Hugging Face's logo

TienAnh
/
stage3-moe-ft-llavaqwen1.5-0.5b-Viet-ShareGPT-2ep_moe

PyTorch
Safetensors
moe_llava_qwen1_5
Generated from Trainer
Model card Files Community
TienAnh commited on
Commit
7004902
·
verified ·
1 Parent(s): 77d09af

Model save

Browse files
Files changed (12) hide show
  1. README.md +58 -0
  2. added_tokens.json +6 -0
  3. config.json +93 -0
  4. generation_config.json +6 -0
  5. merges.txt +0 -0
  6. model.safetensors +3 -0
  7. pytorch_model.bin +3 -0
  8. special_tokens_map.json +21 -0
  9. tokenizer_config.json +52 -0
  10. trainer_state.json +2988 -0
  11. training_args.bin +3 -0
  12. vocab.json +0 -0
README.md ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: TienAnh/stage2-ft-llavaqwen1.5-0.5B-Viet-ShareGPT-2ep
3
+ tags:
4
+ - generated_from_trainer
5
+ model-index:
6
+ - name: stage3-moe-ft-llavaqwen1.5-0.5b-Viet-ShareGPT-2ep_moe
7
+ results: []
8
+ ---
9
+
10
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
11
+ should probably proofread and complete it, then remove this comment. -->
12
+
13
+ # stage3-moe-ft-llavaqwen1.5-0.5b-Viet-ShareGPT-2ep_moe
14
+
15
+ This model is a fine-tuned version of [TienAnh/stage2-ft-llavaqwen1.5-0.5B-Viet-ShareGPT-2ep](https://huggingface.co/TienAnh/stage2-ft-llavaqwen1.5-0.5B-Viet-ShareGPT-2ep) on an unknown dataset.
16
+
17
+ ## Model description
18
+
19
+ More information needed
20
+
21
+ ## Intended uses & limitations
22
+
23
+ More information needed
24
+
25
+ ## Training and evaluation data
26
+
27
+ More information needed
28
+
29
+ ## Training procedure
30
+
31
+ ### Training hyperparameters
32
+
33
+ The following hyperparameters were used during training:
34
+ - learning_rate: 2e-06
35
+ - train_batch_size: 4
36
+ - eval_batch_size: 2
37
+ - seed: 42
38
+ - distributed_type: multi-GPU
39
+ - num_devices: 2
40
+ - gradient_accumulation_steps: 8
41
+ - total_train_batch_size: 64
42
+ - total_eval_batch_size: 4
43
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
+ - lr_scheduler_type: cosine
45
+ - lr_scheduler_warmup_ratio: 0.03
46
+ - num_epochs: 2.0
47
+ - mixed_precision_training: Native AMP
48
+
49
+ ### Training results
50
+
51
+
52
+
53
+ ### Framework versions
54
+
55
+ - Transformers 4.37.0
56
+ - Pytorch 2.4.1+cu121
57
+ - Datasets 2.21.0
58
+ - Tokenizers 0.15.1
added_tokens.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "<|endoftext|>": 151643,
3
+ "<|extra_0|>": 151646,
4
+ "<|im_end|>": 151645,
5
+ "<|im_start|>": 151644
6
+ }
config.json ADDED
@@ -0,0 +1,93 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "TienAnh/stage2-ft-llavaqwen1.5-0.5B-Viet-ShareGPT-2ep",
3
+ "architectures": [
4
+ "MoELLaVAQwen1_5ForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "eos_token_id": 151643,
9
+ "freeze_mm_mlp_adapter": false,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 1024,
12
+ "image_aspect_ratio": "pad",
13
+ "image_projector_type": "mlp2x_gelu",
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 2816,
16
+ "lora": {},
17
+ "max_position_embeddings": 32768,
18
+ "max_window_layers": 21,
19
+ "mm_hidden_size": 768,
20
+ "mm_image_tower": "google/siglip-base-patch16-256-multilingual",
21
+ "mm_projector_lr": null,
22
+ "mm_use_im_patch_token": false,
23
+ "mm_use_im_start_end": false,
24
+ "mm_video_tower": null,
25
+ "mm_vision_select_feature": "patch",
26
+ "mm_vision_select_layer": -2,
27
+ "model_type": "moe_llava_qwen1_5",
28
+ "moe": {
29
+ "capacity_factor": 1.5,
30
+ "ep_size": 1,
31
+ "eval_capacity_factor": 2.0,
32
+ "min_capacity": 0,
33
+ "moe_enable": true,
34
+ "moe_layers_idx": [
35
+ 0,
36
+ 2,
37
+ 4,
38
+ 6,
39
+ 8,
40
+ 10,
41
+ 12,
42
+ 14,
43
+ 16,
44
+ 18,
45
+ 20,
46
+ 22
47
+ ],
48
+ "moe_mode": "sparse",
49
+ "num_experts": [
50
+ 4,
51
+ 4,
52
+ 4,
53
+ 4,
54
+ 4,
55
+ 4,
56
+ 4,
57
+ 4,
58
+ 4,
59
+ 4,
60
+ 4,
61
+ 4
62
+ ],
63
+ "router_aux_loss_coef": 0.01,
64
+ "top_k_experts": 2,
65
+ "train_modules": [
66
+ "mlp.gate_proj",
67
+ "mlp.up_proj",
68
+ "mlp.down_proj",
69
+ "wg"
70
+ ],
71
+ "use_residual": false
72
+ },
73
+ "num_attention_heads": 16,
74
+ "num_hidden_layers": 24,
75
+ "num_key_value_heads": 16,
76
+ "pad_token_id": 151646,
77
+ "rms_norm_eps": 1e-06,
78
+ "rope_theta": 1000000.0,
79
+ "sliding_window": 32768,
80
+ "tie_word_embeddings": true,
81
+ "tokenizer_padding_side": "right",
82
+ "torch_dtype": "float16",
83
+ "transformers_version": "4.37.0",
84
+ "tune_mm_mlp_adapter": false,
85
+ "use_cache": true,
86
+ "use_mm_proj": true,
87
+ "use_sliding_window": false,
88
+ "video_global_proj": false,
89
+ "video_projector_type": "linear",
90
+ "video_spatial_proj": false,
91
+ "video_temproal_proj": false,
92
+ "vocab_size": 151646
93
+ }
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 151643,
3
+ "eos_token_id": 151643,
4
+ "max_new_tokens": 2048,
5
+ "transformers_version": "4.37.0"
6
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4aee94ec61a0293519bd7026246bc1a8771bfad5b3543588bb2c66a7fff7a412
3
+ size 4100964168
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:db7dd81cb49b0b5cb77c01dc4b228021dc4fdde4fea91ca406533bff6836f205
3
+ size 1740053574
special_tokens_map.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>"
5
+ ],
6
+ "eos_token": {
7
+ "content": "<|endoftext|>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false
12
+ },
13
+ "pad_token": "<|extra_0|>",
14
+ "unk_token": {
15
+ "content": "<|extra_0|>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false
20
+ }
21
+ }
tokenizer_config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "151643": {
5
+ "content": "<|endoftext|>",
6
+ "lstrip": false,
7
+ "normalized": false,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "151644": {
13
+ "content": "<|im_start|>",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "151645": {
21
+ "content": "<|im_end|>",
22
+ "lstrip": false,
23
+ "normalized": false,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "151646": {
29
+ "content": "<|extra_0|>",
30
+ "lstrip": false,
31
+ "normalized": false,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ }
36
+ },
37
+ "additional_special_tokens": [
38
+ "<|im_start|>",
39
+ "<|im_end|>"
40
+ ],
41
+ "bos_token": null,
42
+ "chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
43
+ "clean_up_tokenization_spaces": false,
44
+ "eos_token": "<|endoftext|>",
45
+ "errors": "replace",
46
+ "model_max_length": 1024,
47
+ "pad_token": "<|extra_0|>",
48
+ "padding_side": "right",
49
+ "split_special_tokens": false,
50
+ "tokenizer_class": "Qwen2Tokenizer",
51
+ "unk_token": "<|extra_0|>"
52
+ }
trainer_state.json ADDED
@@ -0,0 +1,2988 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.998479858120091,
5
+ "eval_steps": 500,
6
+ "global_step": 986,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0,
13
+ "learning_rate": 1.3333333333333334e-07,
14
+ "loss": 2.0093,
15
+ "step": 2
16
+ },
17
+ {
18
+ "epoch": 0.01,
19
+ "learning_rate": 2.6666666666666667e-07,
20
+ "loss": 2.1267,
21
+ "step": 4
22
+ },
23
+ {
24
+ "epoch": 0.01,
25
+ "learning_rate": 4e-07,
26
+ "loss": 2.0596,
27
+ "step": 6
28
+ },
29
+ {
30
+ "epoch": 0.02,
31
+ "learning_rate": 5.333333333333333e-07,
32
+ "loss": 2.0195,
33
+ "step": 8
34
+ },
35
+ {
36
+ "epoch": 0.02,
37
+ "learning_rate": 6.666666666666666e-07,
38
+ "loss": 1.9924,
39
+ "step": 10
40
+ },
41
+ {
42
+ "epoch": 0.02,
43
+ "learning_rate": 8e-07,
44
+ "loss": 2.0199,
45
+ "step": 12
46
+ },
47
+ {
48
+ "epoch": 0.03,
49
+ "learning_rate": 9.333333333333333e-07,
50
+ "loss": 2.1074,
51
+ "step": 14
52
+ },
53
+ {
54
+ "epoch": 0.03,
55
+ "learning_rate": 1.0666666666666667e-06,
56
+ "loss": 2.0768,
57
+ "step": 16
58
+ },
59
+ {
60
+ "epoch": 0.04,
61
+ "learning_rate": 1.2e-06,
62
+ "loss": 2.0471,
63
+ "step": 18
64
+ },
65
+ {
66
+ "epoch": 0.04,
67
+ "learning_rate": 1.3333333333333332e-06,
68
+ "loss": 2.0815,
69
+ "step": 20
70
+ },
71
+ {
72
+ "epoch": 0.04,
73
+ "learning_rate": 1.4666666666666665e-06,
74
+ "loss": 2.0491,
75
+ "step": 22
76
+ },
77
+ {
78
+ "epoch": 0.05,
79
+ "learning_rate": 1.6e-06,
80
+ "loss": 1.9666,
81
+ "step": 24
82
+ },
83
+ {
84
+ "epoch": 0.05,
85
+ "learning_rate": 1.7333333333333334e-06,
86
+ "loss": 2.0742,
87
+ "step": 26
88
+ },
89
+ {
90
+ "epoch": 0.06,
91
+ "learning_rate": 1.8666666666666667e-06,
92
+ "loss": 2.0388,
93
+ "step": 28
94
+ },
95
+ {
96
+ "epoch": 0.06,
97
+ "learning_rate": 2e-06,
98
+ "loss": 2.0142,
99
+ "step": 30
100
+ },
101
+ {
102
+ "epoch": 0.06,
103
+ "learning_rate": 1.9999784020568754e-06,
104
+ "loss": 2.0564,
105
+ "step": 32
106
+ },
107
+ {
108
+ "epoch": 0.07,
109
+ "learning_rate": 1.999913609160443e-06,
110
+ "loss": 2.0268,
111
+ "step": 34
112
+ },
113
+ {
114
+ "epoch": 0.07,
115
+ "learning_rate": 1.999805624109491e-06,
116
+ "loss": 2.0147,
117
+ "step": 36
118
+ },
119
+ {
