Hugging Face's logo

ajaymin28
/
vl-sg-vrd-loras

Safetensors
Model card Files Community
ajaymin28 commited on
Commit
7cb43d0
·
verified ·
1 Parent(s): b3e7362

Upload folder using huggingface_hub

Browse files
Files changed (8) hide show
  1. llavanext-google_siglip-so400m-patch14-384-Qwen_Qwen2-7B-Instruct-VRD_v5_3_split03_all_mm_tune_olora256_512_llm/README.md +11 -0
  2. llavanext-google_siglip-so400m-patch14-384-Qwen_Qwen2-7B-Instruct-VRD_v5_3_split03_all_mm_tune_olora256_512_llm/adapter_config.json +34 -0
  3. llavanext-google_siglip-so400m-patch14-384-Qwen_Qwen2-7B-Instruct-VRD_v5_3_split03_all_mm_tune_olora256_512_llm/adapter_model.bin +3 -0
  4. llavanext-google_siglip-so400m-patch14-384-Qwen_Qwen2-7B-Instruct-VRD_v5_3_split03_all_mm_tune_olora256_512_llm/adapter_model.safetensors +3 -0
  5. llavanext-google_siglip-so400m-patch14-384-Qwen_Qwen2-7B-Instruct-VRD_v5_3_split03_all_mm_tune_olora256_512_llm/config.json +221 -0
  6. llavanext-google_siglip-so400m-patch14-384-Qwen_Qwen2-7B-Instruct-VRD_v5_3_split03_all_mm_tune_olora256_512_llm/generation_config.json +14 -0
  7. llavanext-google_siglip-so400m-patch14-384-Qwen_Qwen2-7B-Instruct-VRD_v5_3_split03_all_mm_tune_olora256_512_llm/non_lora_trainables.bin +3 -0
  8. llavanext-google_siglip-so400m-patch14-384-Qwen_Qwen2-7B-Instruct-VRD_v5_3_split03_all_mm_tune_olora256_512_llm/trainer_state.json +3530 -0
llavanext-google_siglip-so400m-patch14-384-Qwen_Qwen2-7B-Instruct-VRD_v5_3_split03_all_mm_tune_olora256_512_llm/README.md ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ base_model: lmms-lab/llava-onevision-qwen2-7b-si
4
+ ---
5
+ ## Training procedure
6
+
7
+ ### Framework versions
8
+
9
+ - PEFT 0.13.2
10
+
11
+ - PEFT 0.4.0
llavanext-google_siglip-so400m-patch14-384-Qwen_Qwen2-7B-Instruct-VRD_v5_3_split03_all_mm_tune_olora256_512_llm/adapter_config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "lmms-lab/llava-onevision-qwen2-7b-si",
5
+ "bias": "none",
6
+ "fan_in_fan_out": false,
7
+ "inference_mode": true,
8
+ "init_lora_weights": "olora",
9
+ "layer_replication": null,
10
+ "layers_pattern": null,
11
+ "layers_to_transform": null,
12
+ "loftq_config": {},
13
+ "lora_alpha": 512,
14
+ "lora_dropout": 0.05,
15
+ "megatron_config": null,
16
+ "megatron_core": "megatron.core",
17
+ "modules_to_save": null,
18
+ "peft_type": "LORA",
19
+ "r": 256,
20
+ "rank_pattern": {},
21
+ "revision": null,
22
+ "target_modules": [
23
+ "up_proj",
24
+ "gate_proj",
25
+ "o_proj",
26
+ "down_proj",
27
+ "k_proj",
28
+ "v_proj",
29
+ "q_proj"
30
+ ],
31
+ "task_type": "CAUSAL_LM",
32
+ "use_dora": false,
33
+ "use_rslora": false
34
+ }
llavanext-google_siglip-so400m-patch14-384-Qwen_Qwen2-7B-Instruct-VRD_v5_3_split03_all_mm_tune_olora256_512_llm/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ff3ba8af0f29970e3ef307fb941b44ed430b9f62418e3cbe3de04230dc9c081f
3
+ size 1384057050
llavanext-google_siglip-so400m-patch14-384-Qwen_Qwen2-7B-Instruct-VRD_v5_3_split03_all_mm_tune_olora256_512_llm/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ce0ad1e382c4c14c8b2694f3612b62261342e52034a4243041c429ffa1c41030
3
+ size 1383941368
llavanext-google_siglip-so400m-patch14-384-Qwen_Qwen2-7B-Instruct-VRD_v5_3_split03_all_mm_tune_olora256_512_llm/config.json ADDED
@@ -0,0 +1,221 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "lmms-lab/llava-onevision-qwen2-7b-si",
3
+ "add_faster_video": false,
4
+ "add_time_instruction": true,
5
+ "architectures": [
6
+ "LlavaQwenForCausalLM"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "bos_token_id": 151643,
10
+ "eos_token_id": 151645,
11
+ "faster_token_stride": 10,
12
+ "force_sample": true,
13
+ "hidden_act": "silu",
14
+ "hidden_size": 3584,
15
+ "ignore_index": -100,
16
+ "image_aspect_ratio": "anyres_max_9",
17
+ "image_crop_resolution": null,
18
+ "image_grid_pinpoints": [
19
+ [
20
+ 384,
21
+ 384
22
+ ],
23
+ [
24
+ 384,
25
+ 768
26
+ ],
27
+ [
28
+ 384,
29
+ 1152
30
+ ],
31
+ [
32
+ 384,
33
+ 1536
34
+ ],
35
+ [
36
+ 384,
37
+ 1920
38
+ ],
39
+ [
40
+ 384,
41
+ 2304
42
+ ],
43
+ [
44
+ 768,
45
+ 384
46
+ ],
47
+ [
48
+ 768,
49
+ 768
50
+ ],
51
+ [
52
+ 768,
53
+ 1152
54
+ ],
55
+ [
56
+ 768,
57
+ 1536
58
+ ],
59
+ [
60
+ 768,
61
+ 1920
62
+ ],
63
+ [
64
+ 768,
65
+ 2304
66
+ ],
67
+ [
68
+ 1152,
69
+ 384
70
+ ],
71
+ [
72
+ 1152,
73
+ 768
74
+ ],
75
+ [
76
+ 1152,
77
+ 1152
78
+ ],
79
+ [
80
+ 1152,
81
+ 1536
82
+ ],
83
+ [
84
+ 1152,
85
+ 1920
86
+ ],
87
+ [
88
+ 1152,
89
+ 2304
90
+ ],
91
+ [
92
+ 1536,
93
+ 384
94
+ ],
95
+ [
96
+ 1536,
97
+ 768
98
+ ],
99
+ [
100
+ 1536,
101
+ 1152
102
+ ],
103
+ [
104
+ 1536,
105
+ 1536
106
+ ],
107
+ [
108
+ 1536,
109
+ 1920
110
+ ],
111
+ [
112
+ 1536,
113
+ 2304
114
+ ],
115
+ [
116
+ 1920,
117
+ 384
118
+ ],
119
+ [
120
+ 1920,
121
+ 768
122
+ ],
123
+ [
124
+ 1920,
125
+ 1152
126
+ ],
127
+ [
128
+ 1920,
129
+ 1536
130
+ ],
131
+ [
132
+ 1920,
133
+ 1920
134
+ ],
135
+ [
136
+ 1920,
137
+ 2304
138
+ ],
139
+ [
140
+ 2304,
141
+ 384
142
+ ],
143
+ [
144
+ 2304,
145
+ 768
146
+ ],
147
+ [
148
+ 2304,
149
+ 1152
150
+ ],
151
+ [
152
+ 2304,
153
+ 1536
154
+ ],
155
+ [
156
+ 2304,
157
+ 1920
158
+ ],
159
+ [
160
+ 2304,
161
+ 2304
162
+ ]
163
+ ],
164
+ "image_split_resolution": null,
165
+ "image_token_index": 151646,
166
+ "initializer_range": 0.02,
167
+ "intermediate_size": 18944,
168
+ "max_position_embeddings": 32768,
169
+ "max_window_layers": 28,
170
+ "mm_hidden_size": 1152,
171
+ "mm_newline_position": "grid",
172
+ "mm_patch_merge_type": "spatial_unpad",
173
+ "mm_projector_lr": 2e-05,
174
+ "mm_projector_type": "mlp2x_gelu",
175
+ "mm_resampler_type": null,
176
+ "mm_spatial_pool_mode": "bilinear",
177
+ "mm_spatial_pool_stride": 2,
178
+ "mm_tunable_parts": "mm_vision_tower,mm_mlp_adapter,mm_language_model",
179
+ "mm_use_im_patch_token": false,
180
+ "mm_use_im_start_end": false,
181
+ "mm_vision_select_feature": "patch",
182
+ "mm_vision_select_layer": -2,
183
+ "mm_vision_tower": "google/siglip-so400m-patch14-384",
184
+ "mm_vision_tower_lr": null,
185
+ "model_type": "llava",
186
+ "num_attention_heads": 28,
187
+ "num_hidden_layers": 28,
188
+ "num_key_value_heads": 4,
189
+ "pos_skipping_range": 4096,
190
+ "projector_hidden_act": "gelu",
191
+ "rms_norm_eps": 1e-06,
192
+ "rope_scaling": null,
193
+ "rope_theta": 1000000.0,
194
+ "sliding_window": 131072,
195
+ "text_config": {
196
+ "model_type": "llama"
197
+ },
198
+ "tie_word_embeddings": false,
199
+ "tokenizer_model_max_length": 32768,
200
+ "tokenizer_padding_side": "right",
201
+ "torch_dtype": "bfloat16",
202
+ "transformers_version": "4.40.0.dev0",
203
+ "use_cache": true,
204
+ "use_mm_proj": true,
205
+ "use_pos_skipping": false,
206
+ "use_sliding_window": false,
207
+ "vision_config": {
208
+ "hidden_size": 1024,
209
+ "image_size": 336,
210
+ "intermediate_size": 4096,
211
+ "model_type": "clip_vision_model",
212
+ "num_attention_heads": 16,
213
+ "num_hidden_layers": 24,
214
+ "patch_size": 14,
215
+ "projection_dim": 768,
216
+ "vocab_size": 32000
217
+ },
218
+ "vision_feature_layer": -2,
219
+ "vision_feature_select_strategy": "default",
220
+ "vision_tower_pretrained": null
221
+ }
llavanext-google_siglip-so400m-patch14-384-Qwen_Qwen2-7B-Instruct-VRD_v5_3_split03_all_mm_tune_olora256_512_llm/generation_config.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 151643,
3
+ "do_sample": true,
4
+ "eos_token_id": [
5
+ 151645,
6
+ 151643
7
+ ],
8
+ "pad_token_id": 151643,
9
+ "repetition_penalty": 1.05,
10
+ "temperature": 0.7,
11
+ "top_k": 20,
12
+ "top_p": 0.8,
13
+ "transformers_version": "4.40.0.dev0"
14
+ }
llavanext-google_siglip-so400m-patch14-384-Qwen_Qwen2-7B-Instruct-VRD_v5_3_split03_all_mm_tune_olora256_512_llm/non_lora_trainables.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:557026d9b285a379a3e41815200fd16a2bbb08c8be9981e5cd7de3a1f323d90e
3
+ size 33964208
llavanext-google_siglip-so400m-patch14-384-Qwen_Qwen2-7B-Instruct-VRD_v5_3_split03_all_mm_tune_olora256_512_llm/trainer_state.json ADDED
@@ -0,0 +1,3530 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 500,
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
+ "grad_norm": 14282.023983549067,
14
+ "learning_rate": 6.666666666666667e-07,
15
+ "loss": 2.2727,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.0,
20
+ "grad_norm": 3913.9435315966352,
21
+ "learning_rate": 1.3333333333333334e-06,
22
+ "loss": 2.1329,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.01,
27
+ "grad_norm": 1556.7307031010662,
28
+ "learning_rate": 2.0000000000000003e-06,
29
+ "loss": 2.4585,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.01,
34
+ "grad_norm": 2547.418204299765,
35
+ "learning_rate": 2.666666666666667e-06,
36
+ "loss": 2.6093,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.01,
41
+ "grad_norm": 1143.1400626070667,
42
+ "learning_rate": 3.3333333333333333e-06,
43
+ "loss": 2.2147,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.01,
48
+ "grad_norm": 5339.2683171351555,
49
+ "learning_rate": 4.000000000000001e-06,
50
+ "loss": 1.8549,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.01,
55
+ "grad_norm": 1963.5501078045986,
56
+ "learning_rate": 4.666666666666667e-06,
57
+ "loss": 2.3025,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.02,
62
+ "grad_norm": 6029.432267260134,
63
+ "learning_rate": 5.333333333333334e-06,
64
+ "loss": 1.5048,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.02,
69
+ "grad_norm": 1491.712218620925,
70
+ "learning_rate": 6e-06,
71
+ "loss": 1.4955,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 0.02,
76
+ "grad_norm": 14500.123750254728,
77
+ "learning_rate": 6.666666666666667e-06,
78
+ "loss": 1.2686,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 0.02,
83
+ "grad_norm": 666.7779178120796,
84
+ "learning_rate": 7.333333333333333e-06,
85
+ "loss": 1.1677,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 0.02,
90
+ "grad_norm": 354.1811265283398,
91
+ "learning_rate": 8.000000000000001e-06,
92
+ "loss": 1.0633,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 0.03,
97
+ "grad_norm": 360.7997748784865,
98
+ "learning_rate": 8.666666666666668e-06,
99
+ "loss": 0.9443,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 0.03,
104
+ "grad_norm": 269.8006097942352,
105
+ "learning_rate": 9.333333333333334e-06,
106
+ "loss": 0.7588,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 0.03,
111
+ "grad_norm": 277.06733384739704,
112
+ "learning_rate": 1e-05,
113
+ "loss": 0.624,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 0.03,
118
+ "grad_norm": 246.8138375596777,
119
+ "learning_rate": 9.999895105006995e-06,
120
+ "loss": 0.5555,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 0.03,
125
+ "grad_norm": 133.30109659575643,
126
+ "learning_rate": 9.99958042442916e-06,
127
+ "loss": 0.4418,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 0.04,
132
+ "grad_norm": 243.960286154522,
133
+ "learning_rate": 9.999055971469864e-06,
134
+ "loss": 0.3747,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 0.04,
139
+ "grad_norm": 147.26012475618376,
140
+ "learning_rate": 9.998321768134101e-06,
141
+ "loss": 0.3115,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 0.04,
146
+ "grad_norm": 58.1290667574483,
147
+ "learning_rate": 9.997377845227577e-06,
148
+ "loss": 0.2779,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 0.04,
153
+ "grad_norm": 64.85994597307703,
154
+ "learning_rate": 9.9962242423554e-06,
155
+ "loss": 0.2197,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 0.04,
160
+ "grad_norm": 46.28532786855361,
161
