Jackie999 commited on
Commit
2503b9d
1 Parent(s): 6251a8e

Model save

Browse files
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: gemma
3
+ library_name: peft
4
+ tags:
5
+ - trl
6
+ - sft
7
+ - generated_from_trainer
8
+ base_model: google/gemma-7b
9
+ model-index:
10
+ - name: zephyr-7b-gemma-sft-20p
11
+ results: []
12
+ ---
13
+
14
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
+ should probably proofread and complete it, then remove this comment. -->
16
+
17
+ # zephyr-7b-gemma-sft-20p
18
+
19
+ This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the None dataset.
20
+ It achieves the following results on the evaluation set:
21
+ - Loss: 1.1172
22
+
23
+ ## Model description
24
+
25
+ More information needed
26
+
27
+ ## Intended uses & limitations
28
+
29
+ More information needed
30
+
31
+ ## Training and evaluation data
32
+
33
+ More information needed
34
+
35
+ ## Training procedure
36
+
37
+ ### Training hyperparameters
38
+
39
+ The following hyperparameters were used during training:
40
+ - learning_rate: 0.0002
41
+ - train_batch_size: 8
42
+ - eval_batch_size: 8
43
+ - seed: 42
44
+ - distributed_type: multi-GPU
45
+ - num_devices: 4
46
+ - gradient_accumulation_steps: 2
47
+ - total_train_batch_size: 64
48
+ - total_eval_batch_size: 32
49
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
+ - lr_scheduler_type: cosine
51
+ - lr_scheduler_warmup_ratio: 0.1
52
+ - num_epochs: 3
53
+
54
+ ### Training results
55
+
56
+ | Training Loss | Epoch | Step | Validation Loss |
57
+ |:-------------:|:-----:|:----:|:---------------:|
58
+ | 0.9328 | 1.0 | 337 | 1.1520 |
59
+ | 0.8771 | 2.0 | 675 | 1.1151 |
60
+ | 0.8486 | 3.0 | 1011 | 1.1172 |
61
+
62
+
63
+ ### Framework versions
64
+
65
+ - PEFT 0.7.1
66
+ - Transformers 4.39.0.dev0
67
+ - Pytorch 2.1.2
68
+ - Datasets 2.14.6
69
+ - Tokenizers 0.15.2
adapter_config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "google/gemma-7b",
5
+ "bias": "none",
6
+ "fan_in_fan_out": false,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 8,
13
+ "lora_dropout": 0.05,
14
+ "megatron_config": null,
15
+ "megatron_core": "megatron.core",
16
+ "modules_to_save": null,
17
+ "peft_type": "LORA",
18
+ "r": 6,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "o_proj",
23
+ "v_proj",
24
+ "gate_proj",
25
+ "down_proj",
26
+ "k_proj",
27
+ "q_proj",
28
+ "up_proj"
29
+ ],
30
+ "task_type": "CAUSAL_LM"
31
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:895443b2b3202e37d22d60f6414d071e38f9d45ca5a46008f15bed89812c8fa4
3
+ size 37555048
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "train_loss": 1.2570129054000658,
4
+ "train_runtime": 12020.0319,
5
+ "train_samples": 21594,
6
+ "train_samples_per_second": 5.39,
7
+ "train_steps_per_second": 0.084
8
+ }
runs/Apr21_05-34-01_COE-CS-sv003/events.out.tfevents.1713677659.COE-CS-sv003.350753.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0b82f0e3680276362cfa36a1f931c0083d9c16e6bead8737e9dcd08e31a48db2
3
+ size 4790
runs/Apr21_16-10-25_COE-CS-sv003/events.out.tfevents.1713715838.COE-CS-sv003.361543.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e1e2937f2a8c65338618c38b5787a15a6bdf2313a9584ac1c3c531aa1ba22c64
3
+ size 4790
