teowu commited on
Commit
4f7f756
1 Parent(s): 648e603

Upload folder using huggingface_hub

Browse files
README.md ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+ ### Framework versions
7
+
8
+ - PEFT 0.4.0
9
+
10
+ - PEFT 0.4.0
adapter_config.json ADDED
@@ -0,0 +1,83 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "q-future/one-align",
4
+ "bias": "none",
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 256,
11
+ "lora_dropout": 0.05,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 128,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "model.layers.3.self_attn.q_proj",
18
+ "model.layers.19.self_attn.v_proj.multiway.1",
19
+ "model.layers.8.self_attn.v_proj.multiway.1",
20
+ "model.layers.8.self_attn.q_proj",
21
+ "model.layers.15.self_attn.q_proj",
22
+ "model.layers.21.self_attn.v_proj.multiway.1",
23
+ "model.layers.24.self_attn.v_proj.multiway.1",
24
+ "model.layers.31.self_attn.q_proj",
25
+ "model.layers.14.self_attn.q_proj",
26
+ "model.layers.14.self_attn.v_proj.multiway.1",
27
+ "model.layers.13.self_attn.v_proj.multiway.1",
28
+ "model.layers.7.self_attn.q_proj",
29
+ "model.layers.6.self_attn.q_proj",
30
+ "model.layers.0.self_attn.v_proj.multiway.1",
31
+ "model.layers.6.self_attn.v_proj.multiway.1",
32
+ "model.layers.21.self_attn.q_proj",
33
+ "model.layers.12.self_attn.v_proj.multiway.1",
34
+ "model.layers.30.self_attn.q_proj",
35
+ "model.layers.20.self_attn.q_proj",
36
+ "model.layers.27.self_attn.q_proj",
37
+ "model.layers.23.self_attn.q_proj",
38
+ "model.layers.7.self_attn.v_proj.multiway.1",
39
+ "model.layers.12.self_attn.q_proj",
40
+ "model.layers.10.self_attn.v_proj.multiway.1",
41
+ "model.layers.23.self_attn.v_proj.multiway.1",
42
+ "model.layers.16.self_attn.q_proj",
43
+ "model.layers.18.self_attn.q_proj",
44
+ "model.layers.22.self_attn.v_proj.multiway.1",
45
+ "model.layers.0.self_attn.q_proj",
46
+ "model.layers.17.self_attn.v_proj.multiway.1",
47
+ "model.layers.5.self_attn.q_proj",
48
+ "model.layers.9.self_attn.v_proj.multiway.1",
49
+ "model.layers.11.self_attn.q_proj",
50
+ "model.layers.11.self_attn.v_proj.multiway.1",
51
+ "model.layers.19.self_attn.q_proj",
52
+ "model.layers.1.self_attn.q_proj",
53
+ "model.layers.26.self_attn.v_proj.multiway.1",
54
+ "model.layers.30.self_attn.v_proj.multiway.1",
55
+ "model.layers.17.self_attn.q_proj",
56
+ "model.layers.4.self_attn.v_proj.multiway.1",
57
+ "model.layers.2.self_attn.q_proj",
58
+ "model.layers.25.self_attn.v_proj.multiway.1",
59
+ "model.layers.28.self_attn.q_proj",
60
+ "model.layers.27.self_attn.v_proj.multiway.1",
61
+ "model.layers.20.self_attn.v_proj.multiway.1",
62
+ "model.layers.31.self_attn.v_proj.multiway.1",
63
+ "model.layers.15.self_attn.v_proj.multiway.1",
64
+ "model.layers.24.self_attn.q_proj",
65
+ "model.layers.18.self_attn.v_proj.multiway.1",
66
+ "model.layers.22.self_attn.q_proj",
67
+ "model.layers.28.self_attn.v_proj.multiway.1",
68
+ "model.layers.29.self_attn.q_proj",
69
+ "model.layers.13.self_attn.q_proj",
70
+ "model.layers.2.self_attn.v_proj.multiway.1",
71
+ "model.layers.26.self_attn.q_proj",
72
+ "model.layers.5.self_attn.v_proj.multiway.1",
73
+ "model.layers.4.self_attn.q_proj",
74
+ "model.layers.1.self_attn.v_proj.multiway.1",
75
+ "model.layers.16.self_attn.v_proj.multiway.1",
76
+ "model.layers.3.self_attn.v_proj.multiway.1",
77
+ "model.layers.29.self_attn.v_proj.multiway.1",
78
+ "model.layers.9.self_attn.q_proj",
79
+ "model.layers.10.self_attn.q_proj",
80
+ "model.layers.25.self_attn.q_proj"
81
+ ],
82
+ "task_type": "CAUSAL_LM"
83
+ }
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8236acee4e810d0e2c0b54b413718ff603df47f09138829f9c0013efa160ce33
3
+ size 134262669
config.json ADDED
@@ -0,0 +1,192 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "q-future/one-align",
3
+ "architectures": [
4
+ "MPLUGOwl2LlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "auto_map": {
9
+ "AutoConfig": "q-future/one-align--configuration_mplug_owl2.MPLUGOwl2Config",
10
+ "AutoModel": "q-future/one-align--modeling_mplug_owl2.MPLUGOwl2LlamaForCausalLM",
11
+ "AutoModelForCausalLM": "q-future/one-align--modeling_mplug_owl2.MPLUGOwl2LlamaForCausalLM"
12
+ },
13
+ "bos_token_id": 1,
14
+ "eos_token_id": 2,
15
+ "freeze_vision_model": false,
16
+ "hidden_act": "silu",
17
+ "hidden_size": 4096,
18
+ "image_aspect_ratio": "pad",
19
+ "image_grid_pinpoints": null,
20
+ "initializer_range": 0.02,
21
+ "intermediate_size": 11008,
22
+ "max_position_embeddings": 2048,
23
+ "model_type": "mplug_owl2",
24
+ "num_attention_heads": 32,
25
+ "num_hidden_layers": 32,
26
+ "num_key_value_heads": 32,
27
+ "pretraining_tp": 1,
28
+ "rms_norm_eps": 1e-06,
29
+ "rope_scaling": null,
30
+ "rope_theta": 10000.0,
31
+ "tie_word_embeddings": false,
32
+ "torch_dtype": "bfloat16",
33
