},
|
632 |
+
{
|
633 |
+
"epoch": 3.83,
|
634 |
+
"eval_loss": 4.3104963302612305,
|
635 |
+
"eval_runtime": 903.7164,
|
636 |
+
"eval_samples_per_second": 0.324,
|
637 |
+
"eval_steps_per_second": 0.041,
|
638 |
+
"step": 4500
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 3.92,
|
642 |
+
"learning_rate": 1.608177172061329e-05,
|
643 |
+
"loss": 3.961,
|
644 |
+
"step": 4600
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 3.92,
|
648 |
+
"eval_loss": 4.292612552642822,
|
649 |
+
"eval_runtime": 900.6329,
|
650 |
+
"eval_samples_per_second": 0.325,
|
651 |
+
"eval_steps_per_second": 0.041,
|
652 |
+
"step": 4600
|
653 |
+
},
|
654 |
+
{
|
655 |
+
"epoch": 4.0,
|
656 |
+
"learning_rate": 1.5996592844974447e-05,
|
657 |
+
"loss": 4.0047,
|
658 |
+
"step": 4700
|
659 |
+
},
|
660 |
+
{
|
661 |
+
"epoch": 4.0,
|
662 |
+
"eval_loss": 4.3063883781433105,
|
663 |
+
"eval_runtime": 900.2789,
|
664 |
+
"eval_samples_per_second": 0.325,
|
665 |
+
"eval_steps_per_second": 0.041,
|
666 |
+
"step": 4700
|
667 |
+
},
|
668 |
+
{
|
669 |
+
"epoch": 4.09,
|
670 |
+
"learning_rate": 1.5911413969335605e-05,
|
671 |
+
"loss": 3.9058,
|
672 |
+
"step": 4800
|
673 |
+
},
|
674 |
+
{
|
675 |
+
"epoch": 4.09,
|
676 |
+
"eval_loss": 4.303318023681641,
|
677 |
+
"eval_runtime": 906.0598,
|
678 |
+
"eval_samples_per_second": 0.323,
|
679 |
+
"eval_steps_per_second": 0.041,
|
680 |
+
"step": 4800
|
681 |
+
},
|
682 |
+
{
|
683 |
+
"epoch": 4.17,
|
684 |
+
"learning_rate": 1.5826235093696765e-05,
|
685 |
+
"loss": 3.7936,
|
686 |
+
"step": 4900
|
687 |
+
},
|
688 |
+
{
|
689 |
+
"epoch": 4.17,
|
690 |
+
"eval_loss": 4.301506996154785,
|
691 |
+
"eval_runtime": 906.1238,
|
692 |
+
"eval_samples_per_second": 0.323,
|
693 |
+
"eval_steps_per_second": 0.041,
|
694 |
+
"step": 4900
|
695 |
+
},
|
696 |
+
{
|
697 |
+
"epoch": 4.26,
|
698 |
+
"learning_rate": 1.5741056218057923e-05,
|
699 |
+
"loss": 3.8733,
|
700 |
+
"step": 5000
|
701 |
+
},
|
702 |
+
{
|
703 |
+
"epoch": 4.26,
|
704 |
+
"eval_loss": 4.344919681549072,
|
705 |
+
"eval_runtime": 916.0111,
|
706 |
+
"eval_samples_per_second": 0.32,
|
707 |
+
"eval_steps_per_second": 0.04,
|
708 |
+
"step": 5000
|
709 |
+
},
|
710 |
+
{
|
711 |
+
"epoch": 4.34,
|
712 |
+
"learning_rate": 1.565587734241908e-05,
|
713 |
+
"loss": 3.8625,
|
714 |
+
"step": 5100
|
715 |
+
},
|
716 |
+
{
|
717 |
+
"epoch": 4.34,
|
718 |
+
"eval_loss": 4.274160385131836,
|
719 |
+
"eval_runtime": 906.5215,
|
720 |
+
"eval_samples_per_second": 0.323,
|
721 |
+
"eval_steps_per_second": 0.041,
|
722 |
+
"step": 5100
|
723 |
+
},
|
724 |
+
{
|
725 |
+
"epoch": 4.43,
|
726 |
+
"learning_rate": 1.557069846678024e-05,
|
727 |
+
"loss": 4.0015,
|
728 |
+
"step": 5200
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 4.43,
|
732 |
+
"eval_loss": 4.2957258224487305,
|
733 |
+
"eval_runtime": 920.9763,
|
734 |
+
"eval_samples_per_second": 0.318,
|
735 |
+
"eval_steps_per_second": 0.04,
|
736 |
+
"step": 5200
|
737 |
+
},
|
738 |
+
{
|
739 |
+
"epoch": 4.51,
|
740 |
+
"learning_rate": 1.54855195911414e-05,
|
741 |
+
"loss": 4.0173,
|
742 |
+
"step": 5300
|
743 |
+
},
|
744 |
+
{
|
745 |
+
"epoch": 4.51,
|
746 |
+
"eval_loss": 4.30204439163208,
|
747 |
+
"eval_runtime": 900.2302,
|
748 |
+
"eval_samples_per_second": 0.325,
|
749 |
+
"eval_steps_per_second": 0.041,
|
750 |
+
"step": 5300
|
751 |
+
},
|
752 |
+
{
|
753 |
+
"epoch": 4.6,
|
754 |
+
"learning_rate": 1.5400340715502556e-05,
|
755 |
+
"loss": 3.8079,
|
756 |
+
"step": 5400
|
757 |
+
},
|
758 |
+
{
|
759 |
+
"epoch": 4.6,
|
760 |
+
"eval_loss": 4.274052619934082,
|
761 |
+
"eval_runtime": 902.0347,
|
762 |
+
"eval_samples_per_second": 0.325,
|
763 |
+
"eval_steps_per_second": 0.041,
|
764 |
+
"step": 5400
|
765 |
+
},
|
766 |
+
{
|
767 |
+
"epoch": 4.68,
|
768 |
+
"learning_rate": 1.5315161839863713e-05,
|
769 |
+
"loss": 4.0081,
|
770 |
+
"step": 5500
|
771 |
+
},
|
772 |
+
{
|
773 |
+
"epoch": 4.68,
|
774 |
+
"eval_loss": 4.2950968742370605,
|
775 |
+
"eval_runtime": 910.5366,
|
776 |
+
"eval_samples_per_second": 0.322,
|
777 |
+
"eval_steps_per_second": 0.041,
|
778 |
+
"step": 5500
|
779 |
+
},
|
780 |
+
{
|
781 |
+
"epoch": 4.77,
|
782 |
+
"learning_rate": 1.5229982964224874e-05,
|
783 |
+
"loss": 3.8823,