runs/Apr21_23-09-28_COE-CS-sv003/events.out.tfevents.1713740982.COE-CS-sv003.365685.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6494cfea69387a1394f85d00826ec6b0a79c71c2772a5b6e9b7ed765051e9027
3
+ size 48686
special_tokens_map.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ {
4
+ "content": "<|im_start|>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ {
11
+ "content": "<|im_end|>",
12
+ "lstrip": false,
13
+ "normalized": false,
14
+ "rstrip": false,
15
+ "single_word": false
16
+ }
17
+ ],
18
+ "bos_token": "<|im_start|>",
19
+ "eos_token": "<|im_end|>",
20
+ "pad_token": "<|im_end|>",
21
+ "unk_token": {
22
+ "content": "<unk>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false
27
+ }
28
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:22449cb9ef4bad0db7dd93b46ddff7ab7d6a654dd4f903e130ddb6361eac3af5
3
+ size 17477473
tokenizer_config.json ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_eos_token": false,
4
+ "added_tokens_decoder": {
5
+ "0": {
6
+ "content": "<pad>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "1": {
14
+ "content": "<eos>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "2": {
22
+ "content": "<bos>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "3": {
30
+ "content": "<unk>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "106": {
38
+ "content": "<|im_start|>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "107": {
46
+ "content": "<|im_end|>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ }
53
+ },
54
+ "additional_special_tokens": [
55
+ "<|im_start|>",
56
+ "<|im_end|>"
57
+ ],
58
+ "bos_token": "<|im_start|>",
59
+ "chat_template": "{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
60
+ "clean_up_tokenization_spaces": false,
61
+ "eos_token": "<|im_end|>",
62
+ "legacy": null,
63
+ "model_max_length": 2048,
64
+ "pad_token": "<|im_end|>",
65
+ "sp_model_kwargs": {},
66
+ "spaces_between_special_tokens": false,
67
+ "tokenizer_class": "GemmaTokenizer",
68
+ "unk_token": "<unk>",
69
+ "use_default_system_prompt": false
70
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "train_loss": 1.2570129054000658,
4
+ "train_runtime": 12020.0319,
5
+ "train_samples": 21594,
6
+ "train_samples_per_second": 5.39,
7
+ "train_steps_per_second": 0.084
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1475 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.9955555555555557,
5
+ "eval_steps": 500,
6
+ "global_step": 1011,
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": 72.2578351791039,
14
+ "learning_rate": 1.96078431372549e-06,
15
+ "loss": 16.2543,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.01,
20
+ "grad_norm": 69.14793459091129,
21
+ "learning_rate": 9.803921568627451e-06,
22
+ "loss": 16.1546,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.03,
27
+ "grad_norm": 61.19418424850778,
28
+ "learning_rate": 1.9607843137254903e-05,
29
+ "loss": 15.7851,
30
+ "step": 10
31
+ },
32
+ {
33
+ "epoch": 0.04,
34
+ "grad_norm": 50.95192336423662,
35
+ "learning_rate": 2.9411764705882354e-05,
36
+ "loss": 13.304,
37
+ "step": 15
38
+ },
39
+ {
40
+ "epoch": 0.06,
41
+ "grad_norm": 44.16161556810121,
42
+ "learning_rate": 3.9215686274509805e-05,
43