+ "transformers_version": "4.36.1",
34
+ "tune_visual_abstractor": true,
35
+ "use_cache": true,
36
+ "visual_abstractor_lr": 2e-05,
37
+ "visual_config": {
38
+ "visual_abstractor": {
39
+ "_name_or_path": "",
40
+ "add_cross_attention": false,
41
+ "architectures": null,
42
+ "attention_probs_dropout_prob": 0.0,
43
+ "bad_words_ids": null,
44
+ "begin_suppress_tokens": null,
45
+ "bos_token_id": null,
46
+ "chunk_size_feed_forward": 0,
47
+ "cross_attention_hidden_size": null,
48
+ "decoder_start_token_id": null,
49
+ "diversity_penalty": 0.0,
50
+ "do_sample": false,
51
+ "early_stopping": false,
52
+ "encoder_hidden_size": 1024,
53
+ "encoder_no_repeat_ngram_size": 0,
54
+ "eos_token_id": null,
55
+ "exponential_decay_length_penalty": null,
56
+ "finetuning_task": null,
57
+ "forced_bos_token_id": null,
58
+ "forced_eos_token_id": null,
59
+ "grid_size": 32,
60
+ "hidden_size": 1024,
61
+ "id2label": {
62
+ "0": "LABEL_0",
63
+ "1": "LABEL_1"
64
+ },
65
+ "initializer_range": 0.02,
66
+ "intermediate_size": 2816,
67
+ "is_decoder": false,
68
+ "is_encoder_decoder": false,
69
+ "label2id": {
70
+ "LABEL_0": 0,
71
+ "LABEL_1": 1
72
+ },
73
+ "layer_norm_eps": 1e-06,
74
+ "length_penalty": 1.0,
75
+ "max_length": 20,
76
+ "min_length": 0,
77
+ "model_type": "mplug_owl_visual_abstract",
78
+ "no_repeat_ngram_size": 0,
79
+ "num_attention_heads": 16,
80
+ "num_beam_groups": 1,
81
+ "num_beams": 1,
82
+ "num_hidden_layers": 6,
83
+ "num_learnable_queries": 64,
84
+ "num_return_sequences": 1,
85
+ "output_attentions": false,
86
+ "output_hidden_states": false,
87
+ "output_scores": false,
88
+ "pad_token_id": null,
89
+ "prefix": null,
90
+ "problem_type": null,
91
+ "pruned_heads": {},
92
+ "remove_invalid_values": false,
93
+ "repetition_penalty": 1.0,
94
+ "return_dict": true,
95
+ "return_dict_in_generate": false,
96
+ "sep_token_id": null,
97
+ "suppress_tokens": null,
98
+ "task_specific_params": null,
99
+ "temperature": 1.0,
100
+ "tf_legacy_loss": false,
101
+ "tie_encoder_decoder": false,
102
+ "tie_word_embeddings": true,
103
+ "tokenizer_class": null,
104
+ "top_k": 50,
105
+ "top_p": 1.0,
106
+ "torch_dtype": null,
107
+ "torchscript": false,
108
+ "transformers_version": "4.28.1",
109
+ "typical_p": 1.0,
110
+ "use_bfloat16": false
111
+ },
112
+ "visual_model": {
113
+ "_name_or_path": "",
114
+ "add_cross_attention": false,
115
+ "architectures": null,
116
+ "attention_dropout": 0.0,
117
+ "bad_words_ids": null,
118
+ "begin_suppress_tokens": null,
119
+ "bos_token_id": null,
120
+ "chunk_size_feed_forward": 0,
121
+ "cross_attention_hidden_size": null,
122
+ "decoder_start_token_id": null,
123
+ "diversity_penalty": 0.0,
124
+ "do_sample": false,
125
+ "early_stopping": false,
126
+ "encoder_no_repeat_ngram_size": 0,
127
+ "eos_token_id": null,
128
+ "exponential_decay_length_penalty": null,
129
+ "finetuning_task": null,
130
+ "forced_bos_token_id": null,
131
+ "forced_eos_token_id": null,
132
+ "hidden_act": "quick_gelu",
133
+ "hidden_size": 1024,
134
+ "id2label": {
135
+ "0": "LABEL_0",
136
+ "1": "LABEL_1"
137
+ },
138
+ "image_size": 448,
139
+ "initializer_factor": 1.0,
140
+ "initializer_range": 0.02,
141
+ "intermediate_size": 4096,
142
+ "is_decoder": false,
143
+ "is_encoder_decoder": false,
144
+ "label2id": {
145
+ "LABEL_0": 0,
146
+ "LABEL_1": 1
147
+ },
148
+ "layer_norm_eps": 1e-06,
149
+ "length_penalty": 1.0,
150
+ "max_length": 20,
151
+ "min_length": 0,
152
+ "model_type": "mplug_owl_vision_model",
153
+ "no_repeat_ngram_size": 0,
154
+ "num_attention_heads": 16,
155
+ "num_beam_groups": 1,
156
+ "num_beams": 1,
157
+ "num_channels": 3,
158
+ "num_hidden_layers": 24,
159
+ "num_return_sequences": 1,
160
+ "output_attentions": false,
161
+ "output_hidden_states": false,
162
+ "output_scores": false,
163
+ "pad_token_id": null,
164
+ "patch_size": 14,
165
+ "prefix": null,
166
+ "problem_type": null,
167
+ "projection_dim": 768,
168
+ "pruned_heads": {},
169
+ "remove_invalid_values": false,
170
+ "repetition_penalty": 1.0,
171
+ "return_dict": true,
172
+ "return_dict_in_generate": false,
173
+ "sep_token_id": null,
174
+ "suppress_tokens": null,
175
+ "task_specific_params": null,
176
+ "temperature": 1.0,
177
+ "tf_legacy_loss": false,
178
+ "tie_encoder_decoder": false,
179
+ "tie_word_embeddings": true,
180
+ "tokenizer_class": null,
181
+ "top_k": 50,
182
+ "top_p": 1.0,
183
+ "torch_dtype": null,
184
+ "torchscript": false,
185
+ "transformers_version": "4.28.1",
186
+ "typical_p": 1.0,
187
+ "use_bfloat16": false,
188
+ "use_flash_attn": false
189
+ }
190
+ },
191
+ "vocab_size": 32000
192
+ }
non_lora_trainables.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a4b6a43c22c5a3c37e0ce86c612a5b90c5501306a7296fc796c15b48854ca66e
3
+ size 162998185
