kkeezz commited on
Commit
eaa552b
·
verified ·
1 Parent(s): f5efa19

Upload 5 files

Browse files
adapter_config.json ADDED
@@ -0,0 +1,83 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "MAGAer13/mplug-owl2-llama2-7b",
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.10.self_attn.v_proj.multiway.1",
18
+ "model.layers.7.self_attn.v_proj.multiway.1",
19
+ "model.layers.19.self_attn.q_proj",
20
+ "model.layers.21.self_attn.q_proj",
21
+ "model.layers.4.self_attn.v_proj.multiway.1",
22
+ "model.layers.6.self_attn.v_proj.multiway.1",
23
+ "model.layers.25.self_attn.q_proj",
24
+ "model.layers.13.self_attn.v_proj.multiway.1",
25
+ "model.layers.29.self_attn.q_proj",
26
+ "model.layers.30.self_attn.v_proj.multiway.1",
27
+ "model.layers.18.self_attn.v_proj.multiway.1",
28
+ "model.layers.0.self_attn.v_proj.multiway.1",
29
+ "model.layers.12.self_attn.v_proj.multiway.1",
30
+ "model.layers.16.self_attn.q_proj",
31
+ "model.layers.2.self_attn.v_proj.multiway.1",
32
+ "model.layers.27.self_attn.v_proj.multiway.1",
33
+ "model.layers.5.self_attn.q_proj",
34
+ "model.layers.22.self_attn.v_proj.multiway.1",
35
+ "model.layers.1.self_attn.q_proj",
36
+ "model.layers.17.self_attn.v_proj.multiway.1",
37
+ "model.layers.12.self_attn.q_proj",
38
+ "model.layers.18.self_attn.q_proj",
39
+ "model.layers.31.self_attn.q_proj",
40
+ "model.layers.10.self_attn.q_proj",
41
+ "model.layers.14.self_attn.q_proj",
42
+ "model.layers.11.self_attn.q_proj",
43
+ "model.layers.27.self_attn.q_proj",
44
+ "model.layers.6.self_attn.q_proj",
45
+ "model.layers.9.self_attn.v_proj.multiway.1",
46
+ "model.layers.8.self_attn.v_proj.multiway.1",
47
+ "model.layers.26.self_attn.v_proj.multiway.1",
48
+ "model.layers.23.self_attn.v_proj.multiway.1",
49
+ "model.layers.20.self_attn.v_proj.multiway.1",
50
+ "model.layers.30.self_attn.q_proj",
51
+ "model.layers.15.self_attn.v_proj.multiway.1",
52
+ "model.layers.9.self_attn.q_proj",
53
+ "model.layers.11.self_attn.v_proj.multiway.1",
54
+ "model.layers.23.self_attn.q_proj",
55
+ "model.layers.1.self_attn.v_proj.multiway.1",
56
+ "model.layers.16.self_attn.v_proj.multiway.1",
57
+ "model.layers.21.self_attn.v_proj.multiway.1",
58
+ "model.layers.5.self_attn.v_proj.multiway.1",
59
+ "model.layers.3.self_attn.q_proj",
60
+ "model.layers.14.self_attn.v_proj.multiway.1",
61
+ "model.layers.25.self_attn.v_proj.multiway.1",
62
+ "model.layers.15.self_attn.q_proj",
63
+ "model.layers.17.self_attn.q_proj",
64
+ "model.layers.28.self_attn.v_proj.multiway.1",
65
+ "model.layers.13.self_attn.q_proj",
66
+ "model.layers.20.self_attn.q_proj",
67
+ "model.layers.24.self_attn.q_proj",
68
+ "model.layers.19.self_attn.v_proj.multiway.1",
69
+ "model.layers.26.self_attn.q_proj",
70
+ "model.layers.28.self_attn.q_proj",
71
+ "model.layers.31.self_attn.v_proj.multiway.1",
72
+ "model.layers.24.self_attn.v_proj.multiway.1",
73
+ "model.layers.7.self_attn.q_proj",
74
+ "model.layers.22.self_attn.q_proj",
75
+ "model.layers.2.self_attn.q_proj",
76
+ "model.layers.8.self_attn.q_proj",
77
+ "model.layers.0.self_attn.q_proj",
78
+ "model.layers.29.self_attn.v_proj.multiway.1",
79
+ "model.layers.3.self_attn.v_proj.multiway.1",
80
+ "model.layers.4.self_attn.q_proj"
81
+ ],
82
+ "task_type": "CAUSAL_LM"
83
+ }
config.json ADDED
@@ -0,0 +1,188 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "MAGAer13/mplug-owl2-llama2-7b",
3
+ "architectures": [
4
+ "MPLUGOwl2LlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "eos_token_id": 2,
10
+ "freeze_vision_model": false,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 4096,
13
+ "image_aspect_ratio": "pad",
14
+ "image_grid_pinpoints": null,
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 11008,
17
+ "max_position_embeddings": 2048,
18
+ "model_type": "mplug_owl2",
19
+ "num_attention_heads": 32,
20
+ "num_beams": 3,
21
+ "num_hidden_layers": 32,
22
+ "num_key_value_heads": 32,
23
+ "pretraining_tp": 1,
24
+ "rms_norm_eps": 1e-06,
25
+ "rope_scaling": null,
26
+ "rope_theta": 10000.0,
27
+ "tie_word_embeddings": false,
28
+ "torch_dtype": "float32",
29
+ "transformers_version": "4.36.1",
30
+ "tune_visual_abstractor": true,
31
+ "use_cache": true,
32
+ "visual_abstractor_lr": 2e-05,
33
+ "visual_config": {
34
+ "visual_abstractor": {
35
+ "_name_or_path": "",
36
+ "add_cross_attention": false,
37
+ "architectures": null,
38
+ "attention_probs_dropout_prob": 0.0,
39
+ "bad_words_ids": null,
40
+ "begin_suppress_tokens": null,
41
+ "bos_token_id": null,
42
+ "chunk_size_feed_forward": 0,
43
+ "cross_attention_hidden_size": null,
44
+ "decoder_start_token_id": null,
45
+ "diversity_penalty": 0.0,
46
+ "do_sample": false,
47