+ "loss": 10.1286,
44
+ "step": 20
45
+ },
46
+ {
47
+ "epoch": 0.07,
48
+ "grad_norm": 22.013360927764538,
49
+ "learning_rate": 4.901960784313725e-05,
50
+ "loss": 6.5136,
51
+ "step": 25
52
+ },
53
+ {
54
+ "epoch": 0.09,
55
+ "grad_norm": 13.13769527146034,
56
+ "learning_rate": 5.882352941176471e-05,
57
+ "loss": 4.6002,
58
+ "step": 30
59
+ },
60
+ {
61
+ "epoch": 0.1,
62
+ "grad_norm": 9.380927030189634,
63
+ "learning_rate": 6.862745098039216e-05,
64
+ "loss": 3.4775,
65
+ "step": 35
66
+ },
67
+ {
68
+ "epoch": 0.12,
69
+ "grad_norm": 8.047209493969948,
70
+ "learning_rate": 7.843137254901961e-05,
71
+ "loss": 2.5528,
72
+ "step": 40
73
+ },
74
+ {
75
+ "epoch": 0.13,
76
+ "grad_norm": 4.835270040905147,
77
+ "learning_rate": 8.823529411764706e-05,
78
+ "loss": 1.9181,
79
+ "step": 45
80
+ },
81
+ {
82
+ "epoch": 0.15,
83
+ "grad_norm": 4.461126160581573,
84
+ "learning_rate": 9.80392156862745e-05,
85
+ "loss": 1.7189,
86
+ "step": 50
87
+ },
88
+ {
89
+ "epoch": 0.16,
90
+ "grad_norm": 2.4713562833452647,
91
+ "learning_rate": 0.00010784313725490196,
92
+ "loss": 1.4816,
93
+ "step": 55
94
+ },
95
+ {
96
+ "epoch": 0.18,
97
+ "grad_norm": 3.006847832744875,
98
+ "learning_rate": 0.00011764705882352942,
99
+ "loss": 1.3855,
100
+ "step": 60
101
+ },
102
+ {
103
+ "epoch": 0.19,
104
+ "grad_norm": 1.6003996046085147,
105
+ "learning_rate": 0.00012745098039215687,
106
+ "loss": 1.2728,
107
+ "step": 65
108
+ },
109
+ {
110
+ "epoch": 0.21,
111
+ "grad_norm": 1.6221503977277323,
112
+ "learning_rate": 0.0001372549019607843,
113
+ "loss": 1.2497,
114
+ "step": 70
115
+ },
116
+ {
117
+ "epoch": 0.22,
118
+ "grad_norm": 1.584552829167114,
119
+ "learning_rate": 0.00014705882352941178,
120
+ "loss": 1.2432,
121
+ "step": 75
122
+ },
123
+ {
124
+ "epoch": 0.24,
125
+ "grad_norm": 1.5050158166503569,
126
+ "learning_rate": 0.00015686274509803922,
127
+ "loss": 1.1723,
128
+ "step": 80
129
+ },
130
+ {
131
+ "epoch": 0.25,
132
+ "grad_norm": 1.7107243802135035,
133
+ "learning_rate": 0.0001666666666666667,
134
+ "loss": 1.083,
135
+ "step": 85
136
+ },
137
+ {
138
+ "epoch": 0.27,
139
+ "grad_norm": 1.7381735421551534,
140
+ "learning_rate": 0.00017647058823529413,
141
+ "loss": 1.0726,
142
+ "step": 90
143
+ },
144
+ {
145
+ "epoch": 0.28,
146
+ "grad_norm": 1.4981023359065837,
147
+ "learning_rate": 0.00018627450980392157,
148
+ "loss": 1.1057,
149
+ "step": 95
150
+ },
151
+ {
152
+ "epoch": 0.3,
153
+ "grad_norm": 1.7877995266746756,
154
+ "learning_rate": 0.000196078431372549,
155
+ "loss": 1.0783,
156
+ "step": 100
157
+ },
158
+ {
159
+ "epoch": 0.31,
160
+ "grad_norm": 1.5140220512101432,
161
+ "learning_rate": 0.00019999462497359466,
162
+ "loss": 1.1228,
163
+ "step": 105
164
+ },
165
+ {
166
+ "epoch": 0.33,
167
+ "grad_norm": 1.329498665050575,
168
+ "learning_rate": 0.00019996177968249334,
169
+ "loss": 1.0441,
170
+ "step": 110
171
+ },
172
+ {
173
+ "epoch": 0.34,
174
+ "grad_norm": 1.7238891644878542,
175
+ "learning_rate": 0.00019989908502204292,
176
+ "loss": 1.1001,
177
+ "step": 115
178
+ },
179
+ {
180
+ "epoch": 0.36,
181
+ "grad_norm": 1.8266342354047433,
182
+ "learning_rate": 0.00019980655971335945,
183
+ "loss": 1.0601,
184