preprocessor_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": 448,
3
+ "do_center_crop": true,
4
+ "do_normalize": true,
5
+ "do_resize": true,
6
+ "feature_extractor_type": "CLIPFeatureExtractor",
7
+ "image_mean": [
8
+ 0.48145466,
9
+ 0.4578275,
10
+ 0.40821073
11
+ ],
12
+ "image_std": [
13
+ 0.26862954,
14
+ 0.26130258,
15
+ 0.27577711
16
+ ],
17
+ "resample": 3,
18
+ "size": 448
19
+ }
20
+
trainer_state.json ADDED
@@ -0,0 +1,1362 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.0,
5
+ "eval_steps": 500,
6
+ "global_step": 222,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01,
13
+ "learning_rate": 2.857142857142857e-05,
14
+ "loss": 0.2358,
15
+ "step": 1
16
+ },
17
+ {
18
+ "epoch": 0.03,
19
+ "learning_rate": 5.714285714285714e-05,
20
+ "loss": 0.21,
21
+ "step": 2
22
+ },
23
+ {
24
+ "epoch": 0.04,
25
+ "learning_rate": 8.571428571428571e-05,
26
+ "loss": 0.1914,
27
+ "step": 3
28
+ },
29
+ {
30
+ "epoch": 0.05,
31
+ "learning_rate": 0.00011428571428571428,
32
+ "loss": 0.2148,
33
+ "step": 4
34
+ },
35
+ {
36
+ "epoch": 0.07,
37
+ "learning_rate": 0.00014285714285714287,
38
+ "loss": 0.1733,
39
+ "step": 5
40
+ },
41
+ {
42
+ "epoch": 0.08,
43
+ "learning_rate": 0.00017142857142857143,
44
+ "loss": 0.1304,
45
+ "step": 6
46
+ },
47
+ {
48
+ "epoch": 0.09,
49
+ "learning_rate": 0.0002,
50
+ "loss": 0.1372,
51
+ "step": 7
52
+ },
53
+ {
54
+ "epoch": 0.11,
55
+ "learning_rate": 0.00019998932457674902,
56
+ "loss": 0.124,
57
+ "step": 8
58
+ },
59
+ {
60
+ "epoch": 0.12,
61
+ "learning_rate": 0.00019995730058628927,
62
+ "loss": 0.1387,
63
+ "step": 9
64
+ },
65
+ {
66
+ "epoch": 0.14,
67
+ "learning_rate": 0.00019990393486601384,
68
+ "loss": 0.125,
69
+ "step": 10
70
+ },
71
+ {
72
+ "epoch": 0.15,
73
+ "learning_rate": 0.00019982923880995572,
74
+ "loss": 0.123,
75
+ "step": 11
76
+ },
77
+ {
78
+ "epoch": 0.16,
79
+ "learning_rate": 0.00019973322836635518,
80
+ "loss": 0.1196,
81
+ "step": 12
82
+ },
83
+ {
84
+ "epoch": 0.18,
85
+ "learning_rate": 0.00019961592403425468,
86
+ "loss": 0.1223,
87
+ "step": 13
88
+ },
89
+ {
90
+ "epoch": 0.19,
91
+ "learning_rate": 0.00019947735085912207,
92
+ "loss": 0.1208,
93
+ "step": 14
94
+ },
95
+ {
96
+ "epoch": 0.2,
97
+ "learning_rate": 0.0001993175384275033,
98
+ "loss": 0.1211,
99
+ "step": 15
100
+ },
101
+ {
102
+ "epoch": 0.22,
103
+ "learning_rate": 0.00019913652086070535,
104
+ "loss": 0.1084,
105
+ "step": 16
106
+ },
107
+ {
108
+ "epoch": 0.23,
109
+ "learning_rate": 0.00019893433680751103,
110
+ "loss": 0.1169,
111
+ "step": 17
112
+ },
113
+ {
114
+ "epoch": 0.24,
115
+ "learning_rate": 0.00019871102943592715,
116
+ "loss": 0.1125,
117
+ "step": 18
118
+ },
119
+ {
120
+ "epoch": 0.26,
121
+ "learning_rate": 0.0001984666464239679,
122
+ "loss": 0.1206,
123
+ "step": 19
124
+ },
125
+ {
126
+ "epoch": 0.27,
127
+ "learning_rate": 0.00019820123994947504,
128
+ "loss": 0.124,
129
+ "step": 20
130
+ },
131
+ {
132
+ "epoch": 0.28,
133
+ "learning_rate": 0.0001979148666789775,
134
+ "loss": 0.116,
135
+ "step": 21
136
+ },
137
+ {
138
+ "epoch": 0.3,
139
+ "learning_rate": 0.00019760758775559274,
140
+ "loss": 0.0994,
141
+ "step": 22
142
+ },
143
+ {
144
+ "epoch": 0.31,
145
+ "learning_rate": 0.00019727946878597195,
146
+ "loss": 0.1128,
147
+ "step": 23
148
+ },
149
+ {
150
+ "epoch": 0.32,
151
+ "learning_rate": 0.00019693057982629277,
152
+ "loss": 0.1267,
153
+ "step": 24
154
+ },
155
+ {
156
+ "epoch": 0.34,
157
+ "learning_rate": 0.00019656099536730136,
158
+ "loss": 0.1096,
159
+ "step": 25
160
+ },
161
+ {
162
+ "epoch": 0.35,
163
+ "learning_rate": 0.0001961707943184083,
164
+ "loss": 0.1226,
165
+ "step": 26
166
+ },
167
+ {
168
+ "epoch": 0.36,
169
+ "learning_rate": 0.0001957600599908406,
170
+ "loss": 0.1084,
171
+ "step": 27
172
+ },
173
+ {
174
+ "epoch": 0.38,
175
+ "learning_rate": 0.00019532888007985406,
176
+ "loss": 0.1018,
177
+ "step": 28
178
+ },
179
+ {
180
+ "epoch": 0.39,
181
+ "learning_rate": 0.00019487734664600957,
182
+ "loss": 0.1099,
183
+ "step": 29
184
+ },
185
+ {
186
+ "epoch": 0.41,
187
+ "learning_rate": 0.0001944055560955176,
188
+ "loss": 0.116,
189
+ "step": 30
190
+ },
191
+ {
192
+ "epoch": 0.42,
193
+ "learning_rate": 0.00019391360915965426,
194
+ "loss": 0.1101,
195
+ "step": 31
196
+ },
197
+ {
198
+ "epoch": 0.43,
199
+ "learning_rate": 0.0001934016108732548,
200
+ "loss": 0.1052,
201
+ "step": 32
202
+ },
203
+ {
204
+ "epoch": 0.45,
205
+ "learning_rate": 0.00019286967055228744,
206
+ "loss": 0.1208,
207
+ "step": 33
208
+ },
209
+ {
210
+ "epoch": 0.46,
211
+ "learning_rate": 0.00019231790177051355,
212
+ "loss": 0.0981,
213
+ "step": 34
214
+ },
215
+ {
216
+ "epoch": 0.47,
217
+ "learning_rate": 0.00019174642233523877,
218
+ "loss": 0.1147,
219
+ "step": 35
220
+ },
221
+ {
222
+ "epoch": 0.49,
223