+ "learning_rate": 9.99486100792044e-06,
162
+ "loss": 0.2309,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 0.05,
167
+ "grad_norm": 54.55193868442643,
168
+ "learning_rate": 9.993288199121283e-06,
169
+ "loss": 0.1816,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 0.05,
174
+ "grad_norm": 50.342133496951135,
175
+ "learning_rate": 9.991505881949837e-06,
176
+ "loss": 0.1768,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 0.05,
181
+ "grad_norm": 61.05704418486633,
182
+ "learning_rate": 9.98951413118856e-06,
183
+ "loss": 0.1558,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 0.05,
188
+ "grad_norm": 93.76122667188282,
189
+ "learning_rate": 9.987313030407325e-06,
190
+ "loss": 0.136,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 0.05,
195
+ "grad_norm": 22.631097297368704,
196
+ "learning_rate": 9.984902671959911e-06,
197
+ "loss": 0.1423,
198
+ "step": 27
199
+ },
200
+ {
201
+ "epoch": 0.06,
202
+ "grad_norm": 30.916549400619537,
203
+ "learning_rate": 9.982283156980133e-06,
204
+ "loss": 0.1327,
205
+ "step": 28
206
+ },
207
+ {
208
+ "epoch": 0.06,
209
+ "grad_norm": 11.739747630325619,
210
+ "learning_rate": 9.979454595377594e-06,
211
+ "loss": 0.1115,
212
+ "step": 29
213
+ },
214
+ {
215
+ "epoch": 0.06,
216
+ "grad_norm": 15.642038390205942,
217
+ "learning_rate": 9.97641710583307e-06,
218
+ "loss": 0.0819,
219
+ "step": 30
220
+ },
221
+ {
222
+ "epoch": 0.06,
223
+ "grad_norm": 24.02762304821651,
224
+ "learning_rate": 9.973170815793543e-06,
225
+ "loss": 0.071,
226
+ "step": 31
227
+ },
228
+ {
229
+ "epoch": 0.06,
230
+ "grad_norm": 8.378995335655844,
231
+ "learning_rate": 9.969715861466839e-06,
232
+ "loss": 0.082,
233
+ "step": 32
234
+ },
235
+ {
236
+ "epoch": 0.07,
237
+ "grad_norm": 4.524270645803727,
238
+ "learning_rate": 9.966052387815923e-06,
239
+ "loss": 0.0727,
240
+ "step": 33
241
+ },
242
+ {
243
+ "epoch": 0.07,
244
+ "grad_norm": 10.497933511429377,
245
+ "learning_rate": 9.962180548552812e-06,
246
+ "loss": 0.066,
247
+ "step": 34
248
+ },
249
+ {
250
+ "epoch": 0.07,
251
+ "grad_norm": 5.531735304853509,
252
+ "learning_rate": 9.958100506132127e-06,
253
+ "loss": 0.0735,
254
+ "step": 35
255
+ },
256
+ {
257
+ "epoch": 0.07,
258
+ "grad_norm": 8.832691471107868,
259
+ "learning_rate": 9.953812431744274e-06,
260
+ "loss": 0.0602,
261
+ "step": 36
262
+ },
263
+ {
264
+ "epoch": 0.07,
265
+ "grad_norm": 24.006955458894293,
266
+ "learning_rate": 9.94931650530827e-06,
267
+ "loss": 0.0491,
268
+ "step": 37
269
+ },
270
+ {
271
+ "epoch": 0.08,
272
+ "grad_norm": 16.05398939158306,
273
+ "learning_rate": 9.944612915464183e-06,
274
+ "loss": 0.0643,
275
+ "step": 38
276
+ },
277
+ {
278
+ "epoch": 0.08,
279
+ "grad_norm": 5.514973964264771,
280
+ "learning_rate": 9.93970185956522e-06,
281
+ "loss": 0.0506,
282
+ "step": 39
283
+ },
284
+ {
285
+ "epoch": 0.08,
286
+ "grad_norm": 4.557232044301031,
287
+ "learning_rate": 9.934583543669454e-06,
288
+ "loss": 0.0476,
289
+ "step": 40
290
+ },
291
+ {
292
+ "epoch": 0.08,
293
+ "grad_norm": 4.959056070104521,
294
+ "learning_rate": 9.929258182531167e-06,
295
+ "loss": 0.0531,
296
+ "step": 41
297
+ },
298
+ {
299
+ "epoch": 0.08,
300
+ "grad_norm": 5.9109247547510995,
301
+ "learning_rate": 9.923725999591846e-06,
302
+ "loss": 0.052,
303
+ "step": 42
304
+ },
305
+ {
306
+ "epoch": 0.09,
307
+ "grad_norm": 4.311982090832518,
308
+ "learning_rate": 9.917987226970811e-06,
309
+ "loss": 0.0397,
310
+ "step": 43
311
+ },
312
+ {
313
+ "epoch": 0.09,
314
+ "grad_norm": 43.46699793941009,
315
+ "learning_rate": 9.912042105455462e-06,
316
+ "loss": 0.0461,
317
+ "step": 44
318
+ },
319
+ {
320
+ "epoch": 0.09,
321
+ "grad_norm": 6.028634235669971,
322
+ "learning_rate": 9.905890884491196e-06,
323
+ "loss": 0.0567,
324
+ "step": 45
325
+ },
326
+ {
327
+ "epoch": 0.09,
328
+ "grad_norm": 5.795721040949649,
329
+ "learning_rate": 9.899533822170922e-06,
330
+ "loss": 0.074,
331
+ "step": 46
332
+ },
333
+ {
334
+ "epoch": 0.09,
335
+ "grad_norm": 3.7095197385099876,
336
+ "learning_rate": 9.892971185224244e-06,
337
+ "loss": 0.044,
338
+ "step": 47
339
+ },
340
+ {
341
+ "epoch": 0.1,
342
+ "grad_norm": 3.507371285993653,
343
+ "learning_rate": 9.886203249006265e-06,
344
+ "loss": 0.0404,
345
+ "step": 48
346
+ },
347
+ {
348
+ "epoch": 0.1,
349
+ "grad_norm": 4.228937568189676,
350
+ "learning_rate": 9.879230297486034e-06,
351
+ "loss": 0.0357,
352
+ "step": 49
353
+ },
354
+ {
355
+ "epoch": 0.1,
356
+ "grad_norm": 4.096159750934641,
357
+ "learning_rate": 9.872052623234632e-06,
358
+ "loss": 0.0625,
359
+ "step": 50
360
+ },
361
+ {
362
+ "epoch": 0.1,
363
+ "grad_norm": 4.257040030842638,
364
+ "learning_rate": 9.864670527412891e-06,
365
+ "loss": 0.0414,
366
+ "step": 51
367
+ },
368
+ {
369
+ "epoch": 0.1,
370
+ "grad_norm": 2.723624584753764,
371
+ "learning_rate": 9.857084319758772e-06,
372
+ "loss": 0.035,
373
+ "step": 52
374
+ },
375
+ {
376
+ "epoch": 0.11,
377
+ "grad_norm": 3.164806797828908,
378
+ "learning_rate": 9.849294318574353e-06,
379
+ "loss": 0.0429,
380
+ "step": 53
381
+ },
382
+ {
383
+ "epoch": 0.11,
384
+ "grad_norm": 4.125557906252407,
385
+ "learning_rate": 9.841300850712479e-06,
386
+ "loss": 0.042,
387
+ "step": 54
388
+ },
389
+ {
390
+ "epoch": 0.11,
391
+ "grad_norm": 3.954954046766278,
392
+ "learning_rate": 9.833104251563058e-06,
393
+ "loss": 0.0427,
394
+ "step": 55
395
+ },
396
+ {
397
+ "epoch": 0.11,
398
+ "grad_norm": 3.067844501816898,
399
+ "learning_rate": 9.824704865038967e-06,
400
+ "loss": 0.041,
401
+ "step": 56
402
+ },
403
+ {
404
+ "epoch": 0.11,
405
+ "grad_norm": 2.0741521332632233,
406
+ "learning_rate": 9.816103043561648e-06,
407
+ "loss": 0.0386,
408
+ "step": 57
409
+ },
410
+ {
411
+ "epoch": 0.12,
412
+ "grad_norm": 3.343297841373388,
413
+ "learning_rate": 9.807299148046301e-06,
414
+ "loss": 0.0552,
415
+ "step": 58
416
+ },
417
+ {
418
+ "epoch": 0.12,
419
+ "grad_norm": 5.60989625436747,
420
+ "learning_rate": 9.798293547886748e-06,
421
+ "loss": 0.0395,
422
+ "step": 59
423
+ },
424
+ {
425
+ "epoch": 0.12,
426
+ "grad_norm": 2.7372057907904788,
427
+ "learning_rate": 9.789086620939936e-06,
428
+ "loss": 0.0409,
429
+ "step": 60
430
+ },
431
+ {
432
+ "epoch": 0.12,
433
+ "grad_norm": 2.2005162459541308,
434
+ "learning_rate": 9.779678753510082e-06,
435
+ "loss": 0.0421,
436
+ "step": 61
437
+ },
438
+ {
439
+ "epoch": 0.12,
440
+ "grad_norm": 1.902244054983275,
441
+ "learning_rate": 9.770070340332457e-06,
442
+ "loss": 0.0298,
443
+ "step": 62
444
+ },
445
+ {
446
+ "epoch": 0.13,
447
+ "grad_norm": 3.25538472961429,
448
+ "learning_rate": 9.76026178455684e-06,
449
+ "loss": 0.0415,
450
+ "step": 63
451
+ },
452
+ {
453
+ "epoch": 0.13,
454
+ "grad_norm": 4.343441279191469,
455
+ "learning_rate": 9.75025349773058e-06,
456
+ "loss": 0.043,
457
+ "step": 64
458
+ },
459
+ {
460
+ "epoch": 0.13,
461
+ "grad_norm": 2.6231467187214945,
462
+ "learning_rate": 9.740045899781353e-06,
463
+ "loss": 0.0532,
464
+ "step": 65
465
+ },
466
+ {
467
+ "epoch": 0.13,
468
+ "grad_norm": 3.1475042135239044,
469
+ "learning_rate": 9.729639418999524e-06,
470
+ "loss": 0.0662,
471
+ "step": 66
472
+ },
473
+ {
474
+ "epoch": 0.13,
475
+ "grad_norm": 2.5322954252153815,
476
+ "learning_rate": 9.719034492020183e-06,
477
+ "loss": 0.0335,
478
+ "step": 67
479
+ },
480
+ {
481
+ "epoch": 0.14,
482
+ "grad_norm": 2.1861278077823028,
483
+ "learning_rate": 9.708231563804828e-06,
484
+ "loss": 0.0362,
485
+ "step": 68
486
+ },
487
+ {
488
+ "epoch": 0.14,
489
+ "grad_norm": 2.276055259027938,
490
+ "learning_rate": 9.697231087622691e-06,
491
+ "loss": 0.0343,
492
+ "step": 69
493
+ },
494
+ {
495
+ "epoch": 0.14,
496
+ "grad_norm": 3.1293382310265563,
497
+ "learning_rate": 9.68603352503172e-06,
498
+ "loss": 0.0375,
499
+ "step": 70
500
+ },
501
+ {
502
+ "epoch": 0.14,
503
+ "grad_norm": 8.052851654782353,
504
+ "learning_rate": 9.674639345859213e-06,
505
+ "loss": 0.0271,
506
+ "step": 71
507
+ },
508
+ {
509
+ "epoch": 0.14,
510
+ "grad_norm": 2.9101275827906434,
511
+ "learning_rate": 9.663049028182112e-06,
512
+ "loss": 0.0313,
513
+ "step": 72
514
+ },
515
+ {
516
+ "epoch": 0.15,
517
+ "grad_norm": 1.74761893823187,
518
+ "learning_rate": 9.651263058306932e-06,
519
+ "loss": 0.0279,
520
+ "step": 73
521
+ },
522
+ {
523
+ "epoch": 0.15,
524
+ "grad_norm": 2.630906107065737,
525
+ "learning_rate": 9.639281930749363e-06,
526
+ "loss": 0.0454,
527
+ "step": 74
528
+ },
529
+ {
530
+ "epoch": 0.15,
531
+ "grad_norm": 4.474405168150697,
532
+ "learning_rate": 9.627106148213521e-06,
533
+ "loss": 0.0451,
534
+ "step": 75
535
+ },
536
+ {
537
+ "epoch": 0.15,
538
+ "grad_norm": 2.507853919415977,
539
+ "learning_rate": 9.61473622157086e-06,
540
+ "loss": 0.0365,
541
+ "step": 76
542
+ },
543
+ {
544
+ "epoch": 0.15,
545
+ "grad_norm": 1.8348352914396244,
546
+ "learning_rate": 9.602172669838721e-06,
547
+ "loss": 0.0371,
548
+ "step": 77
549
+ },
550
+ {
551
+ "epoch": 0.16,
552
+ "grad_norm": 3.0019483386400565,
553
+ "learning_rate": 9.589416020158577e-06,
554
+ "loss": 0.0308,
555
+ "step": 78
556
+ },
557
+ {
558
+ "epoch": 0.16,
559
+ "grad_norm": 1.8298893052093932,
560
+ "learning_rate": 9.5764668077739e-06,
561
+ "loss": 0.0308,
562
+ "step": 79
563
+ },
564
+ {
565
+ "epoch": 0.16,
566
+ "grad_norm": 1.972040536793431,
567
+ "learning_rate": 9.563325576007702e-06,
568
+ "loss": 0.0332,
569
+ "step": 80
570
+ },
571
+ {
572
+ "epoch": 0.16,
573
+ "grad_norm": 2.9666307325696173,
574
+ "learning_rate": 9.549992876239753e-06,
575
+ "loss": 0.0489,
576
+ "step": 81
577
+ },
578
+ {
579
+ "epoch": 0.16,
580
+ "grad_norm": 3.46053160443569,
581
+ "learning_rate": 9.536469267883432e-06,
582
+ "loss": 0.0455,
583
+ "step": 82
584
+ },
585
+ {
586
+ "epoch": 0.17,
587
+ "grad_norm": 4.530510206346124,
588
+ "learning_rate": 9.52275531836226e-06,
589
+ "loss": 0.0423,
590
+ "step": 83
591
+ },
592
+ {
593
+ "epoch": 0.17,
594
+ "grad_norm": 4.536423488068978,
595
+ "learning_rate": 9.508851603086094e-06,
596
+ "loss": 0.0439,
597
+ "step": 84
598
+ },
599
+ {
600
+ "epoch": 0.17,
601
+ "grad_norm": 2.6937927272920934,
602
+ "learning_rate": 9.494758705426978e-06,
603
+ "loss": 0.0352,
604
+ "step": 85
605
+ },
606
+ {
607
+ "epoch": 0.17,
608
+ "grad_norm": 2.0111594842098097,
609
+ "learning_rate": 9.480477216694674e-06,
610
+ "loss": 0.0338,
611
+ "step": 86
612
+ },
613
+ {
614
+ "epoch": 0.17,
615
+ "grad_norm": 6.613348245756698,
616
+ "learning_rate": 9.466007736111846e-06,
617
+ "loss": 0.0432,
618
+ "step": 87
619
+ },
620
+ {
621
+ "epoch": 0.18,
622
+ "grad_norm": 1.865997650182757,
623
+ "learning_rate": 9.451350870788922e-06,
624
+ "loss": 0.0317,
625
+ "step": 88
626
+ },
627
+ {
628
+ "epoch": 0.18,
629
+ "grad_norm": 2.067050675497705,
630
+ "learning_rate": 9.436507235698613e-06,
631
+ "loss": 0.035,
632
+ "step": 89
633
+ },
634
+ {
635
+ "epoch": 0.18,
636
+ "grad_norm": 3.2579257717581247,
637
+ "learning_rate": 9.421477453650118e-06,
638
+ "loss": 0.0409,
639
+ "step": 90
640
+ },
641
+ {
642
+ "epoch": 0.18,
643
+ "grad_norm": 2.027502055576434,
644
+ "learning_rate": 9.406262155262995e-06,
645
+ "loss": 0.0347,
646
+ "step": 91
647
+ },
648
+ {
649
+ "epoch": 0.18,
650
+ "grad_norm": 1.4784309174935295,
651
+ "learning_rate": 9.390861978940687e-06,
652
+ "loss": 0.0277,
653
+ "step": 92
654
+ },
655
+ {
656
+ "epoch": 0.19,
657
+ "grad_norm": 1.669647168263475,
658
+ "learning_rate": 9.37527757084375e-06,
659
+ "loss": 0.041,
660
+ "step": 93
661
+ },
662
+ {
663