+ "step": 120
185
+ },
186
+ {
187
+ "epoch": 0.37,
188
+ "grad_norm": 1.6200107092414249,
189
+ "learning_rate": 0.0001996842313852238,
190
+ "loss": 1.0045,
191
+ "step": 125
192
+ },
193
+ {
194
+ "epoch": 0.39,
195
+ "grad_norm": 1.5172675508165814,
196
+ "learning_rate": 0.00019953213656583168,
197
+ "loss": 1.006,
198
+ "step": 130
199
+ },
200
+ {
201
+ "epoch": 0.4,
202
+ "grad_norm": 1.478313589538863,
203
+ "learning_rate": 0.0001993503206718859,
204
+ "loss": 0.9678,
205
+ "step": 135
206
+ },
207
+ {
208
+ "epoch": 0.41,
209
+ "grad_norm": 1.8797845296046498,
210
+ "learning_rate": 0.0001991388379950346,
211
+ "loss": 1.0181,
212
+ "step": 140
213
+ },
214
+ {
215
+ "epoch": 0.43,
216
+ "grad_norm": 1.8037331639091116,
217
+ "learning_rate": 0.00019889775168565943,
218
+ "loss": 1.0096,
219
+ "step": 145
220
+ },
221
+ {
222
+ "epoch": 0.44,
223
+ "grad_norm": 1.4294890809484986,
224
+ "learning_rate": 0.0001986271337340182,
225
+ "loss": 0.9842,
226
+ "step": 150
227
+ },
228
+ {
229
+ "epoch": 0.46,
230
+ "grad_norm": 1.4637597540944496,
231
+ "learning_rate": 0.0001983270649487481,
232
+ "loss": 0.9992,
233
+ "step": 155
234
+ },
235
+ {
236
+ "epoch": 0.47,
237
+ "grad_norm": 1.519557023555869,
238
+ "learning_rate": 0.0001979976349327357,
239
+ "loss": 1.036,
240
+ "step": 160
241
+ },
242
+ {
243
+ "epoch": 0.49,
244
+ "grad_norm": 1.6521810842892943,
245
+ "learning_rate": 0.00019763894205636072,
246
+ "loss": 0.9961,
247
+ "step": 165
248
+ },
249
+ {
250
+ "epoch": 0.5,
251
+ "grad_norm": 1.7888048328412252,
252
+ "learning_rate": 0.0001972510934281218,
253
+ "loss": 0.9796,
254
+ "step": 170
255
+ },
256
+ {
257
+ "epoch": 0.52,
258
+ "grad_norm": 1.4704141321403639,
259
+ "learning_rate": 0.00019683420486265327,
260
+ "loss": 0.9766,
261
+ "step": 175
262
+ },
263
+ {
264
+ "epoch": 0.53,
265
+ "grad_norm": 1.1929169794016272,
266
+ "learning_rate": 0.00019638840084614182,
267
+ "loss": 1.0146,
268
+ "step": 180
269
+ },
270
+ {
271
+ "epoch": 0.55,
272
+ "grad_norm": 1.1785071553887134,
273
+ "learning_rate": 0.00019591381449915397,
274
+ "loss": 1.0056,
275
+ "step": 185
276
+ },
277
+ {
278
+ "epoch": 0.56,
279
+ "grad_norm": 1.2250942438326087,
280
+ "learning_rate": 0.00019541058753688538,
281
+ "loss": 0.9809,
282
+ "step": 190
283
+ },
284
+ {
285
+ "epoch": 0.58,
286
+ "grad_norm": 1.4932596567461078,
287
+ "learning_rate": 0.00019487887022684336,
288
+ "loss": 0.9827,
289
+ "step": 195
290
+ },
291
+ {
292
+ "epoch": 0.59,
293
+ "grad_norm": 1.2885727190352654,
294
+ "learning_rate": 0.00019431882134397598,
295
+ "loss": 1.0027,
296
+ "step": 200
297
+ },
298
+ {
299
+ "epoch": 0.61,
300
+ "grad_norm": 2.57327759834574,
301
+ "learning_rate": 0.00019373060812326052,
302
+ "loss": 0.9351,
303
+ "step": 205
304
+ },
305
+ {
306
+ "epoch": 0.62,
307
+ "grad_norm": 1.3232509112686246,
308
+ "learning_rate": 0.00019311440620976597,
309
+ "loss": 0.9811,
310
+ "step": 210
311
+ },
312
+ {
313
+ "epoch": 0.64,
314
+ "grad_norm": 1.3160771870305465,
315
+ "learning_rate": 0.0001924703996062038,
316
+ "loss": 0.9925,
317
+ "step": 215
318
+ },
319
+ {
320
+ "epoch": 0.65,
321
+ "grad_norm": 1.2477864155169314,
322