+ "learning_rate": 0.00019115535426216017,
224
+ "loss": 0.1003,
225
+ "step": 36
226
+ },
227
+ {
228
+ "epoch": 0.5,
229
+ "learning_rate": 0.00019054482374931467,
230
+ "loss": 0.0989,
231
+ "step": 37
232
+ },
233
+ {
234
+ "epoch": 0.51,
235
+ "learning_rate": 0.000189914961150135,
236
+ "loss": 0.1091,
237
+ "step": 38
238
+ },
239
+ {
240
+ "epoch": 0.53,
241
+ "learning_rate": 0.00018926590094561782,
242
+ "loss": 0.104,
243
+ "step": 39
244
+ },
245
+ {
246
+ "epoch": 0.54,
247
+ "learning_rate": 0.00018859778171561114,
248
+ "loss": 0.0955,
249
+ "step": 40
250
+ },
251
+ {
252
+ "epoch": 0.55,
253
+ "learning_rate": 0.00018791074610922622,
254
+ "loss": 0.1384,
255
+ "step": 41
256
+ },
257
+ {
258
+ "epoch": 0.57,
259
+ "learning_rate": 0.00018720494081438078,
260
+ "loss": 0.0906,
261
+ "step": 42
262
+ },
263
+ {
264
+ "epoch": 0.58,
265
+ "learning_rate": 0.0001864805165264799,
266
+ "loss": 0.0945,
267
+ "step": 43
268
+ },
269
+ {
270
+ "epoch": 0.59,
271
+ "learning_rate": 0.00018573762791624133,
272
+ "loss": 0.0991,
273
+ "step": 44
274
+ },
275
+ {
276
+ "epoch": 0.61,
277
+ "learning_rate": 0.0001849764335966719,
278
+ "loss": 0.0876,
279
+ "step": 45
280
+ },
281
+ {
282
+ "epoch": 0.62,
283
+ "learning_rate": 0.00018419709608920242,
284
+ "loss": 0.0918,
285
+ "step": 46
286
+ },
287
+ {
288
+ "epoch": 0.64,
289
+ "learning_rate": 0.0001833997817889878,
290
+ "loss": 0.1074,
291
+ "step": 47
292
+ },
293
+ {
294
+ "epoch": 0.65,
295
+ "learning_rate": 0.00018258466092938044,
296
+ "loss": 0.1099,
297
+ "step": 48
298
+ },
299
+ {
300
+ "epoch": 0.66,
301
+ "learning_rate": 0.00018175190754558384,
302
+ "loss": 0.1045,
303
+ "step": 49
304
+ },
305
+ {
306
+ "epoch": 0.68,
307
+ "learning_rate": 0.00018090169943749476,
308
+ "loss": 0.1025,
309
+ "step": 50
310
+ },
311
+ {
312
+ "epoch": 0.69,
313
+ "learning_rate": 0.00018003421813174127,
314
+ "loss": 0.0847,
315
+ "step": 51
316
+ },
317
+ {
318
+ "epoch": 0.7,
319
+ "learning_rate": 0.00017914964884292544,
320
+ "loss": 0.0842,
321
+ "step": 52
322
+ },
323
+ {
324
+ "epoch": 0.72,
325
+ "learning_rate": 0.00017824818043407826,
326
+ "loss": 0.0991,
327
+ "step": 53
328
+ },
329
+ {
330
+ "epoch": 0.73,
331
+ "learning_rate": 0.00017733000537633605,
332
+ "loss": 0.1055,
333
+ "step": 54
334
+ },
335
+ {
336
+ "epoch": 0.74,
337
+ "learning_rate": 0.00017639531970784593,
338
+ "loss": 0.092,
339
+ "step": 55
340
+ },
341
+ {
342
+ "epoch": 0.76,
343
+ "learning_rate": 0.00017544432299191026,
344
+ "loss": 0.0923,
345
+ "step": 56
346
+ },
347
+ {
348
+ "epoch": 0.77,
349
+ "learning_rate": 0.0001744772182743782,
350
+ "loss": 0.1121,
351
+ "step": 57
352
+ },
353
+ {
354
+ "epoch": 0.78,
355
+ "learning_rate": 0.00017349421204029342,
356
+ "loss": 0.0942,
357
+ "step": 58
358
+ },
359
+ {
360
+ "epoch": 0.8,
361
+ "learning_rate": 0.00017249551416980804,
362
+ "loss": 0.1079,
363
+ "step": 59
364
+ },
365
+ {
366
+ "epoch": 0.81,
367
+ "learning_rate": 0.00017148133789337145,
368
+ "loss": 0.0967,
369
+ "step": 60
370
+ },
371
+ {
372
+ "epoch": 0.82,
373
+ "learning_rate": 0.0001704518997462037,
374
+ "loss": 0.1094,
375
+ "step": 61
376
+ },
377
+ {
378
+ "epoch": 0.84,
379
+ "learning_rate": 0.0001694074195220634,
380
+ "loss": 0.0942,
381
+ "step": 62
382
+ },
383
+ {
384
+ "epoch": 0.85,
385
+ "learning_rate": 0.00016834812022631997,
386
+ "loss": 0.0894,
387
+ "step": 63
388
+ },
389
+ {
390
+ "epoch": 0.86,
391
+ "learning_rate": 0.00016727422802834,
392
+ "loss": 0.0972,
393
+ "step": 64
394
+ },
395
+ {
396
+ "epoch": 0.88,
397
+ "learning_rate": 0.00016618597221319833,
398
+ "loss": 0.0894,
399
+ "step": 65
400
+ },
401
+ {
402
+ "epoch": 0.89,
403
+ "learning_rate": 0.00016508358513272358,
404
+ "loss": 0.0791,
405
+ "step": 66
406
+ },
407
+ {
408
+ "epoch": 0.91,
409
+ "learning_rate": 0.00016396730215588915,
410
+ "loss": 0.1182,
411
+ "step": 67
412
+ },
413
+ {
414
+ "epoch": 0.92,
415
+ "learning_rate": 0.00016283736161855993,
416
+ "loss": 0.1042,
417
+ "step": 68
418
+ },
419
+ {
420
+ "epoch": 0.93,
421
+ "learning_rate": 0.00016169400477260566,
422
+ "loss": 0.1021,
423
+ "step": 69
424
+ },
425
+ {
426
+ "epoch": 0.95,
427
+ "learning_rate": 0.00016053747573439145,
428
+ "loss": 0.0808,
429
+ "step": 70
430
+ },
431
+ {
432
+ "epoch": 0.96,
433
+ "learning_rate": 0.0001593680214326571,
434
+ "loss": 0.0967,
435
+ "step": 71
436
+ },
437
+ {
438
+ "epoch": 0.97,
439
+ "learning_rate": 0.0001581858915557953,
440
+ "loss": 0.0884,
441
+ "step": 72
442
+ },
443
+ {
444
+ "epoch": 0.99,
445
+ "learning_rate": 0.00015699133849854163,
446
+ "loss": 0.0903,
447
+ "step": 73
448
+ },
449
+ {
450
+ "epoch": 1.0,
451
+ "learning_rate": 0.00015578461730808574,
452
+ "loss": 0.0989,
453
+ "step": 74
454
+ },
455
+ {
456