+ "epoch": 0.19,
664
+ "grad_norm": 3.599102454624084,
665
+ "learning_rate": 9.359509584862735e-06,
666
+ "loss": 0.0495,
667
+ "step": 94
668
+ },
669
+ {
670
+ "epoch": 0.19,
671
+ "grad_norm": 1.8698147311224507,
672
+ "learning_rate": 9.343558682590757e-06,
673
+ "loss": 0.0267,
674
+ "step": 95
675
+ },
676
+ {
677
+ "epoch": 0.19,
678
+ "grad_norm": 2.7639925272838943,
679
+ "learning_rate": 9.327425533295725e-06,
680
+ "loss": 0.0322,
681
+ "step": 96
682
+ },
683
+ {
684
+ "epoch": 0.19,
685
+ "grad_norm": 2.434158064219475,
686
+ "learning_rate": 9.31111081389227e-06,
687
+ "loss": 0.0356,
688
+ "step": 97
689
+ },
690
+ {
691
+ "epoch": 0.2,
692
+ "grad_norm": 2.6832589514971086,
693
+ "learning_rate": 9.29461520891335e-06,
694
+ "loss": 0.0419,
695
+ "step": 98
696
+ },
697
+ {
698
+ "epoch": 0.2,
699
+ "grad_norm": 1.5604134450196618,
700
+ "learning_rate": 9.277939410481507e-06,
701
+ "loss": 0.0297,
702
+ "step": 99
703
+ },
704
+ {
705
+ "epoch": 0.2,
706
+ "grad_norm": 2.132194370810084,
707
+ "learning_rate": 9.261084118279846e-06,
708
+ "loss": 0.0366,
709
+ "step": 100
710
+ },
711
+ {
712
+ "epoch": 0.2,
713
+ "grad_norm": 1.7339858658559861,
714
+ "learning_rate": 9.244050039522673e-06,
715
+ "loss": 0.0259,
716
+ "step": 101
717
+ },
718
+ {
719
+ "epoch": 0.2,
720
+ "grad_norm": 3.0828984043833123,
721
+ "learning_rate": 9.226837888925813e-06,
722
+ "loss": 0.0358,
723
+ "step": 102
724
+ },
725
+ {
726
+ "epoch": 0.21,
727
+ "grad_norm": 1.9839047275114943,
728
+ "learning_rate": 9.209448388676636e-06,
729
+ "loss": 0.0295,
730
+ "step": 103
731
+ },
732
+ {
733
+ "epoch": 0.21,
734
+ "grad_norm": 2.0978517118041586,
735
+ "learning_rate": 9.191882268403743e-06,
736
+ "loss": 0.0284,
737
+ "step": 104
738
+ },
739
+ {
740
+ "epoch": 0.21,
741
+ "grad_norm": 2.0924942585883395,
742
+ "learning_rate": 9.174140265146356e-06,
743
+ "loss": 0.0336,
744
+ "step": 105
745
+ },
746
+ {
747
+ "epoch": 0.21,
748
+ "grad_norm": 1.9495366399981342,
749
+ "learning_rate": 9.156223123323405e-06,
750
+ "loss": 0.0378,
751
+ "step": 106
752
+ },
753
+ {
754
+ "epoch": 0.21,
755
+ "grad_norm": 1.8628540581298327,
756
+ "learning_rate": 9.13813159470227e-06,
757
+ "loss": 0.0302,
758
+ "step": 107
759
+ },
760
+ {
761
+ "epoch": 0.22,
762
+ "grad_norm": 2.07311798592702,
763
+ "learning_rate": 9.119866438367263e-06,
764
+ "loss": 0.0418,
765
+ "step": 108
766
+ },
767
+ {
768
+ "epoch": 0.22,
769
+ "grad_norm": 2.022648087929557,
770
+ "learning_rate": 9.101428420687759e-06,
771
+ "loss": 0.0448,
772
+ "step": 109
773
+ },
774
+ {
775
+ "epoch": 0.22,
776
+ "grad_norm": 8.182322737909697,
777
+ "learning_rate": 9.082818315286054e-06,
778
+ "loss": 0.0312,
779
+ "step": 110
780
+ },
781
+ {
782
+ "epoch": 0.22,
783
+ "grad_norm": 2.102798698987175,
784
+ "learning_rate": 9.0640369030049e-06,
785
+ "loss": 0.0336,
786
+ "step": 111
787
+ },
788
+ {
789
+ "epoch": 0.22,
790
+ "grad_norm": 1.7842549462877166,
791
+ "learning_rate": 9.045084971874738e-06,
792
+ "loss": 0.0327,
793
+ "step": 112
794
+ },
795
+ {
796
+ "epoch": 0.23,
797
+ "grad_norm": 2.1420067865524453,
798
+ "learning_rate": 9.025963317080641e-06,
799
+ "loss": 0.0316,
800
+ "step": 113
801
+ },
802
+ {
803
+ "epoch": 0.23,
804
+ "grad_norm": 1.4091043313029292,
805
+ "learning_rate": 9.006672740928952e-06,
806
+ "loss": 0.0253,
807
+ "step": 114
808
+ },
809
+ {
810
+ "epoch": 0.23,
811
+ "grad_norm": 5.641658831511388,
812
+ "learning_rate": 8.987214052813605e-06,
813
+ "loss": 0.0364,
814
+ "step": 115
815
+ },
816
+ {
817
+ "epoch": 0.23,
818
+ "grad_norm": 1.8365981933506732,
819
+ "learning_rate": 8.967588069182184e-06,
820
+ "loss": 0.0308,
821
+ "step": 116
822
+ },
823
+ {
824
+ "epoch": 0.23,
825
+ "grad_norm": 1.9206686733316154,
826
+ "learning_rate": 8.947795613501658e-06,
827
+ "loss": 0.0363,
828
+ "step": 117
829
+ },
830
+ {
831
+ "epoch": 0.24,
832
+ "grad_norm": 2.2303169644502514,
833
+ "learning_rate": 8.927837516223824e-06,
834
+ "loss": 0.0304,
835
+ "step": 118
836
+ },
837
+ {
838
+ "epoch": 0.24,
839
+ "grad_norm": 2.803903256974414,
840
+ "learning_rate": 8.907714614750473e-06,
841
+ "loss": 0.0364,
842
+ "step": 119
843
+ },
844
+ {
845
+ "epoch": 0.24,
846
+ "grad_norm": 1.8227774059382604,
847
+ "learning_rate": 8.887427753398249e-06,
848
+ "loss": 0.0356,
849
+ "step": 120
850
+ },
851
+ {
852
+ "epoch": 0.24,
853
+ "grad_norm": 3.8780357498558695,
854
+ "learning_rate": 8.866977783363219e-06,
855
+ "loss": 0.0398,
856
+ "step": 121
857
+ },
858
+ {
859
+ "epoch": 0.24,
860
+ "grad_norm": 2.644426904247864,
861
+ "learning_rate": 8.846365562685178e-06,
862
+ "loss": 0.0432,
863
+ "step": 122
864
+ },
865
+ {
866
+ "epoch": 0.25,
867
+ "grad_norm": 3.4352605499875137,
868
+ "learning_rate": 8.825591956211614e-06,
869
+ "loss": 0.0324,
870
+ "step": 123
871
+ },
872
+ {
873
+ "epoch": 0.25,
874
+ "grad_norm": 2.3350009805151037,
875
+ "learning_rate": 8.804657835561456e-06,
876
+ "loss": 0.038,
877
+ "step": 124
878
+ },
879
+ {
880
+ "epoch": 0.25,
881
+ "grad_norm": 2.5930387001732016,
882
+ "learning_rate": 8.783564079088478e-06,
883
+ "loss": 0.0303,
884
+ "step": 125
885
+ },
886
+ {
887
+ "epoch": 0.25,
888
+ "grad_norm": 3.342284462566408,
889
+ "learning_rate": 8.762311571844453e-06,
890
+ "loss": 0.0414,
891
+ "step": 126
892
+ },
893
+ {
894
+ "epoch": 0.25,
895
+ "grad_norm": 1.7767928014356122,
896
+ "learning_rate": 8.74090120554202e-06,
897
+ "loss": 0.0248,
898
+ "step": 127
899
+ },
900
+ {
901
+ "epoch": 0.26,
902
+ "grad_norm": 1.9411629493717453,
903
+ "learning_rate": 8.719333878517274e-06,
904
+ "loss": 0.0394,
905
+ "step": 128
906
+ },
907
+ {
908
+ "epoch": 0.26,
909
+ "grad_norm": 1.2311149579511578,
910
+ "learning_rate": 8.697610495692055e-06,
911
+ "loss": 0.03,
912
+ "step": 129
913
+ },
914
+ {
915
+ "epoch": 0.26,
916
+ "grad_norm": 1.7853017028883622,
917
+ "learning_rate": 8.675731968536004e-06,
918
+ "loss": 0.0272,
919
+ "step": 130
920
+ },
921
+ {
922
+ "epoch": 0.26,
923
+ "grad_norm": 10.405182353924877,
924
+ "learning_rate": 8.653699215028298e-06,
925
+ "loss": 0.0347,
926
+ "step": 131
927
+ },
928
+ {
929
+ "epoch": 0.26,
930
+ "grad_norm": 2.267411645090359,
931
+ "learning_rate": 8.63151315961915e-06,
932
+ "loss": 0.0344,
933
+ "step": 132
934
+ },
935
+ {
936
+ "epoch": 0.27,
937
+ "grad_norm": 2.7194599487459032,
938
+ "learning_rate": 8.609174733191012e-06,
939
+ "loss": 0.0325,
940
+ "step": 133
941
+ },
942
+ {
943
+ "epoch": 0.27,
944
+ "grad_norm": 3.472929645040217,
945
+ "learning_rate": 8.586684873019513e-06,
946
+ "loss": 0.033,
947
+ "step": 134
948
+ },
949
+ {
950
+ "epoch": 0.27,
951
+ "grad_norm": 2.2526538872121744,
952
+ "learning_rate": 8.564044522734147e-06,
953
+ "loss": 0.0299,
954
+ "step": 135
955
+ },
956
+ {
957
+ "epoch": 0.27,
958
+ "grad_norm": 1.7638640349688237,
959
+ "learning_rate": 8.541254632278667e-06,
960
+ "loss": 0.016,
961
+ "step": 136
962
+ },
963
+ {
964
+ "epoch": 0.27,
965
+ "grad_norm": 1.9705238267627823,
966
+ "learning_rate": 8.518316157871232e-06,
967
+ "loss": 0.0317,
968
+ "step": 137
969
+ },
970
+ {
971
+ "epoch": 0.28,
972
+ "grad_norm": 2.494620447668877,
973
+ "learning_rate": 8.495230061964289e-06,
974
+ "loss": 0.0321,
975
+ "step": 138
976
+ },
977
+ {
978
+ "epoch": 0.28,
979
+ "grad_norm": 2.1156870399464065,
980
+ "learning_rate": 8.471997313204183e-06,
981
+ "loss": 0.029,
982
+ "step": 139
983
+ },
984
+ {
985
+ "epoch": 0.28,
986
+ "grad_norm": 1.2309211661930353,
987
+ "learning_rate": 8.448618886390523e-06,
988
+ "loss": 0.0308,
989
+ "step": 140
990
+ },
991
+ {
992
+ "epoch": 0.28,
993
+ "grad_norm": 1.6655674091834316,
994
+ "learning_rate": 8.425095762435274e-06,
995
+ "loss": 0.0344,
996
+ "step": 141
997
+ },
998
+ {
999
+ "epoch": 0.28,
1000
+ "grad_norm": 1.3826341872992225,
1001
+ "learning_rate": 8.401428928321607e-06,
1002
+ "loss": 0.0192,
1003
+ "step": 142
1004
+ },
1005
+ {
1006
+ "epoch": 0.29,
1007
+ "grad_norm": 2.4208404672055863,
1008
+ "learning_rate": 8.377619377062483e-06,
1009
+ "loss": 0.0248,
1010
+ "step": 143
1011
+ },
1012
+ {
1013
+ "epoch": 0.29,
1014
+ "grad_norm": 1.9339923918812472,
1015
+ "learning_rate": 8.353668107658984e-06,
1016
+ "loss": 0.0307,
1017
+ "step": 144
1018
+ },
1019
+ {
1020
+ "epoch": 0.29,
1021
+ "grad_norm": 3.8159235347863647,
1022
+ "learning_rate": 8.329576125058406e-06,
1023
+ "loss": 0.0418,
1024
+ "step": 145
1025
+ },
1026
+ {
1027
+ "epoch": 0.29,
1028
+ "grad_norm": 1.7247843475045765,
1029
+ "learning_rate": 8.305344440112089e-06,
1030
+ "loss": 0.0276,
1031
+ "step": 146
1032
+ },
1033
+ {
1034
+ "epoch": 0.29,
1035
+ "grad_norm": 1.1868436201631507,
1036
+ "learning_rate": 8.280974069532999e-06,
1037
+ "loss": 0.0176,
1038
+ "step": 147
1039
+ },
1040
+ {
1041
+ "epoch": 0.3,
1042
+ "grad_norm": 1.4067807831089105,
1043
+ "learning_rate": 8.256466035853077e-06,
1044
+ "loss": 0.0221,
1045
+ "step": 148
1046
+ },
1047
+ {
1048
+ "epoch": 0.3,
1049
+ "grad_norm": 2.3813609861072975,
1050
+ "learning_rate": 8.231821367380335e-06,
1051
+ "loss": 0.0429,
1052
+ "step": 149
1053
+ },
1054
+ {
1055
+ "epoch": 0.3,
1056
+ "grad_norm": 3.4226557570351135,
1057
+ "learning_rate": 8.207041098155701e-06,
1058
+ "loss": 0.0279,
1059
+ "step": 150
1060
+ },
1061
+ {
1062
+ "epoch": 0.3,
1063
+ "grad_norm": 1.5164723715962871,
1064
+ "learning_rate": 8.182126267909642e-06,
1065
+ "loss": 0.035,
1066
+ "step": 151
1067
+ },
1068
+ {
1069
+ "epoch": 0.3,
1070
+ "grad_norm": 1.4157729072312213,
1071
+ "learning_rate": 8.157077922018537e-06,
1072
+ "loss": 0.031,
1073
+ "step": 152
1074
+ },
1075
+ {
1076
+ "epoch": 0.31,
1077
+ "grad_norm": 2.412999834705328,
1078
+ "learning_rate": 8.13189711146081e-06,
1079
+ "loss": 0.036,
1080
+ "step": 153
1081
+ },
1082
+ {
1083
+ "epoch": 0.31,
1084
+ "grad_norm": 2.1604098304834403,
1085
+ "learning_rate": 8.106584892772844e-06,
1086
+ "loss": 0.0446,
1087
+ "step": 154
1088
+ },
1089
+ {
1090
+ "epoch": 0.31,
1091
+ "grad_norm": 1.8099187169129658,
1092
+ "learning_rate": 8.081142328004638e-06,
1093
+ "loss": 0.0362,
1094
+ "step": 155
1095
+ },
1096
+ {
1097
+ "epoch": 0.31,
1098
+ "grad_norm": 1.3500376860976162,
1099
+ "learning_rate": 8.055570484675252e-06,
1100
+ "loss": 0.0292,
1101
+ "step": 156
1102
+ },
1103
+ {
1104
+ "epoch": 0.31,
1105
+ "grad_norm": 1.2969042450546118,
1106
+ "learning_rate": 8.029870435728018e-06,
1107
+ "loss": 0.0269,
1108
+ "step": 157
1109
+ },
1110
+ {
1111
+ "epoch": 0.32,
1112
+ "grad_norm": 1.443481544762842,
1113
+ "learning_rate": 8.004043259485519e-06,
1114
+ "loss": 0.0292,
1115
+ "step": 158
1116
+ },
1117
+ {
1118
+ "epoch": 0.32,
1119
+ "grad_norm": 1.4221828995737216,
1120
+ "learning_rate": 7.978090039604342e-06,
1121
+ "loss": 0.0299,
1122
+ "step": 159
1123
+ },
1124
+ {
1125
+ "epoch": 0.32,
1126
+ "grad_norm": 2.0486570056177893,
1127
+ "learning_rate": 7.952011865029614e-06,
1128
+ "loss": 0.0344,
1129
+ "step": 160
1130
+ },
1131
+ {
1132
+ "epoch": 0.32,
1133
+ "grad_norm": 1.3663452087626575,
1134
+ "learning_rate": 7.925809829949312e-06,
1135
+ "loss": 0.02,
1136
+ "step": 161
1137
+ },
1138
+ {
1139
+ "epoch": 0.32,
1140
+ "grad_norm": 2.5157316115454513,
1141
+ "learning_rate": 7.89948503374835e-06,
1142
+ "loss": 0.0557,
1143
+ "step": 162
1144
+ },
1145
+ {
1146
+ "epoch": 0.33,
1147
+ "grad_norm": 1.6577501474024066,
1148
+ "learning_rate": 7.873038580962453e-06,
1149
+ "loss": 0.0246,
1150
+ "step": 163
1151
+ },
1152
+ {
1153
+ "epoch": 0.33,
1154
+ "grad_norm": 1.8748680034247138,