|
784 |
+
"step": 5600
|
785 |
+
},
|
786 |
+
{
|
787 |
+
"epoch": 4.77,
|
788 |
+
"eval_loss": 4.277224540710449,
|
789 |
+
"eval_runtime": 901.0907,
|
790 |
+
"eval_samples_per_second": 0.325,
|
791 |
+
"eval_steps_per_second": 0.041,
|
792 |
+
"step": 5600
|
793 |
+
},
|
794 |
+
{
|
795 |
+
"epoch": 4.86,
|
796 |
+
"learning_rate": 1.5144804088586031e-05,
|
797 |
+
"loss": 3.9056,
|
798 |
+
"step": 5700
|
799 |
+
},
|
800 |
+
{
|
801 |
+
"epoch": 4.86,
|
802 |
+
"eval_loss": 4.272303581237793,
|
803 |
+
"eval_runtime": 911.2022,
|
804 |
+
"eval_samples_per_second": 0.322,
|
805 |
+
"eval_steps_per_second": 0.041,
|
806 |
+
"step": 5700
|
807 |
+
},
|
808 |
+
{
|
809 |
+
"epoch": 4.94,
|
810 |
+
"learning_rate": 1.505962521294719e-05,
|
811 |
+
"loss": 3.813,
|
812 |
+
"step": 5800
|
813 |
+
},
|
814 |
+
{
|
815 |
+
"epoch": 4.94,
|
816 |
+
"eval_loss": 4.275490760803223,
|
817 |
+
"eval_runtime": 913.2487,
|
818 |
+
"eval_samples_per_second": 0.321,
|
819 |
+
"eval_steps_per_second": 0.041,
|
820 |
+
"step": 5800
|
821 |
+
},
|
822 |
+
{
|
823 |
+
"epoch": 5.03,
|
824 |
+
"learning_rate": 1.4974446337308348e-05,
|
825 |
+
"loss": 3.9263,
|
826 |
+
"step": 5900
|
827 |
+
},
|
828 |
+
{
|
829 |
+
"epoch": 5.03,
|
830 |
+
"eval_loss": 4.2567267417907715,
|
831 |
+
"eval_runtime": 922.3278,
|
832 |
+
"eval_samples_per_second": 0.318,
|
833 |
+
"eval_steps_per_second": 0.04,
|
834 |
+
"step": 5900
|
835 |
+
},
|
836 |
+
{
|
837 |
+
"epoch": 5.11,
|
838 |
+
"learning_rate": 1.4889267461669507e-05,
|
839 |
+
"loss": 3.7405,
|
840 |
+
"step": 6000
|
841 |
+
},
|
842 |
+
{
|
843 |
+
"epoch": 5.11,
|
844 |
+
"eval_loss": 4.279594421386719,
|
845 |
+
"eval_runtime": 902.112,
|
846 |
+
"eval_samples_per_second": 0.325,
|
847 |
+
"eval_steps_per_second": 0.041,
|
848 |
+
"step": 6000
|
849 |
+
},
|
850 |
+
{
|
851 |
+
"epoch": 5.2,
|
852 |
+
"learning_rate": 1.4804088586030664e-05,
|
853 |
+
"loss": 3.8432,
|
854 |
+
"step": 6100
|
855 |
+
},
|
856 |
+
{
|
857 |
+
"epoch": 5.2,
|
858 |
+
"eval_loss": 4.277362823486328,
|
859 |
+
"eval_runtime": 909.7644,
|
860 |
+
"eval_samples_per_second": 0.322,
|
861 |
+
"eval_steps_per_second": 0.041,
|
862 |
+
"step": 6100
|
863 |
+
},
|
864 |
+
{
|
865 |
+
"epoch": 5.28,
|
866 |
+
"learning_rate": 1.4718909710391824e-05,
|
867 |
+
"loss": 3.797,
|
868 |
+
"step": 6200
|
869 |
+
},
|
870 |
+
{
|
871 |
+
"epoch": 5.28,
|
872 |
+
"eval_loss": 4.2573161125183105,
|
873 |
+
"eval_runtime": 900.2164,
|
874 |
+
"eval_samples_per_second": 0.325,
|
875 |
+
"eval_steps_per_second": 0.041,
|
876 |
+
"step": 6200
|
877 |
+
},
|
878 |
+
{
|
879 |
+
"epoch": 5.37,
|
880 |
+
"learning_rate": 1.4633730834752981e-05,
|
881 |
+
"loss": 3.7348,
|
882 |
+
"step": 6300
|
883 |
+
},
|
884 |
+
{
|
885 |
+
"epoch": 5.37,
|
886 |
+
"eval_loss": 4.261054515838623,
|
887 |
+
"eval_runtime": 899.5894,
|
888 |
+
"eval_samples_per_second": 0.326,
|
889 |
+
"eval_steps_per_second": 0.041,
|
890 |
+
"step": 6300
|
891 |
+
},
|
892 |
+
{
|
893 |
+
"epoch": 5.45,
|
894 |
+
"learning_rate": 1.4548551959114142e-05,
|
895 |
+
"loss": 3.7363,
|
896 |
+
"step": 6400
|
897 |
+
},
|
898 |
+
{
|
899 |
+
"epoch": 5.45,
|
900 |
+
"eval_loss": 4.2909698486328125,
|
901 |
+
"eval_runtime": 914.0387,
|
902 |
+
"eval_samples_per_second": 0.321,
|
903 |
+
"eval_steps_per_second": 0.04,
|
904 |
+
"step": 6400
|
905 |
+
},
|
906 |
+
{
|
907 |
+
"epoch": 5.54,
|
908 |
+
"learning_rate": 1.44633730834753e-05,
|
909 |
+
"loss": 3.8448,
|
910 |
+
"step": 6500
|
911 |
+
},
|
912 |
+
{
|
913 |
+
"epoch": 5.54,
|
914 |
+
"eval_loss": 4.257021903991699,
|
915 |
+
"eval_runtime": 906.6331,
|
916 |
+
"eval_samples_per_second": 0.323,
|
917 |
+
"eval_steps_per_second": 0.041,
|
918 |
+
"step": 6500
|
919 |
+
},
|
920 |
+
{
|
921 |
+
"epoch": 5.62,
|
922 |
+
"learning_rate": 1.4378194207836458e-05,
|
923 |
+
"loss": 3.8023,
|
924 |
+
"step": 6600
|
925 |
+
},
|
926 |
+
{
|
927 |
+
"epoch": 5.62,
|
928 |
+
"eval_loss": 4.263571262359619,
|
929 |
+
"eval_runtime": 903.3527,
|
930 |
+
"eval_samples_per_second": 0.324,
|
931 |
+
"eval_steps_per_second": 0.041,
|
932 |
+
"step": 6600
|
933 |
+
},
|
934 |
+
{
|
935 |
+
"epoch": 5.71,
|
936 |
+
"learning_rate": 1.4293015332197616e-05,
|
937 |
+