+ "epoch": 1.01,
457
+ "learning_rate": 0.00015456598562961667,
458
+ "loss": 0.0784,
459
+ "step": 75
460
+ },
461
+ {
462
+ "epoch": 1.03,
463
+ "learning_rate": 0.00015333570365131353,
464
+ "loss": 0.0815,
465
+ "step": 76
466
+ },
467
+ {
468
+ "epoch": 1.04,
469
+ "learning_rate": 0.00015209403404879303,
470
+ "loss": 0.0815,
471
+ "step": 77
472
+ },
473
+ {
474
+ "epoch": 1.05,
475
+ "learning_rate": 0.0001508412419290261,
476
+ "loss": 0.0786,
477
+ "step": 78
478
+ },
479
+ {
480
+ "epoch": 1.07,
481
+ "learning_rate": 0.0001495775947737352,
482
+ "loss": 0.0776,
483
+ "step": 79
484
+ },
485
+ {
486
+ "epoch": 1.08,
487
+ "learning_rate": 0.0001483033623822848,
488
+ "loss": 0.0925,
489
+ "step": 80
490
+ },
491
+ {
492
+ "epoch": 1.09,
493
+ "learning_rate": 0.00014701881681407683,
494
+ "loss": 0.0713,
495
+ "step": 81
496
+ },
497
+ {
498
+ "epoch": 1.11,
499
+ "learning_rate": 0.00014572423233046386,
500
+ "loss": 0.0918,
501
+ "step": 82
502
+ },
503
+ {
504
+ "epoch": 1.12,
505
+ "learning_rate": 0.00014441988533619182,
506
+ "loss": 0.0981,
507
+ "step": 83
508
+ },
509
+ {
510
+ "epoch": 1.14,
511
+ "learning_rate": 0.00014310605432038526,
512
+ "loss": 0.0781,
513
+ "step": 84
514
+ },
515
+ {
516
+ "epoch": 1.15,
517
+ "learning_rate": 0.0001417830197970877,
518
+ "loss": 0.0811,
519
+ "step": 85
520
+ },
521
+ {
522
+ "epoch": 1.16,
523
+ "learning_rate": 0.00014045106424536937,
524
+ "loss": 0.0811,
525
+ "step": 86
526
+ },
527
+ {
528
+ "epoch": 1.18,
529
+ "learning_rate": 0.0001391104720490156,
530
+ "loss": 0.1033,
531
+ "step": 87
532
+ },
533
+ {
534
+ "epoch": 1.19,
535
+ "learning_rate": 0.00013776152943580848,
536
+ "loss": 0.0828,
537
+ "step": 88
538
+ },
539
+ {
540
+ "epoch": 1.2,
541
+ "learning_rate": 0.00013640452441641463,
542
+ "loss": 0.0933,
543
+ "step": 89
544
+ },
545
+ {
546
+ "epoch": 1.22,
547
+ "learning_rate": 0.00013503974672289296,
548
+ "loss": 0.0818,
549
+ "step": 90
550
+ },
551
+ {
552
+ "epoch": 1.23,
553
+ "learning_rate": 0.00013366748774683375,
554
+ "loss": 0.0764,
555
+ "step": 91
556
+ },
557
+ {
558
+ "epoch": 1.24,
559
+ "learning_rate": 0.00013228804047714463,
560
+ "loss": 0.0774,
561
+ "step": 92
562
+ },
563
+ {
564
+ "epoch": 1.26,
565
+ "learning_rate": 0.00013090169943749476,
566
+ "loss": 0.1003,
567
+ "step": 93
568
+ },
569
+ {
570
+ "epoch": 1.27,
571
+ "learning_rate": 0.00012950876062343148,
572
+ "loss": 0.0674,
573
+ "step": 94
574
+ },
575
+ {
576
+ "epoch": 1.28,
577
+ "learning_rate": 0.00012810952143918285,
578
+ "loss": 0.0684,
579
+ "step": 95
580
+ },
581
+ {
582
+ "epoch": 1.3,
583
+ "learning_rate": 0.0001267042806341593,
584
+ "loss": 0.0815,
585
+ "step": 96
586
+ },
587
+ {
588
+ "epoch": 1.31,
589
+ "learning_rate": 0.00012529333823916807,
590
+ "loss": 0.0938,
591
+ "step": 97
592
+ },
593
+ {
594
+ "epoch": 1.32,
595
+ "learning_rate": 0.0001238769955023542,
596
+ "loss": 0.0791,
597
+ "step": 98
598
+ },
599
+ {
600
+ "epoch": 1.34,
601
+ "learning_rate": 0.00012245555482488134,
602
+ "loss": 0.0828,
603
+ "step": 99
604
+ },
605
+ {
606
+ "epoch": 1.35,
607
+ "learning_rate": 0.00012102931969636664,
608
+ "loss": 0.0693,
609
+ "step": 100
610
+ },
611
+ {
612
+ "epoch": 1.36,
613
+ "learning_rate": 0.00011959859463008316,
614
+ "loss": 0.0979,
615
+ "step": 101
616
+ },
617
+ {
618
+ "epoch": 1.38,
619
+ "learning_rate": 0.00011816368509794364,
620
+ "loss": 0.0996,
621
+ "step": 102
622
+ },
623
+ {
624
+ "epoch": 1.39,
625
+ "learning_rate": 0.00011672489746527979,
626
+ "loss": 0.0774,
627
+ "step": 103
628
+ },
629
+ {
630
+ "epoch": 1.41,
631
+ "learning_rate": 0.00011528253892543053,
632
+ "loss": 0.0852,
633
+ "step": 104
634
+ },
635
+ {
636
+ "epoch": 1.42,
637
+ "learning_rate": 0.00011383691743415363,
638
+ "loss": 0.0701,
639
+ "step": 105
640
+ },
641
+ {
642
+ "epoch": 1.43,
643
+ "learning_rate": 0.0001123883416438748,
644
+ "loss": 0.0618,
645
+ "step": 106
646
+ },
647
+ {
648
+ "epoch": 1.45,
649
+ "learning_rate": 0.00011093712083778746,
650
+ "loss": 0.0669,
651
+ "step": 107
652
+ },
653
+ {
654
+ "epoch": 1.46,
655
+ "learning_rate": 0.00010948356486381827,
656
+ "loss": 0.103,
657
+ "step": 108
658
+ },
659
+ {
660
+ "epoch": 1.47,
661
+ "learning_rate": 0.00010802798406847212,
662
+ "loss": 0.0793,
663
+ "step": 109
664
+ },
665
+ {
666
+ "epoch": 1.49,
667
+ "learning_rate": 0.0001065706892305703,
668
+ "loss": 0.0837,
669
+ "step": 110
670
+ },
671
+ {
672
+ "epoch": 1.5,
673
+ "learning_rate": 0.00010511199149489673,
674
+ "loss": 0.0659,
675
+ "step": 111
676
+ },
677
+ {
678
+ "epoch": 1.51,
679
+ "learning_rate": 0.0001036522023057659,
680
+ "loss": 0.0945,
681
+ "step": 112
682
+ },
683
+ {
684
+ "epoch": 1.53,
685
+ "learning_rate": 0.00010219163334052682,
686
+ "loss": 0.0901,
687