+ "early_stopping": false,
48
+ "encoder_hidden_size": 1024,
49
+ "encoder_no_repeat_ngram_size": 0,
50
+ "eos_token_id": null,
51
+ "exponential_decay_length_penalty": null,
52
+ "finetuning_task": null,
53
+ "forced_bos_token_id": null,
54
+ "forced_eos_token_id": null,
55
+ "grid_size": 32,
56
+ "hidden_size": 1024,
57
+ "id2label": {
58
+ "0": "LABEL_0",
59
+ "1": "LABEL_1"
60
+ },
61
+ "initializer_range": 0.02,
62
+ "intermediate_size": 2816,
63
+ "is_decoder": false,
64
+ "is_encoder_decoder": false,
65
+ "label2id": {
66
+ "LABEL_0": 0,
67
+ "LABEL_1": 1
68
+ },
69
+ "layer_norm_eps": 1e-06,
70
+ "length_penalty": 1.0,
71
+ "max_length": 20,
72
+ "min_length": 0,
73
+ "model_type": "mplug_owl_visual_abstract",
74
+ "no_repeat_ngram_size": 0,
75
+ "num_attention_heads": 16,
76
+ "num_beam_groups": 1,
77
+ "num_beams": 1,
78
+ "num_hidden_layers": 6,
79
+ "num_learnable_queries": 64,
80
+ "num_return_sequences": 1,
81
+ "output_attentions": false,
82
+ "output_hidden_states": false,
83
+ "output_scores": false,
84
+ "pad_token_id": null,
85
+ "prefix": null,
86
+ "problem_type": null,
87
+ "pruned_heads": {},
88
+ "remove_invalid_values": false,
89
+ "repetition_penalty": 1.0,
90
+ "return_dict": true,
91
+ "return_dict_in_generate": false,
92
+ "sep_token_id": null,
93
+ "suppress_tokens": null,
94
+ "task_specific_params": null,
95
+ "temperature": 1.0,
96
+ "tf_legacy_loss": false,
97
+ "tie_encoder_decoder": false,
98
+ "tie_word_embeddings": true,
99
+ "tokenizer_class": null,
100
+ "top_k": 50,
101
+ "top_p": 1.0,
102
+ "torch_dtype": null,
103
+ "torchscript": false,
104
+ "transformers_version": "4.28.1",
105
+ "typical_p": 1.0,
106
+ "use_bfloat16": false
107
+ },
108
+ "visual_model": {
109
+ "_name_or_path": "",
110
+ "add_cross_attention": false,
111
+ "architectures": null,
112
+ "attention_dropout": 0.0,
113
+ "bad_words_ids": null,
114
+ "begin_suppress_tokens": null,
115
+ "bos_token_id": null,
116
+ "chunk_size_feed_forward": 0,
117
+ "cross_attention_hidden_size": null,
118
+ "decoder_start_token_id": null,
119
+ "diversity_penalty": 0.0,
120
+ "do_sample": false,
121
+ "early_stopping": false,
122
+ "encoder_no_repeat_ngram_size": 0,
123
+ "eos_token_id": null,
124
+ "exponential_decay_length_penalty": null,
125
+ "finetuning_task": null,
126
+ "forced_bos_token_id": null,
127
+ "forced_eos_token_id": null,
128
+ "hidden_act": "quick_gelu",
129
+ "hidden_size": 1024,
130
+ "id2label": {
131
+ "0": "LABEL_0",
132
+ "1": "LABEL_1"
133
+ },
134
+ "image_size": 448,
135
+ "initializer_factor": 1.0,
136
+ "initializer_range": 0.02,
137
+ "intermediate_size": 4096,
138
+ "is_decoder": false,
139
+ "is_encoder_decoder": false,
140
+ "label2id": {
141
+ "LABEL_0": 0,
142
+ "LABEL_1": 1
143
+ },
144
+ "layer_norm_eps": 1e-06,
145
+ "length_penalty": 1.0,
146
+ "max_length": 20,
147
+ "min_length": 0,
148
+ "model_type": "mplug_owl_vision_model",
149
+ "no_repeat_ngram_size": 0,
150
+ "num_attention_heads": 16,
151
+ "num_beam_groups": 1,
152
+ "num_beams": 1,
153
+ "num_channels": 3,
154
+ "num_hidden_layers": 24,
155
+ "num_return_sequences": 1,
156
+ "output_attentions": false,
157
+ "output_hidden_states": false,
158
+ "output_scores": false,
159
+ "pad_token_id": null,
160
+ "patch_size": 14,
161
+ "prefix": null,
162
+ "problem_type": null,
163
+ "projection_dim": 768,
164
+ "pruned_heads": {},
165
+ "remove_invalid_values": false,
166
+ "repetition_penalty": 1.0,
167
+ "return_dict": true,
168
+ "return_dict_in_generate": false,
169
+ "sep_token_id": null,
170
+ "suppress_tokens": null,
171
+ "task_specific_params": null,
172
+ "temperature": 1.0,
173
+ "tf_legacy_loss": false,
174
+ "tie_encoder_decoder": false,
175
+ "tie_word_embeddings": true,
176
+ "tokenizer_class": null,
177
+ "top_k": 50,
178
+ "top_p": 1.0,
179
+ "torch_dtype": null,
180
+ "torchscript": false,
181
+ "transformers_version": "4.28.1",
182
+ "typical_p": 1.0,
183
+ "use_bfloat16": false,
184
+ "use_flash_attn": false
185
+ }
186
+ },
187
+ "vocab_size": 32000
188
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e44ce263e6fd885f50d82ca515b9325375b43ee36ededb75acf161ce88bc2e41
3
+ size 48
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": 0.05011286681715576,
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": 0.0,
13
+ "learning_rate": 1.1428571428571427e-07,
14
+ "loss": 0.0007,
15
+ "step": 1
16
+ },
17
+ {
18
+ "epoch": 0.0,
19
+ "learning_rate": 2.2857142857142855e-07,
20
+ "loss": -0.0016,
21
+ "step": 2
22
+ },
23
+ {
24
+ "epoch": 0.0,
25
+ "learning_rate": 3.428571428571428e-07,
26
+ "loss": -0.0009,
27
+ "step": 3
28
+ },
29
+ {
30
+ "epoch": 0.0,
31