120
+ "epoch": 0.08,
121
+ "learning_rate": 1.999654451568528e-06,
122
+ "loss": 2.0643,
123
+ "step": 38
124
+ },
125
+ {
126
+ "epoch": 0.08,
127
+ "learning_rate": 1.999460098067586e-06,
128
+ "loss": 2.1189,
129
+ "step": 40
130
+ },
131
+ {
132
+ "epoch": 0.09,
133
+ "learning_rate": 1.999222572001937e-06,
134
+ "loss": 2.1375,
135
+ "step": 42
136
+ },
137
+ {
138
+ "epoch": 0.09,
139
+ "learning_rate": 1.9989418836317303e-06,
140
+ "loss": 2.1305,
141
+ "step": 44
142
+ },
143
+ {
144
+ "epoch": 0.09,
145
+ "learning_rate": 1.9986180450815485e-06,
146
+ "loss": 1.9724,
147
+ "step": 46
148
+ },
149
+ {
150
+ "epoch": 0.1,
151
+ "learning_rate": 1.9982510703398844e-06,
152
+ "loss": 1.9929,
153
+ "step": 48
154
+ },
155
+ {
156
+ "epoch": 0.1,
157
+ "learning_rate": 1.9978409752585376e-06,
158
+ "loss": 2.0284,
159
+ "step": 50
160
+ },
161
+ {
162
+ "epoch": 0.11,
163
+ "learning_rate": 1.9973877775519285e-06,
164
+ "loss": 2.0599,
165
+ "step": 52
166
+ },
167
+ {
168
+ "epoch": 0.11,
169
+ "learning_rate": 1.9968914967963333e-06,
170
+ "loss": 2.0289,
171
+ "step": 54
172
+ },
173
+ {
174
+ "epoch": 0.11,
175
+ "learning_rate": 1.99635215442904e-06,
176
+ "loss": 1.971,
177
+ "step": 56
178
+ },
179
+ {
180
+ "epoch": 0.12,
181
+ "learning_rate": 1.9957697737474197e-06,
182
+ "loss": 2.1077,
183
+ "step": 58
184
+ },
185
+ {
186
+ "epoch": 0.12,
187
+ "learning_rate": 1.9951443799079214e-06,
188
+ "loss": 1.9899,
189
+ "step": 60
190
+ },
191
+ {
192
+ "epoch": 0.13,
193
+ "learning_rate": 1.994475999924987e-06,
194
+ "loss": 2.0886,
195
+ "step": 62
196
+ },
197
+ {
198
+ "epoch": 0.13,
199
+ "learning_rate": 1.993764662669882e-06,
200
+ "loss": 2.0682,
201
+ "step": 64
202
+ },
203
+ {
204
+ "epoch": 0.13,
205
+ "learning_rate": 1.9930103988694497e-06,
206
+ "loss": 2.0396,
207
+ "step": 66
208
+ },
209
+ {
210
+ "epoch": 0.14,
211
+ "learning_rate": 1.9922132411047833e-06,
212
+ "loss": 2.0508,
213
+ "step": 68
214
+ },
215
+ {
216
+ "epoch": 0.14,
217
+ "learning_rate": 1.991373223809819e-06,
218
+ "loss": 2.0468,
219
+ "step": 70
220
+ },
221
+ {
222
+ "epoch": 0.15,
223
+ "learning_rate": 1.990490383269848e-06,
224
+ "loss": 2.1025,
225
+ "step": 72
226
+ },
227
+ {
228
+ "epoch": 0.15,
229
+ "learning_rate": 1.9895647576199505e-06,
230
+ "loss": 2.1965,
231
+ "step": 74
232
+ },
233
+ {
234
+ "epoch": 0.15,
235
+ "learning_rate": 1.988596386843346e-06,
236
+ "loss": 2.0172,
237
+ "step": 76
238
+ },
239
+ {
240
+ "epoch": 0.16,
241
+ "learning_rate": 1.987585312769669e-06,
242
+ "loss": 2.0095,
243
+ "step": 78
244
+ },
245
+ {
246
+ "epoch": 0.16,
247
+ "learning_rate": 1.9865315790731598e-06,
248
+ "loss": 1.9972,
249
+ "step": 80
250
+ },
251
+ {
252
+ "epoch": 0.17,
253
+ "learning_rate": 1.9854352312707796e-06,
254
+ "loss": 2.115,
255
+ "step": 82
256
+ },
257
+ {
258
+ "epoch": 0.17,
259
+ "learning_rate": 1.984296316720243e-06,
260
+ "loss": 2.0605,
261
+ "step": 84
262
+ },
263
+ {
264
+ "epoch": 0.17,
265
+ "learning_rate": 1.983114884617974e-06,
266
+ "loss": 1.9956,
267
+ "step": 86
268
+ },
269
+ {
270
+ "epoch": 0.18,
271
+ "learning_rate": 1.9818909859969784e-06,
272
+ "loss": 2.0327,
273
+ "step": 88
274
+ },
275
+ {
276
+ "epoch": 0.18,
277
+ "learning_rate": 1.980624673724643e-06,
278
+ "loss": 2.0218,
279
+ "step": 90
280
+ },
281
+ {
282
+ "epoch": 0.19,
283
+ "learning_rate": 1.9793160025004475e-06,
284
+ "loss": 1.9757,
285
+ "step": 92
286
+ },
287
+ {
288
+ "epoch": 0.19,
289
+ "learning_rate": 1.9779650288536056e-06,
290
+ "loss": 2.0011,
291
+ "step": 94
292
+ },
293
+ {
294
+ "epoch": 0.19,
295
+ "learning_rate": 1.9765718111406216e-06,
296
+ "loss": 2.0784,
297
+ "step": 96
298
+ },
299
+ {
300
+ "epoch": 0.2,
301
+ "learning_rate": 1.975136409542769e-06,
302
+ "loss": 2.0673,
303
+ "step": 98
304
+ },
305
+ {
306
+ "epoch": 0.2,
307
+ "learning_rate": 1.9736588860634923e-06,
308
+ "loss": 2.0109,
309
+ "step": 100
310
+ },
311
+ {
312
+ "epoch": 0.21,
313
+ "learning_rate": 1.9721393045257275e-06,
314
+ "loss": 2.0554,
315
+ "step": 102
316
+ },
317
+ {
318
+ "epoch": 0.21,
319
+ "learning_rate": 1.9705777305691457e-06,
320
+ "loss": 1.9808,
321
+ "step": 104
322
+ },
323
+ {
324
+ "epoch": 0.21,
325
+ "learning_rate": 1.9689742316473177e-06,
326
+ "loss": 1.9741,
327
+ "step": 106
328
+ },
329
+ {
330
+ "epoch": 0.22,
331
+ "learning_rate": 1.967328877024801e-06,
332
+ "loss": 2.0922,
333
+ "step": 108
334
+ },
335
+ {
336
+ "epoch": 0.22,
337
+ "learning_rate": 1.9656417377741466e-06,
338
+ "loss": 2.0362,
339
+ "step": 110
340
+ },
341
+ {
342
+ "epoch": 0.23,
343
+ "learning_rate": 1.96391288677283e-06,
344
+ "loss": 2.0535,
345
+ "step": 112
346
+ },
347
+ {
348
+ "epoch": 0.23,
349
+ "learning_rate": 1.962142398700101e-06,
350
+ "loss": 2.0159,
351
+ "step": 114
352
+ },
353
+ {
354
+ "epoch": 0.24,
355
+ "learning_rate": 1.9603303500337626e-06,
356
+ "loss": 1.9948,
357
+ "step": 116
358
+ },
359
+ {
360
+ "epoch": 0.24,
361
+ "learning_rate": 1.9584768190468624e-06,
362
+ "loss": 2.0756,
363
+ "step": 118
364
+ },
365
+ {
366
+ "epoch": 0.24,
367
+ "learning_rate": 1.9565818858043136e-06,
368
+ "loss": 1.9603,
369
+ "step": 120
370
+ },
371
+ {
372
+ "epoch": 0.25,
373
+ "learning_rate": 1.9546456321594373e-06,
374
+ "loss": 2.0958,
375
+ "step": 122
376
+ },
377
+ {
378
+ "epoch": 0.25,
379
+ "learning_rate": 1.9526681417504258e-06,
380
+ "loss": 2.0459,
381
+ "step": 124
382
+ },
383
+ {
384
+ "epoch": 0.26,
385
+ "learning_rate": 1.9506494999967296e-06,
386
+ "loss": 2.0352,
387
+ "step": 126
388
+ },
389
+ {
390
+ "epoch": 0.26,
391
+ "learning_rate": 1.9485897940953686e-06,
392
+ "loss": 2.0477,
393
+ "step": 128
394
+ },
395
+ {
396
+ "epoch": 0.26,
397
+ "learning_rate": 1.9464891130171644e-06,
398
+ "loss": 2.0226,
399
+ "step": 130
400
+ },
401
+ {
402
+ "epoch": 0.27,
403
+ "learning_rate": 1.9443475475028982e-06,
404
+ "loss": 2.0586,
405
+ "step": 132
406
+ },
407
+ {
408
+ "epoch": 0.27,
409
+ "learning_rate": 1.9421651900593897e-06,
410
+ "loss": 2.0146,
411
+ "step": 134
412
+ },
413
+ {
414
+ "epoch": 0.28,
415
+ "learning_rate": 1.9399421349555036e-06,
416
+ "loss": 2.0293,
417
+ "step": 136
418
+ },
419
+ {
420
+ "epoch": 0.28,
421
+ "learning_rate": 1.9376784782180743e-06,
422
+ "loss": 2.0364,
423
+ "step": 138
424
+ },
425
+ {
426
+ "epoch": 0.28,
427
+ "learning_rate": 1.935374317627762e-06,
428
+ "loss": 2.0551,
429
+ "step": 140
430
+ },
431
+ {
432
+ "epoch": 0.29,
433
+ "learning_rate": 1.9330297527148245e-06,
434
+ "loss": 2.035,
435
+ "step": 142
436
+ },
437
+ {
438
+ "epoch": 0.29,
439
+ "learning_rate": 1.9306448847548215e-06,
440
+ "loss": 2.063,
441
+ "step": 144
442
+ },
443
+ {
444
+ "epoch": 0.3,
445
+ "learning_rate": 1.928219816764238e-06,
446
+ "loss": 2.0036,
447
+ "step": 146
448
+ },
449
+ {
450
+ "epoch": 0.3,
451
+ "learning_rate": 1.9257546534960348e-06,
452
+ "loss": 2.0141,
453
+ "step": 148
454
+ },
455
+ {
456
+ "epoch": 0.3,
457
+ "learning_rate": 1.9232495014351246e-06,
458
+ "loss": 2.0572,
459
+ "step": 150
460
+ },
461
+ {
462
+ "epoch": 0.31,
463
+ "learning_rate": 1.92070446879377e-06,
464
+ "loss": 1.9757,
465
+ "step": 152
466
+ },
467
+ {
468
+ "epoch": 0.31,
469
+ "learning_rate": 1.9181196655069123e-06,
470
+ "loss": 2.054,
471
+ "step": 154
472
+ },
473
+ {
474
+ "epoch": 0.32,
475
+ "learning_rate": 1.9154952032274205e-06,
476
+ "loss": 2.0053,
477
+ "step": 156
478
+ },
479
+ {
480
+ "epoch": 0.32,
481
+ "learning_rate": 1.9128311953212676e-06,
482
+ "loss": 2.0291,
483
+ "step": 158
484
+ },
485
+ {
486
+ "epoch": 0.32,
487
+ "learning_rate": 1.910127756862637e-06,
488
+ "loss": 2.0518,
489
+ "step": 160
490
+ },
491
+ {
492
+ "epoch": 0.33,
493
+ "learning_rate": 1.907385004628948e-06,
494
+ "loss": 2.014,
495
+ "step": 162
496
+ },
497
+ {
498
+ "epoch": 0.33,
499
+ "learning_rate": 1.904603057095815e-06,
500
+ "loss": 1.9771,
501
+ "step": 164
502
+ },
503
+ {
504
+ "epoch": 0.34,
505
+ "learning_rate": 1.9017820344319269e-06,
506
+ "loss": 2.0273,
507
+ "step": 166
508
+ },
509
+ {
510
+ "epoch": 0.34,
511
+ "learning_rate": 1.8989220584938572e-06,
512
+ "loss": 1.9516,
513
+ "step": 168
514
+ },
515
+ {
516
+ "epoch": 0.34,
517
+ "learning_rate": 1.896023252820802e-06,
518
+ "loss": 1.9544,
519
+ "step": 170
520
+ },
521
+ {
522
+ "epoch": 0.35,
523
+ "learning_rate": 1.893085742629241e-06,
524
+ "loss": 2.0229,
525
+ "step": 172
526
+ },
527
+ {
528
+ "epoch": 0.35,
529
+ "learning_rate": 1.8901096548075303e-06,
530
+ "loss": 2.0656,
531
+ "step": 174
532
+ },
533
+ {
534
+ "epoch": 0.36,
535
+ "learning_rate": 1.8870951179104212e-06,
536
+ "loss": 2.0522,
537
+ "step": 176
538
+ },
539
+ {
540
+ "epoch": 0.36,
541
+ "learning_rate": 1.8840422621535064e-06,
542
+ "loss": 1.9912,
543
+ "step": 178
544
+ },
545
+ {
546