+ "step": 113
688
+ },
689
+ {
690
+ "epoch": 1.54,
691
+ "learning_rate": 0.0001007305964430173,
692
+ "loss": 0.0867,
693
+ "step": 114
694
+ },
695
+ {
696
+ "epoch": 1.55,
697
+ "learning_rate": 9.92694035569827e-05,
698
+ "loss": 0.0752,
699
+ "step": 115
700
+ },
701
+ {
702
+ "epoch": 1.57,
703
+ "learning_rate": 9.780836665947319e-05,
704
+ "loss": 0.092,
705
+ "step": 116
706
+ },
707
+ {
708
+ "epoch": 1.58,
709
+ "learning_rate": 9.63477976942341e-05,
710
+ "loss": 0.084,
711
+ "step": 117
712
+ },
713
+ {
714
+ "epoch": 1.59,
715
+ "learning_rate": 9.48880085051033e-05,
716
+ "loss": 0.0732,
717
+ "step": 118
718
+ },
719
+ {
720
+ "epoch": 1.61,
721
+ "learning_rate": 9.342931076942972e-05,
722
+ "loss": 0.1045,
723
+ "step": 119
724
+ },
725
+ {
726
+ "epoch": 1.62,
727
+ "learning_rate": 9.19720159315279e-05,
728
+ "loss": 0.0955,
729
+ "step": 120
730
+ },
731
+ {
732
+ "epoch": 1.64,
733
+ "learning_rate": 9.051643513618175e-05,
734
+ "loss": 0.1003,
735
+ "step": 121
736
+ },
737
+ {
738
+ "epoch": 1.65,
739
+ "learning_rate": 8.906287916221259e-05,
740
+ "loss": 0.0901,
741
+ "step": 122
742
+ },
743
+ {
744
+ "epoch": 1.66,
745
+ "learning_rate": 8.76116583561252e-05,
746
+ "loss": 0.0732,
747
+ "step": 123
748
+ },
749
+ {
750
+ "epoch": 1.68,
751
+ "learning_rate": 8.616308256584637e-05,
752
+ "loss": 0.0852,
753
+ "step": 124
754
+ },
755
+ {
756
+ "epoch": 1.69,
757
+ "learning_rate": 8.47174610745695e-05,
758
+ "loss": 0.0989,
759
+ "step": 125
760
+ },
761
+ {
762
+ "epoch": 1.7,
763
+ "learning_rate": 8.327510253472023e-05,
764
+ "loss": 0.0701,
765
+ "step": 126
766
+ },
767
+ {
768
+ "epoch": 1.72,
769
+ "learning_rate": 8.183631490205637e-05,
770
+ "loss": 0.0898,
771
+ "step": 127
772
+ },
773
+ {
774
+ "epoch": 1.73,
775
+ "learning_rate": 8.040140536991687e-05,
776
+ "loss": 0.0867,
777
+ "step": 128
778
+ },
779
+ {
780
+ "epoch": 1.74,
781
+ "learning_rate": 7.89706803036334e-05,
782
+ "loss": 0.0901,
783
+ "step": 129
784
+ },
785
+ {
786
+ "epoch": 1.76,
787
+ "learning_rate": 7.75444451751187e-05,
788
+ "loss": 0.0697,
789
+ "step": 130
790
+ },
791
+ {
792
+ "epoch": 1.77,
793
+ "learning_rate": 7.61230044976458e-05,
794
+ "loss": 0.0854,
795
+ "step": 131
796
+ },
797
+ {
798
+ "epoch": 1.78,
799
+ "learning_rate": 7.470666176083192e-05,
800
+ "loss": 0.0779,
801
+ "step": 132
802
+ },
803
+ {
804
+ "epoch": 1.8,
805
+ "learning_rate": 7.329571936584072e-05,
806
+ "loss": 0.0852,
807
+ "step": 133
808
+ },
809
+ {
810
+ "epoch": 1.81,
811
+ "learning_rate": 7.189047856081719e-05,
812
+ "loss": 0.0747,
813
+ "step": 134
814
+ },
815
+ {
816
+ "epoch": 1.82,
817
+ "learning_rate": 7.049123937656855e-05,
818
+ "loss": 0.0928,
819
+ "step": 135
820
+ },
821
+ {
822
+ "epoch": 1.84,
823
+ "learning_rate": 6.909830056250527e-05,
824
+ "loss": 0.0786,
825
+ "step": 136
826
+ },
827
+ {
828
+ "epoch": 1.85,
829
+ "learning_rate": 6.77119595228554e-05,
830
+ "loss": 0.0933,
831
+ "step": 137
832
+ },
833
+ {
834
+ "epoch": 1.86,
835
+ "learning_rate": 6.63325122531663e-05,
836
+ "loss": 0.0662,
837
+ "step": 138
838
+ },
839
+ {
840
+ "epoch": 1.88,
841
+ "learning_rate": 6.496025327710707e-05,
842
+ "loss": 0.0847,
843
+ "step": 139
844
+ },
845
+ {
846
+ "epoch": 1.89,
847
+ "learning_rate": 6.359547558358533e-05,
848
+ "loss": 0.0808,
849
+ "step": 140
850
+ },
851
+ {
852
+ "epoch": 1.91,
853
+ "learning_rate": 6.223847056419154e-05,
854
+ "loss": 0.0776,
855
+ "step": 141
856
+ },
857
+ {
858
+ "epoch": 1.92,
859
+ "learning_rate": 6.0889527950984416e-05,
860
+ "loss": 0.1028,
861
+ "step": 142
862
+ },
863
+ {
864
+ "epoch": 1.93,
865
+ "learning_rate": 5.954893575463064e-05,
866
+ "loss": 0.0725,
867
+ "step": 143
868
+ },
869
+ {
870
+ "epoch": 1.95,
871
+ "learning_rate": 5.821698020291234e-05,
872
+ "loss": 0.072,
873
+ "step": 144
874
+ },
875
+ {
876
+ "epoch": 1.96,
877
+ "learning_rate": 5.689394567961477e-05,
878
+ "loss": 0.0823,
879
+ "step": 145
880
+ },
881
+ {
882
+ "epoch": 1.97,
883
+ "learning_rate": 5.558011466380823e-05,
884
+ "loss": 0.083,
885
+ "step": 146
886
+ },
887
+ {
888
+ "epoch": 1.99,
889
+ "learning_rate": 5.4275767669536146e-05,
890
+ "loss": 0.0752,
891
+ "step": 147
892
+ },
893
+ {
894
+ "epoch": 2.0,
895
+ "learning_rate": 5.2981183185923156e-05,
896
+ "loss": 0.0928,
897
+ "step": 148
898
+ },
899
+ {
900
+ "epoch": 2.01,
901
+ "learning_rate": 5.169663761771521e-05,
902
+ "loss": 0.0654,
903
+ "step": 149
904
+ },
905
+ {
906
+ "epoch": 2.03,
907
+ "learning_rate": 5.042240522626482e-05,
908
+ "loss": 0.0669,
909
+ "step": 150
910
+ },
911
+ {
912
+ "epoch": 2.04,
913
+ "learning_rate": 4.91587580709739e-05,
914
+ "loss": 0.079,
915
+ "step": 151
916
+ },
917
+ {
918
+ "epoch": 2.05,
919
+ "learning_rate": 4.790596595120699e-05,
920