+ "learning_rate": 0.00019179878061798347,
323
+ "loss": 0.9212,
324
+ "step": 220
325
+ },
326
+ {
327
+ "epoch": 0.67,
328
+ "grad_norm": 1.093292440232411,
329
+ "learning_rate": 0.0001910997497957885,
330
+ "loss": 0.9851,
331
+ "step": 225
332
+ },
333
+ {
334
+ "epoch": 0.68,
335
+ "grad_norm": 1.362427572762746,
336
+ "learning_rate": 0.0001903735158756905,
337
+ "loss": 0.9893,
338
+ "step": 230
339
+ },
340
+ {
341
+ "epoch": 0.7,
342
+ "grad_norm": 1.346631495980947,
343
+ "learning_rate": 0.00018962029571681886,
344
+ "loss": 0.945,
345
+ "step": 235
346
+ },
347
+ {
348
+ "epoch": 0.71,
349
+ "grad_norm": 1.090827112192976,
350
+ "learning_rate": 0.0001888403142366049,
351
+ "loss": 1.0224,
352
+ "step": 240
353
+ },
354
+ {
355
+ "epoch": 0.73,
356
+ "grad_norm": 1.6700272825279434,
357
+ "learning_rate": 0.00018803380434362,
358
+ "loss": 0.9305,
359
+ "step": 245
360
+ },
361
+ {
362
+ "epoch": 0.74,
363
+ "grad_norm": 1.0096253684038474,
364
+ "learning_rate": 0.00018720100686802694,
365
+ "loss": 1.0128,
366
+ "step": 250
367
+ },
368
+ {
369
+ "epoch": 0.76,
370
+ "grad_norm": 1.7873981051539485,
371
+ "learning_rate": 0.00018634217048966637,
372
+ "loss": 0.953,
373
+ "step": 255
374
+ },
375
+ {
376
+ "epoch": 0.77,
377
+ "grad_norm": 1.3129933811195174,
378
+ "learning_rate": 0.000185457551663799,
379
+ "loss": 0.9363,
380
+ "step": 260
381
+ },
382
+ {
383
+ "epoch": 0.79,
384
+ "grad_norm": 1.2927375510712122,
385
+ "learning_rate": 0.00018454741454452603,
386
+ "loss": 0.9862,
387
+ "step": 265
388
+ },
389
+ {
390
+ "epoch": 0.8,
391
+ "grad_norm": 1.1686920395458062,
392
+ "learning_rate": 0.00018361203090591071,
393
+ "loss": 0.967,
394
+ "step": 270
395
+ },
396
+ {
397
+ "epoch": 0.81,
398
+ "grad_norm": 1.1863754899949552,
399
+ "learning_rate": 0.00018265168006082437,
400
+ "loss": 0.9056,
401
+ "step": 275
402
+ },
403
+ {
404
+ "epoch": 0.83,
405
+ "grad_norm": 1.1941161746530229,
406
+ "learning_rate": 0.0001816666487775416,
407
+ "loss": 0.9791,
408
+ "step": 280
409
+ },
410
+ {
411
+ "epoch": 0.84,
412
+ "grad_norm": 1.1930577483488491,
413
+ "learning_rate": 0.00018065723119410884,
414
+ "loss": 0.9537,
415
+ "step": 285
416
+ },
417
+ {
418
+ "epoch": 0.86,
419
+ "grad_norm": 1.4589361109870367,
420
+ "learning_rate": 0.00017962372873051252,
421
+ "loss": 0.9954,
422
+ "step": 290
423
+ },
424
+ {
425
+ "epoch": 0.87,
426
+ "grad_norm": 1.3290947156035955,
427
+ "learning_rate": 0.00017856644999867264,
428
+ "loss": 0.922,
429
+ "step": 295
430
+ },
431
+ {
432
+ "epoch": 0.89,
433
+ "grad_norm": 1.2441967467176946,
434
+ "learning_rate": 0.000177485710710289,
435
+ "loss": 0.9639,
436
+ "step": 300
437
+ },
438
+ {
439
+ "epoch": 0.9,
440
+ "grad_norm": 1.1382757452000494,
441
+ "learning_rate": 0.00017638183358256696,
442
+ "loss": 0.9811,
443
+ "step": 305
444
+ },
445
+ {
446
+ "epoch": 0.92,
447
+ "grad_norm": 1.147116712031671,
448
+ "learning_rate": 0.00017525514824185185,
449
+ "loss": 0.9457,
450
+ "step": 310
451
+ },
452
+ {
453
+ "epoch": 0.93,
454
+ "grad_norm": 1.0220320338471351,
455
+ "learning_rate": 0.0001741059911251997,
456
+ "loss": 0.9233,
457
+ "step": 315
458
+ },
459
+ {
460
+ "epoch": 0.95,
461