+ "epoch": 0.36,
547
+ "learning_rate": 1.8809512194075957e-06,
548
+ "loss": 1.9998,
549
+ "step": 180
550
+ },
551
+ {
552
+ "epoch": 0.37,
553
+ "learning_rate": 1.8778221231930203e-06,
554
+ "loss": 2.0262,
555
+ "step": 182
556
+ },
557
+ {
558
+ "epoch": 0.37,
559
+ "learning_rate": 1.8746551086738639e-06,
560
+ "loss": 1.9977,
561
+ "step": 184
562
+ },
563
+ {
564
+ "epoch": 0.38,
565
+ "learning_rate": 1.871450312652126e-06,
566
+ "loss": 2.0313,
567
+ "step": 186
568
+ },
569
+ {
570
+ "epoch": 0.38,
571
+ "learning_rate": 1.8682078735618107e-06,
572
+ "loss": 2.0137,
573
+ "step": 188
574
+ },
575
+ {
576
+ "epoch": 0.39,
577
+ "learning_rate": 1.8649279314629483e-06,
578
+ "loss": 2.1067,
579
+ "step": 190
580
+ },
581
+ {
582
+ "epoch": 0.39,
583
+ "learning_rate": 1.8616106280355442e-06,
584
+ "loss": 2.0302,
585
+ "step": 192
586
+ },
587
+ {
588
+ "epoch": 0.39,
589
+ "learning_rate": 1.85825610657346e-06,
590
+ "loss": 2.0871,
591
+ "step": 194
592
+ },
593
+ {
594
+ "epoch": 0.4,
595
+ "learning_rate": 1.8548645119782237e-06,
596
+ "loss": 2.0279,
597
+ "step": 196
598
+ },
599
+ {
600
+ "epoch": 0.4,
601
+ "learning_rate": 1.851435990752769e-06,
602
+ "loss": 2.098,
603
+ "step": 198
604
+ },
605
+ {
606
+ "epoch": 0.41,
607
+ "learning_rate": 1.8479706909951095e-06,
608
+ "loss": 1.9829,
609
+ "step": 200
610
+ },
611
+ {
612
+ "epoch": 0.41,
613
+ "learning_rate": 1.8444687623919383e-06,
614
+ "loss": 2.0382,
615
+ "step": 202
616
+ },
617
+ {
618
+ "epoch": 0.41,
619
+ "learning_rate": 1.8409303562121661e-06,
620
+ "loss": 2.083,
621
+ "step": 204
622
+ },
623
+ {
624
+ "epoch": 0.42,
625
+ "learning_rate": 1.837355625300383e-06,
626
+ "loss": 1.9887,
627
+ "step": 206
628
+ },
629
+ {
630
+ "epoch": 0.42,
631
+ "learning_rate": 1.8337447240702592e-06,
632
+ "loss": 1.9671,
633
+ "step": 208
634
+ },
635
+ {
636
+ "epoch": 0.43,
637
+ "learning_rate": 1.8300978084978734e-06,
638
+ "loss": 1.9845,
639
+ "step": 210
640
+ },
641
+ {
642
+ "epoch": 0.43,
643
+ "learning_rate": 1.8264150361149757e-06,
644
+ "loss": 2.0818,
645
+ "step": 212
646
+ },
647
+ {
648
+ "epoch": 0.43,
649
+ "learning_rate": 1.8226965660021834e-06,
650
+ "loss": 1.932,
651
+ "step": 214
652
+ },
653
+ {
654
+ "epoch": 0.44,
655
+ "learning_rate": 1.818942558782108e-06,
656
+ "loss": 2.0247,
657
+ "step": 216
658
+ },
659
+ {
660
+ "epoch": 0.44,
661
+ "learning_rate": 1.8151531766124184e-06,
662
+ "loss": 1.9921,
663
+ "step": 218
664
+ },
665
+ {
666
+ "epoch": 0.45,
667
+ "learning_rate": 1.8113285831788363e-06,
668
+ "loss": 2.1025,
669
+ "step": 220
670
+ },
671
+ {
672
+ "epoch": 0.45,
673
+ "learning_rate": 1.807468943688064e-06,
674
+ "loss": 2.0222,
675
+ "step": 222
676
+ },
677
+ {
678
+ "epoch": 0.45,
679
+ "learning_rate": 1.8035744248606509e-06,
680
+ "loss": 2.066,
681
+ "step": 224
682
+ },
683
+ {
684
+ "epoch": 0.46,
685
+ "learning_rate": 1.7996451949237876e-06,
686
+ "loss": 2.0426,
687
+ "step": 226
688
+ },
689
+ {
690
+ "epoch": 0.46,
691
+ "learning_rate": 1.7956814236040449e-06,
692
+ "loss": 2.0251,
693
+ "step": 228
694
+ },
695
+ {
696
+ "epoch": 0.47,
697
+ "learning_rate": 1.7916832821200373e-06,
698
+ "loss": 1.9788,
699
+ "step": 230
700
+ },
701
+ {
702
+ "epoch": 0.47,
703
+ "learning_rate": 1.78765094317503e-06,
704
+ "loss": 2.0021,
705
+ "step": 232
706
+ },
707
+ {
708
+ "epoch": 0.47,
709
+ "learning_rate": 1.7835845809494766e-06,
710
+ "loss": 2.0346,
711
+ "step": 234
712
+ },
713
+ {
714
+ "epoch": 0.48,
715
+ "learning_rate": 1.7794843710934978e-06,
716
+ "loss": 2.1158,
717
+ "step": 236
718
+ },
719
+ {
720
+ "epoch": 0.48,
721
+ "learning_rate": 1.7753504907192923e-06,
722
+ "loss": 2.0334,
723
+ "step": 238
724
+ },
725
+ {
726
+ "epoch": 0.49,
727
+ "learning_rate": 1.7711831183934858e-06,
728
+ "loss": 2.0871,
729
+ "step": 240
730
+ },
731
+ {
732
+ "epoch": 0.49,
733
+ "learning_rate": 1.7669824341294199e-06,
734
+ "loss": 2.0769,
735
+ "step": 242
736
+ },
737
+ {
738
+ "epoch": 0.49,
739
+ "learning_rate": 1.7627486193793741e-06,
740
+ "loss": 2.0154,
741
+ "step": 244
742
+ },
743
+ {
744
+ "epoch": 0.5,
745
+ "learning_rate": 1.7584818570267283e-06,
746
+ "loss": 2.0399,
747
+ "step": 246
748
+ },
749
+ {
750
+ "epoch": 0.5,
751
+ "learning_rate": 1.7541823313780645e-06,
752
+ "loss": 2.0746,
753
+ "step": 248
754
+ },
755
+ {
756
+ "epoch": 0.51,
757
+ "learning_rate": 1.749850228155203e-06,
758
+ "loss": 2.0178,
759
+ "step": 250
760
+ },
761
+ {
762
+ "epoch": 0.51,
763
+ "learning_rate": 1.745485734487182e-06,
764
+ "loss": 2.0285,
765
+ "step": 252
766
+ },
767
+ {
768
+ "epoch": 0.51,
769
+ "learning_rate": 1.7410890389021735e-06,
770
+ "loss": 1.9656,
771
+ "step": 254
772
+ },
773
+ {
774
+ "epoch": 0.52,
775
+ "learning_rate": 1.7366603313193397e-06,
776
+ "loss": 2.0397,
777
+ "step": 256
778
+ },
779
+ {
780
+ "epoch": 0.52,
781
+ "learning_rate": 1.73219980304063e-06,
782
+ "loss": 2.0359,
783
+ "step": 258
784
+ },
785
+ {
786
+ "epoch": 0.53,
787
+ "learning_rate": 1.727707646742516e-06,
788
+ "loss": 2.0021,
789
+ "step": 260
790
+ },
791
+ {
792
+ "epoch": 0.53,
793
+ "learning_rate": 1.7231840564676707e-06,
794
+ "loss": 2.0783,
795
+ "step": 262
796
+ },
797
+ {
798
+ "epoch": 0.54,
799
+ "learning_rate": 1.7186292276165847e-06,
800
+ "loss": 2.0932,
801
+ "step": 264
802
+ },
803
+ {
804
+ "epoch": 0.54,
805
+ "learning_rate": 1.7140433569391272e-06,
806
+ "loss": 2.0216,
807
+ "step": 266
808
+ },
809
+ {
810
+ "epoch": 0.54,
811
+ "learning_rate": 1.709426642526046e-06,
812
+ "loss": 1.9481,
813
+ "step": 268
814
+ },
815
+ {
816
+ "epoch": 0.55,
817
+ "learning_rate": 1.7047792838004117e-06,
818
+ "loss": 1.9999,
819
+ "step": 270
820
+ },
821
+ {
822
+ "epoch": 0.55,
823
+ "learning_rate": 1.7001014815090037e-06,
824
+ "loss": 2.0074,
825
+ "step": 272
826
+ },
827
+ {
828
+ "epoch": 0.56,
829
+ "learning_rate": 1.6953934377136375e-06,
830
+ "loss": 2.0013,
831
+ "step": 274
832
+ },
833
+ {
834
+ "epoch": 0.56,
835
+ "learning_rate": 1.690655355782437e-06,
836
+ "loss": 2.0775,
837
+ "step": 276
838
+ },
839
+ {
840
+ "epoch": 0.56,
841
+ "learning_rate": 1.6858874403810505e-06,
842
+ "loss": 2.0042,
843
+ "step": 278
844
+ },
845
+ {
846
+ "epoch": 0.57,
847
+ "learning_rate": 1.6810898974638094e-06,
848
+ "loss": 1.997,
849
+ "step": 280
850
+ },
851
+ {
852
+ "epoch": 0.57,
853
+ "learning_rate": 1.6762629342648318e-06,
854
+ "loss": 1.9919,
855
+ "step": 282
856
+ },
857
+ {
858
+ "epoch": 0.58,
859
+ "learning_rate": 1.671406759289071e-06,
860
+ "loss": 2.0189,
861
+ "step": 284
862
+ },
863
+ {
864
+ "epoch": 0.58,
865
+ "learning_rate": 1.666521582303309e-06,
866
+ "loss": 1.9571,
867
+ "step": 286
868
+ },
869
+ {
870
+ "epoch": 0.58,
871
+ "learning_rate": 1.6616076143270947e-06,
872
+ "loss": 1.989,
873
+ "step": 288
874
+ },
875
+ {
876
+ "epoch": 0.59,
877
+ "learning_rate": 1.6566650676236305e-06,
878
+ "loss": 2.0321,
879
+ "step": 290
880
+ },
881
+ {
882
+ "epoch": 0.59,
883
+ "learning_rate": 1.6516941556906009e-06,
884
+ "loss": 2.0197,
885
+ "step": 292
886
+ },
887
+ {
888
+ "epoch": 0.6,
889
+ "learning_rate": 1.646695093250953e-06,
890
+ "loss": 2.0096,
891
+ "step": 294
892
+ },
893
+ {
894
+ "epoch": 0.6,
895
+ "learning_rate": 1.6416680962436187e-06,
896
+ "loss": 1.9316,
897
+ "step": 296
898
+ },
899
+ {
900
+ "epoch": 0.6,
901
+ "learning_rate": 1.6366133818141891e-06,
902
+ "loss": 2.0131,
903
+ "step": 298
904
+ },
905
+ {
906
+ "epoch": 0.61,
907
+ "learning_rate": 1.631531168305534e-06,
908
+ "loss": 2.0016,
909
+ "step": 300
910
+ },
911
+ {
912
+ "epoch": 0.61,
913
+ "learning_rate": 1.6264216752483695e-06,
914
+ "loss": 2.0862,
915
+ "step": 302
916
+ },
917
+ {
918
+ "epoch": 0.62,
919
+ "learning_rate": 1.621285123351777e-06,
920
+ "loss": 1.9881,
921
+ "step": 304
922
+ },
923
+ {
924
+ "epoch": 0.62,
925
+ "learning_rate": 1.6161217344936677e-06,
926
+ "loss": 2.0715,
927
+ "step": 306
928
+ },
929
+ {
930
+ "epoch": 0.62,
931
+ "learning_rate": 1.6109317317111995e-06,
932
+ "loss": 2.0205,
933
+ "step": 308
934
+ },
935
+ {
936
+ "epoch": 0.63,
937
+ "learning_rate": 1.605715339191142e-06,
938
+ "loss": 2.0036,
939
+ "step": 310
940
+ },
941
+ {
942
+ "epoch": 0.63,
943
+ "learning_rate": 1.6004727822601932e-06,
944
+ "loss": 2.0367,
945
+ "step": 312
946
+ },
947
+ {
948
+ "epoch": 0.64,
949
+ "learning_rate": 1.595204287375246e-06,
950
+ "loss": 1.9935,
951
+ "step": 314
952
+ },
953
+ {
954
+ "epoch": 0.64,
955
+ "learning_rate": 1.5899100821136061e-06,
956
+ "loss": 2.0275,
957
+ "step": 316
958
+ },
959
+ {
960
+ "epoch": 0.64,
961
+ "learning_rate": 1.584590395163162e-06,
962
+ "loss": 1.947,
963
+ "step": 318
964
+ },
965
+ {
966
+ "epoch": 0.65,
967
+ "learning_rate": 1.5792454563125058e-06,
968
+ "loss": 2.009,
969
+ "step": 320