"loss": 3.8202,
|
938 |
+
"step": 6700
|
939 |
+
},
|
940 |
+
{
|
941 |
+
"epoch": 5.71,
|
942 |
+
"eval_loss": 4.226749897003174,
|
943 |
+
"eval_runtime": 913.4737,
|
944 |
+
"eval_samples_per_second": 0.321,
|
945 |
+
"eval_steps_per_second": 0.041,
|
946 |
+
"step": 6700
|
947 |
+
},
|
948 |
+
{
|
949 |
+
"epoch": 5.79,
|
950 |
+
"learning_rate": 1.4207836456558775e-05,
|
951 |
+
"loss": 3.731,
|
952 |
+
"step": 6800
|
953 |
+
},
|
954 |
+
{
|
955 |
+
"epoch": 5.79,
|
956 |
+
"eval_loss": 4.232919216156006,
|
957 |
+
"eval_runtime": 905.4134,
|
958 |
+
"eval_samples_per_second": 0.324,
|
959 |
+
"eval_steps_per_second": 0.041,
|
960 |
+
"step": 6800
|
961 |
+
},
|
962 |
+
{
|
963 |
+
"epoch": 5.88,
|
964 |
+
"learning_rate": 1.4122657580919934e-05,
|
965 |
+
"loss": 3.7507,
|
966 |
+
"step": 6900
|
967 |
+
},
|
968 |
+
{
|
969 |
+
"epoch": 5.88,
|
970 |
+
"eval_loss": 4.2412004470825195,
|
971 |
+
"eval_runtime": 917.4159,
|
972 |
+
"eval_samples_per_second": 0.319,
|
973 |
+
"eval_steps_per_second": 0.04,
|
974 |
+
"step": 6900
|
975 |
+
},
|
976 |
+
{
|
977 |
+
"epoch": 5.96,
|
978 |
+
"learning_rate": 1.4037478705281091e-05,
|
979 |
+
"loss": 3.8812,
|
980 |
+
"step": 7000
|
981 |
+
},
|
982 |
+
{
|
983 |
+
"epoch": 5.96,
|
984 |
+
"eval_loss": 4.239901542663574,
|
985 |
+
"eval_runtime": 904.0296,
|
986 |
+
"eval_samples_per_second": 0.324,
|
987 |
+
"eval_steps_per_second": 0.041,
|
988 |
+
"step": 7000
|
989 |
+
},
|
990 |
+
{
|
991 |
+
"epoch": 6.05,
|
992 |
+
"learning_rate": 1.395229982964225e-05,
|
993 |
+
"loss": 3.6882,
|
994 |
+
"step": 7100
|
995 |
+
},
|
996 |
+
{
|
997 |
+
"epoch": 6.05,
|
998 |
+
"eval_loss": 4.252227306365967,
|
999 |
+
"eval_runtime": 909.685,
|
1000 |
+
"eval_samples_per_second": 0.322,
|
1001 |
+
"eval_steps_per_second": 0.041,
|
1002 |
+
"step": 7100
|
1003 |
+
},
|
1004 |
+
{
|
1005 |
+
"epoch": 6.13,
|
1006 |
+
"learning_rate": 1.3867120954003408e-05,
|
1007 |
+
"loss": 3.6485,
|
1008 |
+
"step": 7200
|
1009 |
+
},
|
1010 |
+
{
|
1011 |
+
"epoch": 6.13,
|
1012 |
+
"eval_loss": 4.240085601806641,
|
1013 |
+
"eval_runtime": 917.8845,
|
1014 |
+
"eval_samples_per_second": 0.319,
|
1015 |
+
"eval_steps_per_second": 0.04,
|
1016 |
+
"step": 7200
|
1017 |
+
},
|
1018 |
+
{
|
1019 |
+
"epoch": 6.22,
|
1020 |
+
"learning_rate": 1.3781942078364567e-05,
|
1021 |
+
"loss": 3.6561,
|
1022 |
+
"step": 7300
|
1023 |
+
},
|
1024 |
+
{
|
1025 |
+
"epoch": 6.22,
|
1026 |
+
"eval_loss": 4.261867046356201,
|
1027 |
+
"eval_runtime": 912.6991,
|
1028 |
+
"eval_samples_per_second": 0.321,
|
1029 |
+
"eval_steps_per_second": 0.041,
|
1030 |
+
"step": 7300
|
1031 |
+
},
|
1032 |
+
{
|
1033 |
+
"epoch": 6.3,
|
1034 |
+
"learning_rate": 1.3696763202725724e-05,
|
1035 |
+
"loss": 3.6851,
|
1036 |
+
"step": 7400
|
1037 |
+
},
|
1038 |
+
{
|
1039 |
+
"epoch": 6.3,
|
1040 |
+
"eval_loss": 4.274002552032471,
|
1041 |
+
"eval_runtime": 911.5799,
|
1042 |
+
"eval_samples_per_second": 0.321,
|
1043 |
+
"eval_steps_per_second": 0.041,
|
1044 |
+
"step": 7400
|
1045 |
+
},
|
1046 |
+
{
|
1047 |
+
"epoch": 6.39,
|
1048 |
+
"learning_rate": 1.3611584327086883e-05,
|
1049 |
+
"loss": 3.7842,
|
1050 |
+
"step": 7500
|
1051 |
+
},
|
1052 |
+
{
|
1053 |
+
"epoch": 6.39,
|
1054 |
+
"eval_loss": 4.243954658508301,
|
1055 |
+
"eval_runtime": 906.9325,
|
1056 |
+
"eval_samples_per_second": 0.323,
|
1057 |
+
"eval_steps_per_second": 0.041,
|
1058 |
+
"step": 7500
|
1059 |
+
},
|
1060 |
+
{
|
1061 |
+
"epoch": 6.47,
|
1062 |
+
"learning_rate": 1.352640545144804e-05,
|
1063 |
+
"loss": 3.6484,
|
1064 |
+
"step": 7600
|
1065 |
+
},
|
1066 |
+
{
|
1067 |
+
"epoch": 6.47,
|
1068 |
+
"eval_loss": 4.238761901855469,
|
1069 |
+
"eval_runtime": 910.5622,
|
1070 |
+
"eval_samples_per_second": 0.322,
|
1071 |
+
"eval_steps_per_second": 0.041,
|
1072 |
+
"step": 7600
|
1073 |
+
},
|
1074 |
+
{
|
1075 |
+
"epoch": 6.56,
|
1076 |
+
"learning_rate": 1.3441226575809202e-05,
|
1077 |
+
"loss": 3.7509,
|
1078 |
+
"step": 7700
|
1079 |
+
},
|
1080 |
+
{
|
1081 |
+
"epoch": 6.56,
|
1082 |
+
"eval_loss": 4.240095138549805,
|
1083 |
+
"eval_runtime": 935.7714,
|
1084 |
+
"eval_samples_per_second": 0.313,
|
1085 |
+
"eval_steps_per_second": 0.04,
|
1086 |
+
"step": 7700
|
1087 |
+
},
|
1088 |