+ "learning_rate": 4.571428571428571e-07,
32
+ "loss": -0.0002,
33
+ "step": 4
34
+ },
35
+ {
36
+ "epoch": 0.0,
37
+ "learning_rate": 5.714285714285714e-07,
38
+ "loss": 0.0005,
39
+ "step": 5
40
+ },
41
+ {
42
+ "epoch": 0.0,
43
+ "learning_rate": 6.857142857142856e-07,
44
+ "loss": -0.0004,
45
+ "step": 6
46
+ },
47
+ {
48
+ "epoch": 0.0,
49
+ "learning_rate": 8e-07,
50
+ "loss": -0.0013,
51
+ "step": 7
52
+ },
53
+ {
54
+ "epoch": 0.0,
55
+ "learning_rate": 7.999572983069959e-07,
56
+ "loss": -0.0003,
57
+ "step": 8
58
+ },
59
+ {
60
+ "epoch": 0.0,
61
+ "learning_rate": 7.99829202345157e-07,
62
+ "loss": -0.0009,
63
+ "step": 9
64
+ },
65
+ {
66
+ "epoch": 0.0,
67
+ "learning_rate": 7.996157394640553e-07,
68
+ "loss": -0.0003,
69
+ "step": 10
70
+ },
71
+ {
72
+ "epoch": 0.0,
73
+ "learning_rate": 7.993169552398228e-07,
74
+ "loss": -0.0001,
75
+ "step": 11
76
+ },
77
+ {
78
+ "epoch": 0.0,
79
+ "learning_rate": 7.989329134654206e-07,
80
+ "loss": -0.0005,
81
+ "step": 12
82
+ },
83
+ {
84
+ "epoch": 0.0,
85
+ "learning_rate": 7.984636961370187e-07,
86
+ "loss": -0.0007,
87
+ "step": 13
88
+ },
89
+ {
90
+ "epoch": 0.0,
91
+ "learning_rate": 7.979094034364882e-07,
92
+ "loss": 0.0007,
93
+ "step": 14
94
+ },
95
+ {
96
+ "epoch": 0.0,
97
+ "learning_rate": 7.972701537100132e-07,
98
+ "loss": -0.0001,
99
+ "step": 15
100
+ },
101
+ {
102
+ "epoch": 0.0,
103
+ "learning_rate": 7.965460834428213e-07,
104
+ "loss": -0.0011,
105
+ "step": 16
106
+ },
107
+ {
108
+ "epoch": 0.0,
109
+ "learning_rate": 7.957373472300441e-07,
110
+ "loss": -0.0019,
111
+ "step": 17
112
+ },
113
+ {
114
+ "epoch": 0.0,
115
+ "learning_rate": 7.948441177437086e-07,
116
+ "loss": -0.0009,
117
+ "step": 18
118
+ },
119
+ {
120
+ "epoch": 0.0,
121
+ "learning_rate": 7.938665856958715e-07,
122
+ "loss": 0.0006,
123
+ "step": 19
124
+ },
125
+ {
126
+ "epoch": 0.0,
127
+ "learning_rate": 7.928049597979001e-07,
128
+ "loss": -0.0015,
129
+ "step": 20
130
+ },
131
+ {
132
+ "epoch": 0.0,
133
+ "learning_rate": 7.916594667159099e-07,
134
+ "loss": -0.0008,
135
+ "step": 21
136
+ },
137
+ {
138
+ "epoch": 0.0,
139
+ "learning_rate": 7.904303510223708e-07,
140
+ "loss": -0.0003,
141
+ "step": 22
142
+ },
143
+ {
144
+ "epoch": 0.01,
145
+ "learning_rate": 7.891178751438877e-07,
146
+ "loss": 0.0001,
147
+ "step": 23
148
+ },
149
+ {
150
+ "epoch": 0.01,
151
+ "learning_rate": 7.87722319305171e-07,
152
+ "loss": -0.0004,
153
+ "step": 24
154
+ },
155
+ {
156
+ "epoch": 0.01,
157
+ "learning_rate": 7.862439814692054e-07,
158
+ "loss": -0.0007,
159
+ "step": 25
160
+ },
161
+ {
162
+ "epoch": 0.01,
163
+ "learning_rate": 7.846831772736331e-07,
164
+ "loss": -0.0009,
165
+ "step": 26
166
+ },
167
+ {
168
+ "epoch": 0.01,
169
+ "learning_rate": 7.830402399633623e-07,
170
+ "loss": -0.0017,
171
+ "step": 27
172
+ },
173
+ {
174
+ "epoch": 0.01,
175
+ "learning_rate": 7.813155203194161e-07,
176
+ "loss": -0.0016,
177
+ "step": 28
178
+ },
179
+ {
180
+ "epoch": 0.01,
181
+ "learning_rate": 7.795093865840382e-07,
182
+ "loss": -0.0009,
183
+ "step": 29
184
+ },
185
+ {
186
+ "epoch": 0.01,
187
+ "learning_rate": 7.776222243820702e-07,
188
+ "loss": -0.001,
189
+ "step": 30
190
+ },
191
+ {
192
+ "epoch": 0.01,
193
+ "learning_rate": 7.75654436638617e-07,
194
+ "loss": -0.0018,
195
+ "step": 31
196
+ },
197
+ {
198
+ "epoch": 0.01,
199
+ "learning_rate": 7.736064434930192e-07,
200
+ "loss": -0.0006,
201
+ "step": 32
202
+ },
203
+ {
204
+ "epoch": 0.01,
205
+ "learning_rate": 7.714786822091497e-07,
206
+ "loss": -0.0024,
207
+ "step": 33
208
+ },
209
+ {
210
+ "epoch": 0.01,
211
+ "learning_rate": 7.692716070820541e-07,
212
+ "loss": -0.0011,
213
+ "step": 34
214
+ },
215
+ {
216
+ "epoch": 0.01,
217
+ "learning_rate": 7.66985689340955e-07,
218
+ "loss": -0.0024,
219
+ "step": 35
220
+ },
221
+ {
222
+ "epoch": 0.01,
223
+ "learning_rate": 7.646214170486406e-07,
224
+ "loss": -0.0011,
225
+ "step": 36
226
+ },
227
+ {
228
+ "epoch": 0.01,
229
+ "learning_rate": 7.621792949972586e-07,
230
+ "loss": -0.0007,
231
+ "step": 37
232
+ },
233
+ {
234
+ "epoch": 0.01,
235
+ "learning_rate": 7.596598446005399e-07,
236
+ "loss": -0.0017,
237
+ "step": 38
238
+ },
239
+ {
240
+ "epoch": 0.01,
241
+ "learning_rate": 7.570636037824713e-07,
242
+ "loss": 0.001,
243
+ "step": 39
244
+ },
245
+ {
246
+ "epoch": 0.01,
247
+ "learning_rate": 7.543911268624445e-07,
248
+ "loss": -0.001,
249
+ "step": 40
250
+ },
251
+ {
252
+ "epoch": 0.01,
253
+ "learning_rate": 7.516429844369049e-07,
254
+ "loss": -0.0003,
255
+ "step": 41
256
+ },
257
+ {
258
+ "epoch": 0.01,
259
+ "learning_rate": 7.48819763257523e-07,
260
+ "loss": -0.0002,
261
+ "step": 42
262
+ },