970
+ },
971
+ {
972
+ "epoch": 0.65,
973
+ "learning_rate": 1.573875496441008e-06,
974
+ "loss": 2.0075,
975
+ "step": 322
976
+ },
977
+ {
978
+ "epoch": 0.66,
979
+ "learning_rate": 1.5684807475088451e-06,
980
+ "loss": 1.9395,
981
+ "step": 324
982
+ },
983
+ {
984
+ "epoch": 0.66,
985
+ "learning_rate": 1.5630614425469775e-06,
986
+ "loss": 2.0678,
987
+ "step": 326
988
+ },
989
+ {
990
+ "epoch": 0.66,
991
+ "learning_rate": 1.557617815647086e-06,
992
+ "loss": 2.0026,
993
+ "step": 328
994
+ },
995
+ {
996
+ "epoch": 0.67,
997
+ "learning_rate": 1.5521501019514595e-06,
998
+ "loss": 2.0108,
999
+ "step": 330
1000
+ },
1001
+ {
1002
+ "epoch": 0.67,
1003
+ "learning_rate": 1.5466585376428362e-06,
1004
+ "loss": 2.0776,
1005
+ "step": 332
1006
+ },
1007
+ {
1008
+ "epoch": 0.68,
1009
+ "learning_rate": 1.5411433599342037e-06,
1010
+ "loss": 1.9468,
1011
+ "step": 334
1012
+ },
1013
+ {
1014
+ "epoch": 0.68,
1015
+ "learning_rate": 1.535604807058551e-06,
1016
+ "loss": 1.9065,
1017
+ "step": 336
1018
+ },
1019
+ {
1020
+ "epoch": 0.69,
1021
+ "learning_rate": 1.5300431182585777e-06,
1022
+ "loss": 2.0186,
1023
+ "step": 338
1024
+ },
1025
+ {
1026
+ "epoch": 0.69,
1027
+ "learning_rate": 1.524458533776361e-06,
1028
+ "loss": 2.0968,
1029
+ "step": 340
1030
+ },
1031
+ {
1032
+ "epoch": 0.69,
1033
+ "learning_rate": 1.5188512948429764e-06,
1034
+ "loss": 2.0716,
1035
+ "step": 342
1036
+ },
1037
+ {
1038
+ "epoch": 0.7,
1039
+ "learning_rate": 1.5132216436680797e-06,
1040
+ "loss": 2.0505,
1041
+ "step": 344
1042
+ },
1043
+ {
1044
+ "epoch": 0.7,
1045
+ "learning_rate": 1.5075698234294422e-06,
1046
+ "loss": 2.035,
1047
+ "step": 346
1048
+ },
1049
+ {
1050
+ "epoch": 0.71,
1051
+ "learning_rate": 1.5018960782624484e-06,
1052
+ "loss": 1.999,
1053
+ "step": 348
1054
+ },
1055
+ {
1056
+ "epoch": 0.71,
1057
+ "learning_rate": 1.4962006532495487e-06,
1058
+ "loss": 1.975,
1059
+ "step": 350
1060
+ },
1061
+ {
1062
+ "epoch": 0.71,
1063
+ "learning_rate": 1.4904837944096743e-06,
1064
+ "loss": 2.0454,
1065
+ "step": 352
1066
+ },
1067
+ {
1068
+ "epoch": 0.72,
1069
+ "learning_rate": 1.4847457486876096e-06,
1070
+ "loss": 2.0372,
1071
+ "step": 354
1072
+ },
1073
+ {
1074
+ "epoch": 0.72,
1075
+ "learning_rate": 1.4789867639433246e-06,
1076
+ "loss": 1.9876,
1077
+ "step": 356
1078
+ },
1079
+ {
1080
+ "epoch": 0.73,
1081
+ "learning_rate": 1.4732070889412692e-06,
1082
+ "loss": 2.018,
1083
+ "step": 358
1084
+ },
1085
+ {
1086
+ "epoch": 0.73,
1087
+ "learning_rate": 1.4674069733396275e-06,
1088
+ "loss": 2.0072,
1089
+ "step": 360
1090
+ },
1091
+ {
1092
+ "epoch": 0.73,
1093
+ "learning_rate": 1.4615866676795332e-06,
1094
+ "loss": 2.0242,
1095
+ "step": 362
1096
+ },
1097
+ {
1098
+ "epoch": 0.74,
1099
+ "learning_rate": 1.4557464233742476e-06,
1100
+ "loss": 1.992,
1101
+ "step": 364
1102
+ },
1103
+ {
1104
+ "epoch": 0.74,
1105
+ "learning_rate": 1.4498864926982995e-06,
1106
+ "loss": 2.01,
1107
+ "step": 366
1108
+ },
1109
+ {
1110
+ "epoch": 0.75,
1111
+ "learning_rate": 1.4440071287765874e-06,
1112
+ "loss": 1.9977,
1113
+ "step": 368
1114
+ },
1115
+ {
1116
+ "epoch": 0.75,
1117
+ "learning_rate": 1.4381085855734468e-06,
1118
+ "loss": 2.0828,
1119
+ "step": 370
1120
+ },
1121
+ {
1122
+ "epoch": 0.75,
1123
+ "learning_rate": 1.4321911178816788e-06,
1124
+ "loss": 1.9816,
1125
+ "step": 372
1126
+ },
1127
+ {
1128
+ "epoch": 0.76,
1129
+ "learning_rate": 1.4262549813115447e-06,
1130
+ "loss": 2.0288,
1131
+ "step": 374
1132
+ },
1133
+ {
1134
+ "epoch": 0.76,
1135
+ "learning_rate": 1.4203004322797249e-06,
1136
+ "loss": 2.075,
1137
+ "step": 376
1138
+ },
1139
+ {
1140
+ "epoch": 0.77,
1141
+ "learning_rate": 1.4143277279982413e-06,
1142
+ "loss": 2.0101,
1143
+ "step": 378
1144
+ },
1145
+ {
1146
+ "epoch": 0.77,
1147
+ "learning_rate": 1.4083371264633494e-06,
1148
+ "loss": 1.9774,
1149
+ "step": 380
1150
+ },
1151
+ {
1152
+ "epoch": 0.77,
1153
+ "learning_rate": 1.4023288864443915e-06,
1154
+ "loss": 2.0611,
1155
+ "step": 382
1156
+ },
1157
+ {
1158
+ "epoch": 0.78,
1159
+ "learning_rate": 1.3963032674726196e-06,
1160
+ "loss": 2.0358,
1161
+ "step": 384
1162
+ },
1163
+ {
1164
+ "epoch": 0.78,
1165
+ "learning_rate": 1.390260529829986e-06,
1166
+ "loss": 1.9747,
1167
+ "step": 386
1168
+ },
1169
+ {
1170
+ "epoch": 0.79,
1171
+ "learning_rate": 1.3842009345378975e-06,
1172
+ "loss": 2.0558,
1173
+ "step": 388
1174
+ },
1175
+ {
1176
+ "epoch": 0.79,
1177
+ "learning_rate": 1.3781247433459446e-06,
1178
+ "loss": 2.0173,
1179
+ "step": 390
1180
+ },
1181
+ {
1182
+ "epoch": 0.79,
1183
+ "learning_rate": 1.3720322187205895e-06,
1184
+ "loss": 2.0066,
1185
+ "step": 392
1186
+ },
1187
+ {
1188
+ "epoch": 0.8,
1189
+ "learning_rate": 1.3659236238338336e-06,
1190
+ "loss": 2.0284,
1191
+ "step": 394
1192
+ },
1193
+ {
1194
+ "epoch": 0.8,
1195
+ "learning_rate": 1.3597992225518464e-06,
1196
+ "loss": 2.0022,
1197
+ "step": 396
1198
+ },
1199
+ {
1200
+ "epoch": 0.81,
1201
+ "learning_rate": 1.3536592794235695e-06,
1202
+ "loss": 2.0356,
1203
+ "step": 398
1204
+ },
1205
+ {
1206
+ "epoch": 0.81,
1207
+ "learning_rate": 1.3475040596692876e-06,
1208
+ "loss": 2.0387,
1209
+ "step": 400
1210
+ },
1211
+ {
1212
+ "epoch": 0.81,
1213
+ "learning_rate": 1.3413338291691726e-06,
1214
+ "loss": 2.0006,
1215
+ "step": 402
1216
+ },
1217
+ {
1218
+ "epoch": 0.82,
1219
+ "learning_rate": 1.3351488544518e-06,
1220
+ "loss": 2.0543,
1221
+ "step": 404
1222
+ },
1223
+ {
1224
+ "epoch": 0.82,
1225
+ "learning_rate": 1.3289494026826335e-06,
1226
+ "loss": 2.0021,
1227
+ "step": 406
1228
+ },
1229
+ {
1230
+ "epoch": 0.83,
1231
+ "learning_rate": 1.3227357416524875e-06,
1232
+ "loss": 2.0066,
1233
+ "step": 408
1234
+ },
1235
+ {
1236
+ "epoch": 0.83,
1237
+ "learning_rate": 1.316508139765956e-06,
1238
+ "loss": 1.9943,
1239
+ "step": 410
1240
+ },
1241
+ {
1242
+ "epoch": 0.84,
1243
+ "learning_rate": 1.3102668660298226e-06,
1244
+ "loss": 1.9659,
1245
+ "step": 412
1246
+ },
1247
+ {
1248
+ "epoch": 0.84,
1249
+ "learning_rate": 1.304012190041437e-06,
1250
+ "loss": 2.0319,
1251
+ "step": 414
1252
+ },
1253
+ {
1254
+ "epoch": 0.84,
1255
+ "learning_rate": 1.2977443819770714e-06,
1256
+ "loss": 2.0505,
1257
+ "step": 416
1258
+ },
1259
+ {
1260
+ "epoch": 0.85,
1261
+ "learning_rate": 1.2914637125802511e-06,
1262
+ "loss": 1.9475,
1263
+ "step": 418
1264
+ },
1265
+ {
1266
+ "epoch": 0.85,
1267
+ "learning_rate": 1.2851704531500562e-06,
1268
+ "loss": 2.0061,
1269
+ "step": 420
1270
+ },
1271
+ {
1272
+ "epoch": 0.86,
1273
+ "learning_rate": 1.2788648755294054e-06,
1274
+ "loss": 1.9459,
1275
+ "step": 422
1276
+ },
1277
+ {
1278
+ "epoch": 0.86,
1279
+ "learning_rate": 1.272547252093312e-06,
1280
+ "loss": 2.003,
1281
+ "step": 424
1282
+ },
1283
+ {
1284
+ "epoch": 0.86,
1285
+ "learning_rate": 1.2662178557371197e-06,
1286
+ "loss": 2.0195,
1287
+ "step": 426
1288
+ },
1289
+ {
1290
+ "epoch": 0.87,
1291
+ "learning_rate": 1.2598769598647134e-06,
1292
+ "loss": 2.0393,
1293
+ "step": 428
1294
+ },
1295
+ {
1296
+ "epoch": 0.87,
1297
+ "learning_rate": 1.2535248383767101e-06,
1298
+ "loss": 2.0154,
1299
+ "step": 430
1300
+ },
1301
+ {
1302
+ "epoch": 0.88,
1303
+ "learning_rate": 1.2471617656586268e-06,
1304
+ "loss": 2.0338,
1305
+ "step": 432
1306
+ },
1307
+ {
1308
+ "epoch": 0.88,
1309
+ "learning_rate": 1.2407880165690287e-06,
1310
+ "loss": 1.8944,
1311
+ "step": 434
1312
+ },
1313
+ {
1314
+ "epoch": 0.88,
1315
+ "learning_rate": 1.2344038664276568e-06,
1316
+ "loss": 2.0415,
1317
+ "step": 436
1318
+ },
1319
+ {
1320
+ "epoch": 0.89,
1321
+ "learning_rate": 1.2280095910035341e-06,
1322
+ "loss": 1.9654,
1323
+ "step": 438
1324
+ },
1325
+ {
1326
+ "epoch": 0.89,
1327
+ "learning_rate": 1.221605466503055e-06,
1328
+ "loss": 2.0289,
1329
+ "step": 440
1330
+ },
1331
+ {
1332
+ "epoch": 0.9,
1333
+ "learning_rate": 1.2151917695580523e-06,
1334
+ "loss": 2.0046,
1335
+ "step": 442
1336
+ },
1337
+ {
1338
+ "epoch": 0.9,
1339
+ "learning_rate": 1.2087687772138499e-06,
1340
+ "loss": 2.0994,
1341
+ "step": 444
1342
+ },
1343
+ {
1344
+ "epoch": 0.9,
1345
+ "learning_rate": 1.2023367669172946e-06,
1346
+ "loss": 2.0389,
1347
+ "step": 446
1348
+ },
1349
+ {
1350
+ "epoch": 0.91,
1351
+ "learning_rate": 1.1958960165047715e-06,
1352
+ "loss": 2.0108,
1353
+ "step": 448
1354
+ },
1355
+ {
1356
+ "epoch": 0.91,
1357
+ "learning_rate": 1.1894468041902029e-06,
1358
+ "loss": 2.1058,
1359
+ "step": 450
1360
+ },
1361
+ {
1362
+ "epoch": 0.92,
1363
+ "learning_rate": 1.1829894085530296e-06,
1364
+ "loss": 2.0575,
1365
+ "step": 452
1366
+ },
1367
+ {
1368
+ "epoch": 0.92,
1369
+ "learning_rate": 1.1765241085261801e-06,
1370
+ "loss": 2.0362,
1371
+ "step": 454
1372
+ },
1373
+ {
1374
+ "epoch": 0.92,
1375
+ "learning_rate": 1.1700511833840186e-06,