1155
+ "learning_rate": 7.846471581231814e-06,
1156
+ "loss": 0.0317,
1157
+ "step": 164
1158
+ },
1159
+ {
1160
+ "epoch": 0.33,
1161
+ "grad_norm": 1.7027027975464297,
1162
+ "learning_rate": 7.819785149254534e-06,
1163
+ "loss": 0.0289,
1164
+ "step": 165
1165
+ },
1166
+ {
1167
+ "epoch": 0.33,
1168
+ "grad_norm": 1.259459625809648,
1169
+ "learning_rate": 7.792980404739849e-06,
1170
+ "loss": 0.0246,
1171
+ "step": 166
1172
+ },
1173
+ {
1174
+ "epoch": 0.33,
1175
+ "grad_norm": 10.120734456359045,
1176
+ "learning_rate": 7.766058472361154e-06,
1177
+ "loss": 0.0186,
1178
+ "step": 167
1179
+ },
1180
+ {
1181
+ "epoch": 0.34,
1182
+ "grad_norm": 1.4588855987929235,
1183
+ "learning_rate": 7.739020481708816e-06,
1184
+ "loss": 0.0272,
1185
+ "step": 168
1186
+ },
1187
+ {
1188
+ "epoch": 0.34,
1189
+ "grad_norm": 1.4436715547493166,
1190
+ "learning_rate": 7.711867567242769e-06,
1191
+ "loss": 0.022,
1192
+ "step": 169
1193
+ },
1194
+ {
1195
+ "epoch": 0.34,
1196
+ "grad_norm": 2.1772398518344986,
1197
+ "learning_rate": 7.68460086824492e-06,
1198
+ "loss": 0.0259,
1199
+ "step": 170
1200
+ },
1201
+ {
1202
+ "epoch": 0.34,
1203
+ "grad_norm": 1.4623740515910426,
1204
+ "learning_rate": 7.657221528771352e-06,
1205
+ "loss": 0.0271,
1206
+ "step": 171
1207
+ },
1208
+ {
1209
+ "epoch": 0.34,
1210
+ "grad_norm": 1.2982882763912305,
1211
+ "learning_rate": 7.629730697604314e-06,
1212
+ "loss": 0.0238,
1213
+ "step": 172
1214
+ },
1215
+ {
1216
+ "epoch": 0.35,
1217
+ "grad_norm": 2.2479187226730795,
1218
+ "learning_rate": 7.602129528204023e-06,
1219
+ "loss": 0.0355,
1220
+ "step": 173
1221
+ },
1222
+ {
1223
+ "epoch": 0.35,
1224
+ "grad_norm": 1.421101292379166,
1225
+ "learning_rate": 7.574419178660269e-06,
1226
+ "loss": 0.0315,
1227
+ "step": 174
1228
+ },
1229
+ {
1230
+ "epoch": 0.35,
1231
+ "grad_norm": 1.603145131464548,
1232
+ "learning_rate": 7.546600811643816e-06,
1233
+ "loss": 0.0263,
1234
+ "step": 175
1235
+ },
1236
+ {
1237
+ "epoch": 0.35,
1238
+ "grad_norm": 1.8144742918636079,
1239
+ "learning_rate": 7.5186755943576324e-06,
1240
+ "loss": 0.0262,
1241
+ "step": 176
1242
+ },
1243
+ {
1244
+ "epoch": 0.35,
1245
+ "grad_norm": 1.3485160117723676,
1246
+ "learning_rate": 7.490644698487909e-06,
1247
+ "loss": 0.0289,
1248
+ "step": 177
1249
+ },
1250
+ {
1251
+ "epoch": 0.36,
1252
+ "grad_norm": 1.6412473032382502,
1253
+ "learning_rate": 7.462509300154892e-06,
1254
+ "loss": 0.0302,
1255
+ "step": 178
1256
+ },
1257
+ {
1258
+ "epoch": 0.36,
1259
+ "grad_norm": 1.435826544232547,
1260
+ "learning_rate": 7.434270579863549e-06,
1261
+ "loss": 0.0193,
1262
+ "step": 179
1263
+ },
1264
+ {
1265
+ "epoch": 0.36,
1266
+ "grad_norm": 1.0572028598963985,
1267
+ "learning_rate": 7.405929722454026e-06,
1268
+ "loss": 0.0279,
1269
+ "step": 180
1270
+ },
1271
+ {
1272
+ "epoch": 0.36,
1273
+ "grad_norm": 1.304962775251496,
1274
+ "learning_rate": 7.3774879170519386e-06,
1275
+ "loss": 0.0198,
1276
+ "step": 181
1277
+ },
1278
+ {
1279
+ "epoch": 0.36,
1280
+ "grad_norm": 1.5033278833312405,
1281
+ "learning_rate": 7.348946357018479e-06,
1282
+ "loss": 0.0176,
1283
+ "step": 182
1284
+ },
1285
+ {
1286
+ "epoch": 0.37,
1287
+ "grad_norm": 4.109861542090897,
1288
+ "learning_rate": 7.320306239900343e-06,
1289
+ "loss": 0.0318,
1290
+ "step": 183
1291
+ },
1292
+ {
1293
+ "epoch": 0.37,
1294
+ "grad_norm": 1.8635309828936057,
1295
+ "learning_rate": 7.291568767379484e-06,
1296
+ "loss": 0.0298,
1297
+ "step": 184
1298
+ },
1299
+ {
1300
+ "epoch": 0.37,
1301
+ "grad_norm": 1.700064241715689,
1302
+ "learning_rate": 7.262735145222696e-06,
1303
+ "loss": 0.0206,
1304
+ "step": 185
1305
+ },
1306
+ {
1307
+ "epoch": 0.37,
1308
+ "grad_norm": 1.3327471500230215,
1309
+ "learning_rate": 7.233806583231012e-06,
1310
+ "loss": 0.0194,
1311
+ "step": 186
1312
+ },
1313
+ {
1314
+ "epoch": 0.37,
1315
+ "grad_norm": 1.3563536670135046,
1316
+ "learning_rate": 7.204784295188959e-06,
1317
+ "loss": 0.0281,
1318
+ "step": 187
1319
+ },
1320
+ {
1321
+ "epoch": 0.38,
1322
+ "grad_norm": 1.8054652956349042,
1323
+ "learning_rate": 7.1756694988136165e-06,
1324
+ "loss": 0.0222,
1325
+ "step": 188
1326
+ },
1327
+ {
1328
+ "epoch": 0.38,
1329
+ "grad_norm": 1.6558037455318473,
1330
+ "learning_rate": 7.14646341570353e-06,
1331
+ "loss": 0.0341,
1332
+ "step": 189
1333
+ },
1334
+ {
1335
+ "epoch": 0.38,
1336
+ "grad_norm": 1.2664567109533194,
1337
+ "learning_rate": 7.117167271287453e-06,
1338
+ "loss": 0.0223,
1339
+ "step": 190
1340
+ },
1341
+ {
1342
+ "epoch": 0.38,
1343
+ "grad_norm": 1.8611199264470106,
1344
+ "learning_rate": 7.0877822947729265e-06,
1345
+ "loss": 0.0161,
1346
+ "step": 191
1347
+ },
1348
+ {
1349
+ "epoch": 0.38,
1350
+ "grad_norm": 9.864612348900984,
1351
+ "learning_rate": 7.05830971909472e-06,
1352
+ "loss": 0.0218,
1353
+ "step": 192
1354
+ },
1355
+ {
1356
+ "epoch": 0.39,
1357
+ "grad_norm": 2.0371148673208745,
1358
+ "learning_rate": 7.028750780863078e-06,
1359
+ "loss": 0.0241,
1360
+ "step": 193
1361
+ },
1362
+ {
1363
+ "epoch": 0.39,
1364
+ "grad_norm": 1.9981953987747059,
1365
+ "learning_rate": 6.999106720311846e-06,
1366
+ "loss": 0.0269,
1367
+ "step": 194
1368
+ },
1369
+ {
1370
+ "epoch": 0.39,
1371
+ "grad_norm": 1.8989854958128511,
1372
+ "learning_rate": 6.969378781246436e-06,
1373
+ "loss": 0.0233,
1374
+ "step": 195
1375
+ },
1376
+ {
1377
+ "epoch": 0.39,
1378
+ "grad_norm": 2.730664172508983,
1379
+ "learning_rate": 6.939568210991633e-06,
1380
+ "loss": 0.0211,
1381
+ "step": 196
1382
+ },
1383
+ {
1384
+ "epoch": 0.39,
1385
+ "grad_norm": 2.380611095389141,
1386
+ "learning_rate": 6.9096762603392595e-06,
1387
+ "loss": 0.024,
1388
+ "step": 197
1389
+ },
1390
+ {
1391
+ "epoch": 0.4,
1392
+ "grad_norm": 3.0440854236811385,
1393
+ "learning_rate": 6.8797041834956955e-06,
1394
+ "loss": 0.0258,
1395
+ "step": 198
1396
+ },
1397
+ {
1398
+ "epoch": 0.4,
1399
+ "grad_norm": 1.5115073349959407,
1400
+ "learning_rate": 6.849653238029261e-06,
1401
+ "loss": 0.0229,
1402
+ "step": 199
1403
+ },
1404
+ {
1405
+ "epoch": 0.4,
1406
+ "grad_norm": 1.6974089932476533,
1407
+ "learning_rate": 6.819524684817439e-06,
1408
+ "loss": 0.0289,
1409
+ "step": 200
1410
+ },
1411
+ {
1412
+ "epoch": 0.4,
1413
+ "grad_norm": 1.2243125057970263,
1414
+ "learning_rate": 6.78931978799398e-06,
1415
+ "loss": 0.0246,
1416
+ "step": 201
1417
+ },
1418
+ {
1419
+ "epoch": 0.4,
1420
+ "grad_norm": 2.6717717416764137,
1421
+ "learning_rate": 6.7590398148958625e-06,
1422
+ "loss": 0.0308,
1423
+ "step": 202
1424
+ },
1425
+ {
1426
+ "epoch": 0.41,
1427
+ "grad_norm": 1.1598888317399834,
1428
+ "learning_rate": 6.728686036010115e-06,
1429
+ "loss": 0.0209,
1430
+ "step": 203
1431
+ },
1432
+ {
1433
+ "epoch": 0.41,
1434
+ "grad_norm": 1.3831321362305298,
1435
+ "learning_rate": 6.698259724920503e-06,
1436
+ "loss": 0.0249,
1437
+ "step": 204
1438
+ },
1439
+ {
1440
+ "epoch": 0.41,
1441
+ "grad_norm": 1.5323574831848763,
1442
+ "learning_rate": 6.667762158254104e-06,
1443
+ "loss": 0.0237,
1444
+ "step": 205
1445
+ },
1446
+ {
1447
+ "epoch": 0.41,
1448
+ "grad_norm": 0.7042346300795984,
1449
+ "learning_rate": 6.637194615627733e-06,
1450
+ "loss": 0.0155,
1451
+ "step": 206
1452
+ },
1453
+ {
1454
+ "epoch": 0.41,
1455
+ "grad_norm": 1.1596200971614221,
1456
+ "learning_rate": 6.6065583795942625e-06,
1457
+ "loss": 0.0258,
1458
+ "step": 207
1459
+ },
1460
+ {
1461
+ "epoch": 0.42,
1462
+ "grad_norm": 1.5336354027758974,
1463
+ "learning_rate": 6.5758547355887944e-06,
1464
+ "loss": 0.0292,
1465
+ "step": 208
1466
+ },
1467
+ {
1468
+ "epoch": 0.42,
1469
+ "grad_norm": 1.320381578649728,
1470
+ "learning_rate": 6.545084971874738e-06,
1471
+ "loss": 0.0204,
1472
+ "step": 209
1473
+ },
1474
+ {
1475
+ "epoch": 0.42,
1476
+ "grad_norm": 3.459139728829293,
1477
+ "learning_rate": 6.514250379489754e-06,
1478
+ "loss": 0.029,
1479
+ "step": 210
1480
+ },
1481
+ {
1482
+ "epoch": 0.42,
1483
+ "grad_norm": 3.0546943664425226,
1484
+ "learning_rate": 6.483352252191585e-06,
1485
+ "loss": 0.0261,
1486
+ "step": 211
1487
+ },
1488
+ {
1489
+ "epoch": 0.42,
1490
+ "grad_norm": 1.8383132757525742,
1491
+ "learning_rate": 6.452391886403767e-06,
1492
+ "loss": 0.0279,
1493
+ "step": 212
1494
+ },
1495
+ {
1496
+ "epoch": 0.43,
1497
+ "grad_norm": 1.9964104722460956,
1498
+ "learning_rate": 6.421370581161244e-06,
1499
+ "loss": 0.0289,
1500
+ "step": 213
1501
+ },
1502
+ {
1503
+ "epoch": 0.43,
1504
+ "grad_norm": 2.6130915844073965,
1505
+ "learning_rate": 6.390289638055851e-06,
1506
+ "loss": 0.032,
1507
+ "step": 214
1508
+ },
1509
+ {
1510
+ "epoch": 0.43,
1511
+ "grad_norm": 1.2827262595043651,
1512
+ "learning_rate": 6.3591503611817155e-06,
1513
+ "loss": 0.0227,
1514
+ "step": 215
1515
+ },
1516
+ {
1517
+ "epoch": 0.43,
1518
+ "grad_norm": 1.5393360377089653,
1519
+ "learning_rate": 6.3279540570805265e-06,
1520
+ "loss": 0.0188,
1521
+ "step": 216
1522
+ },
1523
+ {
1524
+ "epoch": 0.43,
1525
+ "grad_norm": 2.178353143951541,
1526
+ "learning_rate": 6.296702034686726e-06,
1527
+ "loss": 0.0321,
1528
+ "step": 217
1529
+ },
1530
+ {
1531
+ "epoch": 0.44,
1532
+ "grad_norm": 1.4351230515403046,
1533
+ "learning_rate": 6.265395605272581e-06,
1534
+ "loss": 0.0286,
1535
+ "step": 218
1536
+ },
1537
+ {
1538
+ "epoch": 0.44,
1539
+ "grad_norm": 1.3522571682917552,
1540
+ "learning_rate": 6.234036082393171e-06,
1541
+ "loss": 0.0254,
1542
+ "step": 219
1543
+ },
1544
+ {
1545
+ "epoch": 0.44,
1546
+ "grad_norm": 1.1047431847306042,
1547
+ "learning_rate": 6.202624781831269e-06,
1548
+ "loss": 0.0311,
1549
+ "step": 220
1550
+ },
1551
+ {
1552
+ "epoch": 0.44,
1553
+ "grad_norm": 1.3562256309208196,
1554
+ "learning_rate": 6.171163021542134e-06,
1555
+ "loss": 0.0414,
1556
+ "step": 221
1557
+ },
1558
+ {
1559
+ "epoch": 0.44,
1560
+ "grad_norm": 1.4751253895069956,
1561
+ "learning_rate": 6.139652121598219e-06,
1562
+ "loss": 0.0295,
1563
+ "step": 222
1564
+ },
1565
+ {
1566
+ "epoch": 0.45,
1567
+ "grad_norm": 0.8904175856102439,
1568
+ "learning_rate": 6.108093404133772e-06,
1569
+ "loss": 0.014,
1570
+ "step": 223
1571
+ },
1572
+ {
1573
+ "epoch": 0.45,
1574
+ "grad_norm": 1.341503639464688,
1575
+ "learning_rate": 6.076488193289375e-06,
1576
+ "loss": 0.0245,
1577
+ "step": 224
1578
+ },
1579
+ {
1580
+ "epoch": 0.45,
1581
+ "grad_norm": 1.1260598919776161,
1582
+ "learning_rate": 6.044837815156377e-06,
1583
+ "loss": 0.0205,
1584
+ "step": 225
1585
+ },
1586
+ {
1587
+ "epoch": 0.45,
1588
+ "grad_norm": 1.2316754707144313,
1589
+ "learning_rate": 6.013143597721252e-06,
1590
+ "loss": 0.0219,
1591
+ "step": 226
1592
+ },
1593
+ {
1594
+ "epoch": 0.45,
1595
+ "grad_norm": 1.244353012719773,
1596
+ "learning_rate": 5.981406870809889e-06,
1597
+ "loss": 0.0225,
1598
+ "step": 227
1599
+ },
1600
+ {
1601
+ "epoch": 0.46,
1602
+ "grad_norm": 1.2067458253203331,
1603
+ "learning_rate": 5.949628966031785e-06,
1604
+ "loss": 0.0247,
1605
+ "step": 228
1606
+ },
1607
+ {
1608
+ "epoch": 0.46,
1609
+ "grad_norm": 1.7525447372471177,
1610
+ "learning_rate": 5.9178112167241805e-06,
1611
+ "loss": 0.0267,
1612
+ "step": 229
1613
+ },
1614
+ {
1615
+ "epoch": 0.46,
1616
+ "grad_norm": 1.285892308265446,
1617
+ "learning_rate": 5.885954957896115e-06,
1618
+ "loss": 0.0211,
1619
+ "step": 230
1620
+ },
1621
+ {
1622
+ "epoch": 0.46,
1623
+ "grad_norm": 1.2655044932217716,
1624
+ "learning_rate": 5.854061526172402e-06,
1625
+ "loss": 0.0215,
1626
+ "step": 231
1627
+ },
1628
+ {
1629
+ "epoch": 0.46,
1630
+ "grad_norm": 0.8450183732481381,