263
+ {
264
+ "epoch": 0.01,
265
+ "learning_rate": 7.459220661059195e-07,
266
+ "loss": -0.0007,
267
+ "step": 43
268
+ },
269
+ {
270
+ "epoch": 0.01,
271
+ "learning_rate": 7.429505116649652e-07,
272
+ "loss": 0.0002,
273
+ "step": 44
274
+ },
275
+ {
276
+ "epoch": 0.01,
277
+ "learning_rate": 7.399057343866876e-07,
278
+ "loss": -0.0021,
279
+ "step": 45
280
+ },
281
+ {
282
+ "epoch": 0.01,
283
+ "learning_rate": 7.367883843568095e-07,
284
+ "loss": -0.0019,
285
+ "step": 46
286
+ },
287
+ {
288
+ "epoch": 0.01,
289
+ "learning_rate": 7.335991271559511e-07,
290
+ "loss": 0.0003,
291
+ "step": 47
292
+ },
293
+ {
294
+ "epoch": 0.01,
295
+ "learning_rate": 7.303386437175216e-07,
296
+ "loss": -0.0016,
297
+ "step": 48
298
+ },
299
+ {
300
+ "epoch": 0.01,
301
+ "learning_rate": 7.270076301823353e-07,
302
+ "loss": -0.0018,
303
+ "step": 49
304
+ },
305
+ {
306
+ "epoch": 0.01,
307
+ "learning_rate": 7.23606797749979e-07,
308
+ "loss": -0.0007,
309
+ "step": 50
310
+ },
311
+ {
312
+ "epoch": 0.01,
313
+ "learning_rate": 7.201368725269651e-07,
314
+ "loss": -0.0004,
315
+ "step": 51
316
+ },
317
+ {
318
+ "epoch": 0.01,
319
+ "learning_rate": 7.165985953717016e-07,
320
+ "loss": -0.0012,
321
+ "step": 52
322
+ },
323
+ {
324
+ "epoch": 0.01,
325
+ "learning_rate": 7.12992721736313e-07,
326
+ "loss": -0.0005,
327
+ "step": 53
328
+ },
329
+ {
330
+ "epoch": 0.01,
331
+ "learning_rate": 7.093200215053441e-07,
332
+ "loss": -0.0008,
333
+ "step": 54
334
+ },
335
+ {
336
+ "epoch": 0.01,
337
+ "learning_rate": 7.055812788313836e-07,
338
+ "loss": -0.0008,
339
+ "step": 55
340
+ },
341
+ {
342
+ "epoch": 0.01,
343
+ "learning_rate": 7.01777291967641e-07,
344
+ "loss": 0.0004,
345
+ "step": 56
346
+ },
347
+ {
348
+ "epoch": 0.01,
349
+ "learning_rate": 6.979088730975127e-07,
350
+ "loss": -0.0011,
351
+ "step": 57
352
+ },
353
+ {
354
+ "epoch": 0.01,
355
+ "learning_rate": 6.939768481611736e-07,
356
+ "loss": -0.0003,
357
+ "step": 58
358
+ },
359
+ {
360
+ "epoch": 0.01,
361
+ "learning_rate": 6.899820566792321e-07,
362
+ "loss": -0.0006,
363
+ "step": 59
364
+ },
365
+ {
366
+ "epoch": 0.01,
367
+ "learning_rate": 6.859253515734858e-07,
368
+ "loss": -0.001,
369
+ "step": 60
370
+ },
371
+ {
372
+ "epoch": 0.01,
373
+ "learning_rate": 6.818075989848146e-07,
374
+ "loss": -0.0008,
375
+ "step": 61
376
+ },
377
+ {
378
+ "epoch": 0.01,
379
+ "learning_rate": 6.776296780882536e-07,
380
+ "loss": -0.0009,
381
+ "step": 62
382
+ },
383
+ {
384
+ "epoch": 0.01,
385
+ "learning_rate": 6.733924809052798e-07,
386
+ "loss": -0.0007,
387
+ "step": 63
388
+ },
389
+ {
390
+ "epoch": 0.01,
391
+ "learning_rate": 6.6909691211336e-07,
392
+ "loss": -0.0007,
393
+ "step": 64
394
+ },
395
+ {
396
+ "epoch": 0.01,
397
+ "learning_rate": 6.647438888527933e-07,
398
+ "loss": -0.0011,
399
+ "step": 65
400
+ },
401
+ {
402
+ "epoch": 0.01,
403
+ "learning_rate": 6.603343405308942e-07,
404
+ "loss": -0.0007,
405
+ "step": 66
406
+ },
407
+ {
408
+ "epoch": 0.02,
409
+ "learning_rate": 6.558692086235565e-07,
410
+ "loss": -0.0011,
411
+ "step": 67
412
+ },
413
+ {
414
+ "epoch": 0.02,
415
+ "learning_rate": 6.513494464742397e-07,
416
+ "loss": -0.0001,
417
+ "step": 68
418
+ },
419
+ {
420
+ "epoch": 0.02,
421
+ "learning_rate": 6.467760190904226e-07,
422
+ "loss": -0.0017,
423
+ "step": 69
424
+ },
425
+ {
426
+ "epoch": 0.02,
427
+ "learning_rate": 6.421499029375658e-07,
428
+ "loss": -0.0009,
429
+ "step": 70
430
+ },
431
+ {
432
+ "epoch": 0.02,
433
+ "learning_rate": 6.374720857306283e-07,
434
+ "loss": 0.0003,
435
+ "step": 71
436
+ },
437
+ {
438
+ "epoch": 0.02,
439
+ "learning_rate": 6.327435662231812e-07,
440
+ "loss": -0.0018,
441
+ "step": 72
442
+ },
443
+ {
444
+ "epoch": 0.02,
445
+ "learning_rate": 6.279653539941665e-07,
446
+ "loss": -0.0006,
447
+ "step": 73
448
+ },
449
+ {
450
+ "epoch": 0.02,
451
+ "learning_rate": 6.231384692323428e-07,
452
+ "loss": -0.0002,
453
+ "step": 74
454
+ },
455
+ {
456
+ "epoch": 0.02,
457
+ "learning_rate": 6.182639425184666e-07,
458
+ "loss": -0.0009,
459
+ "step": 75
460
+ },
461
+ {
462
+ "epoch": 0.02,
463
+ "learning_rate": 6.133428146052541e-07,
464
+ "loss": -0.0007,
465
+ "step": 76
466
+ },
467
+ {
468
+ "epoch": 0.02,
469
+ "learning_rate": 6.083761361951721e-07,
470
+ "loss": 0.0001,
471
+ "step": 77
472
+ },
473
+ {
474
+ "epoch": 0.02,
475
+ "learning_rate": 6.033649677161044e-07,
476
+ "loss": -0.0009,
477
+ "step": 78
478
+ },
479
+ {
480
+ "epoch": 0.02,
481
+ "learning_rate": 5.983103790949407e-07,
482
+ "loss": -0.0,
483
+ "step": 79
484
+ },
485
+ {
486
+ "epoch": 0.02,
487
+ "learning_rate": 5.932134495291391e-07,
488
+ "loss": 0.0005,