1376
+ "loss": 1.9836,
1377
+ "step": 456
1378
+ },
1379
+ {
1380
+ "epoch": 0.93,
1381
+ "learning_rate": 1.163570912730283e-06,
1382
+ "loss": 2.0067,
1383
+ "step": 458
1384
+ },
1385
+ {
1386
+ "epoch": 0.93,
1387
+ "learning_rate": 1.157083576486007e-06,
1388
+ "loss": 2.099,
1389
+ "step": 460
1390
+ },
1391
+ {
1392
+ "epoch": 0.94,
1393
+ "learning_rate": 1.1505894548774293e-06,
1394
+ "loss": 2.0218,
1395
+ "step": 462
1396
+ },
1397
+ {
1398
+ "epoch": 0.94,
1399
+ "learning_rate": 1.1440888284238889e-06,
1400
+ "loss": 2.0159,
1401
+ "step": 464
1402
+ },
1403
+ {
1404
+ "epoch": 0.94,
1405
+ "learning_rate": 1.1375819779257057e-06,
1406
+ "loss": 2.0044,
1407
+ "step": 466
1408
+ },
1409
+ {
1410
+ "epoch": 0.95,
1411
+ "learning_rate": 1.1310691844520542e-06,
1412
+ "loss": 2.098,
1413
+ "step": 468
1414
+ },
1415
+ {
1416
+ "epoch": 0.95,
1417
+ "learning_rate": 1.1245507293288204e-06,
1418
+ "loss": 2.0625,
1419
+ "step": 470
1420
+ },
1421
+ {
1422
+ "epoch": 0.96,
1423
+ "learning_rate": 1.1180268941264498e-06,
1424
+ "loss": 2.0096,
1425
+ "step": 472
1426
+ },
1427
+ {
1428
+ "epoch": 0.96,
1429
+ "learning_rate": 1.1114979606477865e-06,
1430
+ "loss": 1.999,
1431
+ "step": 474
1432
+ },
1433
+ {
1434
+ "epoch": 0.96,
1435
+ "learning_rate": 1.104964210915898e-06,
1436
+ "loss": 2.0256,
1437
+ "step": 476
1438
+ },
1439
+ {
1440
+ "epoch": 0.97,
1441
+ "learning_rate": 1.0984259271618945e-06,
1442
+ "loss": 2.0035,
1443
+ "step": 478
1444
+ },
1445
+ {
1446
+ "epoch": 0.97,
1447
+ "learning_rate": 1.0918833918127375e-06,
1448
+ "loss": 2.0277,
1449
+ "step": 480
1450
+ },
1451
+ {
1452
+ "epoch": 0.98,
1453
+ "learning_rate": 1.0853368874790392e-06,
1454
+ "loss": 2.0701,
1455
+ "step": 482
1456
+ },
1457
+ {
1458
+ "epoch": 0.98,
1459
+ "learning_rate": 1.0787866969428567e-06,
1460
+ "loss": 1.9819,
1461
+ "step": 484
1462
+ },
1463
+ {
1464
+ "epoch": 0.99,
1465
+ "learning_rate": 1.0722331031454747e-06,
1466
+ "loss": 2.0453,
1467
+ "step": 486
1468
+ },
1469
+ {
1470
+ "epoch": 0.99,
1471
+ "learning_rate": 1.0656763891751863e-06,
1472
+ "loss": 2.059,
1473
+ "step": 488
1474
+ },
1475
+ {
1476
+ "epoch": 0.99,
1477
+ "learning_rate": 1.0591168382550616e-06,
1478
+ "loss": 2.0781,
1479
+ "step": 490
1480
+ },
1481
+ {
1482
+ "epoch": 1.0,
1483
+ "learning_rate": 1.052554733730716e-06,
1484
+ "loss": 2.0133,
1485
+ "step": 492
1486
+ },
1487
+ {
1488
+ "epoch": 1.0,
1489
+ "learning_rate": 1.0459903590580706e-06,
1490
+ "loss": 1.9779,
1491
+ "step": 494
1492
+ },
1493
+ {
1494
+ "epoch": 1.01,
1495
+ "learning_rate": 1.0394239977911068e-06,
1496
+ "loss": 1.9705,
1497
+ "step": 496
1498
+ },
1499
+ {
1500
+ "epoch": 1.01,
1501
+ "learning_rate": 1.0328559335696187e-06,
1502
+ "loss": 1.9889,
1503
+ "step": 498
1504
+ },
1505
+ {
1506
+ "epoch": 1.01,
1507
+ "learning_rate": 1.0262864501069616e-06,
1508
+ "loss": 2.0143,
1509
+ "step": 500
1510
+ },
1511
+ {
1512
+ "epoch": 1.02,
1513
+ "learning_rate": 1.0197158311777955e-06,
1514
+ "loss": 1.9436,
1515
+ "step": 502
1516
+ },
1517
+ {
1518
+ "epoch": 1.02,
1519
+ "learning_rate": 1.013144360605829e-06,
1520
+ "loss": 2.0313,
1521
+ "step": 504
1522
+ },
1523
+ {
1524
+ "epoch": 1.03,
1525
+ "learning_rate": 1.0065723222515565e-06,
1526
+ "loss": 1.9854,
1527
+ "step": 506
1528
+ },
1529
+ {
1530
+ "epoch": 1.03,
1531
+ "learning_rate": 1e-06,
1532
+ "loss": 1.9626,
1533
+ "step": 508
1534
+ },
1535
+ {
1536
+ "epoch": 1.03,
1537
+ "learning_rate": 9.934276777484434e-07,
1538
+ "loss": 2.0107,
1539
+ "step": 510
1540
+ },
1541
+ {
1542
+ "epoch": 1.04,
1543
+ "learning_rate": 9.868556393941712e-07,
1544
+ "loss": 1.9761,
1545
+ "step": 512
1546
+ },
1547
+ {
1548
+ "epoch": 1.04,
1549
+ "learning_rate": 9.802841688222042e-07,
1550
+ "loss": 1.9393,
1551
+ "step": 514
1552
+ },
1553
+ {
1554
+ "epoch": 1.05,
1555
+ "learning_rate": 9.737135498930383e-07,
1556
+ "loss": 2.0417,
1557
+ "step": 516
1558
+ },
1559
+ {
1560
+ "epoch": 1.05,
1561
+ "learning_rate": 9.671440664303812e-07,
1562
+ "loss": 1.9319,
1563
+ "step": 518
1564
+ },
1565
+ {
1566
+ "epoch": 1.05,
1567
+ "learning_rate": 9.605760022088933e-07,
1568
+ "loss": 1.8757,
1569
+ "step": 520
1570
+ },
1571
+ {
1572
+ "epoch": 1.06,
1573
+ "learning_rate": 9.540096409419296e-07,
1574
+ "loss": 1.9148,
1575
+ "step": 522
1576
+ },
1577
+ {
1578
+ "epoch": 1.06,
1579
+ "learning_rate": 9.474452662692837e-07,
1580
+ "loss": 1.9374,
1581
+ "step": 524
1582
+ },
1583
+ {
1584
+ "epoch": 1.07,
1585
+ "learning_rate": 9.408831617449384e-07,
1586
+ "loss": 1.9556,
1587
+ "step": 526
1588
+ },
1589
+ {
1590
+ "epoch": 1.07,
1591
+ "learning_rate": 9.343236108248138e-07,
1592
+ "loss": 1.9933,
1593
+ "step": 528
1594
+ },
1595
+ {
1596
+ "epoch": 1.07,
1597
+ "learning_rate": 9.277668968545253e-07,
1598
+ "loss": 2.0201,
1599
+ "step": 530
1600
+ },
1601
+ {
1602
+ "epoch": 1.08,
1603
+ "learning_rate": 9.212133030571436e-07,
1604
+ "loss": 1.9669,
1605
+ "step": 532
1606
+ },
1607
+ {
1608
+ "epoch": 1.08,
1609
+ "learning_rate": 9.146631125209607e-07,
1610
+ "loss": 1.9788,
1611
+ "step": 534
1612
+ },
1613
+ {
1614
+ "epoch": 1.09,
1615
+ "learning_rate": 9.081166081872625e-07,
1616
+ "loss": 1.9948,
1617
+ "step": 536
1618
+ },
1619
+ {
1620
+ "epoch": 1.09,
1621
+ "learning_rate": 9.015740728381054e-07,
1622
+ "loss": 1.9575,
1623
+ "step": 538
1624
+ },
1625
+ {
1626
+ "epoch": 1.09,
1627
+ "learning_rate": 8.950357890841019e-07,
1628
+ "loss": 2.0099,
1629
+ "step": 540
1630
+ },
1631
+ {
1632
+ "epoch": 1.1,
1633
+ "learning_rate": 8.885020393522135e-07,
1634
+ "loss": 1.9485,
1635
+ "step": 542
1636
+ },
1637
+ {
1638
+ "epoch": 1.1,
1639
+ "learning_rate": 8.8197310587355e-07,
1640
+ "loss": 1.9656,
1641
+ "step": 544
1642
+ },
1643
+ {
1644
+ "epoch": 1.11,
1645
+ "learning_rate": 8.754492706711797e-07,
1646
+ "loss": 1.9441,
1647
+ "step": 546
1648
+ },
1649
+ {
1650
+ "epoch": 1.11,
1651
+ "learning_rate": 8.689308155479459e-07,
1652
+ "loss": 1.9749,
1653
+ "step": 548
1654
+ },
1655
+ {
1656
+ "epoch": 1.11,
1657
+ "learning_rate": 8.624180220742945e-07,
1658
+ "loss": 1.9914,
1659
+ "step": 550
1660
+ },
1661
+ {
1662
+ "epoch": 1.12,
1663
+ "learning_rate": 8.559111715761114e-07,
1664
+ "loss": 1.9579,
1665
+ "step": 552
1666
+ },
1667
+ {
1668
+ "epoch": 1.12,
1669
+ "learning_rate": 8.494105451225703e-07,
1670
+ "loss": 1.9853,
1671
+ "step": 554
1672
+ },
1673
+ {
1674
+ "epoch": 1.13,
1675
+ "learning_rate": 8.42916423513993e-07,
1676
+ "loss": 2.002,
1677
+ "step": 556
1678
+ },
1679
+ {
1680
+ "epoch": 1.13,
1681
+ "learning_rate": 8.364290872697173e-07,
1682
+ "loss": 1.9817,
1683
+ "step": 558
1684
+ },
1685
+ {
1686
+ "epoch": 1.14,
1687
+ "learning_rate": 8.299488166159816e-07,
1688
+ "loss": 1.946,
1689
+ "step": 560
1690
+ },
1691
+ {
1692
+ "epoch": 1.14,
1693
+ "learning_rate": 8.234758914738198e-07,
1694
+ "loss": 1.9846,
1695
+ "step": 562
1696
+ },
1697
+ {
1698
+ "epoch": 1.14,
1699
+ "learning_rate": 8.1701059144697e-07,
1700
+ "loss": 2.0105,
1701
+ "step": 564
1702
+ },
1703
+ {
1704
+ "epoch": 1.15,
1705
+ "learning_rate": 8.105531958097972e-07,
1706
+ "loss": 2.0266,
1707
+ "step": 566
1708
+ },
1709
+ {
1710
+ "epoch": 1.15,
1711
+ "learning_rate": 8.041039834952286e-07,
1712
+ "loss": 2.0291,
1713
+ "step": 568
1714
+ },
1715
+ {
1716
+ "epoch": 1.16,
1717
+ "learning_rate": 7.976632330827054e-07,
1718
+ "loss": 2.0173,
1719
+ "step": 570
1720
+ },
1721
+ {
1722
+ "epoch": 1.16,
1723
+ "learning_rate": 7.912312227861502e-07,
1724
+ "loss": 1.8973,
1725
+ "step": 572
1726
+ },
1727
+ {
1728
+ "epoch": 1.16,
1729
+ "learning_rate": 7.848082304419477e-07,
1730
+ "loss": 1.9388,
1731
+ "step": 574
1732
+ },
1733
+ {
1734
+ "epoch": 1.17,
1735
+ "learning_rate": 7.78394533496945e-07,
1736
+ "loss": 1.9652,
1737
+ "step": 576
1738
+ },
1739
+ {
1740
+ "epoch": 1.17,
1741
+ "learning_rate": 7.719904089964658e-07,
1742
+ "loss": 1.974,
1743
+ "step": 578
1744
+ },
1745
+ {
1746
+ "epoch": 1.18,
1747
+ "learning_rate": 7.655961335723432e-07,
1748
+ "loss": 1.9029,
1749
+ "step": 580
1750
+ },
1751
+ {
1752
+ "epoch": 1.18,
1753
+ "learning_rate": 7.592119834309713e-07,
1754
+ "loss": 1.9993,
1755
+ "step": 582
1756
+ },
1757
+ {
1758
+ "epoch": 1.18,
1759
+ "learning_rate": 7.528382343413732e-07,
1760
+ "loss": 1.9763,
1761
+ "step": 584
1762
+ },
1763
+ {
1764
+ "epoch": 1.19,
1765
+ "learning_rate": 7.4647516162329e-07,
1766
+ "loss": 1.9842,
1767
+ "step": 586
1768
+ },
1769
+ {
1770
+ "epoch": 1.19,
1771
+ "learning_rate": 7.401230401352866e-07,
1772
+ "loss": 1.9444,
1773
+ "step": 588
1774
+ },
1775
+ {
1776
+ "epoch": 1.2,
1777
+ "learning_rate": 7.337821442628805e-07,
1778
+ "loss": 2.0126,
1779
+ "step": 590
1780
+ },
1781
+ {
1782
+ "epoch": 1.2,
1783
+ "learning_rate": 7.274527479066882e-07,