1631
+ "learning_rate": 5.822132259737565e-06,
1632
+ "loss": 0.0139,
1633
+ "step": 232
1634
+ },
1635
+ {
1636
+ "epoch": 0.47,
1637
+ "grad_norm": 1.231038573711708,
1638
+ "learning_rate": 5.7901684982796716e-06,
1639
+ "loss": 0.0237,
1640
+ "step": 233
1641
+ },
1642
+ {
1643
+ "epoch": 0.47,
1644
+ "grad_norm": 0.8616252932972882,
1645
+ "learning_rate": 5.75817158293414e-06,
1646
+ "loss": 0.0198,
1647
+ "step": 234
1648
+ },
1649
+ {
1650
+ "epoch": 0.47,
1651
+ "grad_norm": 1.097591281668511,
1652
+ "learning_rate": 5.726142856227453e-06,
1653
+ "loss": 0.0199,
1654
+ "step": 235
1655
+ },
1656
+ {
1657
+ "epoch": 0.47,
1658
+ "grad_norm": 0.9026637389293156,
1659
+ "learning_rate": 5.694083662020835e-06,
1660
+ "loss": 0.0165,
1661
+ "step": 236
1662
+ },
1663
+ {
1664
+ "epoch": 0.47,
1665
+ "grad_norm": 1.1371871944037466,
1666
+ "learning_rate": 5.661995345453867e-06,
1667
+ "loss": 0.015,
1668
+ "step": 237
1669
+ },
1670
+ {
1671
+ "epoch": 0.48,
1672
+ "grad_norm": 1.1194551430640876,
1673
+ "learning_rate": 5.629879252888046e-06,
1674
+ "loss": 0.025,
1675
+ "step": 238
1676
+ },
1677
+ {
1678
+ "epoch": 0.48,
1679
+ "grad_norm": 1.5692980288918033,
1680
+ "learning_rate": 5.597736731850295e-06,
1681
+ "loss": 0.0193,
1682
+ "step": 239
1683
+ },
1684
+ {
1685
+ "epoch": 0.48,
1686
+ "grad_norm": 1.2144907091588857,
1687
+ "learning_rate": 5.5655691309764225e-06,
1688
+ "loss": 0.0246,
1689
+ "step": 240
1690
+ },
1691
+ {
1692
+ "epoch": 0.48,
1693
+ "grad_norm": 1.125742395479265,
1694
+ "learning_rate": 5.533377799954532e-06,
1695
+ "loss": 0.0243,
1696
+ "step": 241
1697
+ },
1698
+ {
1699
+ "epoch": 0.48,
1700
+ "grad_norm": 1.2148479718729075,
1701
+ "learning_rate": 5.501164089468406e-06,
1702
+ "loss": 0.0266,
1703
+ "step": 242
1704
+ },
1705
+ {
1706
+ "epoch": 0.49,
1707
+ "grad_norm": 0.9618960534219361,
1708
+ "learning_rate": 5.4689293511408155e-06,
1709
+ "loss": 0.0206,
1710
+ "step": 243
1711
+ },
1712
+ {
1713
+ "epoch": 0.49,
1714
+ "grad_norm": 1.0858785521346364,
1715
+ "learning_rate": 5.43667493747682e-06,
1716
+ "loss": 0.0152,
1717
+ "step": 244
1718
+ },
1719
+ {
1720
+ "epoch": 0.49,
1721
+ "grad_norm": 1.9632568319741832,
1722
+ "learning_rate": 5.404402201807022e-06,
1723
+ "loss": 0.021,
1724
+ "step": 245
1725
+ },
1726
+ {
1727
+ "epoch": 0.49,
1728
+ "grad_norm": 1.3825751435750187,
1729
+ "learning_rate": 5.372112498230771e-06,
1730
+ "loss": 0.0277,
1731
+ "step": 246
1732
+ },
1733
+ {
1734
+ "epoch": 0.49,
1735
+ "grad_norm": 1.467739760194593,
1736
+ "learning_rate": 5.339807181559359e-06,
1737
+ "loss": 0.0275,
1738
+ "step": 247
1739
+ },
1740
+ {
1741
+ "epoch": 0.5,
1742
+ "grad_norm": 1.1142452086061956,
1743
+ "learning_rate": 5.307487607259175e-06,
1744
+ "loss": 0.0265,
1745
+ "step": 248
1746
+ },
1747
+ {
1748
+ "epoch": 0.5,
1749
+ "grad_norm": 2.0131362553768017,
1750
+ "learning_rate": 5.275155131394825e-06,
1751
+ "loss": 0.033,
1752
+ "step": 249
1753
+ },
1754
+ {
1755
+ "epoch": 0.5,
1756
+ "grad_norm": 1.7106656017553448,
1757
+ "learning_rate": 5.242811110572243e-06,
1758
+ "loss": 0.0254,
1759
+ "step": 250
1760
+ },
1761
+ {
1762
+ "epoch": 0.5,
1763
+ "grad_norm": 1.3827754086909076,
1764
+ "learning_rate": 5.210456901881761e-06,
1765
+ "loss": 0.0275,
1766
+ "step": 251
1767
+ },
1768
+ {
1769
+ "epoch": 0.5,
1770
+ "grad_norm": 1.1060302836748128,
1771
+ "learning_rate": 5.1780938628411795e-06,
1772
+ "loss": 0.0201,
1773
+ "step": 252
1774
+ },
1775
+ {
1776
+ "epoch": 0.51,
1777
+ "grad_norm": 1.3070388963094217,
1778
+ "learning_rate": 5.145723351338799e-06,
1779
+ "loss": 0.0248,
1780
+ "step": 253
1781
+ },
1782
+ {
1783
+ "epoch": 0.51,
1784
+ "grad_norm": 1.6188089855502,
1785
+ "learning_rate": 5.11334672557645e-06,
1786
+ "loss": 0.0251,
1787
+ "step": 254
1788
+ },
1789
+ {
1790
+ "epoch": 0.51,
1791
+ "grad_norm": 1.1025069914783028,
1792
+ "learning_rate": 5.080965344012509e-06,
1793
+ "loss": 0.0214,
1794
+ "step": 255
1795
+ },
1796
+ {
1797
+ "epoch": 0.51,
1798
+ "grad_norm": 1.4639797117222222,
1799
+ "learning_rate": 5.048580565304887e-06,
1800
+ "loss": 0.0309,
1801
+ "step": 256
1802
+ },
1803
+ {
1804
+ "epoch": 0.51,
1805
+ "grad_norm": 1.2634085559909505,
1806
+ "learning_rate": 5.016193748254045e-06,
1807
+ "loss": 0.0263,
1808
+ "step": 257
1809
+ },
1810
+ {
1811
+ "epoch": 0.52,
1812
+ "grad_norm": 2.401826264360897,
1813
+ "learning_rate": 4.983806251745958e-06,
1814
+ "loss": 0.0245,
1815
+ "step": 258
1816
+ },
1817
+ {
1818
+ "epoch": 0.52,
1819
+ "grad_norm": 1.2131016102910568,
1820
+ "learning_rate": 4.951419434695115e-06,
1821
+ "loss": 0.022,
1822
+ "step": 259
1823
+ },
1824
+ {
1825
+ "epoch": 0.52,
1826
+ "grad_norm": 1.387179441243972,
1827
+ "learning_rate": 4.919034655987493e-06,
1828
+ "loss": 0.0228,
1829
+ "step": 260
1830
+ },
1831
+ {
1832
+ "epoch": 0.52,
1833
+ "grad_norm": 1.5014106023675926,
1834
+ "learning_rate": 4.886653274423551e-06,
1835
+ "loss": 0.0266,
1836
+ "step": 261
1837
+ },
1838
+ {
1839
+ "epoch": 0.52,
1840
+ "grad_norm": 1.62238675696772,
1841
+ "learning_rate": 4.8542766486612035e-06,
1842
+ "loss": 0.0254,
1843
+ "step": 262
1844
+ },
1845
+ {
1846
+ "epoch": 0.53,
1847
+ "grad_norm": 1.2417979776067387,
1848
+ "learning_rate": 4.821906137158822e-06,
1849
+ "loss": 0.0206,
1850
+ "step": 263
1851
+ },
1852
+ {
1853
+ "epoch": 0.53,
1854
+ "grad_norm": 1.0223301156801394,
1855
+ "learning_rate": 4.7895430981182415e-06,
1856
+ "loss": 0.0206,
1857
+ "step": 264
1858
+ },
1859
+ {
1860
+ "epoch": 0.53,
1861
+ "grad_norm": 1.2193380002647791,
1862
+ "learning_rate": 4.757188889427761e-06,
1863
+ "loss": 0.0217,
1864
+ "step": 265
1865
+ },
1866
+ {
1867
+ "epoch": 0.53,
1868
+ "grad_norm": 1.006885095941849,
1869
+ "learning_rate": 4.724844868605176e-06,
1870
+ "loss": 0.0191,
1871
+ "step": 266
1872
+ },
1873
+ {
1874
+ "epoch": 0.53,
1875
+ "grad_norm": 0.9226097277140832,
1876
+ "learning_rate": 4.6925123927408265e-06,
1877
+ "loss": 0.0128,
1878
+ "step": 267
1879
+ },
1880
+ {
1881
+ "epoch": 0.54,
1882
+ "grad_norm": 1.5498107181284657,
1883
+ "learning_rate": 4.660192818440642e-06,
1884
+ "loss": 0.0228,
1885
+ "step": 268
1886
+ },
1887
+ {
1888
+ "epoch": 0.54,
1889
+ "grad_norm": 0.7821729716320834,
1890
+ "learning_rate": 4.627887501769231e-06,
1891
+ "loss": 0.0149,
1892
+ "step": 269
1893
+ },
1894
+ {
1895
+ "epoch": 0.54,
1896
+ "grad_norm": 1.4801219995657608,
1897
+ "learning_rate": 4.59559779819298e-06,
1898
+ "loss": 0.031,
1899
+ "step": 270
1900
+ },
1901
+ {
1902
+ "epoch": 0.54,
1903
+ "grad_norm": 1.3517242351761702,
1904
+ "learning_rate": 4.5633250625231806e-06,
1905
+ "loss": 0.0258,
1906
+ "step": 271
1907
+ },
1908
+ {
1909
+ "epoch": 0.54,
1910
+ "grad_norm": 1.7399328000837162,
1911
+ "learning_rate": 4.531070648859186e-06,
1912
+ "loss": 0.0317,
1913
+ "step": 272
1914
+ },
1915
+ {
1916
+ "epoch": 0.55,
1917
+ "grad_norm": 0.8514451456821471,
1918
+ "learning_rate": 4.498835910531595e-06,
1919
+ "loss": 0.0176,
1920
+ "step": 273
1921
+ },
1922
+ {
1923
+ "epoch": 0.55,
1924
+ "grad_norm": 0.7340657635665374,
1925
+ "learning_rate": 4.4666222000454685e-06,
1926
+ "loss": 0.0146,
1927
+ "step": 274
1928
+ },
1929
+ {
1930
+ "epoch": 0.55,
1931
+ "grad_norm": 1.6655543591379138,
1932
+ "learning_rate": 4.434430869023579e-06,
1933
+ "loss": 0.0242,
1934
+ "step": 275
1935
+ },
1936
+ {
1937
+ "epoch": 0.55,
1938
+ "grad_norm": 1.379561159409268,
1939
+ "learning_rate": 4.402263268149707e-06,
1940
+ "loss": 0.0247,
1941
+ "step": 276
1942
+ },
1943
+ {
1944
+ "epoch": 0.55,
1945
+ "grad_norm": 1.108331162162277,
1946
+ "learning_rate": 4.370120747111956e-06,
1947
+ "loss": 0.0214,
1948
+ "step": 277
1949
+ },
1950
+ {
1951
+ "epoch": 0.56,
1952
+ "grad_norm": 1.202164182947932,
1953
+ "learning_rate": 4.338004654546136e-06,
1954
+ "loss": 0.0208,
1955
+ "step": 278
1956
+ },
1957
+ {
1958
+ "epoch": 0.56,
1959
+ "grad_norm": 1.6219618221437264,
1960
+ "learning_rate": 4.3059163379791676e-06,
1961
+ "loss": 0.025,
1962
+ "step": 279
1963
+ },
1964
+ {
1965
+ "epoch": 0.56,
1966
+ "grad_norm": 0.9339521278582009,
1967
+ "learning_rate": 4.27385714377255e-06,
1968
+ "loss": 0.0165,
1969
+ "step": 280
1970
+ },
1971
+ {
1972
+ "epoch": 0.56,
1973
+ "grad_norm": 1.0200494412107808,
1974
+ "learning_rate": 4.24182841706586e-06,
1975
+ "loss": 0.0213,
1976
+ "step": 281
1977
+ },
1978
+ {
1979
+ "epoch": 0.56,
1980
+ "grad_norm": 0.8815703419251237,
1981
+ "learning_rate": 4.209831501720328e-06,
1982
+ "loss": 0.0193,
1983
+ "step": 282
1984
+ },
1985
+ {
1986
+ "epoch": 0.57,
1987
+ "grad_norm": 1.202793491421607,
1988
+ "learning_rate": 4.177867740262437e-06,
1989
+ "loss": 0.0185,
1990
+ "step": 283
1991
+ },
1992
+ {
1993
+ "epoch": 0.57,
1994
+ "grad_norm": 1.2225404005197098,
1995
+ "learning_rate": 4.145938473827598e-06,
1996
+ "loss": 0.0214,
1997
+ "step": 284
1998
+ },
1999
+ {
2000
+ "epoch": 0.57,
2001
+ "grad_norm": 1.0364703340755124,
2002
+ "learning_rate": 4.1140450421038865e-06,
2003
+ "loss": 0.0244,
2004
+ "step": 285
2005
+ },
2006
+ {
2007
+ "epoch": 0.57,
2008
+ "grad_norm": 1.026305772288425,
2009
+ "learning_rate": 4.08218878327582e-06,
2010
+ "loss": 0.0161,
2011
+ "step": 286
2012
+ },
2013
+ {
2014
+ "epoch": 0.57,
2015
+ "grad_norm": 1.1221495322408541,
2016
+ "learning_rate": 4.050371033968216e-06,
2017
+ "loss": 0.0192,
2018
+ "step": 287
2019
+ },
2020
+ {
2021
+ "epoch": 0.58,
2022
+ "grad_norm": 0.8817095091748988,
2023
+ "learning_rate": 4.018593129190113e-06,
2024
+ "loss": 0.0134,
2025
+ "step": 288
2026
+ },
2027
+ {
2028
+ "epoch": 0.58,
2029
+ "grad_norm": 0.9747345763267818,
2030
+ "learning_rate": 3.98685640227875e-06,
2031
+ "loss": 0.0226,
2032
+ "step": 289
2033
+ },
2034
+ {
2035
+ "epoch": 0.58,
2036
+ "grad_norm": 0.9982684850063999,
2037
+ "learning_rate": 3.955162184843625e-06,
2038
+ "loss": 0.0195,
2039
+ "step": 290
2040
+ },
2041
+ {
2042
+ "epoch": 0.58,
2043
+ "grad_norm": 0.9695335229320834,
2044
+ "learning_rate": 3.9235118067106255e-06,
2045
+ "loss": 0.0153,
2046
+ "step": 291
2047
+ },
2048
+ {
2049
+ "epoch": 0.58,
2050
+ "grad_norm": 1.304247925221665,
2051
+ "learning_rate": 3.89190659586623e-06,
2052
+ "loss": 0.0208,
2053
+ "step": 292
2054
+ },
2055
+ {
2056
+ "epoch": 0.59,
2057
+ "grad_norm": 0.7190780015296302,
2058
+ "learning_rate": 3.8603478784017845e-06,
2059
+ "loss": 0.0135,
2060
+ "step": 293
2061
+ },
2062
+ {
2063
+ "epoch": 0.59,
2064
+ "grad_norm": 0.9132734546691439,
2065
+ "learning_rate": 3.828836978457868e-06,
2066
+ "loss": 0.0187,
2067
+ "step": 294
2068
+ },
2069
+ {
2070
+ "epoch": 0.59,
2071
+ "grad_norm": 1.138182220158514,
2072
+ "learning_rate": 3.7973752181687336e-06,
2073
+ "loss": 0.0207,
2074
+ "step": 295
2075
+ },
2076
+ {
2077
+ "epoch": 0.59,
2078
+ "grad_norm": 1.2239401605335127,
2079
+ "learning_rate": 3.7659639176068287e-06,
2080
+ "loss": 0.017,
2081
+ "step": 296
2082
+ },
2083
+ {
2084
+ "epoch": 0.59,
2085
+ "grad_norm": 1.0911081008130064,
2086
+ "learning_rate": 3.734604394727419e-06,
2087
+ "loss": 0.0236,
2088
+ "step": 297
2089
+ },
2090
+ {
2091
+ "epoch": 0.6,
2092
+ "grad_norm": 1.1234055740466902,
2093
+ "learning_rate": 3.703297965313275e-06,
2094
+ "loss": 0.0164,
2095
+ "step": 298
2096
+ },
2097
+ {
2098
+ "epoch": 0.6,
2099
+ "grad_norm": 1.6245699602433148,
2100
+ "learning_rate": 3.6720459429194743e-06,
2101
+ "loss": 0.0206,
2102
+ "step": 299
2103
+ },
2104
+ {
2105
+ "epoch": 0.6,
2106
+ "grad_norm": 0.8456189844420401,