+
{
|
1089 |
+
"epoch": 6.64,
|
1090 |
+
"learning_rate": 1.3356047700170357e-05,
|
1091 |
+
"loss": 3.7597,
|
1092 |
+
"step": 7800
|
1093 |
+
},
|
1094 |
+
{
|
1095 |
+
"epoch": 6.64,
|
1096 |
+
"eval_loss": 4.256958484649658,
|
1097 |
+
"eval_runtime": 915.2208,
|
1098 |
+
"eval_samples_per_second": 0.32,
|
1099 |
+
"eval_steps_per_second": 0.04,
|
1100 |
+
"step": 7800
|
1101 |
+
},
|
1102 |
+
{
|
1103 |
+
"epoch": 6.73,
|
1104 |
+
"learning_rate": 1.3270868824531518e-05,
|
1105 |
+
"loss": 3.7253,
|
1106 |
+
"step": 7900
|
1107 |
+
},
|
1108 |
+
{
|
1109 |
+
"epoch": 6.73,
|
1110 |
+
"eval_loss": 4.291894435882568,
|
1111 |
+
"eval_runtime": 915.8862,
|
1112 |
+
"eval_samples_per_second": 0.32,
|
1113 |
+
"eval_steps_per_second": 0.04,
|
1114 |
+
"step": 7900
|
1115 |
+
},
|
1116 |
+
{
|
1117 |
+
"epoch": 6.81,
|
1118 |
+
"learning_rate": 1.3185689948892676e-05,
|
1119 |
+
"loss": 3.7335,
|
1120 |
+
"step": 8000
|
1121 |
+
},
|
1122 |
+
{
|
1123 |
+
"epoch": 6.81,
|
1124 |
+
"eval_loss": 4.206986427307129,
|
1125 |
+
"eval_runtime": 916.8897,
|
1126 |
+
"eval_samples_per_second": 0.32,
|
1127 |
+
"eval_steps_per_second": 0.04,
|
1128 |
+
"step": 8000
|
1129 |
+
},
|
1130 |
+
{
|
1131 |
+
"epoch": 6.9,
|
1132 |
+
"learning_rate": 1.3100511073253835e-05,
|
1133 |
+
"loss": 3.6213,
|
1134 |
+
"step": 8100
|
1135 |
+
},
|
1136 |
+
{
|
1137 |
+
"epoch": 6.9,
|
1138 |
+
"eval_loss": 4.203628063201904,
|
1139 |
+
"eval_runtime": 910.5741,
|
1140 |
+
"eval_samples_per_second": 0.322,
|
1141 |
+
"eval_steps_per_second": 0.041,
|
1142 |
+
"step": 8100
|
1143 |
+
},
|
1144 |
+
{
|
1145 |
+
"epoch": 6.98,
|
1146 |
+
"learning_rate": 1.3015332197614992e-05,
|
1147 |
+
"loss": 3.588,
|
1148 |
+
"step": 8200
|
1149 |
+
},
|
1150 |
+
{
|
1151 |
+
"epoch": 6.98,
|
1152 |
+
"eval_loss": 4.219061851501465,
|
1153 |
+
"eval_runtime": 910.3103,
|
1154 |
+
"eval_samples_per_second": 0.322,
|
1155 |
+
"eval_steps_per_second": 0.041,
|
1156 |
+
"step": 8200
|
1157 |
+
},
|
1158 |
+
{
|
1159 |
+
"epoch": 7.07,
|
1160 |
+
"learning_rate": 1.2930153321976151e-05,
|
1161 |
+
"loss": 3.6381,
|
1162 |
+
"step": 8300
|
1163 |
+
},
|
1164 |
+
{
|
1165 |
+
"epoch": 7.07,
|
1166 |
+
"eval_loss": 4.232492923736572,
|
1167 |
+
"eval_runtime": 903.8691,
|
1168 |
+
"eval_samples_per_second": 0.324,
|
1169 |
+
"eval_steps_per_second": 0.041,
|
1170 |
+
"step": 8300
|
1171 |
+
},
|
1172 |
+
{
|
1173 |
+
"epoch": 7.16,
|
1174 |
+
"learning_rate": 1.284497444633731e-05,
|
1175 |
+
"loss": 3.6023,
|
1176 |
+
"step": 8400
|
1177 |
+
},
|
1178 |
+
{
|
1179 |
+
"epoch": 7.16,
|
1180 |
+
"eval_loss": 4.247186183929443,
|
1181 |
+
"eval_runtime": 910.7714,
|
1182 |
+
"eval_samples_per_second": 0.322,
|
1183 |
+
"eval_steps_per_second": 0.041,
|
1184 |
+
"step": 8400
|
1185 |
+
},
|
1186 |
+
{
|
1187 |
+
"epoch": 7.24,
|
1188 |
+
"learning_rate": 1.2759795570698468e-05,
|
1189 |
+
"loss": 3.5601,
|
1190 |
+
"step": 8500
|
1191 |
+
},
|
1192 |
+
{
|
1193 |
+
"epoch": 7.24,
|
1194 |
+
"eval_loss": 4.2428107261657715,
|
1195 |
+
"eval_runtime": 926.441,
|
1196 |
+
"eval_samples_per_second": 0.316,
|
1197 |
+
"eval_steps_per_second": 0.04,
|
1198 |
+
"step": 8500
|
1199 |
+
},
|
1200 |
+
{
|
1201 |
+
"epoch": 7.33,
|
1202 |
+
"learning_rate": 1.2674616695059627e-05,
|
1203 |
+
"loss": 3.6893,
|
1204 |
+
"step": 8600
|
1205 |
+
},
|
1206 |
+
{
|
1207 |
+
"epoch": 7.33,
|
1208 |
+
"eval_loss": 4.235583305358887,
|
1209 |
+
"eval_runtime": 905.7471,
|
1210 |
+
"eval_samples_per_second": 0.323,
|
1211 |
+
"eval_steps_per_second": 0.041,
|
1212 |
+
"step": 8600
|
1213 |
+
},
|
1214 |
+
{
|
1215 |
+
"epoch": 7.41,
|
1216 |
+
"learning_rate": 1.2589437819420784e-05,
|
1217 |
+
"loss": 3.4977,
|
1218 |
+
"step": 8700
|
1219 |
+
},
|
1220 |
+
{
|
1221 |
+
"epoch": 7.41,
|
1222 |
+
"eval_loss": 4.210697650909424,
|
1223 |
+
"eval_runtime": 909.4116,
|
1224 |
+
"eval_samples_per_second": 0.322,
|
1225 |
+
"eval_steps_per_second": 0.041,
|
1226 |
+
"step": 8700
|
1227 |
+
},
|
1228 |
+
{
|
1229 |
+
"epoch": 7.5,
|
1230 |
+
"learning_rate": 1.2504258943781943e-05,
|
1231 |
+
"loss": 3.5489,
|
1232 |
+
"step": 8800
|
1233 |
+
},
|
1234 |
+
{
|
1235 |
+
"epoch": 7.5,
|
1236 |
+
"eval_loss": 4.2333455085754395,
|
1237 |
+