1784
+ "loss": 1.9759,
1785
+ "step": 592
1786
+ },
1787
+ {
1788
+ "epoch": 1.2,
1789
+ "learning_rate": 7.211351244705946e-07,
1790
+ "loss": 2.0648,
1791
+ "step": 594
1792
+ },
1793
+ {
1794
+ "epoch": 1.21,
1795
+ "learning_rate": 7.148295468499437e-07,
1796
+ "loss": 1.9903,
1797
+ "step": 596
1798
+ },
1799
+ {
1800
+ "epoch": 1.21,
1801
+ "learning_rate": 7.085362874197489e-07,
1802
+ "loss": 1.9218,
1803
+ "step": 598
1804
+ },
1805
+ {
1806
+ "epoch": 1.22,
1807
+ "learning_rate": 7.022556180229284e-07,
1808
+ "loss": 2.0351,
1809
+ "step": 600
1810
+ },
1811
+ {
1812
+ "epoch": 1.22,
1813
+ "learning_rate": 6.959878099585633e-07,
1814
+ "loss": 1.9695,
1815
+ "step": 602
1816
+ },
1817
+ {
1818
+ "epoch": 1.22,
1819
+ "learning_rate": 6.897331339701775e-07,
1820
+ "loss": 1.9638,
1821
+ "step": 604
1822
+ },
1823
+ {
1824
+ "epoch": 1.23,
1825
+ "learning_rate": 6.834918602340438e-07,
1826
+ "loss": 1.9558,
1827
+ "step": 606
1828
+ },
1829
+ {
1830
+ "epoch": 1.23,
1831
+ "learning_rate": 6.772642583475125e-07,
1832
+ "loss": 1.9801,
1833
+ "step": 608
1834
+ },
1835
+ {
1836
+ "epoch": 1.24,
1837
+ "learning_rate": 6.710505973173664e-07,
1838
+ "loss": 1.938,
1839
+ "step": 610
1840
+ },
1841
+ {
1842
+ "epoch": 1.24,
1843
+ "learning_rate": 6.648511455482003e-07,
1844
+ "loss": 2.0556,
1845
+ "step": 612
1846
+ },
1847
+ {
1848
+ "epoch": 1.24,
1849
+ "learning_rate": 6.586661708308272e-07,
1850
+ "loss": 1.984,
1851
+ "step": 614
1852
+ },
1853
+ {
1854
+ "epoch": 1.25,
1855
+ "learning_rate": 6.524959403307125e-07,
1856
+ "loss": 1.9922,
1857
+ "step": 616
1858
+ },
1859
+ {
1860
+ "epoch": 1.25,
1861
+ "learning_rate": 6.463407205764305e-07,
1862
+ "loss": 1.9744,
1863
+ "step": 618
1864
+ },
1865
+ {
1866
+ "epoch": 1.26,
1867
+ "learning_rate": 6.402007774481535e-07,
1868
+ "loss": 1.9974,
1869
+ "step": 620
1870
+ },
1871
+ {
1872
+ "epoch": 1.26,
1873
+ "learning_rate": 6.340763761661665e-07,
1874
+ "loss": 1.9169,
1875
+ "step": 622
1876
+ },
1877
+ {
1878
+ "epoch": 1.26,
1879
+ "learning_rate": 6.279677812794102e-07,
1880
+ "loss": 1.9615,
1881
+ "step": 624
1882
+ },
1883
+ {
1884
+ "epoch": 1.27,
1885
+ "learning_rate": 6.218752566540554e-07,
1886
+ "loss": 2.0275,
1887
+ "step": 626
1888
+ },
1889
+ {
1890
+ "epoch": 1.27,
1891
+ "learning_rate": 6.157990654621023e-07,
1892
+ "loss": 1.9945,
1893
+ "step": 628
1894
+ },
1895
+ {
1896
+ "epoch": 1.28,
1897
+ "learning_rate": 6.097394701700144e-07,
1898
+ "loss": 1.9583,
1899
+ "step": 630
1900
+ },
1901
+ {
1902
+ "epoch": 1.28,
1903
+ "learning_rate": 6.036967325273806e-07,
1904
+ "loss": 1.919,
1905
+ "step": 632
1906
+ },
1907
+ {
1908
+ "epoch": 1.29,
1909
+ "learning_rate": 5.976711135556086e-07,
1910
+ "loss": 2.0058,
1911
+ "step": 634
1912
+ },
1913
+ {
1914
+ "epoch": 1.29,
1915
+ "learning_rate": 5.916628735366504e-07,
1916
+ "loss": 1.8788,
1917
+ "step": 636
1918
+ },
1919
+ {
1920
+ "epoch": 1.29,
1921
+ "learning_rate": 5.856722720017586e-07,
1922
+ "loss": 2.0375,
1923
+ "step": 638
1924
+ },
1925
+ {
1926
+ "epoch": 1.3,
1927
+ "learning_rate": 5.796995677202753e-07,
1928
+ "loss": 2.0462,
1929
+ "step": 640
1930
+ },
1931
+ {
1932
+ "epoch": 1.3,
1933
+ "learning_rate": 5.737450186884554e-07,
1934
+ "loss": 1.979,
1935
+ "step": 642
1936
+ },
1937
+ {
1938
+ "epoch": 1.31,
1939
+ "learning_rate": 5.678088821183211e-07,
1940
+ "loss": 1.9989,
1941
+ "step": 644
1942
+ },
1943
+ {
1944
+ "epoch": 1.31,
1945
+ "learning_rate": 5.618914144265531e-07,
1946
+ "loss": 1.913,
1947
+ "step": 646
1948
+ },
1949
+ {
1950
+ "epoch": 1.31,
1951
+ "learning_rate": 5.559928712234126e-07,
1952
+ "loss": 1.9717,
1953
+ "step": 648
1954
+ },
1955
+ {
1956
+ "epoch": 1.32,
1957
+ "learning_rate": 5.501135073017007e-07,
1958
+ "loss": 1.9494,
1959
+ "step": 650
1960
+ },
1961
+ {
1962
+ "epoch": 1.32,
1963
+ "learning_rate": 5.442535766257524e-07,
1964
+ "loss": 1.9811,
1965
+ "step": 652
1966
+ },
1967
+ {
1968
+ "epoch": 1.33,
1969
+ "learning_rate": 5.384133323204665e-07,
1970
+ "loss": 2.0013,
1971
+ "step": 654
1972
+ },
1973
+ {
1974
+ "epoch": 1.33,
1975
+ "learning_rate": 5.325930266603723e-07,
1976
+ "loss": 1.993,
1977
+ "step": 656
1978
+ },
1979
+ {
1980
+ "epoch": 1.33,
1981
+ "learning_rate": 5.267929110587307e-07,
1982
+ "loss": 2.0253,
1983
+ "step": 658
1984
+ },
1985
+ {
1986
+ "epoch": 1.34,
1987
+ "learning_rate": 5.210132360566754e-07,
1988
+ "loss": 1.9134,
1989
+ "step": 660
1990
+ },
1991
+ {
1992
+ "epoch": 1.34,
1993
+ "learning_rate": 5.152542513123905e-07,
1994
+ "loss": 1.9898,
1995
+ "step": 662
1996
+ },
1997
+ {
1998
+ "epoch": 1.35,
1999
+ "learning_rate": 5.095162055903257e-07,
2000
+ "loss": 2.0297,
2001
+ "step": 664
2002
+ },
2003
+ {
2004
+ "epoch": 1.35,
2005
+ "learning_rate": 5.037993467504514e-07,
2006
+ "loss": 1.9936,
2007
+ "step": 666
2008
+ },
2009
+ {
2010
+ "epoch": 1.35,
2011
+ "learning_rate": 4.981039217375519e-07,
2012
+ "loss": 1.9283,
2013
+ "step": 668
2014
+ },
2015
+ {
2016
+ "epoch": 1.36,
2017
+ "learning_rate": 4.924301765705579e-07,
2018
+ "loss": 2.0013,
2019
+ "step": 670
2020
+ },
2021
+ {
2022
+ "epoch": 1.36,
2023
+ "learning_rate": 4.867783563319205e-07,
2024
+ "loss": 2.0661,
2025
+ "step": 672
2026
+ },
2027
+ {
2028
+ "epoch": 1.37,
2029
+ "learning_rate": 4.811487051570234e-07,
2030
+ "loss": 1.9584,
2031
+ "step": 674
2032
+ },
2033
+ {
2034
+ "epoch": 1.37,
2035
+ "learning_rate": 4.755414662236391e-07,
2036
+ "loss": 1.9999,
2037
+ "step": 676
2038
+ },
2039
+ {
2040
+ "epoch": 1.37,
2041
+ "learning_rate": 4.6995688174142236e-07,
2042
+ "loss": 1.9383,
2043
+ "step": 678
2044
+ },
2045
+ {
2046
+ "epoch": 1.38,
2047
+ "learning_rate": 4.6439519294144925e-07,
2048
+ "loss": 2.0078,
2049
+ "step": 680
2050
+ },
2051
+ {
2052
+ "epoch": 1.38,
2053
+ "learning_rate": 4.5885664006579636e-07,
2054
+ "loss": 1.9662,
2055
+ "step": 682
2056
+ },
2057
+ {
2058
+ "epoch": 1.39,
2059
+ "learning_rate": 4.5334146235716364e-07,
2060
+ "loss": 2.0411,
2061
+ "step": 684
2062
+ },
2063
+ {
2064
+ "epoch": 1.39,
2065
+ "learning_rate": 4.4784989804854047e-07,
2066
+ "loss": 1.9178,
2067
+ "step": 686
2068
+ },
2069
+ {
2070
+ "epoch": 1.39,
2071
+ "learning_rate": 4.4238218435291385e-07,
2072
+ "loss": 1.971,
2073
+ "step": 688
2074
+ },
2075
+ {
2076
+ "epoch": 1.4,
2077
+ "learning_rate": 4.3693855745302265e-07,
2078
+ "loss": 1.9665,
2079
+ "step": 690
2080
+ },
2081
+ {
2082
+ "epoch": 1.4,
2083
+ "learning_rate": 4.3151925249115505e-07,
2084
+ "loss": 1.9915,
2085
+ "step": 692
2086
+ },
2087
+ {
2088
+ "epoch": 1.41,
2089
+ "learning_rate": 4.2612450355899165e-07,
2090
+ "loss": 2.0694,
2091
+ "step": 694
2092
+ },
2093
+ {
2094
+ "epoch": 1.41,
2095
+ "learning_rate": 4.207545436874941e-07,
2096
+ "loss": 1.9952,
2097
+ "step": 696
2098
+ },
2099
+ {
2100
+ "epoch": 1.41,
2101
+ "learning_rate": 4.1540960483683793e-07,
2102
+ "loss": 1.9204,
2103
+ "step": 698
2104
+ },
2105
+ {
2106
+ "epoch": 1.42,
2107
+ "learning_rate": 4.100899178863938e-07,
2108
+ "loss": 2.0088,
2109
+ "step": 700
2110
+ },
2111
+ {
2112
+ "epoch": 1.42,
2113
+ "learning_rate": 4.047957126247541e-07,
2114
+ "loss": 1.9492,
2115
+ "step": 702
2116
+ },
2117
+ {
2118
+ "epoch": 1.43,
2119
+ "learning_rate": 3.9952721773980694e-07,
2120
+ "loss": 1.9639,
2121
+ "step": 704
2122
+ },
2123
+ {
2124
+ "epoch": 1.43,
2125
+ "learning_rate": 3.9428466080885825e-07,
2126
+ "loss": 2.045,
2127
+ "step": 706
2128
+ },
2129
+ {
2130
+ "epoch": 1.44,
2131
+ "learning_rate": 3.890682682888008e-07,
2132
+ "loss": 1.9398,
2133
+ "step": 708
2134
+ },
2135
+ {
2136
+ "epoch": 1.44,
2137
+ "learning_rate": 3.838782655063325e-07,
2138
+ "loss": 1.9746,
2139
+ "step": 710
2140
+ },
2141
+ {
2142
+ "epoch": 1.44,
2143
+ "learning_rate": 3.7871487664822324e-07,
2144
+ "loss": 1.9819,
2145
+ "step": 712
2146
+ },
2147
+ {
2148
+ "epoch": 1.45,
2149
+ "learning_rate": 3.735783247516304e-07,
2150
+ "loss": 1.9442,
2151
+ "step": 714
2152
+ },
2153
+ {
2154
+ "epoch": 1.45,
2155
+ "learning_rate": 3.684688316944662e-07,
2156
+ "loss": 2.0245,
2157
+ "step": 716
2158
+ },
2159
+ {
2160
+ "epoch": 1.46,
2161
+ "learning_rate": 3.6338661818581096e-07,
2162
+ "loss": 1.9864,
2163
+ "step": 718
2164
+ },
2165
+ {
2166
+ "epoch": 1.46,
2167
+ "learning_rate": 3.5833190375638155e-07,
2168
+ "loss": 1.9861,
2169
+ "step": 720
2170
+ },
2171
+ {
2172
+ "epoch": 1.46,
2173
+ "learning_rate": 3.5330490674904736e-07,
2174
+ "loss": 1.9732,
2175
+ "step": 722
2176
+ },
2177
+ {
2178
+ "epoch": 1.47,
2179
+ "learning_rate": 3.483058443093989e-07,
2180
+ "loss": 2.0337,
2181
+ "step": 724
2182
+ },
2183
+ {
2184
+ "epoch": 1.47,
2185
+ "learning_rate": 3.4333493237636955e-07,
2186
+ "loss": 1.9916,
2187
+ "step": 726
2188
+ },
2189
+ {
2190
+ "epoch": 1.48,
2191