489
+ "step": 80
490
+ },
491
+ {
492
+ "epoch": 0.02,
493
+ "learning_rate": 5.880752672563073e-07,
494
+ "loss": -0.0008,
495
+ "step": 81
496
+ },
497
+ {
498
+ "epoch": 0.02,
499
+ "learning_rate": 5.828969293218554e-07,
500
+ "loss": -0.0006,
501
+ "step": 82
502
+ },
503
+ {
504
+ "epoch": 0.02,
505
+ "learning_rate": 5.776795413447672e-07,
506
+ "loss": -0.0008,
507
+ "step": 83
508
+ },
509
+ {
510
+ "epoch": 0.02,
511
+ "learning_rate": 5.72424217281541e-07,
512
+ "loss": -0.0005,
513
+ "step": 84
514
+ },
515
+ {
516
+ "epoch": 0.02,
517
+ "learning_rate": 5.671320791883507e-07,
518
+ "loss": -0.0013,
519
+ "step": 85
520
+ },
521
+ {
522
+ "epoch": 0.02,
523
+ "learning_rate": 5.618042569814775e-07,
524
+ "loss": -0.0006,
525
+ "step": 86
526
+ },
527
+ {
528
+ "epoch": 0.02,
529
+ "learning_rate": 5.564418881960623e-07,
530
+ "loss": -0.0007,
531
+ "step": 87
532
+ },
533
+ {
534
+ "epoch": 0.02,
535
+ "learning_rate": 5.510461177432338e-07,
536
+ "loss": -0.0006,
537
+ "step": 88
538
+ },
539
+ {
540
+ "epoch": 0.02,
541
+ "learning_rate": 5.456180976656585e-07,
542
+ "loss": -0.0007,
543
+ "step": 89
544
+ },
545
+ {
546
+ "epoch": 0.02,
547
+ "learning_rate": 5.401589868915718e-07,
548
+ "loss": 0.0001,
549
+ "step": 90
550
+ },
551
+ {
552
+ "epoch": 0.02,
553
+ "learning_rate": 5.346699509873349e-07,
554
+ "loss": -0.0007,
555
+ "step": 91
556
+ },
557
+ {
558
+ "epoch": 0.02,
559
+ "learning_rate": 5.291521619085784e-07,
560
+ "loss": -0.0007,
561
+ "step": 92
562
+ },
563
+ {
564
+ "epoch": 0.02,
565
+ "learning_rate": 5.23606797749979e-07,
566
+ "loss": -0.0002,
567
+ "step": 93
568
+ },
569
+ {
570
+ "epoch": 0.02,
571
+ "learning_rate": 5.180350424937258e-07,
572
+ "loss": 0.0002,
573
+ "step": 94
574
+ },
575
+ {
576
+ "epoch": 0.02,
577
+ "learning_rate": 5.124380857567313e-07,
578
+ "loss": -0.0006,
579
+ "step": 95
580
+ },
581
+ {
582
+ "epoch": 0.02,
583
+ "learning_rate": 5.068171225366372e-07,
584
+ "loss": -0.0015,
585
+ "step": 96
586
+ },
587
+ {
588
+ "epoch": 0.02,
589
+ "learning_rate": 5.011733529566722e-07,
590
+ "loss": -0.0004,
591
+ "step": 97
592
+ },
593
+ {
594
+ "epoch": 0.02,
595
+ "learning_rate": 4.955079820094167e-07,
596
+ "loss": -0.001,
597
+ "step": 98
598
+ },
599
+ {
600
+ "epoch": 0.02,
601
+ "learning_rate": 4.898222192995253e-07,
602
+ "loss": 0.0002,
603
+ "step": 99
604
+ },
605
+ {
606
+ "epoch": 0.02,
607
+ "learning_rate": 4.841172787854665e-07,
608
+ "loss": -0.0012,
609
+ "step": 100
610
+ },
611
+ {
612
+ "epoch": 0.02,
613
+ "learning_rate": 4.783943785203326e-07,
614
+ "loss": -0.0026,
615
+ "step": 101
616
+ },
617
+ {
618
+ "epoch": 0.02,
619
+ "learning_rate": 4.726547403917745e-07,
620
+ "loss": -0.0004,
621
+ "step": 102
622
+ },
623
+ {
624
+ "epoch": 0.02,
625
+ "learning_rate": 4.6689958986111913e-07,
626
+ "loss": -0.0005,
627
+ "step": 103
628
+ },
629
+ {
630
+ "epoch": 0.02,
631
+ "learning_rate": 4.6113015570172204e-07,
632
+ "loss": 0.0,
633
+ "step": 104
634
+ },
635
+ {
636
+ "epoch": 0.02,
637
+ "learning_rate": 4.553476697366145e-07,
638
+ "loss": -0.0001,
639
+ "step": 105
640
+ },
641
+ {
642
+ "epoch": 0.02,
643
+ "learning_rate": 4.4955336657549915e-07,
644
+ "loss": -0.0026,
645
+ "step": 106
646
+ },
647
+ {
648
+ "epoch": 0.02,
649
+ "learning_rate": 4.4374848335114983e-07,
650
+ "loss": 0.0018,
651
+ "step": 107
652
+ },
653
+ {
654
+ "epoch": 0.02,
655
+ "learning_rate": 4.3793425945527306e-07,
656
+ "loss": 0.0016,
657
+ "step": 108
658
+ },
659
+ {
660
+ "epoch": 0.02,
661
+ "learning_rate": 4.321119362738884e-07,
662
+ "loss": -0.0006,
663
+ "step": 109
664
+ },
665
+ {
666
+ "epoch": 0.02,
667
+ "learning_rate": 4.2628275692228115e-07,
668
+ "loss": -0.0004,
669
+ "step": 110
670
+ },
671
+ {
672
+ "epoch": 0.03,
673
+ "learning_rate": 4.2044796597958684e-07,
674
+ "loss": -0.0009,
675
+ "step": 111
676
+ },
677
+ {
678
+ "epoch": 0.03,
679
+ "learning_rate": 4.146088092230636e-07,
680
+ "loss": -0.0007,
681
+ "step": 112
682
+ },
683
+ {
684
+ "epoch": 0.03,
685
+ "learning_rate": 4.087665333621072e-07,
686
+ "loss": -0.0003,
687
+ "step": 113
688
+ },
689
+ {
690
+ "epoch": 0.03,
691
+ "learning_rate": 4.029223857720692e-07,
692
+ "loss": -0.0003,
693
+ "step": 114
694
+ },
695
+ {
696
+ "epoch": 0.03,
697
+ "learning_rate": 3.970776142279308e-07,
698
+ "loss": 0.0,
699
+ "step": 115
700
+ },
701
+ {
702
+ "epoch": 0.03,
703
+ "learning_rate": 3.9123346663789275e-07,
704
+ "loss": -0.0006,
705
+ "step": 116
706
+ },
707
+ {
708
+ "epoch": 0.03,
709
+ "learning_rate": 3.853911907769364e-07,
710
+ "loss": -0.0015,
711
+ "step": 117
712
+ },
713
+ {
714
+ "epoch": 0.03,
715