2107
+ "learning_rate": 3.6408496388182857e-06,
2108
+ "loss": 0.0139,
2109
+ "step": 300
2110
+ },
2111
+ {
2112
+ "epoch": 0.6,
2113
+ "grad_norm": 1.1037140258093119,
2114
+ "learning_rate": 3.6097103619441505e-06,
2115
+ "loss": 0.0182,
2116
+ "step": 301
2117
+ },
2118
+ {
2119
+ "epoch": 0.6,
2120
+ "grad_norm": 1.0958235436621413,
2121
+ "learning_rate": 3.578629418838757e-06,
2122
+ "loss": 0.0144,
2123
+ "step": 302
2124
+ },
2125
+ {
2126
+ "epoch": 0.61,
2127
+ "grad_norm": 0.9343050348703398,
2128
+ "learning_rate": 3.5476081135962335e-06,
2129
+ "loss": 0.0135,
2130
+ "step": 303
2131
+ },
2132
+ {
2133
+ "epoch": 0.61,
2134
+ "grad_norm": 0.989208316982701,
2135
+ "learning_rate": 3.516647747808417e-06,
2136
+ "loss": 0.0184,
2137
+ "step": 304
2138
+ },
2139
+ {
2140
+ "epoch": 0.61,
2141
+ "grad_norm": 1.0323252072696174,
2142
+ "learning_rate": 3.4857496205102475e-06,
2143
+ "loss": 0.0214,
2144
+ "step": 305
2145
+ },
2146
+ {
2147
+ "epoch": 0.61,
2148
+ "grad_norm": 0.8853633933322875,
2149
+ "learning_rate": 3.4549150281252635e-06,
2150
+ "loss": 0.0112,
2151
+ "step": 306
2152
+ },
2153
+ {
2154
+ "epoch": 0.61,
2155
+ "grad_norm": 0.8554219852805254,
2156
+ "learning_rate": 3.4241452644112085e-06,
2157
+ "loss": 0.0138,
2158
+ "step": 307
2159
+ },
2160
+ {
2161
+ "epoch": 0.62,
2162
+ "grad_norm": 0.9473250325217796,
2163
+ "learning_rate": 3.3934416204057396e-06,
2164
+ "loss": 0.0137,
2165
+ "step": 308
2166
+ },
2167
+ {
2168
+ "epoch": 0.62,
2169
+ "grad_norm": 1.0821994682023992,
2170
+ "learning_rate": 3.3628053843722674e-06,
2171
+ "loss": 0.0148,
2172
+ "step": 309
2173
+ },
2174
+ {
2175
+ "epoch": 0.62,
2176
+ "grad_norm": 0.8933304986459344,
2177
+ "learning_rate": 3.3322378417458985e-06,
2178
+ "loss": 0.0133,
2179
+ "step": 310
2180
+ },
2181
+ {
2182
+ "epoch": 0.62,
2183
+ "grad_norm": 1.5645419502824816,
2184
+ "learning_rate": 3.3017402750794976e-06,
2185
+ "loss": 0.0295,
2186
+ "step": 311
2187
+ },
2188
+ {
2189
+ "epoch": 0.62,
2190
+ "grad_norm": 1.4216161849435738,
2191
+ "learning_rate": 3.271313963989886e-06,
2192
+ "loss": 0.0232,
2193
+ "step": 312
2194
+ },
2195
+ {
2196
+ "epoch": 0.63,
2197
+ "grad_norm": 1.487170298191098,
2198
+ "learning_rate": 3.240960185104137e-06,
2199
+ "loss": 0.0252,
2200
+ "step": 313
2201
+ },
2202
+ {
2203
+ "epoch": 0.63,
2204
+ "grad_norm": 0.9851878637626099,
2205
+ "learning_rate": 3.2106802120060197e-06,
2206
+ "loss": 0.0269,
2207
+ "step": 314
2208
+ },
2209
+ {
2210
+ "epoch": 0.63,
2211
+ "grad_norm": 0.955699263132902,
2212
+ "learning_rate": 3.180475315182563e-06,
2213
+ "loss": 0.0192,
2214
+ "step": 315
2215
+ },
2216
+ {
2217
+ "epoch": 0.63,
2218
+ "grad_norm": 1.9701528271155129,
2219
+ "learning_rate": 3.1503467619707407e-06,
2220
+ "loss": 0.0227,
2221
+ "step": 316
2222
+ },
2223
+ {
2224
+ "epoch": 0.63,
2225
+ "grad_norm": 0.8929326235012384,
2226
+ "learning_rate": 3.1202958165043053e-06,
2227
+ "loss": 0.016,
2228
+ "step": 317
2229
+ },
2230
+ {
2231
+ "epoch": 0.64,
2232
+ "grad_norm": 1.0969735717706128,
2233
+ "learning_rate": 3.090323739660742e-06,
2234
+ "loss": 0.0216,
2235
+ "step": 318
2236
+ },
2237
+ {
2238
+ "epoch": 0.64,
2239
+ "grad_norm": 0.981178343847294,
2240
+ "learning_rate": 3.060431789008368e-06,
2241
+ "loss": 0.0201,
2242
+ "step": 319
2243
+ },
2244
+ {
2245
+ "epoch": 0.64,
2246
+ "grad_norm": 0.9175091725694041,
2247
+ "learning_rate": 3.0306212187535653e-06,
2248
+ "loss": 0.0152,
2249
+ "step": 320
2250
+ },
2251
+ {
2252
+ "epoch": 0.64,
2253
+ "grad_norm": 1.009966112131104,
2254
+ "learning_rate": 3.000893279688155e-06,
2255
+ "loss": 0.0151,
2256
+ "step": 321
2257
+ },
2258
+ {
2259
+ "epoch": 0.64,
2260
+ "grad_norm": 1.0855300898909286,
2261
+ "learning_rate": 2.9712492191369245e-06,
2262
+ "loss": 0.0292,
2263
+ "step": 322
2264
+ },
2265
+ {
2266
+ "epoch": 0.65,
2267
+ "grad_norm": 0.5530939065899547,
2268
+ "learning_rate": 2.9416902809052817e-06,
2269
+ "loss": 0.0089,
2270
+ "step": 323
2271
+ },
2272
+ {
2273
+ "epoch": 0.65,
2274
+ "grad_norm": 1.3372492730498717,
2275
+ "learning_rate": 2.912217705227075e-06,
2276
+ "loss": 0.0305,
2277
+ "step": 324
2278
+ },
2279
+ {
2280
+ "epoch": 0.65,
2281
+ "grad_norm": 1.048733318259191,
2282
+ "learning_rate": 2.882832728712551e-06,
2283
+ "loss": 0.02,
2284
+ "step": 325
2285
+ },
2286
+ {
2287
+ "epoch": 0.65,
2288
+ "grad_norm": 0.9645076040351853,
2289
+ "learning_rate": 2.8535365842964713e-06,
2290
+ "loss": 0.0251,
2291
+ "step": 326
2292
+ },
2293
+ {
2294
+ "epoch": 0.65,
2295
+ "grad_norm": 1.1128671372421184,
2296
+ "learning_rate": 2.8243305011863843e-06,
2297
+ "loss": 0.0187,
2298
+ "step": 327
2299
+ },
2300
+ {
2301
+ "epoch": 0.66,
2302
+ "grad_norm": 17.009134889048685,
2303
+ "learning_rate": 2.7952157048110406e-06,
2304
+ "loss": 0.0211,
2305
+ "step": 328
2306
+ },
2307
+ {
2308
+ "epoch": 0.66,
2309
+ "grad_norm": 1.0871994251715402,
2310
+ "learning_rate": 2.7661934167689887e-06,
2311
+ "loss": 0.0249,
2312
+ "step": 329
2313
+ },
2314
+ {
2315
+ "epoch": 0.66,
2316
+ "grad_norm": 0.7371909101378508,
2317
+ "learning_rate": 2.7372648547773063e-06,
2318
+ "loss": 0.0146,
2319
+ "step": 330
2320
+ },
2321
+ {
2322
+ "epoch": 0.66,
2323
+ "grad_norm": 1.6891869650412596,
2324
+ "learning_rate": 2.7084312326205164e-06,
2325
+ "loss": 0.0136,
2326
+ "step": 331
2327
+ },
2328
+ {
2329
+ "epoch": 0.66,
2330
+ "grad_norm": 0.9061970697066741,
2331
+ "learning_rate": 2.6796937600996587e-06,
2332
+ "loss": 0.0211,
2333
+ "step": 332
2334
+ },
2335
+ {
2336
+ "epoch": 0.67,
2337
+ "grad_norm": 1.0101714099989707,
2338
+ "learning_rate": 2.6510536429815224e-06,
2339
+ "loss": 0.0227,
2340
+ "step": 333
2341
+ },
2342
+ {
2343
+ "epoch": 0.67,
2344
+ "grad_norm": 1.1309311830125126,
2345
+ "learning_rate": 2.622512082948063e-06,
2346
+ "loss": 0.0214,
2347
+ "step": 334
2348
+ },
2349
+ {
2350
+ "epoch": 0.67,
2351
+ "grad_norm": 1.0755473033191116,
2352
+ "learning_rate": 2.594070277545975e-06,
2353
+ "loss": 0.0182,
2354
+ "step": 335
2355
+ },
2356
+ {
2357
+ "epoch": 0.67,
2358
+ "grad_norm": 1.135660923583282,
2359
+ "learning_rate": 2.5657294201364526e-06,
2360
+ "loss": 0.0173,
2361
+ "step": 336
2362
+ },
2363
+ {
2364
+ "epoch": 0.67,
2365
+ "grad_norm": 1.0090599161717255,
2366
+ "learning_rate": 2.5374906998451094e-06,
2367
+ "loss": 0.0201,
2368
+ "step": 337
2369
+ },
2370
+ {
2371
+ "epoch": 0.68,
2372
+ "grad_norm": 0.9550564502045064,
2373
+ "learning_rate": 2.5093553015120937e-06,
2374
+ "loss": 0.0135,
2375
+ "step": 338
2376
+ },
2377
+ {
2378
+ "epoch": 0.68,
2379
+ "grad_norm": 0.9458478228553405,
2380
+ "learning_rate": 2.4813244056423692e-06,
2381
+ "loss": 0.0134,
2382
+ "step": 339
2383
+ },
2384
+ {
2385
+ "epoch": 0.68,
2386
+ "grad_norm": 0.9743892317954727,
2387
+ "learning_rate": 2.4533991883561868e-06,
2388
+ "loss": 0.0195,
2389
+ "step": 340
2390
+ },
2391
+ {
2392
+ "epoch": 0.68,
2393
+ "grad_norm": 0.9473591718416459,
2394
+ "learning_rate": 2.425580821339733e-06,
2395
+ "loss": 0.0171,
2396
+ "step": 341
2397
+ },
2398
+ {
2399
+ "epoch": 0.68,
2400
+ "grad_norm": 0.7116865130877778,
2401
+ "learning_rate": 2.3978704717959777e-06,
2402
+ "loss": 0.0197,
2403
+ "step": 342
2404
+ },
2405
+ {
2406
+ "epoch": 0.69,
2407
+ "grad_norm": 0.8029963236091671,
2408
+ "learning_rate": 2.3702693023956853e-06,
2409
+ "loss": 0.014,
2410
+ "step": 343
2411
+ },
2412
+ {
2413
+ "epoch": 0.69,
2414
+ "grad_norm": 1.109432862530169,
2415
+ "learning_rate": 2.342778471228648e-06,
2416
+ "loss": 0.0279,
2417
+ "step": 344
2418
+ },
2419
+ {
2420
+ "epoch": 0.69,
2421
+ "grad_norm": 0.847793101735334,
2422
+ "learning_rate": 2.315399131755081e-06,
2423
+ "loss": 0.0162,
2424
+ "step": 345
2425
+ },
2426
+ {
2427
+ "epoch": 0.69,
2428
+ "grad_norm": 0.9182253928103133,
2429
+ "learning_rate": 2.2881324327572336e-06,
2430
+ "loss": 0.0113,
2431
+ "step": 346
2432
+ },
2433
+ {
2434
+ "epoch": 0.69,
2435
+ "grad_norm": 1.138127214726683,
2436
+ "learning_rate": 2.260979518291186e-06,
2437
+ "loss": 0.0185,
2438
+ "step": 347
2439
+ },
2440
+ {
2441
+ "epoch": 0.7,
2442
+ "grad_norm": 1.0677655325748332,
2443
+ "learning_rate": 2.233941527638848e-06,
2444
+ "loss": 0.0245,
2445
+ "step": 348
2446
+ },
2447
+ {
2448
+ "epoch": 0.7,
2449
+ "grad_norm": 2.833590018156483,
2450
+ "learning_rate": 2.207019595260154e-06,
2451
+ "loss": 0.037,
2452
+ "step": 349
2453
+ },
2454
+ {
2455
+ "epoch": 0.7,
2456
+ "grad_norm": 1.0291093768106434,
2457
+ "learning_rate": 2.1802148507454675e-06,
2458
+ "loss": 0.0264,
2459
+ "step": 350
2460
+ },
2461
+ {
2462
+ "epoch": 0.7,
2463
+ "grad_norm": 0.7675894706009592,
2464
+ "learning_rate": 2.1535284187681866e-06,
2465
+ "loss": 0.0169,
2466
+ "step": 351
2467
+ },
2468
+ {
2469
+ "epoch": 0.7,
2470
+ "grad_norm": 0.8132130131859872,
2471
+ "learning_rate": 2.1269614190375477e-06,
2472
+ "loss": 0.013,
2473
+ "step": 352
2474
+ },
2475
+ {
2476
+ "epoch": 0.71,
2477
+ "grad_norm": 0.672727908610309,
2478
+ "learning_rate": 2.1005149662516517e-06,
2479
+ "loss": 0.0152,
2480
+ "step": 353
2481
+ },
2482
+ {
2483
+ "epoch": 0.71,
2484
+ "grad_norm": 0.9846816967902501,
2485
+ "learning_rate": 2.07419017005069e-06,
2486
+ "loss": 0.0206,
2487
+ "step": 354
2488
+ },
2489
+ {
2490
+ "epoch": 0.71,
2491
+ "grad_norm": 1.3340176182138648,
2492
+ "learning_rate": 2.0479881349703885e-06,
2493
+ "loss": 0.0293,
2494
+ "step": 355
2495
+ },
2496
+ {
2497
+ "epoch": 0.71,
2498
+ "grad_norm": 1.0306882613750554,
2499
+ "learning_rate": 2.021909960395661e-06,
2500
+ "loss": 0.0128,
2501
+ "step": 356
2502
+ },
2503
+ {
2504
+ "epoch": 0.71,
2505
+ "grad_norm": 0.7605285931094184,
2506
+ "learning_rate": 1.9959567405144825e-06,
2507
+ "loss": 0.009,
2508
+ "step": 357
2509
+ },
2510
+ {
2511
+ "epoch": 0.72,
2512
+ "grad_norm": 8.374413266050672,
2513
+ "learning_rate": 1.9701295642719836e-06,
2514
+ "loss": 0.026,
2515
+ "step": 358
2516
+ },
2517
+ {
2518
+ "epoch": 0.72,
2519
+ "grad_norm": 0.8180537175745802,
2520
+ "learning_rate": 1.944429515324749e-06,
2521
+ "loss": 0.0127,
2522
+ "step": 359
2523
+ },
2524
+ {
2525
+ "epoch": 0.72,
2526
+ "grad_norm": 0.7164418383093515,
2527
+ "learning_rate": 1.9188576719953635e-06,
2528
+ "loss": 0.016,
2529
+ "step": 360
2530
+ },
2531
+ {
2532
+ "epoch": 0.72,
2533
+ "grad_norm": 0.959247372968875,
2534
+ "learning_rate": 1.8934151072271573e-06,
2535
+ "loss": 0.0163,
2536
+ "step": 361
2537
+ },
2538
+ {
2539
+ "epoch": 0.72,
2540
+ "grad_norm": 0.814220548865861,
2541
+ "learning_rate": 1.8681028885391905e-06,
2542
+ "loss": 0.018,
2543
+ "step": 362
2544
+ },
2545
+ {
2546
+ "epoch": 0.73,
2547
+ "grad_norm": 0.9308896360037523,
2548
+ "learning_rate": 1.8429220779814654e-06,
2549
+ "loss": 0.0129,
2550
+ "step": 363
2551
+ },
2552
+ {
2553
+ "epoch": 0.73,
2554
+ "grad_norm": 0.8043086851217663,
2555
+ "learning_rate": 1.81787373209036e-06,
2556
+ "loss": 0.0129,
2557
+ "step": 364
2558
+ },
2559
+ {
2560
+ "epoch": 0.73,
2561
+ "grad_norm": 0.7286690129547162,
2562
+ "learning_rate": 1.7929589018443016e-06,
2563
+ "loss": 0.0132,
2564
+ "step": 365
2565
+ },
2566
+ {
2567
+ "epoch": 0.73,
2568
+ "grad_norm": 1.73486611889635,
2569
+ "learning_rate": 1.7681786326196665e-06,
2570
+ "loss": 0.0236,
2571
+ "step": 366
2572
+ },
2573
+ {
2574
+ "epoch": 0.73,
2575
+ "grad_norm": 1.1453325639681513,
2576
+ "learning_rate": 1.743533964146924e-06,
2577
+ "loss": 0.0208,
2578
+ "step": 367
2579
+ },
2580
+ {
2581
+ "epoch": 0.74,
2582