+ "loss": 0.0623,
921
+ "step": 152
922
+ },
923
+ {
924
+ "epoch": 2.07,
925
+ "learning_rate": 4.666429634868651e-05,
926
+ "loss": 0.0542,
927
+ "step": 153
928
+ },
929
+ {
930
+ "epoch": 2.08,
931
+ "learning_rate": 4.543401437038335e-05,
932
+ "loss": 0.0703,
933
+ "step": 154
934
+ },
935
+ {
936
+ "epoch": 2.09,
937
+ "learning_rate": 4.4215382691914266e-05,
938
+ "loss": 0.0664,
939
+ "step": 155
940
+ },
941
+ {
942
+ "epoch": 2.11,
943
+ "learning_rate": 4.300866150145837e-05,
944
+ "loss": 0.0723,
945
+ "step": 156
946
+ },
947
+ {
948
+ "epoch": 2.12,
949
+ "learning_rate": 4.181410844420474e-05,
950
+ "loss": 0.0591,
951
+ "step": 157
952
+ },
953
+ {
954
+ "epoch": 2.14,
955
+ "learning_rate": 4.0631978567342946e-05,
956
+ "loss": 0.0537,
957
+ "step": 158
958
+ },
959
+ {
960
+ "epoch": 2.15,
961
+ "learning_rate": 3.946252426560855e-05,
962
+ "loss": 0.0554,
963
+ "step": 159
964
+ },
965
+ {
966
+ "epoch": 2.16,
967
+ "learning_rate": 3.8305995227394365e-05,
968
+ "loss": 0.0703,
969
+ "step": 160
970
+ },
971
+ {
972
+ "epoch": 2.18,
973
+ "learning_rate": 3.716263838144007e-05,
974
+ "loss": 0.0525,
975
+ "step": 161
976
+ },
977
+ {
978
+ "epoch": 2.19,
979
+ "learning_rate": 3.60326978441109e-05,
980
+ "loss": 0.0798,
981
+ "step": 162
982
+ },
983
+ {
984
+ "epoch": 2.2,
985
+ "learning_rate": 3.491641486727645e-05,
986
+ "loss": 0.054,
987
+ "step": 163
988
+ },
989
+ {
990
+ "epoch": 2.22,
991
+ "learning_rate": 3.3814027786801673e-05,
992
+ "loss": 0.0657,
993
+ "step": 164
994
+ },
995
+ {
996
+ "epoch": 2.23,
997
+ "learning_rate": 3.272577197166e-05,
998
+ "loss": 0.0546,
999
+ "step": 165
1000
+ },
1001
+ {
1002
+ "epoch": 2.24,
1003
+ "learning_rate": 3.1651879773680074e-05,
1004
+ "loss": 0.0648,
1005
+ "step": 166
1006
+ },
1007
+ {
1008
+ "epoch": 2.26,
1009
+ "learning_rate": 3.059258047793661e-05,
1010
+ "loss": 0.0575,
1011
+ "step": 167
1012
+ },
1013
+ {
1014
+ "epoch": 2.27,
1015
+ "learning_rate": 2.954810025379633e-05,
1016
+ "loss": 0.0476,
1017
+ "step": 168
1018
+ },
1019
+ {
1020
+ "epoch": 2.28,
1021
+ "learning_rate": 2.851866210662858e-05,
1022
+ "loss": 0.0433,
1023
+ "step": 169
1024
+ },
1025
+ {
1026
+ "epoch": 2.3,
1027
+ "learning_rate": 2.7504485830191984e-05,
1028
+ "loss": 0.0784,
1029
+ "step": 170
1030
+ },
1031
+ {
1032
+ "epoch": 2.31,
1033
+ "learning_rate": 2.650578795970661e-05,
1034
+ "loss": 0.0466,
1035
+ "step": 171
1036
+ },
1037
+ {
1038
+ "epoch": 2.32,
1039
+ "learning_rate": 2.5522781725621813e-05,
1040
+ "loss": 0.0612,
1041
+ "step": 172
1042
+ },
1043
+ {
1044
+ "epoch": 2.34,
1045
+ "learning_rate": 2.455567700808974e-05,
1046
+ "loss": 0.0498,
1047
+ "step": 173
1048
+ },
1049
+ {
1050
+ "epoch": 2.35,
1051
+ "learning_rate": 2.360468029215409e-05,
1052
+ "loss": 0.0426,
1053
+ "step": 174
1054
+ },
1055
+ {
1056
+ "epoch": 2.36,
1057
+ "learning_rate": 2.2669994623664005e-05,
1058
+ "loss": 0.049,
1059
+ "step": 175
1060
+ },
1061
+ {
1062
+ "epoch": 2.38,
1063
+ "learning_rate": 2.1751819565921773e-05,
1064
+ "loss": 0.047,
1065
+ "step": 176
1066
+ },
1067
+ {
1068
+ "epoch": 2.39,
1069
+ "learning_rate": 2.0850351157074598e-05,
1070
+ "loss": 0.0652,
1071
+ "step": 177
1072
+ },
1073
+ {
1074
+ "epoch": 2.41,
1075
+ "learning_rate": 1.996578186825876e-05,
1076
+ "loss": 0.0532,
1077
+ "step": 178
1078
+ },
1079
+ {
1080
+ "epoch": 2.42,
1081
+ "learning_rate": 1.9098300562505266e-05,
1082
+ "loss": 0.0613,
1083
+ "step": 179
1084
+ },
1085
+ {
1086
+ "epoch": 2.43,
1087
+ "learning_rate": 1.8248092454416165e-05,
1088
+ "loss": 0.0614,
1089
+ "step": 180
1090
+ },
1091
+ {
1092
+ "epoch": 2.45,
1093
+ "learning_rate": 1.7415339070619586e-05,
1094
+ "loss": 0.048,
1095
+ "step": 181
1096
+ },
1097
+ {
1098
+ "epoch": 2.46,
1099
+ "learning_rate": 1.660021821101222e-05,
1100
+ "loss": 0.0685,
1101
+ "step": 182
1102
+ },
1103
+ {
1104
+ "epoch": 2.47,
1105
+ "learning_rate": 1.5802903910797583e-05,
1106
+ "loss": 0.0409,
1107
+ "step": 183
1108
+ },
1109
+ {
1110
+ "epoch": 2.49,
1111
+ "learning_rate": 1.5023566403328104e-05,
1112
+ "loss": 0.0531,
1113
+ "step": 184
1114
+ },
1115
+ {
1116
+ "epoch": 2.5,
1117
+ "learning_rate": 1.4262372083758713e-05,
1118
+ "loss": 0.0634,
1119
+ "step": 185
1120
+ },
1121
+ {
1122
+ "epoch": 2.51,
1123
+ "learning_rate": 1.3519483473520122e-05,
1124
+ "loss": 0.0593,
1125
+ "step": 186
1126
+ },
1127
+ {
1128
+ "epoch": 2.53,
1129
+ "learning_rate": 1.2795059185619229e-05,
1130
+ "loss": 0.0687,
1131
+ "step": 187
1132
+ },
1133
+ {
1134
+ "epoch": 2.54,
1135
+ "learning_rate": 1.2089253890773789e-05,
1136
+ "loss": 0.0696,
1137
+ "step": 188
1138
+ },
1139
+ {
1140
+ "epoch": 2.55,
1141
+ "learning_rate": 1.1402218284388844e-05,
1142
+ "loss": 0.0641,
1143
+ "step": 189
1144
+ },
1145
+ {
1146