+ "learning_rate": 3.795520340204132e-07,
716
+ "loss": -0.0006,
717
+ "step": 118
718
+ },
719
+ {
720
+ "epoch": 0.03,
721
+ "learning_rate": 3.7371724307771886e-07,
722
+ "loss": -0.0001,
723
+ "step": 119
724
+ },
725
+ {
726
+ "epoch": 0.03,
727
+ "learning_rate": 3.6788806372611154e-07,
728
+ "loss": 0.0003,
729
+ "step": 120
730
+ },
731
+ {
732
+ "epoch": 0.03,
733
+ "learning_rate": 3.6206574054472696e-07,
734
+ "loss": -0.0003,
735
+ "step": 121
736
+ },
737
+ {
738
+ "epoch": 0.03,
739
+ "learning_rate": 3.562515166488503e-07,
740
+ "loss": -0.0002,
741
+ "step": 122
742
+ },
743
+ {
744
+ "epoch": 0.03,
745
+ "learning_rate": 3.5044663342450076e-07,
746
+ "loss": -0.0007,
747
+ "step": 123
748
+ },
749
+ {
750
+ "epoch": 0.03,
751
+ "learning_rate": 3.446523302633854e-07,
752
+ "loss": -0.0015,
753
+ "step": 124
754
+ },
755
+ {
756
+ "epoch": 0.03,
757
+ "learning_rate": 3.3886984429827797e-07,
758
+ "loss": -0.0004,
759
+ "step": 125
760
+ },
761
+ {
762
+ "epoch": 0.03,
763
+ "learning_rate": 3.331004101388809e-07,
764
+ "loss": -0.0008,
765
+ "step": 126
766
+ },
767
+ {
768
+ "epoch": 0.03,
769
+ "learning_rate": 3.2734525960822544e-07,
770
+ "loss": -0.0003,
771
+ "step": 127
772
+ },
773
+ {
774
+ "epoch": 0.03,
775
+ "learning_rate": 3.2160562147966743e-07,
776
+ "loss": -0.001,
777
+ "step": 128
778
+ },
779
+ {
780
+ "epoch": 0.03,
781
+ "learning_rate": 3.158827212145336e-07,
782
+ "loss": -0.0001,
783
+ "step": 129
784
+ },
785
+ {
786
+ "epoch": 0.03,
787
+ "learning_rate": 3.1017778070047474e-07,
788
+ "loss": -0.0016,
789
+ "step": 130
790
+ },
791
+ {
792
+ "epoch": 0.03,
793
+ "learning_rate": 3.044920179905832e-07,
794
+ "loss": -0.0015,
795
+ "step": 131
796
+ },
797
+ {
798
+ "epoch": 0.03,
799
+ "learning_rate": 2.9882664704332764e-07,
800
+ "loss": -0.0005,
801
+ "step": 132
802
+ },
803
+ {
804
+ "epoch": 0.03,
805
+ "learning_rate": 2.9318287746336285e-07,
806
+ "loss": -0.0003,
807
+ "step": 133
808
+ },
809
+ {
810
+ "epoch": 0.03,
811
+ "learning_rate": 2.8756191424326873e-07,
812
+ "loss": -0.0009,
813
+ "step": 134
814
+ },
815
+ {
816
+ "epoch": 0.03,
817
+ "learning_rate": 2.819649575062742e-07,
818
+ "loss": -0.0007,
819
+ "step": 135
820
+ },
821
+ {
822
+ "epoch": 0.03,
823
+ "learning_rate": 2.7639320225002107e-07,
824
+ "loss": -0.001,
825
+ "step": 136
826
+ },
827
+ {
828
+ "epoch": 0.03,
829
+ "learning_rate": 2.708478380914216e-07,
830
+ "loss": 0.0002,
831
+ "step": 137
832
+ },
833
+ {
834
+ "epoch": 0.03,
835
+ "learning_rate": 2.653300490126652e-07,
836
+ "loss": -0.0004,
837
+ "step": 138
838
+ },
839
+ {
840
+ "epoch": 0.03,
841
+ "learning_rate": 2.598410131084282e-07,
842
+ "loss": 0.0005,
843
+ "step": 139
844
+ },
845
+ {
846
+ "epoch": 0.03,
847
+ "learning_rate": 2.543819023343413e-07,
848
+ "loss": -0.0003,
849
+ "step": 140
850
+ },
851
+ {
852
+ "epoch": 0.03,
853
+ "learning_rate": 2.4895388225676614e-07,
854
+ "loss": -0.0001,
855
+ "step": 141
856
+ },
857
+ {
858
+ "epoch": 0.03,
859
+ "learning_rate": 2.4355811180393763e-07,
860
+ "loss": -0.0005,
861
+ "step": 142
862
+ },
863
+ {
864
+ "epoch": 0.03,
865
+ "learning_rate": 2.3819574301852254e-07,
866
+ "loss": -0.0001,
867
+ "step": 143
868
+ },
869
+ {
870
+ "epoch": 0.03,
871
+ "learning_rate": 2.3286792081164936e-07,
872
+ "loss": -0.0015,
873
+ "step": 144
874
+ },
875
+ {
876
+ "epoch": 0.03,
877
+ "learning_rate": 2.2757578271845907e-07,
878
+ "loss": -0.0007,
879
+ "step": 145
880
+ },
881
+ {
882
+ "epoch": 0.03,
883
+ "learning_rate": 2.223204586552329e-07,
884
+ "loss": -0.0004,
885
+ "step": 146
886
+ },
887
+ {
888
+ "epoch": 0.03,
889
+ "learning_rate": 2.1710307067814458e-07,
890
+ "loss": -0.0013,
891
+ "step": 147
892
+ },
893
+ {
894
+ "epoch": 0.03,
895
+ "learning_rate": 2.119247327436926e-07,
896
+ "loss": 0.0006,
897
+ "step": 148
898
+ },
899
+ {
900
+ "epoch": 0.03,
901
+ "learning_rate": 2.0678655047086084e-07,
902
+ "loss": -0.0008,
903
+ "step": 149
904
+ },
905
+ {
906
+ "epoch": 0.03,
907
+ "learning_rate": 2.0168962090505928e-07,
908
+ "loss": -0.0007,
909
+ "step": 150
910
+ },
911
+ {
912
+ "epoch": 0.03,
913
+ "learning_rate": 1.9663503228389558e-07,
914
+ "loss": -0.0016,
915
+ "step": 151
916
+ },
917
+ {
918
+ "epoch": 0.03,
919
+ "learning_rate": 1.9162386380482794e-07,
920
+ "loss": 0.0003,
921
+ "step": 152
922
+ },
923
+ {
924
+ "epoch": 0.03,
925
+ "learning_rate": 1.8665718539474603e-07,
926
+ "loss": -0.0011,
927
+ "step": 153
928
+ },
929
+ {
930
+ "epoch": 0.03,
931
+ "learning_rate": 1.817360574815334e-07,
932
+ "loss": -0.0008,
933
+ "step": 154
934
+ },
935
+ {
936
+ "epoch": 0.03,
937
+ "learning_rate": 1.7686153076765704e-07,