+ "learning_rate": 3.383923856729052e-07,
2192
+ "loss": 1.9695,
2193
+ "step": 728
2194
+ },
2195
+ {
2196
+ "epoch": 1.48,
2197
+ "learning_rate": 3.334784176966912e-07,
2198
+ "loss": 1.9782,
2199
+ "step": 730
2200
+ },
2201
+ {
2202
+ "epoch": 1.48,
2203
+ "learning_rate": 3.28593240710929e-07,
2204
+ "loss": 1.9212,
2205
+ "step": 732
2206
+ },
2207
+ {
2208
+ "epoch": 1.49,
2209
+ "learning_rate": 3.237370657351679e-07,
2210
+ "loss": 1.9699,
2211
+ "step": 734
2212
+ },
2213
+ {
2214
+ "epoch": 1.49,
2215
+ "learning_rate": 3.189101025361904e-07,
2216
+ "loss": 2.0059,
2217
+ "step": 736
2218
+ },
2219
+ {
2220
+ "epoch": 1.5,
2221
+ "learning_rate": 3.1411255961894933e-07,
2222
+ "loss": 2.01,
2223
+ "step": 738
2224
+ },
2225
+ {
2226
+ "epoch": 1.5,
2227
+ "learning_rate": 3.09344644217563e-07,
2228
+ "loss": 2.0103,
2229
+ "step": 740
2230
+ },
2231
+ {
2232
+ "epoch": 1.5,
2233
+ "learning_rate": 3.046065622863625e-07,
2234
+ "loss": 2.0671,
2235
+ "step": 742
2236
+ },
2237
+ {
2238
+ "epoch": 1.51,
2239
+ "learning_rate": 2.998985184909959e-07,
2240
+ "loss": 1.9484,
2241
+ "step": 744
2242
+ },
2243
+ {
2244
+ "epoch": 1.51,
2245
+ "learning_rate": 2.952207161995879e-07,
2246
+ "loss": 1.9604,
2247
+ "step": 746
2248
+ },
2249
+ {
2250
+ "epoch": 1.52,
2251
+ "learning_rate": 2.9057335747395415e-07,
2252
+ "loss": 2.0014,
2253
+ "step": 748
2254
+ },
2255
+ {
2256
+ "epoch": 1.52,
2257
+ "learning_rate": 2.859566430608731e-07,
2258
+ "loss": 2.0554,
2259
+ "step": 750
2260
+ },
2261
+ {
2262
+ "epoch": 1.52,
2263
+ "learning_rate": 2.813707723834152e-07,
2264
+ "loss": 2.0169,
2265
+ "step": 752
2266
+ },
2267
+ {
2268
+ "epoch": 1.53,
2269
+ "learning_rate": 2.768159435323293e-07,
2270
+ "loss": 1.9582,
2271
+ "step": 754
2272
+ },
2273
+ {
2274
+ "epoch": 1.53,
2275
+ "learning_rate": 2.722923532574839e-07,
2276
+ "loss": 2.0139,
2277
+ "step": 756
2278
+ },
2279
+ {
2280
+ "epoch": 1.54,
2281
+ "learning_rate": 2.678001969593701e-07,
2282
+ "loss": 1.9763,
2283
+ "step": 758
2284
+ },
2285
+ {
2286
+ "epoch": 1.54,
2287
+ "learning_rate": 2.6333966868066036e-07,
2288
+ "loss": 1.9817,
2289
+ "step": 760
2290
+ },
2291
+ {
2292
+ "epoch": 1.54,
2293
+ "learning_rate": 2.589109610978264e-07,
2294
+ "loss": 1.9206,
2295
+ "step": 762
2296
+ },
2297
+ {
2298
+ "epoch": 1.55,
2299
+ "learning_rate": 2.5451426551281797e-07,
2300
+ "loss": 2.0391,
2301
+ "step": 764
2302
+ },
2303
+ {
2304
+ "epoch": 1.55,
2305
+ "learning_rate": 2.501497718447969e-07,
2306
+ "loss": 1.9825,
2307
+ "step": 766
2308
+ },
2309
+ {
2310
+ "epoch": 1.56,
2311
+ "learning_rate": 2.4581766862193555e-07,
2312
+ "loss": 1.9787,
2313
+ "step": 768
2314
+ },
2315
+ {
2316
+ "epoch": 1.56,
2317
+ "learning_rate": 2.4151814297327155e-07,
2318
+ "loss": 1.9006,
2319
+ "step": 770
2320
+ },
2321
+ {
2322
+ "epoch": 1.56,
2323
+ "learning_rate": 2.3725138062062577e-07,
2324
+ "loss": 1.9489,
2325
+ "step": 772
2326
+ },
2327
+ {
2328
+ "epoch": 1.57,
2329
+ "learning_rate": 2.3301756587057987e-07,
2330
+ "loss": 2.0418,
2331
+ "step": 774
2332
+ },
2333
+ {
2334
+ "epoch": 1.57,
2335
+ "learning_rate": 2.2881688160651402e-07,
2336
+ "loss": 1.9554,
2337
+ "step": 776
2338
+ },
2339
+ {
2340
+ "epoch": 1.58,
2341
+ "learning_rate": 2.246495092807077e-07,
2342
+ "loss": 1.9084,
2343
+ "step": 778
2344
+ },
2345
+ {
2346
+ "epoch": 1.58,
2347
+ "learning_rate": 2.20515628906502e-07,
2348
+ "loss": 2.0168,
2349
+ "step": 780
2350
+ },
2351
+ {
2352
+ "epoch": 1.59,
2353
+ "learning_rate": 2.1641541905052306e-07,
2354
+ "loss": 2.0983,
2355
+ "step": 782
2356
+ },
2357
+ {
2358
+ "epoch": 1.59,
2359
+ "learning_rate": 2.1234905682496984e-07,
2360
+ "loss": 1.9285,
2361
+ "step": 784
2362
+ },
2363
+ {
2364
+ "epoch": 1.59,
2365
+ "learning_rate": 2.083167178799623e-07,
2366
+ "loss": 2.0109,
2367
+ "step": 786
2368
+ },
2369
+ {
2370
+ "epoch": 1.6,
2371
+ "learning_rate": 2.0431857639595486e-07,
2372
+ "loss": 1.9614,
2373
+ "step": 788
2374
+ },
2375
+ {
2376
+ "epoch": 1.6,
2377
+ "learning_rate": 2.0035480507621216e-07,
2378
+ "loss": 1.9873,
2379
+ "step": 790
2380
+ },
2381
+ {
2382
+ "epoch": 1.61,
2383
+ "learning_rate": 1.9642557513934932e-07,
2384
+ "loss": 1.9627,
2385
+ "step": 792
2386
+ },
2387
+ {
2388
+ "epoch": 1.61,
2389
+ "learning_rate": 1.9253105631193577e-07,
2390
+ "loss": 2.0673,
2391
+ "step": 794
2392
+ },
2393
+ {
2394
+ "epoch": 1.61,
2395
+ "learning_rate": 1.8867141682116372e-07,
2396
+ "loss": 2.0468,
2397
+ "step": 796
2398
+ },
2399
+ {
2400
+ "epoch": 1.62,
2401
+ "learning_rate": 1.848468233875815e-07,
2402
+ "loss": 1.9262,
2403
+ "step": 798
2404
+ },
2405
+ {
2406
+ "epoch": 1.62,
2407
+ "learning_rate": 1.8105744121789223e-07,
2408
+ "loss": 1.9323,
2409
+ "step": 800
2410
+ },
2411
+ {
2412
+ "epoch": 1.63,
2413
+ "learning_rate": 1.7730343399781666e-07,
2414
+ "loss": 2.0252,
2415
+ "step": 802
2416
+ },
2417
+ {
2418
+ "epoch": 1.63,
2419
+ "learning_rate": 1.735849638850242e-07,
2420
+ "loss": 1.9096,
2421
+ "step": 804
2422
+ },
2423
+ {
2424
+ "epoch": 1.63,
2425
+ "learning_rate": 1.699021915021266e-07,
2426
+ "loss": 1.991,
2427
+ "step": 806
2428
+ },
2429
+ {
2430
+ "epoch": 1.64,
2431
+ "learning_rate": 1.6625527592974076e-07,
2432
+ "loss": 1.9263,
2433
+ "step": 808
2434
+ },
2435
+ {
2436
+ "epoch": 1.64,
2437
+ "learning_rate": 1.62644374699617e-07,
2438
+ "loss": 1.9394,
2439
+ "step": 810
2440
+ },
2441
+ {
2442
+ "epoch": 1.65,
2443
+ "learning_rate": 1.5906964378783372e-07,
2444
+ "loss": 2.0109,
2445
+ "step": 812
2446
+ },
2447
+ {
2448
+ "epoch": 1.65,
2449
+ "learning_rate": 1.5553123760806143e-07,
2450
+ "loss": 1.9834,
2451
+ "step": 814
2452
+ },
2453
+ {
2454
+ "epoch": 1.65,
2455
+ "learning_rate": 1.5202930900489053e-07,
2456
+ "loss": 1.9854,
2457
+ "step": 816
2458
+ },
2459
+ {
2460
+ "epoch": 1.66,
2461
+ "learning_rate": 1.4856400924723077e-07,
2462
+ "loss": 1.9498,
2463
+ "step": 818
2464
+ },
2465
+ {
2466
+ "epoch": 1.66,
2467
+ "learning_rate": 1.4513548802177632e-07,
2468
+ "loss": 1.9421,
2469
+ "step": 820
2470
+ },
2471
+ {
2472
+ "epoch": 1.67,
2473
+ "learning_rate": 1.417438934265397e-07,
2474
+ "loss": 1.9429,
2475
+ "step": 822
2476
+ },
2477
+ {
2478
+ "epoch": 1.67,
2479
+ "learning_rate": 1.3838937196445578e-07,
2480
+ "loss": 1.9836,
2481
+ "step": 824
2482
+ },
2483
+ {
2484
+ "epoch": 1.67,
2485
+ "learning_rate": 1.3507206853705178e-07,
2486
+ "loss": 1.9655,
2487
+ "step": 826
2488
+ },
2489
+ {
2490
+ "epoch": 1.68,
2491
+ "learning_rate": 1.3179212643818926e-07,
2492
+ "loss": 1.9888,
2493
+ "step": 828
2494
+ },
2495
+ {
2496
+ "epoch": 1.68,
2497
+ "learning_rate": 1.2854968734787397e-07,
2498
+ "loss": 1.9259,
2499
+ "step": 830
2500
+ },
2501
+ {
2502
+ "epoch": 1.69,
2503
+ "learning_rate": 1.2534489132613602e-07,
2504
+ "loss": 1.981,
2505
+ "step": 832
2506
+ },
2507
+ {
2508
+ "epoch": 1.69,
2509
+ "learning_rate": 1.2217787680697988e-07,
2510
+ "loss": 1.9655,
2511
+ "step": 834
2512
+ },
2513
+ {
2514
+ "epoch": 1.69,
2515
+ "learning_rate": 1.1904878059240442e-07,
2516
+ "loss": 1.9276,
2517
+ "step": 836
2518
+ },
2519
+ {
2520
+ "epoch": 1.7,
2521
+ "learning_rate": 1.1595773784649387e-07,
2522
+ "loss": 1.9787,
2523
+ "step": 838
2524
+ },
2525
+ {
2526
+ "epoch": 1.7,
2527
+ "learning_rate": 1.1290488208957894e-07,
2528
+ "loss": 2.0179,
2529
+ "step": 840
2530
+ },
2531
+ {
2532
+ "epoch": 1.71,
2533
+ "learning_rate": 1.0989034519246953e-07,
2534
+ "loss": 2.1181,
2535
+ "step": 842
2536
+ },
2537
+ {
2538
+ "epoch": 1.71,
2539
+ "learning_rate": 1.0691425737075899e-07,
2540
+ "loss": 2.0391,
2541
+ "step": 844
2542
+ },
2543
+ {
2544
+ "epoch": 1.71,
2545
+ "learning_rate": 1.0397674717919803e-07,
2546
+ "loss": 1.9924,
2547
+ "step": 846
2548
+ },
2549
+ {
2550
+ "epoch": 1.72,
2551
+ "learning_rate": 1.0107794150614279e-07,
2552
+ "loss": 2.0327,
2553
+ "step": 848
2554
+ },
2555
+ {
2556
+ "epoch": 1.72,
2557
+ "learning_rate": 9.821796556807338e-08,
2558
+ "loss": 2.0083,
2559
+ "step": 850
2560
+ },
2561
+ {
2562
+ "epoch": 1.73,
2563
+ "learning_rate": 9.539694290418487e-08,
2564
+ "loss": 1.9819,
2565
+ "step": 852
2566
+ },
2567
+ {
2568
+ "epoch": 1.73,
2569
+ "learning_rate": 9.261499537105178e-08,
2570
+ "loss": 1.9439,
2571
+ "step": 854
2572
+ },
2573
+ {
2574
+ "epoch": 1.73,
2575
+ "learning_rate": 8.987224313736308e-08,
2576
+ "loss": 1.9482,
2577
+ "step": 856
2578
+ },
2579
+ {
2580
+ "epoch": 1.74,
2581
+ "learning_rate": 8.716880467873233e-08,
2582
+ "loss": 2.0,
2583
+ "step": 858
2584
+ },
2585
+ {
2586
+ "epoch": 1.74,
2587
+ "learning_rate": 8.45047967725796e-08,
2588
+ "loss": 1.8927,
2589
+ "step": 860
2590
+ },
2591
+ {
2592
+ "epoch": 1.75,
2593
+ "learning_rate": 8.188033449308718e-08,
2594
+ "loss": 1.9897,
2595
+ "step": 862