"eval_runtime": 899.4646,
|
1238 |
+
"eval_samples_per_second": 0.326,
|
1239 |
+
"eval_steps_per_second": 0.041,
|
1240 |
+
"step": 8800
|
1241 |
+
},
|
1242 |
+
{
|
1243 |
+
"epoch": 7.58,
|
1244 |
+
"learning_rate": 1.24190800681431e-05,
|
1245 |
+
"loss": 3.5786,
|
1246 |
+
"step": 8900
|
1247 |
+
},
|
1248 |
+
{
|
1249 |
+
"epoch": 7.58,
|
1250 |
+
"eval_loss": 4.2256317138671875,
|
1251 |
+
"eval_runtime": 907.6718,
|
1252 |
+
"eval_samples_per_second": 0.323,
|
1253 |
+
"eval_steps_per_second": 0.041,
|
1254 |
+
"step": 8900
|
1255 |
+
},
|
1256 |
+
{
|
1257 |
+
"epoch": 7.67,
|
1258 |
+
"learning_rate": 1.2333901192504261e-05,
|
1259 |
+
"loss": 3.5991,
|
1260 |
+
"step": 9000
|
1261 |
+
},
|
1262 |
+
{
|
1263 |
+
"epoch": 7.67,
|
1264 |
+
"eval_loss": 4.206646919250488,
|
1265 |
+
"eval_runtime": 928.2897,
|
1266 |
+
"eval_samples_per_second": 0.316,
|
1267 |
+
"eval_steps_per_second": 0.04,
|
1268 |
+
"step": 9000
|
1269 |
+
},
|
1270 |
+
{
|
1271 |
+
"epoch": 7.75,
|
1272 |
+
"learning_rate": 1.2248722316865417e-05,
|
1273 |
+
"loss": 3.6571,
|
1274 |
+
"step": 9100
|
1275 |
+
},
|
1276 |
+
{
|
1277 |
+
"epoch": 7.75,
|
1278 |
+
"eval_loss": 4.213482856750488,
|
1279 |
+
"eval_runtime": 913.2222,
|
1280 |
+
"eval_samples_per_second": 0.321,
|
1281 |
+
"eval_steps_per_second": 0.041,
|
1282 |
+
"step": 9100
|
1283 |
+
},
|
1284 |
+
{
|
1285 |
+
"epoch": 7.84,
|
1286 |
+
"learning_rate": 1.2163543441226578e-05,
|
1287 |
+
"loss": 3.5465,
|
1288 |
+
"step": 9200
|
1289 |
+
},
|
1290 |
+
{
|
1291 |
+
"epoch": 7.84,
|
1292 |
+
"eval_loss": 4.197275638580322,
|
1293 |
+
"eval_runtime": 906.8727,
|
1294 |
+
"eval_samples_per_second": 0.323,
|
1295 |
+
"eval_steps_per_second": 0.041,
|
1296 |
+
"step": 9200
|
1297 |
+
},
|
1298 |
+
{
|
1299 |
+
"epoch": 7.92,
|
1300 |
+
"learning_rate": 1.2078364565587735e-05,
|
1301 |
+
"loss": 3.6476,
|
1302 |
+
"step": 9300
|
1303 |
+
},
|
1304 |
+
{
|
1305 |
+
"epoch": 7.92,
|
1306 |
+
"eval_loss": 4.191103935241699,
|
1307 |
+
"eval_runtime": 910.3991,
|
1308 |
+
"eval_samples_per_second": 0.322,
|
1309 |
+
"eval_steps_per_second": 0.041,
|
1310 |
+
"step": 9300
|
1311 |
+
},
|
1312 |
+
{
|
1313 |
+
"epoch": 8.01,
|
1314 |
+
"learning_rate": 1.1993185689948894e-05,
|
1315 |
+
"loss": 3.6669,
|
1316 |
+
"step": 9400
|
1317 |
+
},
|
1318 |
+
{
|
1319 |
+
"epoch": 8.01,
|
1320 |
+
"eval_loss": 4.2001953125,
|
1321 |
+
"eval_runtime": 908.1025,
|
1322 |
+
"eval_samples_per_second": 0.323,
|
1323 |
+
"eval_steps_per_second": 0.041,
|
1324 |
+
"step": 9400
|
1325 |
+
},
|
1326 |
+
{
|
1327 |
+
"epoch": 8.09,
|
1328 |
+
"learning_rate": 1.1908006814310052e-05,
|
1329 |
+
"loss": 3.4399,
|
1330 |
+
"step": 9500
|
1331 |
+
},
|
1332 |
+
{
|
1333 |
+
"epoch": 8.09,
|
1334 |
+
"eval_loss": 4.230240345001221,
|
1335 |
+
"eval_runtime": 914.9264,
|
1336 |
+
"eval_samples_per_second": 0.32,
|
1337 |
+
"eval_steps_per_second": 0.04,
|
1338 |
+
"step": 9500
|
1339 |
+
},
|
1340 |
+
{
|
1341 |
+
"epoch": 8.18,
|
1342 |
+
"learning_rate": 1.1822827938671211e-05,
|
1343 |
+
"loss": 3.5381,
|
1344 |
+
"step": 9600
|
1345 |
+
},
|
1346 |
+
{
|
1347 |
+
"epoch": 8.18,
|
1348 |
+
"eval_loss": 4.2084150314331055,
|
1349 |
+
"eval_runtime": 928.1244,
|
1350 |
+
"eval_samples_per_second": 0.316,
|
1351 |
+
"eval_steps_per_second": 0.04,
|
1352 |
+
"step": 9600
|
1353 |
+
},
|
1354 |
+
{
|
1355 |
+
"epoch": 8.26,
|
1356 |
+
"learning_rate": 1.1737649063032368e-05,
|
1357 |
+
"loss": 3.5008,
|
1358 |
+
"step": 9700
|
1359 |
+
},
|
1360 |
+
{
|
1361 |
+
"epoch": 8.26,
|
1362 |
+
"eval_loss": 4.216128826141357,
|
1363 |
+
"eval_runtime": 913.9443,
|
1364 |
+
"eval_samples_per_second": 0.321,
|
1365 |
+
"eval_steps_per_second": 0.04,
|
1366 |
+
"step": 9700
|
1367 |
+
},
|
1368 |
+
{
|
1369 |
+
"epoch": 8.35,
|
1370 |
+
"learning_rate": 1.1652470187393527e-05,
|
1371 |
+
"loss": 3.6199,
|
1372 |
+
"step": 9800
|
1373 |
+
},
|
1374 |
+
{
|
1375 |
+
"epoch": 8.35,
|
1376 |
+
"eval_loss": 4.218649387359619,
|
1377 |
+
"eval_runtime": 916.6151,
|
1378 |
+
"eval_samples_per_second": 0.32,
|
1379 |
+
"eval_steps_per_second": 0.04,
|
1380 |
+
"step": 9800
|
1381 |
+
},
|
1382 |
+
{
|
1383 |
+
"epoch": 8.43,
|
1384 |
+
"learning_rate": 1.1567291311754685e-05,
|