938
+ "loss": -0.0022,
939
+ "step": 155
940
+ },
941
+ {
942
+ "epoch": 0.04,
943
+ "learning_rate": 1.7203464600583347e-07,
944
+ "loss": -0.0006,
945
+ "step": 156
946
+ },
947
+ {
948
+ "epoch": 0.04,
949
+ "learning_rate": 1.6725643377681893e-07,
950
+ "loss": -0.0,
951
+ "step": 157
952
+ },
953
+ {
954
+ "epoch": 0.04,
955
+ "learning_rate": 1.6252791426937177e-07,
956
+ "loss": 0.0004,
957
+ "step": 158
958
+ },
959
+ {
960
+ "epoch": 0.04,
961
+ "learning_rate": 1.5785009706243417e-07,
962
+ "loss": -0.001,
963
+ "step": 159
964
+ },
965
+ {
966
+ "epoch": 0.04,
967
+ "learning_rate": 1.5322398090957744e-07,
968
+ "loss": -0.0002,
969
+ "step": 160
970
+ },
971
+ {
972
+ "epoch": 0.04,
973
+ "learning_rate": 1.4865055352576028e-07,
974
+ "loss": -0.0008,
975
+ "step": 161
976
+ },
977
+ {
978
+ "epoch": 0.04,
979
+ "learning_rate": 1.4413079137644356e-07,
980
+ "loss": -0.0012,
981
+ "step": 162
982
+ },
983
+ {
984
+ "epoch": 0.04,
985
+ "learning_rate": 1.3966565946910578e-07,
986
+ "loss": -0.0006,
987
+ "step": 163
988
+ },
989
+ {
990
+ "epoch": 0.04,
991
+ "learning_rate": 1.3525611114720669e-07,
992
+ "loss": 0.0003,
993
+ "step": 164
994
+ },
995
+ {
996
+ "epoch": 0.04,
997
+ "learning_rate": 1.3090308788663999e-07,
998
+ "loss": -0.0004,
999
+ "step": 165
1000
+ },
1001
+ {
1002
+ "epoch": 0.04,
1003
+ "learning_rate": 1.2660751909472027e-07,
1004
+ "loss": -0.0009,
1005
+ "step": 166
1006
+ },
1007
+ {
1008
+ "epoch": 0.04,
1009
+ "learning_rate": 1.223703219117464e-07,
1010
+ "loss": -0.0004,
1011
+ "step": 167
1012
+ },
1013
+ {
1014
+ "epoch": 0.04,
1015
+ "learning_rate": 1.181924010151853e-07,
1016
+ "loss": -0.0004,
1017
+ "step": 168
1018
+ },
1019
+ {
1020
+ "epoch": 0.04,
1021
+ "learning_rate": 1.1407464842651431e-07,
1022
+ "loss": -0.0009,
1023
+ "step": 169
1024
+ },
1025
+ {
1026
+ "epoch": 0.04,
1027
+ "learning_rate": 1.1001794332076793e-07,
1028
+ "loss": -0.0012,
1029
+ "step": 170
1030
+ },
1031
+ {
1032
+ "epoch": 0.04,
1033
+ "learning_rate": 1.0602315183882642e-07,
1034
+ "loss": -0.0002,
1035
+ "step": 171
1036
+ },
1037
+ {
1038
+ "epoch": 0.04,
1039
+ "learning_rate": 1.0209112690248725e-07,
1040
+ "loss": -0.0001,
1041
+ "step": 172
1042
+ },
1043
+ {
1044
+ "epoch": 0.04,
1045
+ "learning_rate": 9.822270803235895e-08,
1046
+ "loss": -0.0013,
1047
+ "step": 173
1048
+ },
1049
+ {
1050
+ "epoch": 0.04,
1051
+ "learning_rate": 9.441872116861636e-08,
1052
+ "loss": 0.0004,
1053
+ "step": 174
1054
+ },
1055
+ {
1056
+ "epoch": 0.04,
1057
+ "learning_rate": 9.0679978494656e-08,
1058
+ "loss": -0.0024,
1059
+ "step": 175
1060
+ },
1061
+ {
1062
+ "epoch": 0.04,
1063
+ "learning_rate": 8.700727826368708e-08,
1064
+ "loss": -0.0006,
1065
+ "step": 176
1066
+ },
1067
+ {
1068
+ "epoch": 0.04,
1069
+ "learning_rate": 8.340140462829839e-08,
1070
+ "loss": -0.0001,
1071
+ "step": 177
1072
+ },
1073
+ {
1074
+ "epoch": 0.04,
1075
+ "learning_rate": 7.986312747303503e-08,
1076
+ "loss": -0.002,
1077
+ "step": 178
1078
+ },
1079
+ {
1080
+ "epoch": 0.04,
1081
+ "learning_rate": 7.639320225002106e-08,
1082
+ "loss": -0.0017,
1083
+ "step": 179
1084
+ },
1085
+ {
1086
+ "epoch": 0.04,
1087
+ "learning_rate": 7.299236981766466e-08,
1088
+ "loss": 0.0003,
1089
+ "step": 180
1090
+ },
1091
+ {
1092
+ "epoch": 0.04,
1093
+ "learning_rate": 6.966135628247834e-08,
1094
+ "loss": -0.0009,
1095
+ "step": 181
1096
+ },
1097
+ {
1098
+ "epoch": 0.04,
1099
+ "learning_rate": 6.640087284404888e-08,
1100
+ "loss": -0.0006,
1101
+ "step": 182
1102
+ },
1103
+ {
1104
+ "epoch": 0.04,
1105
+ "learning_rate": 6.321161564319033e-08,
1106
+ "loss": -0.0012,
1107
+ "step": 183
1108
+ },
1109
+ {
1110
+ "epoch": 0.04,
1111
+ "learning_rate": 6.009426561331241e-08,
1112
+ "loss": -0.0002,
1113
+ "step": 184
1114
+ },
1115
+ {
1116
+ "epoch": 0.04,
1117
+ "learning_rate": 5.704948833503484e-08,
1118
+ "loss": -0.0014,
1119
+ "step": 185
1120
+ },
1121
+ {
1122
+ "epoch": 0.04,
1123
+ "learning_rate": 5.4077933894080484e-08,
1124
+ "loss": -0.0002,
1125
+ "step": 186
1126
+ },
1127
+ {
1128
+ "epoch": 0.04,
1129
+ "learning_rate": 5.118023674247691e-08,
1130
+ "loss": -0.0007,
1131
+ "step": 187
1132
+ },
1133
+ {
1134
+ "epoch": 0.04,
1135
+ "learning_rate": 4.835701556309515e-08,
1136
+ "loss": -0.0012,
1137
+ "step": 188
1138
+ },
1139
+ {
1140
+ "epoch": 0.04,
1141
+ "learning_rate": 4.560887313755537e-08,
1142
+ "loss": -0.0002,
1143
+ "step": 189
1144
+ },
1145
+ {
1146
+ "epoch": 0.04,
1147
+ "learning_rate": 4.293639621752874e-08,
1148
+ "loss": -0.0,
1149
+ "step": 190
1150
+ },
1151
+ {
1152
+ "epoch": 0.04,
1153
+ "learning_rate": 4.03401553994601e-08,