+ "grad_norm": 0.8548180542358728,
2583
+ "learning_rate": 1.7190259304670038e-06,
2584
+ "loss": 0.0132,
2585
+ "step": 368
2586
+ },
2587
+ {
2588
+ "epoch": 0.74,
2589
+ "grad_norm": 1.2250583205056944,
2590
+ "learning_rate": 1.6946555598879138e-06,
2591
+ "loss": 0.0135,
2592
+ "step": 369
2593
+ },
2594
+ {
2595
+ "epoch": 0.74,
2596
+ "grad_norm": 1.0345880771215252,
2597
+ "learning_rate": 1.6704238749415958e-06,
2598
+ "loss": 0.0177,
2599
+ "step": 370
2600
+ },
2601
+ {
2602
+ "epoch": 0.74,
2603
+ "grad_norm": 0.7405117594136527,
2604
+ "learning_rate": 1.6463318923410183e-06,
2605
+ "loss": 0.0129,
2606
+ "step": 371
2607
+ },
2608
+ {
2609
+ "epoch": 0.74,
2610
+ "grad_norm": 0.8360887795991896,
2611
+ "learning_rate": 1.6223806229375182e-06,
2612
+ "loss": 0.0181,
2613
+ "step": 372
2614
+ },
2615
+ {
2616
+ "epoch": 0.75,
2617
+ "grad_norm": 0.9966577189022361,
2618
+ "learning_rate": 1.5985710716783936e-06,
2619
+ "loss": 0.0214,
2620
+ "step": 373
2621
+ },
2622
+ {
2623
+ "epoch": 0.75,
2624
+ "grad_norm": 0.7744910039383206,
2625
+ "learning_rate": 1.5749042375647261e-06,
2626
+ "loss": 0.0146,
2627
+ "step": 374
2628
+ },
2629
+ {
2630
+ "epoch": 0.75,
2631
+ "grad_norm": 0.9445921247832595,
2632
+ "learning_rate": 1.5513811136094786e-06,
2633
+ "loss": 0.0136,
2634
+ "step": 375
2635
+ },
2636
+ {
2637
+ "epoch": 0.75,
2638
+ "grad_norm": 0.8135078772074891,
2639
+ "learning_rate": 1.5280026867958186e-06,
2640
+ "loss": 0.0146,
2641
+ "step": 376
2642
+ },
2643
+ {
2644
+ "epoch": 0.75,
2645
+ "grad_norm": 0.6321845989031962,
2646
+ "learning_rate": 1.5047699380357134e-06,
2647
+ "loss": 0.0143,
2648
+ "step": 377
2649
+ },
2650
+ {
2651
+ "epoch": 0.76,
2652
+ "grad_norm": 1.4013235187615656,
2653
+ "learning_rate": 1.4816838421287693e-06,
2654
+ "loss": 0.0212,
2655
+ "step": 378
2656
+ },
2657
+ {
2658
+ "epoch": 0.76,
2659
+ "grad_norm": 0.8730564359115962,
2660
+ "learning_rate": 1.4587453677213348e-06,
2661
+ "loss": 0.0156,
2662
+ "step": 379
2663
+ },
2664
+ {
2665
+ "epoch": 0.76,
2666
+ "grad_norm": 1.3254762181499105,
2667
+ "learning_rate": 1.4359554772658551e-06,
2668
+ "loss": 0.0227,
2669
+ "step": 380
2670
+ },
2671
+ {
2672
+ "epoch": 0.76,
2673
+ "grad_norm": 0.7795782109779406,
2674
+ "learning_rate": 1.4133151269804873e-06,
2675
+ "loss": 0.0177,
2676
+ "step": 381
2677
+ },
2678
+ {
2679
+ "epoch": 0.76,
2680
+ "grad_norm": 0.9861219267506925,
2681
+ "learning_rate": 1.39082526680899e-06,
2682
+ "loss": 0.0217,
2683
+ "step": 382
2684
+ },
2685
+ {
2686
+ "epoch": 0.77,
2687
+ "grad_norm": 1.2461987132622299,
2688
+ "learning_rate": 1.368486840380851e-06,
2689
+ "loss": 0.0253,
2690
+ "step": 383
2691
+ },
2692
+ {
2693
+ "epoch": 0.77,
2694
+ "grad_norm": 0.9135052292314653,
2695
+ "learning_rate": 1.3463007849717035e-06,
2696
+ "loss": 0.015,
2697
+ "step": 384
2698
+ },
2699
+ {
2700
+ "epoch": 0.77,
2701
+ "grad_norm": 0.8681354507351525,
2702
+ "learning_rate": 1.3242680314639995e-06,
2703
+ "loss": 0.0167,
2704
+ "step": 385
2705
+ },
2706
+ {
2707
+ "epoch": 0.77,
2708
+ "grad_norm": 0.651370947926583,
2709
+ "learning_rate": 1.3023895043079476e-06,
2710
+ "loss": 0.0137,
2711
+ "step": 386
2712
+ },
2713
+ {
2714
+ "epoch": 0.77,
2715
+ "grad_norm": 0.859487086811428,
2716
+ "learning_rate": 1.2806661214827286e-06,
2717
+ "loss": 0.015,
2718
+ "step": 387
2719
+ },
2720
+ {
2721
+ "epoch": 0.78,
2722
+ "grad_norm": 0.8090713872431677,
2723
+ "learning_rate": 1.2590987944579808e-06,
2724
+ "loss": 0.0148,
2725
+ "step": 388
2726
+ },
2727
+ {
2728
+ "epoch": 0.78,
2729
+ "grad_norm": 0.9478543495877076,
2730
+ "learning_rate": 1.2376884281555485e-06,
2731
+ "loss": 0.0152,
2732
+ "step": 389
2733
+ },
2734
+ {
2735
+ "epoch": 0.78,
2736
+ "grad_norm": 0.827410944575496,
2737
+ "learning_rate": 1.2164359209115235e-06,
2738
+ "loss": 0.0101,
2739
+ "step": 390
2740
+ },
2741
+ {
2742
+ "epoch": 0.78,
2743
+ "grad_norm": 0.6695581493060953,
2744
+ "learning_rate": 1.1953421644385444e-06,
2745
+ "loss": 0.0149,
2746
+ "step": 391
2747
+ },
2748
+ {
2749
+ "epoch": 0.78,
2750
+ "grad_norm": 0.9440487033084568,
2751
+ "learning_rate": 1.1744080437883859e-06,
2752
+ "loss": 0.0173,
2753
+ "step": 392
2754
+ },
2755
+ {
2756
+ "epoch": 0.79,
2757
+ "grad_norm": 0.8522096820267036,
2758
+ "learning_rate": 1.1536344373148245e-06,
2759
+ "loss": 0.0149,
2760
+ "step": 393
2761
+ },
2762
+ {
2763
+ "epoch": 0.79,
2764
+ "grad_norm": 0.9890664918452491,
2765
+ "learning_rate": 1.133022216636781e-06,
2766
+ "loss": 0.0177,
2767
+ "step": 394
2768
+ },
2769
+ {
2770
+ "epoch": 0.79,
2771
+ "grad_norm": 0.6227731657859318,
2772
+ "learning_rate": 1.1125722466017547e-06,
2773
+ "loss": 0.0103,
2774
+ "step": 395
2775
+ },
2776
+ {
2777
+ "epoch": 0.79,
2778
+ "grad_norm": 0.9273025159157718,
2779
+ "learning_rate": 1.092285385249528e-06,
2780
+ "loss": 0.0294,
2781
+ "step": 396
2782
+ },
2783
+ {
2784
+ "epoch": 0.79,
2785
+ "grad_norm": 0.8685507295885431,
2786
+ "learning_rate": 1.0721624837761768e-06,
2787
+ "loss": 0.0136,
2788
+ "step": 397
2789
+ },
2790
+ {
2791
+ "epoch": 0.8,
2792
+ "grad_norm": 1.0127796294030704,
2793
+ "learning_rate": 1.0522043864983428e-06,
2794
+ "loss": 0.0192,
2795
+ "step": 398
2796
+ },
2797
+ {
2798
+ "epoch": 0.8,
2799
+ "grad_norm": 1.0537888730981184,
2800
+ "learning_rate": 1.0324119308178166e-06,
2801
+ "loss": 0.0145,
2802
+ "step": 399
2803
+ },
2804
+ {
2805
+ "epoch": 0.8,
2806
+ "grad_norm": 0.7998320097574692,
2807
+ "learning_rate": 1.012785947186397e-06,
2808
+ "loss": 0.0158,
2809
+ "step": 400
2810
+ },
2811
+ {
2812
+ "epoch": 0.8,
2813
+ "grad_norm": 1.2771673609108383,
2814
+ "learning_rate": 9.933272590710508e-07,
2815
+ "loss": 0.0197,
2816
+ "step": 401
2817
+ },
2818
+ {
2819
+ "epoch": 0.8,
2820
+ "grad_norm": 0.7742312289120761,
2821
+ "learning_rate": 9.740366829193587e-07,
2822
+ "loss": 0.01,
2823
+ "step": 402
2824
+ },
2825
+ {
2826
+ "epoch": 0.81,
2827
+ "grad_norm": 0.7686597632856158,
2828
+ "learning_rate": 9.549150281252633e-07,
2829
+ "loss": 0.0116,
2830
+ "step": 403
2831
+ },
2832
+ {
2833
+ "epoch": 0.81,
2834
+ "grad_norm": 1.1269769574429673,
2835
+ "learning_rate": 9.359630969951012e-07,
2836
+ "loss": 0.0142,
2837
+ "step": 404
2838
+ },
2839
+ {
2840
+ "epoch": 0.81,
2841
+ "grad_norm": 1.3902160254287963,
2842
+ "learning_rate": 9.171816847139447e-07,
2843
+ "loss": 0.0265,
2844
+ "step": 405
2845
+ },
2846
+ {
2847
+ "epoch": 0.81,
2848
+ "grad_norm": 1.2531410655034831,
2849
+ "learning_rate": 8.985715793122407e-07,
2850
+ "loss": 0.0274,
2851
+ "step": 406
2852
+ },
2853
+ {
2854
+ "epoch": 0.81,
2855
+ "grad_norm": 0.8014052225720008,
2856
+ "learning_rate": 8.801335616327378e-07,
2857
+ "loss": 0.0128,
2858
+ "step": 407
2859
+ },
2860
+ {
2861
+ "epoch": 0.82,
2862
+ "grad_norm": 0.6073701414297195,
2863
+ "learning_rate": 8.618684052977305e-07,
2864
+ "loss": 0.0119,
2865
+ "step": 408
2866
+ },
2867
+ {
2868
+ "epoch": 0.82,
2869
+ "grad_norm": 1.2968693266660634,
2870
+ "learning_rate": 8.437768766765975e-07,
2871
+ "loss": 0.0266,
2872
+ "step": 409
2873
+ },
2874
+ {
2875
+ "epoch": 0.82,
2876
+ "grad_norm": 0.81042899862227,
2877
+ "learning_rate": 8.258597348536452e-07,
2878
+ "loss": 0.0122,
2879
+ "step": 410
2880
+ },
2881
+ {
2882
+ "epoch": 0.82,
2883
+ "grad_norm": 0.7328896246976655,
2884
+ "learning_rate": 8.081177315962601e-07,
2885
+ "loss": 0.0186,
2886
+ "step": 411
2887
+ },
2888
+ {
2889
+ "epoch": 0.82,
2890
+ "grad_norm": 0.9596594025577012,
2891
+ "learning_rate": 7.905516113233652e-07,
2892
+ "loss": 0.0147,
2893
+ "step": 412
2894
+ },
2895
+ {
2896
+ "epoch": 0.83,
2897
+ "grad_norm": 0.8562958084873945,
2898
+ "learning_rate": 7.731621110741871e-07,
2899
+ "loss": 0.014,
2900
+ "step": 413
2901
+ },
2902
+ {
2903
+ "epoch": 0.83,
2904
+ "grad_norm": 0.9231916716007582,
2905
+ "learning_rate": 7.55949960477328e-07,
2906
+ "loss": 0.0117,
2907
+ "step": 414
2908
+ },
2909
+ {
2910
+ "epoch": 0.83,
2911
+ "grad_norm": 1.4554837378293135,
2912
+ "learning_rate": 7.389158817201541e-07,
2913
+ "loss": 0.0255,
2914
+ "step": 415
2915
+ },
2916
+ {
2917
+ "epoch": 0.83,
2918
+ "grad_norm": 0.7532442609833845,
2919
+ "learning_rate": 7.220605895184946e-07,
2920
+ "loss": 0.0119,
2921
+ "step": 416
2922
+ },
2923
+ {
2924
+ "epoch": 0.83,
2925
+ "grad_norm": 0.6545536104059992,
2926
+ "learning_rate": 7.053847910866513e-07,
2927
+ "loss": 0.0121,
2928
+ "step": 417
2929
+ },
2930
+ {
2931
+ "epoch": 0.84,
2932
+ "grad_norm": 0.9619046426573842,
2933
+ "learning_rate": 6.888891861077301e-07,
2934
+ "loss": 0.0126,
2935
+ "step": 418
2936
+ },
2937
+ {
2938
+ "epoch": 0.84,
2939
+ "grad_norm": 0.810986035332987,
2940
+ "learning_rate": 6.725744667042778e-07,
2941
+ "loss": 0.0175,
2942
+ "step": 419
2943
+ },
2944
+ {
2945
+ "epoch": 0.84,
2946
+ "grad_norm": 1.524457448083235,
2947
+ "learning_rate": 6.564413174092443e-07,
2948
+ "loss": 0.0128,
2949
+ "step": 420
2950
+ },
2951
+ {
2952
+ "epoch": 0.84,
2953
+ "grad_norm": 0.9728314742866659,
2954
+ "learning_rate": 6.404904151372649e-07,
2955
+ "loss": 0.0204,
2956
+ "step": 421
2957
+ },
2958
+ {
2959
+ "epoch": 0.84,
2960
+ "grad_norm": 0.82557912168708,
2961
+ "learning_rate": 6.24722429156251e-07,
2962
+ "loss": 0.0123,
2963
+ "step": 422
2964
+ },
2965
+ {
2966
+ "epoch": 0.85,
2967
+ "grad_norm": 0.8650460143679187,
2968
+ "learning_rate": 6.091380210593145e-07,
2969
+ "loss": 0.0168,
2970
+ "step": 423
2971
+ },
2972
+ {
2973
+ "epoch": 0.85,
2974
+ "grad_norm": 1.1594972392601228,
2975
+ "learning_rate": 5.937378447370068e-07,
2976
+ "loss": 0.0254,
2977
+ "step": 424
2978
+ },
2979
+ {
2980
+ "epoch": 0.85,
2981
+ "grad_norm": 1.1293181344015988,
2982
+ "learning_rate": 5.785225463498828e-07,
2983
+ "loss": 0.0168,
2984
+ "step": 425
2985
+ },
2986
+ {
2987
+ "epoch": 0.85,
2988
+ "grad_norm": 0.9645287783478591,
2989
+ "learning_rate": 5.634927643013899e-07,
2990
+ "loss": 0.0123,
2991
+ "step": 426
2992
+ },
2993
+ {
2994
+ "epoch": 0.85,
2995
+ "grad_norm": 0.7104806347880617,
2996
+ "learning_rate": 5.486491292110796e-07,
2997
+ "loss": 0.0095,
2998
+ "step": 427
2999
+ },
3000
+ {
3001
+ "epoch": 0.86,
3002
+ "grad_norm": 0.5794172109762127,
3003
+ "learning_rate": 5.339922638881545e-07,
3004
+ "loss": 0.0088,
3005
+ "step": 428
3006
+ },
3007
+ {
3008
+ "epoch": 0.86,
3009
+ "grad_norm": 1.102335233337305,
3010
+ "learning_rate": 5.195227833053273e-07,
3011
+ "loss": 0.0168,
3012
+ "step": 429
3013
+ },
3014
+ {
3015
+ "epoch": 0.86,
3016
+ "grad_norm": 1.0087674586204962,
3017
+ "learning_rate": 5.05241294573024e-07,
3018
+ "loss": 0.0228,
3019
+ "step": 430
3020
+ },
3021
+ {
3022
+ "epoch": 0.86,
3023
+ "grad_norm": 0.9314104552367947,
3024
+ "learning_rate": 4.911483969139086e-07,
3025
+ "loss": 0.0164,
3026
+ "step": 431
3027
+ },
3028
+ {
3029
+ "epoch": 0.86,
3030
+ "grad_norm": 9.483187016154936,
3031
+ "learning_rate": 4.772446816377408e-07,
3032
+ "loss": 0.0147,
3033
+ "step": 432
3034
+ },
3035
+ {
3036
+ "epoch": 0.87,
3037
+ "grad_norm": 0.7809689479483307,
3038
+ "learning_rate": 4.6353073211656886e-07,
3039
+ "loss": 0.0145,
3040
+ "step": 433
3041
+ },
3042
+ {
3043
+ "epoch": 0.87,
3044
+ "grad_norm": 0.9704635413624355,
3045
+ "learning_rate": 4.5000712376024826e-07,
3046
+ "loss": 0.0234,
3047
+ "step": 434
3048
+ },
3049
+ {
3050
+ "epoch": 0.87,
3051
+ "grad_norm": 1.1033377817813057,
3052
+ "learning_rate": 4.3667442399229985e-07,
3053
+ "loss": 0.0253,
3054
+ "step": 435
3055
+ },
3056
+ {
3057
+ "epoch": 0.87,