1385 |
+
"loss": 3.5997,
|
1386 |
+
"step": 9900
|
1387 |
+
},
|
1388 |
+
{
|
1389 |
+
"epoch": 8.43,
|
1390 |
+
"eval_loss": 4.269123554229736,
|
1391 |
+
"eval_runtime": 902.9645,
|
1392 |
+
"eval_samples_per_second": 0.324,
|
1393 |
+
"eval_steps_per_second": 0.041,
|
1394 |
+
"step": 9900
|
1395 |
+
},
|
1396 |
+
{
|
1397 |
+
"epoch": 8.52,
|
1398 |
+
"learning_rate": 1.1482112436115844e-05,
|
1399 |
+
"loss": 3.5075,
|
1400 |
+
"step": 10000
|
1401 |
+
},
|
1402 |
+
{
|
1403 |
+
"epoch": 8.52,
|
1404 |
+
"eval_loss": 4.2026848793029785,
|
1405 |
+
"eval_runtime": 903.7139,
|
1406 |
+
"eval_samples_per_second": 0.324,
|
1407 |
+
"eval_steps_per_second": 0.041,
|
1408 |
+
"step": 10000
|
1409 |
+
},
|
1410 |
+
{
|
1411 |
+
"epoch": 8.6,
|
1412 |
+
"learning_rate": 1.1396933560477003e-05,
|
1413 |
+
"loss": 3.5163,
|
1414 |
+
"step": 10100
|
1415 |
+
},
|
1416 |
+
{
|
1417 |
+
"epoch": 8.6,
|
1418 |
+
"eval_loss": 4.2565717697143555,
|
1419 |
+
"eval_runtime": 909.1873,
|
1420 |
+
"eval_samples_per_second": 0.322,
|
1421 |
+
"eval_steps_per_second": 0.041,
|
1422 |
+
"step": 10100
|
1423 |
+
},
|
1424 |
+
{
|
1425 |
+
"epoch": 8.69,
|
1426 |
+
"learning_rate": 1.131175468483816e-05,
|
1427 |
+
"loss": 3.4902,
|
1428 |
+
"step": 10200
|
1429 |
+
},
|
1430 |
+
{
|
1431 |
+
"epoch": 8.69,
|
1432 |
+
"eval_loss": 4.208179473876953,
|
1433 |
+
"eval_runtime": 901.8843,
|
1434 |
+
"eval_samples_per_second": 0.325,
|
1435 |
+
"eval_steps_per_second": 0.041,
|
1436 |
+
"step": 10200
|
1437 |
+
},
|
1438 |
+
{
|
1439 |
+
"epoch": 8.77,
|
1440 |
+
"learning_rate": 1.1226575809199321e-05,
|
1441 |
+
"loss": 3.4829,
|
1442 |
+
"step": 10300
|
1443 |
+
},
|
1444 |
+
{
|
1445 |
+
"epoch": 8.77,
|
1446 |
+
"eval_loss": 4.217052459716797,
|
1447 |
+
"eval_runtime": 913.5755,
|
1448 |
+
"eval_samples_per_second": 0.321,
|
1449 |
+
"eval_steps_per_second": 0.041,
|
1450 |
+
"step": 10300
|
1451 |
+
},
|
1452 |
+
{
|
1453 |
+
"epoch": 8.86,
|
1454 |
+
"learning_rate": 1.1141396933560477e-05,
|
1455 |
+
"loss": 3.599,
|
1456 |
+
"step": 10400
|
1457 |
+
},
|
1458 |
+
{
|
1459 |
+
"epoch": 8.86,
|
1460 |
+
"eval_loss": 4.216422080993652,
|
1461 |
+
"eval_runtime": 904.6206,
|
1462 |
+
"eval_samples_per_second": 0.324,
|
1463 |
+
"eval_steps_per_second": 0.041,
|
1464 |
+
"step": 10400
|
1465 |
+
},
|
1466 |
+
{
|
1467 |
+
"epoch": 8.94,
|
1468 |
+
"learning_rate": 1.1056218057921638e-05,
|
1469 |
+
"loss": 3.5058,
|
1470 |
+
"step": 10500
|
1471 |
+
},
|
1472 |
+
{
|
1473 |
+
"epoch": 8.94,
|
1474 |
+
"eval_loss": 4.200262069702148,
|
1475 |
+
"eval_runtime": 908.4763,
|
1476 |
+
"eval_samples_per_second": 0.323,
|
1477 |
+
"eval_steps_per_second": 0.041,
|
1478 |
+
"step": 10500
|
1479 |
+
},
|
1480 |
+
{
|
1481 |
+
"epoch": 9.03,
|
1482 |
+
"learning_rate": 1.0971039182282795e-05,
|
1483 |
+
"loss": 3.4622,
|
1484 |
+
"step": 10600
|
1485 |
+
},
|
1486 |
+
{
|
1487 |
+
"epoch": 9.03,
|
1488 |
+
"eval_loss": 4.201136112213135,
|
1489 |
+
"eval_runtime": 918.5088,
|
1490 |
+
"eval_samples_per_second": 0.319,
|
1491 |
+
"eval_steps_per_second": 0.04,
|
1492 |
+
"step": 10600
|
1493 |
+
},
|
1494 |
+
{
|
1495 |
+
"epoch": 9.11,
|
1496 |
+
"learning_rate": 1.0885860306643954e-05,
|
1497 |
+
"loss": 3.4836,
|
1498 |
+
"step": 10700
|
1499 |
+
},
|
1500 |
+
{
|
1501 |
+
"epoch": 9.11,
|
1502 |
+
"eval_loss": 4.196112632751465,
|
1503 |
+
"eval_runtime": 907.3086,
|
1504 |
+
"eval_samples_per_second": 0.323,
|
1505 |
+
"eval_steps_per_second": 0.041,
|
1506 |
+
"step": 10700
|
1507 |
+
},
|
1508 |
+
{
|
1509 |
+
"epoch": 9.2,
|
1510 |
+
"learning_rate": 1.0800681431005112e-05,
|
1511 |
+
"loss": 3.4177,
|
1512 |
+
"step": 10800
|
1513 |
+
},
|
1514 |
+
{
|
1515 |
+
"epoch": 9.2,
|
1516 |
+
"eval_loss": 4.2310028076171875,
|
1517 |
+
"eval_runtime": 904.8795,
|
1518 |
+
"eval_samples_per_second": 0.324,
|
1519 |
+
"eval_steps_per_second": 0.041,
|
1520 |
+
"step": 10800
|
1521 |
+
},
|
1522 |
+
{
|
1523 |
+
"epoch": 9.28,
|
1524 |
+
"learning_rate": 1.071550255536627e-05,
|
1525 |
+
"loss": 3.4407,
|
1526 |
+
"step": 10900
|
1527 |
+
},
|
1528 |
+
{
|
1529 |
+
"epoch": 9.28,
|
1530 |
+
"eval_loss": 4.2216362953186035,
|
1531 |
+
"eval_runtime": 910.613,
|
1532 |
+
"eval_samples_per_second": 0.322,