1154
+ "loss": -0.0009,
1155
+ "step": 191
1156
+ },
1157
+ {
1158
+ "epoch": 0.04,
1159
+ "learning_rate": 3.782070500274135e-08,
1160
+ "loss": -0.0004,
1161
+ "step": 192
1162
+ },
1163
+ {
1164
+ "epoch": 0.04,
1165
+ "learning_rate": 3.5378582951359495e-08,
1166
+ "loss": -0.0005,
1167
+ "step": 193
1168
+ },
1169
+ {
1170
+ "epoch": 0.04,
1171
+ "learning_rate": 3.301431065904503e-08,
1172
+ "loss": -0.0004,
1173
+ "step": 194
1174
+ },
1175
+ {
1176
+ "epoch": 0.04,
1177
+ "learning_rate": 3.072839291794591e-08,
1178
+ "loss": -0.0009,
1179
+ "step": 195
1180
+ },
1181
+ {
1182
+ "epoch": 0.04,
1183
+ "learning_rate": 2.852131779085032e-08,
1184
+ "loss": 0.0007,
1185
+ "step": 196
1186
+ },
1187
+ {
1188
+ "epoch": 0.04,
1189
+ "learning_rate": 2.639355650698083e-08,
1190
+ "loss": -0.0003,
1191
+ "step": 197
1192
+ },
1193
+ {
1194
+ "epoch": 0.04,
1195
+ "learning_rate": 2.4345563361383025e-08,
1196
+ "loss": -0.0011,
1197
+ "step": 198
1198
+ },
1199
+ {
1200
+ "epoch": 0.04,
1201
+ "learning_rate": 2.237777561792975e-08,
1202
+ "loss": -0.0012,
1203
+ "step": 199
1204
+ },
1205
+ {
1206
+ "epoch": 0.05,
1207
+ "learning_rate": 2.0490613415961744e-08,
1208
+ "loss": 0.0001,
1209
+ "step": 200
1210
+ },
1211
+ {
1212
+ "epoch": 0.05,
1213
+ "learning_rate": 1.868447968058393e-08,
1214
+ "loss": -0.001,
1215
+ "step": 201
1216
+ },
1217
+ {
1218
+ "epoch": 0.05,
1219
+ "learning_rate": 1.695976003663766e-08,
1220
+ "loss": -0.0016,
1221
+ "step": 202
1222
+ },
1223
+ {
1224
+ "epoch": 0.05,
1225
+ "learning_rate": 1.5316822726366785e-08,
1226
+ "loss": -0.0015,
1227
+ "step": 203
1228
+ },
1229
+ {
1230
+ "epoch": 0.05,
1231
+ "learning_rate": 1.37560185307946e-08,
1232
+ "loss": -0.0008,
1233
+ "step": 204
1234
+ },
1235
+ {
1236
+ "epoch": 0.05,
1237
+ "learning_rate": 1.2277680694828996e-08,
1238
+ "loss": -0.0014,
1239
+ "step": 205
1240
+ },
1241
+ {
1242
+ "epoch": 0.05,
1243
+ "learning_rate": 1.0882124856112218e-08,
1244
+ "loss": -0.0007,
1245
+ "step": 206
1246
+ },
1247
+ {
1248
+ "epoch": 0.05,
1249
+ "learning_rate": 9.569648977629175e-09,
1250
+ "loss": 0.0004,
1251
+ "step": 207
1252
+ },
1253
+ {
1254
+ "epoch": 0.05,
1255
+ "learning_rate": 8.340533284090012e-09,
1256
+ "loss": -0.0015,
1257
+ "step": 208
1258
+ },
1259
+ {
1260
+ "epoch": 0.05,
1261
+ "learning_rate": 7.1950402020998895e-09,
1262
+ "loss": -0.002,
1263
+ "step": 209
1264
+ },
1265
+ {
1266
+ "epoch": 0.05,
1267
+ "learning_rate": 6.133414304128371e-09,
1268
+ "loss": 0.0001,
1269
+ "step": 210
1270
+ },
1271
+ {
1272
+ "epoch": 0.05,
1273
+ "learning_rate": 5.1558822562913774e-09,
1274
+ "loss": -0.0,
1275
+ "step": 211
1276
+ },
1277
+ {
1278
+ "epoch": 0.05,
1279
+ "learning_rate": 4.2626527699558985e-09,
1280
+ "loss": -0.0007,
1281
+ "step": 212
1282
+ },
1283
+ {
1284
+ "epoch": 0.05,
1285
+ "learning_rate": 3.4539165571785713e-09,
1286
+ "loss": -0.0002,
1287
+ "step": 213
1288
+ },
1289
+ {
1290
+ "epoch": 0.05,
1291
+ "learning_rate": 2.729846289986737e-09,
1292
+ "loss": 0.0,
1293
+ "step": 214
1294
+ },
1295
+ {
1296
+ "epoch": 0.05,
1297
+ "learning_rate": 2.0905965635117152e-09,
1298
+ "loss": -0.0003,
1299
+ "step": 215
1300
+ },
1301
+ {
1302
+ "epoch": 0.05,
1303
+ "learning_rate": 1.5363038629813808e-09,
1304
+ "loss": -0.0012,
1305
+ "step": 216
1306
+ },
1307
+ {
1308
+ "epoch": 0.05,
1309
+ "learning_rate": 1.0670865345793423e-09,
1310
+ "loss": -0.0019,
1311
+ "step": 217
1312
+ },
1313
+ {
1314
+ "epoch": 0.05,
1315
+ "learning_rate": 6.830447601771983e-10,
1316
+ "loss": -0.0018,
1317
+ "step": 218
1318
+ },
1319
+ {
1320
+ "epoch": 0.05,
1321
+ "learning_rate": 3.8426053594471375e-10,
1322
+ "loss": -0.0012,
1323
+ "step": 219
1324
+ },
1325
+ {
1326
+ "epoch": 0.05,
1327
+ "learning_rate": 1.7079765484293574e-10,
1328
+ "loss": -0.0003,
1329
+ "step": 220
1330
+ },
1331
+ {
1332
+ "epoch": 0.05,
1333
+ "learning_rate": 4.270169300397697e-11,
1334
+ "loss": 0.0001,
1335
+ "step": 221
1336
+ },
1337
+ {
1338
+ "epoch": 0.05,
1339
+ "learning_rate": 0.0,
1340
+ "loss": -0.0002,
1341
+ "step": 222
1342
+ },
1343
+ {
1344
+ "epoch": 0.05,
1345
+ "step": 222,
1346
+ "total_flos": 6.51669001961472e+16,
1347
+ "train_loss": -0.0006611563352556266,
1348
+ "train_runtime": 59896.0215,
1349
+ "train_samples_per_second": 0.177,
1350
+ "train_steps_per_second": 0.004
1351
+ }
1352
+ ],
1353
+ "logging_steps": 1.0,
1354
+ "max_steps": 222,
1355
+ "num_input_tokens_seen": 0,
1356
+ "num_train_epochs": 1,
1357
+ "save_steps": 200,
1358
+ "total_flos": 6.51669001961472e+16,
1359
+ "train_batch_size": 48,
1360
+ "trial_name": null,
1361
+ "trial_params": null
1362
+ }