3058
+ "grad_norm": 0.8802979964073332,
3059
+ "learning_rate": 4.2353319222610265e-07,
3060
+ "loss": 0.0199,
3061
+ "step": 436
3062
+ },
3063
+ {
3064
+ "epoch": 0.87,
3065
+ "grad_norm": 0.8098932942840738,
3066
+ "learning_rate": 4.1058397984142405e-07,
3067
+ "loss": 0.0129,
3068
+ "step": 437
3069
+ },
3070
+ {
3071
+ "epoch": 0.88,
3072
+ "grad_norm": 1.1707234761916692,
3073
+ "learning_rate": 3.9782733016128006e-07,
3074
+ "loss": 0.0252,
3075
+ "step": 438
3076
+ },
3077
+ {
3078
+ "epoch": 0.88,
3079
+ "grad_norm": 0.6246859280643877,
3080
+ "learning_rate": 3.852637784291424e-07,
3081
+ "loss": 0.0138,
3082
+ "step": 439
3083
+ },
3084
+ {
3085
+ "epoch": 0.88,
3086
+ "grad_norm": 1.1795551671677649,
3087
+ "learning_rate": 3.728938517864794e-07,
3088
+ "loss": 0.0237,
3089
+ "step": 440
3090
+ },
3091
+ {
3092
+ "epoch": 0.88,
3093
+ "grad_norm": 11.307669029312063,
3094
+ "learning_rate": 3.60718069250639e-07,
3095
+ "loss": 0.0197,
3096
+ "step": 441
3097
+ },
3098
+ {
3099
+ "epoch": 0.88,
3100
+ "grad_norm": 0.8789275715882032,
3101
+ "learning_rate": 3.4873694169306915e-07,
3102
+ "loss": 0.0207,
3103
+ "step": 442
3104
+ },
3105
+ {
3106
+ "epoch": 0.89,
3107
+ "grad_norm": 1.120186105461626,
3108
+ "learning_rate": 3.369509718178887e-07,
3109
+ "loss": 0.0259,
3110
+ "step": 443
3111
+ },
3112
+ {
3113
+ "epoch": 0.89,
3114
+ "grad_norm": 1.0120987117144657,
3115
+ "learning_rate": 3.2536065414078724e-07,
3116
+ "loss": 0.0166,
3117
+ "step": 444
3118
+ },
3119
+ {
3120
+ "epoch": 0.89,
3121
+ "grad_norm": 0.91429217255841,
3122
+ "learning_rate": 3.1396647496828245e-07,
3123
+ "loss": 0.0116,
3124
+ "step": 445
3125
+ },
3126
+ {
3127
+ "epoch": 0.89,
3128
+ "grad_norm": 0.865136913141427,
3129
+ "learning_rate": 3.0276891237731085e-07,
3130
+ "loss": 0.0162,
3131
+ "step": 446
3132
+ },
3133
+ {
3134
+ "epoch": 0.89,
3135
+ "grad_norm": 0.8876092437261166,
3136
+ "learning_rate": 2.917684361951728e-07,
3137
+ "loss": 0.0197,
3138
+ "step": 447
3139
+ },
3140
+ {
3141
+ "epoch": 0.9,
3142
+ "grad_norm": 0.703021608869076,
3143
+ "learning_rate": 2.809655079798179e-07,
3144
+ "loss": 0.0086,
3145
+ "step": 448
3146
+ },
3147
+ {
3148
+ "epoch": 0.9,
3149
+ "grad_norm": 1.275162404286764,
3150
+ "learning_rate": 2.7036058100047723e-07,
3151
+ "loss": 0.0183,
3152
+ "step": 449
3153
+ },
3154
+ {
3155
+ "epoch": 0.9,
3156
+ "grad_norm": 1.0083759252664604,
3157
+ "learning_rate": 2.599541002186479e-07,
3158
+ "loss": 0.0196,
3159
+ "step": 450
3160
+ },
3161
+ {
3162
+ "epoch": 0.9,
3163
+ "grad_norm": 1.1046039305427457,
3164
+ "learning_rate": 2.497465022694207e-07,
3165
+ "loss": 0.0247,
3166
+ "step": 451
3167
+ },
3168
+ {
3169
+ "epoch": 0.9,
3170
+ "grad_norm": 0.8516957928921493,
3171
+ "learning_rate": 2.397382154431621e-07,
3172
+ "loss": 0.018,
3173
+ "step": 452
3174
+ },
3175
+ {
3176
+ "epoch": 0.91,
3177
+ "grad_norm": 1.3994562795912273,
3178
+ "learning_rate": 2.2992965966754378e-07,
3179
+ "loss": 0.0314,
3180
+ "step": 453
3181
+ },
3182
+ {
3183
+ "epoch": 0.91,
3184
+ "grad_norm": 0.9890602285333961,
3185
+ "learning_rate": 2.2032124648992015e-07,
3186
+ "loss": 0.0171,
3187
+ "step": 454
3188
+ },
3189
+ {
3190
+ "epoch": 0.91,
3191
+ "grad_norm": 1.0220170674604938,
3192
+ "learning_rate": 2.109133790600648e-07,
3193
+ "loss": 0.0161,
3194
+ "step": 455
3195
+ },
3196
+ {
3197
+ "epoch": 0.91,
3198
+ "grad_norm": 0.6542418248605464,
3199
+ "learning_rate": 2.0170645211325335e-07,
3200
+ "loss": 0.0123,
3201
+ "step": 456
3202
+ },
3203
+ {
3204
+ "epoch": 0.91,
3205
+ "grad_norm": 1.1662835822732398,
3206
+ "learning_rate": 1.9270085195370048e-07,
3207
+ "loss": 0.0192,
3208
+ "step": 457
3209
+ },
3210
+ {
3211
+ "epoch": 0.92,
3212
+ "grad_norm": 1.2727779729278517,
3213
+ "learning_rate": 1.838969564383525e-07,
3214
+ "loss": 0.0197,
3215
+ "step": 458
3216
+ },
3217
+ {
3218
+ "epoch": 0.92,
3219
+ "grad_norm": 1.147074874436642,
3220
+ "learning_rate": 1.7529513496103322e-07,
3221
+ "loss": 0.0238,
3222
+ "step": 459
3223
+ },
3224
+ {
3225
+ "epoch": 0.92,
3226
+ "grad_norm": 0.9679932463174291,
3227
+ "learning_rate": 1.6689574843694433e-07,
3228
+ "loss": 0.0188,
3229
+ "step": 460
3230
+ },
3231
+ {
3232
+ "epoch": 0.92,
3233
+ "grad_norm": 0.7746975756873401,
3234
+ "learning_rate": 1.5869914928752117e-07,
3235
+ "loss": 0.0103,
3236
+ "step": 461
3237
+ },
3238
+ {
3239
+ "epoch": 0.92,
3240
+ "grad_norm": 1.1353747843305269,
3241
+ "learning_rate": 1.5070568142564912e-07,
3242
+ "loss": 0.0211,
3243
+ "step": 462
3244
+ },
3245
+ {
3246
+ "epoch": 0.93,
3247
+ "grad_norm": 0.8839111888105964,
3248
+ "learning_rate": 1.4291568024122848e-07,
3249
+ "loss": 0.0131,
3250
+ "step": 463
3251
+ },
3252
+ {
3253
+ "epoch": 0.93,
3254
+ "grad_norm": 0.7710862516234268,
3255
+ "learning_rate": 1.3532947258710905e-07,
3256
+ "loss": 0.014,
3257
+ "step": 464
3258
+ },
3259
+ {
3260
+ "epoch": 0.93,
3261
+ "grad_norm": 0.8881221990938554,
3262
+ "learning_rate": 1.2794737676536993e-07,
3263
+ "loss": 0.0153,
3264
+ "step": 465
3265
+ },
3266
+ {
3267
+ "epoch": 0.93,
3268
+ "grad_norm": 0.7309604708251977,
3269
+ "learning_rate": 1.2076970251396593e-07,
3270
+ "loss": 0.0127,
3271
+ "step": 466
3272
+ },
3273
+ {
3274
+ "epoch": 0.93,
3275
+ "grad_norm": 0.8892924102101631,
3276
+ "learning_rate": 1.1379675099373489e-07,
3277
+ "loss": 0.0143,
3278
+ "step": 467
3279
+ },
3280
+ {
3281
+ "epoch": 0.94,
3282
+ "grad_norm": 0.840913230172774,
3283
+ "learning_rate": 1.0702881477575589e-07,
3284
+ "loss": 0.0213,
3285
+ "step": 468
3286
+ },
3287
+ {
3288
+ "epoch": 0.94,
3289
+ "grad_norm": 0.8007255565550828,
3290
+ "learning_rate": 1.004661778290783e-07,
3291
+ "loss": 0.0148,
3292
+ "step": 469
3293
+ },
3294
+ {
3295
+ "epoch": 0.94,
3296
+ "grad_norm": 1.0305407323858349,
3297
+ "learning_rate": 9.410911550880474e-08,
3298
+ "loss": 0.0203,
3299
+ "step": 470
3300
+ },
3301
+ {
3302
+ "epoch": 0.94,
3303
+ "grad_norm": 1.748536511532871,
3304
+ "learning_rate": 8.795789454453862e-08,
3305
+ "loss": 0.0293,
3306
+ "step": 471
3307
+ },
3308
+ {
3309
+ "epoch": 0.94,
3310
+ "grad_norm": 1.3175921578082317,
3311
+ "learning_rate": 8.201277302919086e-08,
3312
+ "loss": 0.0106,
3313
+ "step": 472
3314
+ },
3315
+ {
3316
+ "epoch": 0.95,
3317
+ "grad_norm": 0.6936491099060337,
3318
+ "learning_rate": 7.627400040815414e-08,
3319
+ "loss": 0.0162,
3320
+ "step": 473
3321
+ },
3322
+ {
3323
+ "epoch": 0.95,
3324
+ "grad_norm": 0.5940053081104846,
3325
+ "learning_rate": 7.074181746883402e-08,
3326
+ "loss": 0.0137,
3327
+ "step": 474
3328
+ },
3329
+ {
3330
+ "epoch": 0.95,
3331
+ "grad_norm": 0.5984206991026165,
3332
+ "learning_rate": 6.54164563305465e-08,
3333
+ "loss": 0.0083,
3334
+ "step": 475
3335
+ },
3336
+ {
3337
+ "epoch": 0.95,
3338
+ "grad_norm": 0.506267108804762,
3339
+ "learning_rate": 6.029814043478022e-08,
3340
+ "loss": 0.0142,
3341
+ "step": 476
3342
+ },
3343
+ {
3344
+ "epoch": 0.95,
3345
+ "grad_norm": 0.9855883610352663,
3346
+ "learning_rate": 5.538708453581787e-08,
3347
+ "loss": 0.0186,
3348
+ "step": 477
3349
+ },
3350
+ {
3351
+ "epoch": 0.96,
3352
+ "grad_norm": 0.6290945531204232,
3353
+ "learning_rate": 5.068349469173006e-08,
3354
+ "loss": 0.0089,
3355
+ "step": 478
3356
+ },
3357
+ {
3358
+ "epoch": 0.96,
3359
+ "grad_norm": 0.9193751838424389,
3360
+ "learning_rate": 4.618756825572612e-08,
3361
+ "loss": 0.0087,
3362
+ "step": 479
3363
+ },
3364
+ {
3365
+ "epoch": 0.96,
3366
+ "grad_norm": 0.6799544458629045,
3367
+ "learning_rate": 4.189949386787462e-08,
3368
+ "loss": 0.0125,
3369
+ "step": 480
3370
+ },
3371
+ {
3372
+ "epoch": 0.96,
3373
+ "grad_norm": 0.8850223431900668,
3374
+ "learning_rate": 3.781945144718912e-08,
3375
+ "loss": 0.011,
3376
+ "step": 481
3377
+ },
3378
+ {
3379
+ "epoch": 0.96,
3380
+ "grad_norm": 0.8088984405542546,
3381
+ "learning_rate": 3.394761218407705e-08,
3382
+ "loss": 0.0151,
3383
+ "step": 482
3384
+ },
3385
+ {
3386
+ "epoch": 0.97,
3387
+ "grad_norm": 0.9502110200870021,
3388
+ "learning_rate": 3.0284138533160924e-08,
3389
+ "loss": 0.0221,
3390
+ "step": 483
3391
+ },
3392
+ {
3393
+ "epoch": 0.97,
3394
+ "grad_norm": 0.8236390715131118,
3395
+ "learning_rate": 2.6829184206457194e-08,
3396
+ "loss": 0.0103,
3397
+ "step": 484
3398
+ },
3399
+ {
3400
+ "epoch": 0.97,
3401
+ "grad_norm": 0.935804113031717,
3402
+ "learning_rate": 2.358289416693027e-08,
3403
+ "loss": 0.0152,
3404
+ "step": 485
3405
+ },
3406
+ {
3407
+ "epoch": 0.97,
3408
+ "grad_norm": 0.6601206268664995,
3409
+ "learning_rate": 2.0545404622407396e-08,
3410
+ "loss": 0.0119,
3411
+ "step": 486
3412
+ },
3413
+ {
3414
+ "epoch": 0.97,
3415
+ "grad_norm": 0.6123494146804135,
3416
+ "learning_rate": 1.7716843019867646e-08,
3417
+ "loss": 0.0125,
3418
+ "step": 487
3419
+ },
3420
+ {
3421
+ "epoch": 0.98,
3422
+ "grad_norm": 1.006618664807847,
3423
+ "learning_rate": 1.509732804009012e-08,
3424
+ "loss": 0.0165,
3425
+ "step": 488
3426
+ },
3427
+ {
3428
+ "epoch": 0.98,
3429
+ "grad_norm": 1.1887669043417244,
3430
+ "learning_rate": 1.268696959267679e-08,
3431
+ "loss": 0.0199,
3432
+ "step": 489
3433
+ },
3434
+ {
3435
+ "epoch": 0.98,
3436
+ "grad_norm": 0.926176126831444,
3437
+ "learning_rate": 1.0485868811441757e-08,
3438
+ "loss": 0.0186,
3439
+ "step": 490
3440
+ },
3441
+ {
3442
+ "epoch": 0.98,
3443
+ "grad_norm": 1.2602758093343573,
3444
+ "learning_rate": 8.494118050164646e-09,
3445
+ "loss": 0.0193,
3446
+ "step": 491
3447
+ },
3448
+ {
3449
+ "epoch": 0.98,
3450
+ "grad_norm": 0.7669565956768566,
3451
+ "learning_rate": 6.711800878718144e-09,
3452
+ "loss": 0.0122,
3453
+ "step": 492
3454
+ },
3455
+ {
3456
+ "epoch": 0.99,
3457
+ "grad_norm": 1.316388905942753,
3458
+ "learning_rate": 5.138992079561367e-09,
3459
+ "loss": 0.0205,
3460
+ "step": 493
3461
+ },
3462
+ {
3463
+ "epoch": 0.99,
3464
+ "grad_norm": 0.8474199640474874,
3465
+ "learning_rate": 3.775757644601808e-09,
3466
+ "loss": 0.0131,
3467
+ "step": 494
3468
+ },
3469
+ {
3470
+ "epoch": 0.99,
3471
+ "grad_norm": 0.9071529317613827,
3472
+ "learning_rate": 2.6221547724253337e-09,
3473
+ "loss": 0.0172,
3474
+ "step": 495
3475
+ },
3476
+ {
3477
+ "epoch": 0.99,
3478
+ "grad_norm": 0.7952656053128018,
3479
+ "learning_rate": 1.6782318658992159e-09,
3480
+ "loss": 0.0153,
3481
+ "step": 496
3482
+ },
3483
+ {
3484
+ "epoch": 0.99,
3485
+ "grad_norm": 1.0004492487432555,
3486
+ "learning_rate": 9.440285301370865e-10,
3487
+ "loss": 0.016,
3488
+ "step": 497
3489
+ },
3490
+ {
3491
+ "epoch": 1.0,
3492
+ "grad_norm": 0.6579509724399629,
3493
+ "learning_rate": 4.1957557084082447e-10,
3494
+ "loss": 0.0114,
3495
+ "step": 498
3496
+ },
3497
+ {
3498
+ "epoch": 1.0,
3499
+ "grad_norm": 0.9911742983949866,
3500
+ "learning_rate": 1.0489499300603279e-10,
3501
+ "loss": 0.019,
3502
+ "step": 499
3503
+ },
3504
+ {
3505
+ "epoch": 1.0,
3506
+ "grad_norm": 0.7737516449983475,
3507
+ "learning_rate": 0.0,
3508
+ "loss": 0.011,
3509
+ "step": 500
3510
+ },
3511
+ {
3512
+ "epoch": 1.0,
3513
+ "step": 500,
3514
+ "total_flos": 131729809735680.0,
3515
+ "train_loss": 0.0799976912010461,
3516
+ "train_runtime": 2062.5688,
3517
+ "train_samples_per_second": 1.939,
3518
+ "train_steps_per_second": 0.242
3519
+ }
3520
+ ],
3521
+ "logging_steps": 1.0,
3522
+ "max_steps": 500,
3523
+ "num_input_tokens_seen": 0,
3524
+ "num_train_epochs": 1,
3525
+ "save_steps": 50000,
3526
+ "total_flos": 131729809735680.0,
3527
+ "train_batch_size": 1,
3528
+ "trial_name": null,
3529
+ "trial_params": null
3530
+ }