|
1533 |
+
"eval_steps_per_second": 0.041,
|
1534 |
+
"step": 10900
|
1535 |
+
},
|
1536 |
+
{
|
1537 |
+
"epoch": 9.37,
|
1538 |
+
"learning_rate": 1.0630323679727428e-05,
|
1539 |
+
"loss": 3.44,
|
1540 |
+
"step": 11000
|
1541 |
+
},
|
1542 |
+
{
|
1543 |
+
"epoch": 9.37,
|
1544 |
+
"eval_loss": 4.2364935874938965,
|
1545 |
+
"eval_runtime": 907.4387,
|
1546 |
+
"eval_samples_per_second": 0.323,
|
1547 |
+
"eval_steps_per_second": 0.041,
|
1548 |
+
"step": 11000
|
1549 |
+
},
|
1550 |
+
{
|
1551 |
+
"epoch": 9.45,
|
1552 |
+
"learning_rate": 1.0545144804088587e-05,
|
1553 |
+
"loss": 3.5116,
|
1554 |
+
"step": 11100
|
1555 |
+
},
|
1556 |
+
{
|
1557 |
+
"epoch": 9.45,
|
1558 |
+
"eval_loss": 4.201255798339844,
|
1559 |
+
"eval_runtime": 905.3716,
|
1560 |
+
"eval_samples_per_second": 0.324,
|
1561 |
+
"eval_steps_per_second": 0.041,
|
1562 |
+
"step": 11100
|
1563 |
+
},
|
1564 |
+
{
|
1565 |
+
"epoch": 9.54,
|
1566 |
+
"learning_rate": 1.0459965928449745e-05,
|
1567 |
+
"loss": 3.4793,
|
1568 |
+
"step": 11200
|
1569 |
+
},
|
1570 |
+
{
|
1571 |
+
"epoch": 9.54,
|
1572 |
+
"eval_loss": 4.203760147094727,
|
1573 |
+
"eval_runtime": 903.5367,
|
1574 |
+
"eval_samples_per_second": 0.324,
|
1575 |
+
"eval_steps_per_second": 0.041,
|
1576 |
+
"step": 11200
|
1577 |
+
},
|
1578 |
+
{
|
1579 |
+
"epoch": 9.63,
|
1580 |
+
"learning_rate": 1.0374787052810904e-05,
|
1581 |
+
"loss": 3.4414,
|
1582 |
+
"step": 11300
|
1583 |
+
},
|
1584 |
+
{
|
1585 |
+
"epoch": 9.63,
|
1586 |
+
"eval_loss": 4.177506923675537,
|
1587 |
+
"eval_runtime": 904.4055,
|
1588 |
+
"eval_samples_per_second": 0.324,
|
1589 |
+
"eval_steps_per_second": 0.041,
|
1590 |
+
"step": 11300
|
1591 |
+
},
|
1592 |
+
{
|
1593 |
+
"epoch": 9.71,
|
1594 |
+
"learning_rate": 1.0289608177172061e-05,
|
1595 |
+
"loss": 3.509,
|
1596 |
+
"step": 11400
|
1597 |
+
},
|
1598 |
+
{
|
1599 |
+
"epoch": 9.71,
|
1600 |
+
"eval_loss": 4.204129695892334,
|
1601 |
+
"eval_runtime": 904.4565,
|
1602 |
+
"eval_samples_per_second": 0.324,
|
1603 |
+
"eval_steps_per_second": 0.041,
|
1604 |
+
"step": 11400
|
1605 |
+
},
|
1606 |
+
{
|
1607 |
+
"epoch": 9.8,
|
1608 |
+
"learning_rate": 1.020442930153322e-05,
|
1609 |
+
"loss": 3.464,
|
1610 |
+
"step": 11500
|
1611 |
+
},
|
1612 |
+
{
|
1613 |
+
"epoch": 9.8,
|
1614 |
+
"eval_loss": 4.2182393074035645,
|
1615 |
+
"eval_runtime": 915.8218,
|
1616 |
+
"eval_samples_per_second": 0.32,
|
1617 |
+
"eval_steps_per_second": 0.04,
|
1618 |
+
"step": 11500
|
1619 |
+
},
|
1620 |
+
{
|
1621 |
+
"epoch": 9.88,
|
1622 |
+
"learning_rate": 1.0119250425894378e-05,
|
1623 |
+
"loss": 3.3574,
|
1624 |
+
"step": 11600
|
1625 |
+
},
|
1626 |
+
{
|
1627 |
+
"epoch": 9.88,
|
1628 |
+
"eval_loss": 4.222841262817383,
|
1629 |
+
"eval_runtime": 916.715,
|
1630 |
+
"eval_samples_per_second": 0.32,
|
1631 |
+
"eval_steps_per_second": 0.04,
|
1632 |
+
"step": 11600
|
1633 |
+
},
|
1634 |
+
{
|
1635 |
+
"epoch": 9.97,
|
1636 |
+
"learning_rate": 1.0034071550255537e-05,
|
1637 |
+
"loss": 3.4134,
|
1638 |
+
"step": 11700
|
1639 |
+
},
|
1640 |
+
{
|
1641 |
+
"epoch": 9.97,
|
1642 |
+
"eval_loss": 4.198949813842773,
|
1643 |
+
"eval_runtime": 911.1975,
|
1644 |
+
"eval_samples_per_second": 0.322,
|
1645 |
+
"eval_steps_per_second": 0.041,
|
1646 |
+
"step": 11700
|
1647 |
+
},
|
1648 |
+
{
|
1649 |
+
"epoch": 10.05,
|
1650 |
+
"learning_rate": 9.948892674616696e-06,
|
1651 |
+
"loss": 3.4876,
|
1652 |
+
"step": 11800
|
1653 |
+
},
|
1654 |
+
{
|
1655 |
+
"epoch": 10.05,
|
1656 |
+
"eval_loss": 4.1779093742370605,
|
1657 |
+
"eval_runtime": 905.3949,
|
1658 |
+
"eval_samples_per_second": 0.324,
|
1659 |
+
"eval_steps_per_second": 0.041,
|
1660 |
+
"step": 11800
|
1661 |
+
},
|
1662 |
+
{
|
1663 |
+
"epoch": 10.14,
|
1664 |
+
"learning_rate": 9.863713798977853e-06,
|
1665 |
+
"loss": 3.3188,
|
1666 |
+
"step": 11900
|
1667 |
+
},
|
1668 |
+
{
|
1669 |
+
"epoch": 10.14,
|
1670 |
+
"eval_loss": 4.199155330657959,
|
1671 |
+
"eval_runtime": 906.4403,
|
1672 |
+
"eval_samples_per_second": 0.323,
|
1673 |
+
"eval_steps_per_second": 0.041,
|
1674 |
+
"step": 11900
|
1675 |
+
},
|
1676 |
+
{
|
1677 |
+
"epoch": 10.22,
|
1678 |
+
"learning_rate": 9.778534923339012e-06,
|
1679 |
+
"loss": 3.4336,
|
1680 |