2596
+ },
2597
+ {
2598
+ "epoch": 1.75,
2599
+ "learning_rate": 7.929553120622967e-08,
2600
+ "loss": 1.9354,
2601
+ "step": 864
2602
+ },
2603
+ {
2604
+ "epoch": 1.76,
2605
+ "learning_rate": 7.675049856487548e-08,
2606
+ "loss": 2.0003,
2607
+ "step": 866
2608
+ },
2609
+ {
2610
+ "epoch": 1.76,
2611
+ "learning_rate": 7.424534650396519e-08,
2612
+ "loss": 2.0627,
2613
+ "step": 868
2614
+ },
2615
+ {
2616
+ "epoch": 1.76,
2617
+ "learning_rate": 7.178018323576207e-08,
2618
+ "loss": 2.0044,
2619
+ "step": 870
2620
+ },
2621
+ {
2622
+ "epoch": 1.77,
2623
+ "learning_rate": 6.935511524517834e-08,
2624
+ "loss": 1.8565,
2625
+ "step": 872
2626
+ },
2627
+ {
2628
+ "epoch": 1.77,
2629
+ "learning_rate": 6.697024728517531e-08,
2630
+ "loss": 1.917,
2631
+ "step": 874
2632
+ },
2633
+ {
2634
+ "epoch": 1.78,
2635
+ "learning_rate": 6.462568237223786e-08,
2636
+ "loss": 1.9613,
2637
+ "step": 876
2638
+ },
2639
+ {
2640
+ "epoch": 1.78,
2641
+ "learning_rate": 6.232152178192529e-08,
2642
+ "loss": 1.9487,
2643
+ "step": 878
2644
+ },
2645
+ {
2646
+ "epoch": 1.78,
2647
+ "learning_rate": 6.00578650444965e-08,
2648
+ "loss": 1.9888,
2649
+ "step": 880
2650
+ },
2651
+ {
2652
+ "epoch": 1.79,
2653
+ "learning_rate": 5.7834809940610186e-08,
2654
+ "loss": 1.9773,
2655
+ "step": 882
2656
+ },
2657
+ {
2658
+ "epoch": 1.79,
2659
+ "learning_rate": 5.565245249710193e-08,
2660
+ "loss": 1.9982,
2661
+ "step": 884
2662
+ },
2663
+ {
2664
+ "epoch": 1.8,
2665
+ "learning_rate": 5.351088698283557e-08,
2666
+ "loss": 1.9104,
2667
+ "step": 886
2668
+ },
2669
+ {
2670
+ "epoch": 1.8,
2671
+ "learning_rate": 5.141020590463141e-08,
2672
+ "loss": 1.9952,
2673
+ "step": 888
2674
+ },
2675
+ {
2676
+ "epoch": 1.8,
2677
+ "learning_rate": 4.9350500003270456e-08,
2678
+ "loss": 1.9558,
2679
+ "step": 890
2680
+ },
2681
+ {
2682
+ "epoch": 1.81,
2683
+ "learning_rate": 4.73318582495742e-08,
2684
+ "loss": 1.9665,
2685
+ "step": 892
2686
+ },
2687
+ {
2688
+ "epoch": 1.81,
2689
+ "learning_rate": 4.535436784056268e-08,
2690
+ "loss": 1.9709,
2691
+ "step": 894
2692
+ },
2693
+ {
2694
+ "epoch": 1.82,
2695
+ "learning_rate": 4.341811419568653e-08,
2696
+ "loss": 1.9914,
2697
+ "step": 896
2698
+ },
2699
+ {
2700
+ "epoch": 1.82,
2701
+ "learning_rate": 4.1523180953137785e-08,
2702
+ "loss": 2.0053,
2703
+ "step": 898
2704
+ },
2705
+ {
2706
+ "epoch": 1.82,
2707
+ "learning_rate": 3.966964996623734e-08,
2708
+ "loss": 1.9524,
2709
+ "step": 900
2710
+ },
2711
+ {
2712
+ "epoch": 1.83,
2713
+ "learning_rate": 3.785760129989868e-08,
2714
+ "loss": 1.9811,
2715
+ "step": 902
2716
+ },
2717
+ {
2718
+ "epoch": 1.83,
2719
+ "learning_rate": 3.608711322717028e-08,
2720
+ "loss": 2.0287,
2721
+ "step": 904
2722
+ },
2723
+ {
2724
+ "epoch": 1.84,
2725
+ "learning_rate": 3.435826222585325e-08,
2726
+ "loss": 1.9743,
2727
+ "step": 906
2728
+ },
2729
+ {
2730
+ "epoch": 1.84,
2731
+ "learning_rate": 3.267112297519881e-08,
2732
+ "loss": 1.9485,
2733
+ "step": 908
2734
+ },
2735
+ {
2736
+ "epoch": 1.84,
2737
+ "learning_rate": 3.1025768352682114e-08,
2738
+ "loss": 1.9946,
2739
+ "step": 910
2740
+ },
2741
+ {
2742
+ "epoch": 1.85,
2743
+ "learning_rate": 2.942226943085424e-08,
2744
+ "loss": 1.982,
2745
+ "step": 912
2746
+ },
2747
+ {
2748
+ "epoch": 1.85,
2749
+ "learning_rate": 2.7860695474272388e-08,
2750
+ "loss": 1.9575,
2751
+ "step": 914
2752
+ },
2753
+ {
2754
+ "epoch": 1.86,
2755
+ "learning_rate": 2.634111393650751e-08,
2756
+ "loss": 1.9026,
2757
+ "step": 916
2758
+ },
2759
+ {
2760
+ "epoch": 1.86,
2761
+ "learning_rate": 2.4863590457230743e-08,
2762
+ "loss": 1.9581,
2763
+ "step": 918
2764
+ },
2765
+ {
2766
+ "epoch": 1.86,
2767
+ "learning_rate": 2.342818885937825e-08,
2768
+ "loss": 2.0418,
2769
+ "step": 920
2770
+ },
2771
+ {
2772
+ "epoch": 1.87,
2773
+ "learning_rate": 2.2034971146394298e-08,
2774
+ "loss": 2.0184,
2775
+ "step": 922
2776
+ },
2777
+ {
2778
+ "epoch": 1.87,
2779
+ "learning_rate": 2.068399749955263e-08,
2780
+ "loss": 1.9043,
2781
+ "step": 924
2782
+ },
2783
+ {
2784
+ "epoch": 1.88,
2785
+ "learning_rate": 1.9375326275357207e-08,
2786
+ "loss": 1.9647,
2787
+ "step": 926
2788
+ },
2789
+ {
2790
+ "epoch": 1.88,
2791
+ "learning_rate": 1.810901400302145e-08,
2792
+ "loss": 1.9884,
2793
+ "step": 928
2794
+ },
2795
+ {
2796
+ "epoch": 1.88,
2797
+ "learning_rate": 1.6885115382026082e-08,
2798
+ "loss": 1.9841,
2799
+ "step": 930
2800
+ },
2801
+ {
2802
+ "epoch": 1.89,
2803
+ "learning_rate": 1.5703683279756795e-08,
2804
+ "loss": 1.9336,
2805
+ "step": 932
2806
+ },
2807
+ {
2808
+ "epoch": 1.89,
2809
+ "learning_rate": 1.456476872922041e-08,
2810
+ "loss": 1.9851,
2811
+ "step": 934
2812
+ },
2813
+ {
2814
+ "epoch": 1.9,
2815
+ "learning_rate": 1.3468420926840196e-08,
2816
+ "loss": 1.9871,
2817
+ "step": 936
2818
+ },
2819
+ {
2820
+ "epoch": 1.9,
2821
+ "learning_rate": 1.2414687230331122e-08,
2822
+ "loss": 1.9545,
2823
+ "step": 938
2824
+ },
2825
+ {
2826
+ "epoch": 1.91,
2827
+ "learning_rate": 1.1403613156654057e-08,
2828
+ "loss": 1.969,
2829
+ "step": 940
2830
+ },
2831
+ {
2832
+ "epoch": 1.91,
2833
+ "learning_rate": 1.0435242380049558e-08,
2834
+ "loss": 1.9843,
2835
+ "step": 942
2836
+ },
2837
+ {
2838
+ "epoch": 1.91,
2839
+ "learning_rate": 9.509616730151826e-09,
2840
+ "loss": 1.927,
2841
+ "step": 944
2842
+ },
2843
+ {
2844
+ "epoch": 1.92,
2845
+ "learning_rate": 8.62677619018104e-09,
2846
+ "loss": 1.976,
2847
+ "step": 946
2848
+ },
2849
+ {
2850
+ "epoch": 1.92,
2851
+ "learning_rate": 7.786758895216628e-09,
2852
+ "loss": 2.03,
2853
+ "step": 948
2854
+ },
2855
+ {
2856
+ "epoch": 1.93,
2857
+ "learning_rate": 6.989601130550249e-09,
2858
+ "loss": 1.9603,
2859
+ "step": 950
2860
+ },
2861
+ {
2862
+ "epoch": 1.93,
2863
+ "learning_rate": 6.235337330117829e-09,
2864
+ "loss": 1.9866,
2865
+ "step": 952
2866
+ },
2867
+ {
2868
+ "epoch": 1.93,
2869
+ "learning_rate": 5.524000075012969e-09,
2870
+ "loss": 1.9754,
2871
+ "step": 954
2872
+ },
2873
+ {
2874
+ "epoch": 1.94,
2875
+ "learning_rate": 4.8556200920786276e-09,
2876
+ "loss": 1.94,
2877
+ "step": 956
2878
+ },
2879
+ {
2880
+ "epoch": 1.94,
2881
+ "learning_rate": 4.2302262525805155e-09,
2882
+ "loss": 1.9412,
2883
+ "step": 958
2884
+ },
2885
+ {
2886
+ "epoch": 1.95,
2887
+ "learning_rate": 3.647845570959873e-09,
2888
+ "loss": 2.0471,
2889
+ "step": 960
2890
+ },
2891
+ {
2892
+ "epoch": 1.95,
2893
+ "learning_rate": 3.1085032036664015e-09,
2894
+ "loss": 1.9434,
2895
+ "step": 962
2896
+ },
2897
+ {
2898
+ "epoch": 1.95,
2899
+ "learning_rate": 2.612222448071577e-09,
2900
+ "loss": 2.0134,
2901
+ "step": 964
2902
+ },
2903
+ {
2904
+ "epoch": 1.96,
2905
+ "learning_rate": 2.1590247414624564e-09,
2906
+ "loss": 2.006,
2907
+ "step": 966
2908
+ },
2909
+ {
2910
+ "epoch": 1.96,
2911
+ "learning_rate": 1.7489296601156388e-09,
2912
+ "loss": 1.9638,
2913
+ "step": 968
2914
+ },
2915
+ {
2916
+ "epoch": 1.97,
2917
+ "learning_rate": 1.3819549184516111e-09,
2918
+ "loss": 2.0321,
2919
+ "step": 970
2920
+ },
2921
+ {
2922
+ "epoch": 1.97,
2923
+ "learning_rate": 1.0581163682695793e-09,
2924
+ "loss": 1.9233,
2925
+ "step": 972
2926
+ },
2927
+ {
2928
+ "epoch": 1.97,
2929
+ "learning_rate": 7.774279980626853e-10,
2930
+ "loss": 2.0179,
2931
+ "step": 974
2932
+ },
2933
+ {
2934
+ "epoch": 1.98,
2935
+ "learning_rate": 5.399019324139331e-10,
2936
+ "loss": 1.9882,
2937
+ "step": 976
2938
+ },
2939
+ {
2940
+ "epoch": 1.98,
2941
+ "learning_rate": 3.455484314721646e-10,
2942
+ "loss": 1.9149,
2943
+ "step": 978
2944
+ },
2945
+ {
2946
+ "epoch": 1.99,
2947
+ "learning_rate": 1.9437589050907978e-10,
2948
+ "loss": 1.9477,
2949
+ "step": 980
2950
+ },
2951
+ {
2952
+ "epoch": 1.99,
2953
+ "learning_rate": 8.639083955663818e-11,
2954
+ "loss": 1.9924,
2955
+ "step": 982
2956
+ },
2957
+ {
2958
+ "epoch": 1.99,
2959
+ "learning_rate": 2.1597943124729288e-11,
2960
+ "loss": 1.9547,
2961
+ "step": 984
2962
+ },
2963
+ {
2964
+ "epoch": 2.0,
2965
+ "learning_rate": 0.0,
2966
+ "loss": 1.9569,
2967
+ "step": 986
2968
+ },
2969
+ {
2970
+ "epoch": 2.0,
2971
+ "step": 986,
2972
+ "total_flos": 4.368889634042675e+16,
2973
+ "train_loss": 2.002830022247036,
2974
+ "train_runtime": 7223.1483,
2975
+ "train_samples_per_second": 8.741,
2976
+ "train_steps_per_second": 0.137
2977
+ }
2978
+ ],
2979
+ "logging_steps": 2,
2980
+ "max_steps": 986,
2981
+ "num_input_tokens_seen": 0,
2982
+ "num_train_epochs": 2,
2983
+ "save_steps": 2000,
2984
+ "total_flos": 4.368889634042675e+16,
2985
+ "train_batch_size": 4,
2986
+ "trial_name": null,
2987
+ "trial_params": null
2988
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e8bbde777b27a63b3c57640676efab350c757ee526fe7d2ced4526896fe67a9
3
+ size 6392
vocab.json ADDED
The diff for this file is too large to render. See raw diff