+ "epoch": 2.57,
1147
+ "learning_rate": 1.0734099054382185e-05,
1148
+ "loss": 0.059,
1149
+ "step": 190
1150
+ },
1151
+ {
1152
+ "epoch": 2.58,
1153
+ "learning_rate": 1.0085038849865025e-05,
1154
+ "loss": 0.0601,
1155
+ "step": 191
1156
+ },
1157
+ {
1158
+ "epoch": 2.59,
1159
+ "learning_rate": 9.455176250685338e-06,
1160
+ "loss": 0.0424,
1161
+ "step": 192
1162
+ },
1163
+ {
1164
+ "epoch": 2.61,
1165
+ "learning_rate": 8.844645737839873e-06,
1166
+ "loss": 0.0754,
1167
+ "step": 193
1168
+ },
1169
+ {
1170
+ "epoch": 2.62,
1171
+ "learning_rate": 8.253577664761258e-06,
1172
+ "loss": 0.0491,
1173
+ "step": 194
1174
+ },
1175
+ {
1176
+ "epoch": 2.64,
1177
+ "learning_rate": 7.682098229486479e-06,
1178
+ "loss": 0.0425,
1179
+ "step": 195
1180
+ },
1181
+ {
1182
+ "epoch": 2.65,
1183
+ "learning_rate": 7.130329447712581e-06,
1184
+ "loss": 0.0505,
1185
+ "step": 196
1186
+ },
1187
+ {
1188
+ "epoch": 2.66,
1189
+ "learning_rate": 6.598389126745208e-06,
1190
+ "loss": 0.0499,
1191
+ "step": 197
1192
+ },
1193
+ {
1194
+ "epoch": 2.68,
1195
+ "learning_rate": 6.086390840345757e-06,
1196
+ "loss": 0.0457,
1197
+ "step": 198
1198
+ },
1199
+ {
1200
+ "epoch": 2.69,
1201
+ "learning_rate": 5.594443904482438e-06,
1202
+ "loss": 0.0533,
1203
+ "step": 199
1204
+ },
1205
+ {
1206
+ "epoch": 2.7,
1207
+ "learning_rate": 5.122653353990436e-06,
1208
+ "loss": 0.0522,
1209
+ "step": 200
1210
+ },
1211
+ {
1212
+ "epoch": 2.72,
1213
+ "learning_rate": 4.671119920145983e-06,
1214
+ "loss": 0.0596,
1215
+ "step": 201
1216
+ },
1217
+ {
1218
+ "epoch": 2.73,
1219
+ "learning_rate": 4.2399400091594154e-06,
1220
+ "loss": 0.0486,
1221
+ "step": 202
1222
+ },
1223
+ {
1224
+ "epoch": 2.74,
1225
+ "learning_rate": 3.829205681591697e-06,
1226
+ "loss": 0.0425,
1227
+ "step": 203
1228
+ },
1229
+ {
1230
+ "epoch": 2.76,
1231
+ "learning_rate": 3.4390046326986504e-06,
1232
+ "loss": 0.0455,
1233
+ "step": 204
1234
+ },
1235
+ {
1236
+ "epoch": 2.77,
1237
+ "learning_rate": 3.0694201737072492e-06,
1238
+ "loss": 0.0505,
1239
+ "step": 205
1240
+ },
1241
+ {
1242
+ "epoch": 2.78,
1243
+ "learning_rate": 2.720531214028055e-06,
1244
+ "loss": 0.0503,
1245
+ "step": 206
1246
+ },
1247
+ {
1248
+ "epoch": 2.8,
1249
+ "learning_rate": 2.392412244407294e-06,
1250
+ "loss": 0.0674,
1251
+ "step": 207
1252
+ },
1253
+ {
1254
+ "epoch": 2.81,
1255
+ "learning_rate": 2.0851333210225033e-06,
1256
+ "loss": 0.0629,
1257
+ "step": 208
1258
+ },
1259
+ {
1260
+ "epoch": 2.82,
1261
+ "learning_rate": 1.7987600505249724e-06,
1262
+ "loss": 0.0491,
1263
+ "step": 209
1264
+ },
1265
+ {
1266
+ "epoch": 2.84,
1267
+ "learning_rate": 1.5333535760320928e-06,
1268
+ "loss": 0.0413,
1269
+ "step": 210
1270
+ },
1271
+ {
1272
+ "epoch": 2.85,
1273
+ "learning_rate": 1.2889705640728444e-06,
1274
+ "loss": 0.0515,
1275
+ "step": 211
1276
+ },
1277
+ {
1278
+ "epoch": 2.86,
1279
+ "learning_rate": 1.0656631924889749e-06,
1280
+ "loss": 0.0476,
1281
+ "step": 212
1282
+ },
1283
+ {
1284
+ "epoch": 2.88,
1285
+ "learning_rate": 8.634791392946428e-07,
1286
+ "loss": 0.0555,
1287
+ "step": 213
1288
+ },
1289
+ {
1290
+ "epoch": 2.89,
1291
+ "learning_rate": 6.824615724966843e-07,
1292
+ "loss": 0.0582,
1293
+ "step": 214
1294
+ },
1295
+ {
1296
+ "epoch": 2.91,
1297
+ "learning_rate": 5.226491408779288e-07,
1298
+ "loss": 0.0586,
1299
+ "step": 215
1300
+ },
1301
+ {
1302
+ "epoch": 2.92,
1303
+ "learning_rate": 3.840759657453452e-07,
1304
+ "loss": 0.0559,
1305
+ "step": 216
1306
+ },
1307
+ {
1308
+ "epoch": 2.93,
1309
+ "learning_rate": 2.667716336448356e-07,
1310
+ "loss": 0.0583,
1311
+ "step": 217
1312
+ },
1313
+ {
1314
+ "epoch": 2.95,
1315
+ "learning_rate": 1.707611900442996e-07,
1316
+ "loss": 0.0441,
1317
+ "step": 218
1318
+ },
1319
+ {
1320
+ "epoch": 2.96,
1321
+ "learning_rate": 9.606513398617845e-08,
1322
+ "loss": 0.0754,
1323
+ "step": 219
1324
+ },
1325
+ {
1326
+ "epoch": 2.97,
1327
+ "learning_rate": 4.269941371073394e-08,
1328
+ "loss": 0.0579,
1329
+ "step": 220
1330
+ },
1331
+ {
1332
+ "epoch": 2.99,
1333
+ "learning_rate": 1.0675423250994243e-08,
1334
+ "loss": 0.0377,
1335
+ "step": 221
1336
+ },
1337
+ {
1338
+ "epoch": 3.0,
1339
+ "learning_rate": 0.0,
1340
+ "loss": 0.0557,
1341
+ "step": 222
1342
+ },
1343
+ {
1344
+ "epoch": 3.0,
1345
+ "step": 222,
1346
+ "total_flos": 5266094948352.0,
1347
+ "train_loss": 0.08446495812218469,
1348
+ "train_runtime": 2892.0758,
1349
+ "train_samples_per_second": 4.87,
1350
+ "train_steps_per_second": 0.077
1351
+ }
1352
+ ],
1353
+ "logging_steps": 1.0,
1354
+ "max_steps": 222,
1355
+ "num_input_tokens_seen": 0,
1356
+ "num_train_epochs": 3,
1357
+ "save_steps": 800,
1358
+ "total_flos": 5266094948352.0,
1359
+ "train_batch_size": 32,
1360
+ "trial_name": null,
1361
+ "trial_params": null
1362
+ }