+
"step": 12000
|
1681 |
+
},
|
1682 |
+
{
|
1683 |
+
"epoch": 10.22,
|
1684 |
+
"eval_loss": 4.215114593505859,
|
1685 |
+
"eval_runtime": 918.8261,
|
1686 |
+
"eval_samples_per_second": 0.319,
|
1687 |
+
"eval_steps_per_second": 0.04,
|
1688 |
+
"step": 12000
|
1689 |
+
},
|
1690 |
+
{
|
1691 |
+
"epoch": 10.31,
|
1692 |
+
"learning_rate": 9.693356047700172e-06,
|
1693 |
+
"loss": 3.3774,
|
1694 |
+
"step": 12100
|
1695 |
+
},
|
1696 |
+
{
|
1697 |
+
"epoch": 10.31,
|
1698 |
+
"eval_loss": 4.194676876068115,
|
1699 |
+
"eval_runtime": 912.8562,
|
1700 |
+
"eval_samples_per_second": 0.321,
|
1701 |
+
"eval_steps_per_second": 0.041,
|
1702 |
+
"step": 12100
|
1703 |
+
},
|
1704 |
+
{
|
1705 |
+
"epoch": 10.39,
|
1706 |
+
"learning_rate": 9.608177172061329e-06,
|
1707 |
+
"loss": 3.2942,
|
1708 |
+
"step": 12200
|
1709 |
+
},
|
1710 |
+
{
|
1711 |
+
"epoch": 10.39,
|
1712 |
+
"eval_loss": 4.212912082672119,
|
1713 |
+
"eval_runtime": 905.2405,
|
1714 |
+
"eval_samples_per_second": 0.324,
|
1715 |
+
"eval_steps_per_second": 0.041,
|
1716 |
+
"step": 12200
|
1717 |
+
},
|
1718 |
+
{
|
1719 |
+
"epoch": 10.48,
|
1720 |
+
"learning_rate": 9.522998296422488e-06,
|
1721 |
+
"loss": 3.4431,
|
1722 |
+
"step": 12300
|
1723 |
+
},
|
1724 |
+
{
|
1725 |
+
"epoch": 10.48,
|
1726 |
+
"eval_loss": 4.1929144859313965,
|
1727 |
+
"eval_runtime": 908.9828,
|
1728 |
+
"eval_samples_per_second": 0.322,
|
1729 |
+
"eval_steps_per_second": 0.041,
|
1730 |
+
"step": 12300
|
1731 |
+
},
|
1732 |
+
{
|
1733 |
+
"epoch": 10.56,
|
1734 |
+
"learning_rate": 9.437819420783645e-06,
|
1735 |
+
"loss": 3.3895,
|
1736 |
+
"step": 12400
|
1737 |
+
},
|
1738 |
+
{
|
1739 |
+
"epoch": 10.56,
|
1740 |
+
"eval_loss": 4.221463203430176,
|
1741 |
+
"eval_runtime": 905.8419,
|
1742 |
+
"eval_samples_per_second": 0.323,
|
1743 |
+
"eval_steps_per_second": 0.041,
|
1744 |
+
"step": 12400
|
1745 |
+
},
|
1746 |
+
{
|
1747 |
+
"epoch": 10.65,
|
1748 |
+
"learning_rate": 9.352640545144805e-06,
|
1749 |
+
"loss": 3.4624,
|
1750 |
+
"step": 12500
|
1751 |
+
},
|
1752 |
+
{
|
1753 |
+
"epoch": 10.65,
|
1754 |
+
"eval_loss": 4.192135334014893,
|
1755 |
+
"eval_runtime": 915.7075,
|
1756 |
+
"eval_samples_per_second": 0.32,
|
1757 |
+
"eval_steps_per_second": 0.04,
|
1758 |
+
"step": 12500
|
1759 |
+
},
|
1760 |
+
{
|
1761 |
+
"epoch": 10.73,
|
1762 |
+
"learning_rate": 9.267461669505964e-06,
|
1763 |
+
"loss": 3.3823,
|
1764 |
+
"step": 12600
|
1765 |
+
},
|
1766 |
+
{
|
1767 |
+
"epoch": 10.73,
|
1768 |
+
"eval_loss": 4.197402477264404,
|
1769 |
+
"eval_runtime": 902.8153,
|
1770 |
+
"eval_samples_per_second": 0.325,
|
1771 |
+
"eval_steps_per_second": 0.041,
|
1772 |
+
"step": 12600
|
1773 |
+
},
|
1774 |
+
{
|
1775 |
+
"epoch": 10.82,
|
1776 |
+
"learning_rate": 9.182282793867123e-06,
|
1777 |
+
"loss": 3.3671,
|
1778 |
+
"step": 12700
|
1779 |
+
},
|
1780 |
+
{
|
1781 |
+
"epoch": 10.82,
|
1782 |
+
"eval_loss": 4.152112007141113,
|
1783 |
+
"eval_runtime": 903.9495,
|
1784 |
+
"eval_samples_per_second": 0.324,
|
1785 |
+
"eval_steps_per_second": 0.041,
|
1786 |
+
"step": 12700
|
1787 |
+
},
|
1788 |
+
{
|
1789 |
+
"epoch": 10.9,
|
1790 |
+
"learning_rate": 9.09710391822828e-06,
|
1791 |
+
"loss": 3.2883,
|
1792 |
+
"step": 12800
|
1793 |
+
},
|
1794 |
+
{
|
1795 |
+
"epoch": 10.9,
|
1796 |
+
"eval_loss": 4.178409576416016,
|
1797 |
+
"eval_runtime": 904.0493,
|
1798 |
+
"eval_samples_per_second": 0.324,
|
1799 |
+
"eval_steps_per_second": 0.041,
|
1800 |
+
"step": 12800
|
1801 |
+
},
|
1802 |
+
{
|
1803 |
+
"epoch": 10.99,
|
1804 |
+
"learning_rate": 9.01192504258944e-06,
|
1805 |
+
"loss": 3.4145,
|
1806 |
+
"step": 12900
|
1807 |
+
},
|
1808 |
+
{
|
1809 |
+
"epoch": 10.99,
|
1810 |
+
"eval_loss": 4.208373546600342,
|
1811 |
+
"eval_runtime": 903.9191,
|
1812 |
+
"eval_samples_per_second": 0.324,
|
1813 |
+
"eval_steps_per_second": 0.041,
|
1814 |
+
"step": 12900
|
1815 |
+
}
|
1816 |
+
],
|
1817 |
+
"max_steps": 23480,
|
1818 |
+
"num_train_epochs": 20,
|
1819 |
+
"total_flos": 4.542855419394294e+19,
|
1820 |
+
"trial_name": null,
|
1821 |
+
"trial_params": null
|
1822 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ecca8f0d18d0f451b8da7ee09460bb8cd9450be92cd7af6e28969226e3ba75b1
|
3 |
+
size 3771
|