+ "grad_norm": 1.1472331976712749,
462
+ "learning_rate": 0.00017293470537991463,
463
+ "loss": 0.9756,
464
+ "step": 320
465
+ },
466
+ {
467
+ "epoch": 0.96,
468
+ "grad_norm": 1.2410203245777724,
469
+ "learning_rate": 0.0001717416407610824,
470
+ "loss": 0.943,
471
+ "step": 325
472
+ },
473
+ {
474
+ "epoch": 0.98,
475
+ "grad_norm": 1.3981773401938695,
476
+ "learning_rate": 0.00017052715352713075,
477
+ "loss": 0.9477,
478
+ "step": 330
479
+ },
480
+ {
481
+ "epoch": 0.99,
482
+ "grad_norm": 1.2121344807882255,
483
+ "learning_rate": 0.0001692916063334479,
484
+ "loss": 0.9328,
485
+ "step": 335
486
+ },
487
+ {
488
+ "epoch": 1.0,
489
+ "eval_loss": 1.1519585847854614,
490
+ "eval_runtime": 157.1947,
491
+ "eval_samples_per_second": 14.695,
492
+ "eval_steps_per_second": 0.464,
493
+ "step": 337
494
+ },
495
+ {
496
+ "epoch": 1.01,
497
+ "grad_norm": 1.0908262897233953,
498
+ "learning_rate": 0.00016803536812409075,
499
+ "loss": 0.9661,
500
+ "step": 340
501
+ },
502
+ {
503
+ "epoch": 1.02,
504
+ "grad_norm": 1.7756060681085633,
505
+ "learning_rate": 0.00016675881402161536,
506
+ "loss": 0.889,
507
+ "step": 345
508
+ },
509
+ {
510
+ "epoch": 1.04,
511
+ "grad_norm": 1.1136805577794664,
512
+ "learning_rate": 0.0001654623252150624,
513
+ "loss": 0.9091,
514
+ "step": 350
515
+ },
516
+ {
517
+ "epoch": 1.05,
518
+ "grad_norm": 1.3395157303175598,
519
+ "learning_rate": 0.00016414628884613107,
520
+ "loss": 0.9285,
521
+ "step": 355
522
+ },
523
+ {
524
+ "epoch": 1.07,
525
+ "grad_norm": 1.1278036304088261,
526
+ "learning_rate": 0.0001628110978935756,
527
+ "loss": 0.8679,
528
+ "step": 360
529
+ },
530
+ {
531
+ "epoch": 1.08,
532
+ "grad_norm": 1.1866907922967995,
533
+ "learning_rate": 0.0001614571510558588,
534
+ "loss": 0.9031,
535
+ "step": 365
536
+ },
537
+ {
538
+ "epoch": 1.1,
539
+ "grad_norm": 1.1180348411654952,
540
+ "learning_rate": 0.00016008485263209742,
541
+ "loss": 0.9585,
542
+ "step": 370
543
+ },
544
+ {
545
+ "epoch": 1.11,
546
+ "grad_norm": 1.0617311418169892,
547
+ "learning_rate": 0.0001586946124013354,
548
+ "loss": 0.9212,
549
+ "step": 375
550
+ },
551
+ {
552
+ "epoch": 1.13,
553
+ "grad_norm": 0.9259620753456954,
554
+ "learning_rate": 0.00015728684550018064,
555
+ "loss": 0.8746,
556
+ "step": 380
557
+ },
558
+ {
559
+ "epoch": 1.14,
560
+ "grad_norm": 1.1221480241476138,
561
+ "learning_rate": 0.00015586197229884184,
562
+ "loss": 0.924,
563
+ "step": 385
564
+ },
565
+ {
566
+ "epoch": 1.16,
567
+ "grad_norm": 0.9589792763467693,
568
+ "learning_rate": 0.00015442041827560274,
569
+ "loss": 0.8658,
570
+ "step": 390
571
+ },
572
+ {
573
+ "epoch": 1.17,
574
+ "grad_norm": 1.4212740627448033,
575
+ "learning_rate": 0.00015296261388977108,
576
+ "loss": 0.8592,
577
+ "step": 395
578
+ },
579
+ {
580
+ "epoch": 1.19,
581
+ "grad_norm": 1.0480944734011124,
582
+ "learning_rate": 0.00015148899445313981,
583
+ "loss": 0.9497,
584
+ "step": 400
585
+ },
586
+ {
587
+ "epoch": 1.2,
588
+ "grad_norm": 1.0225393084718664,
589
+ "learning_rate": 0.00015000000000000001,
590
+ "loss": 0.9505,
591
+ "step": 405
592
+ },
593
+ {
594
+ "epoch": 1.21,
595
+ "grad_norm": 0.9776923989427987,
596
+ "learning_rate": 0.00014849607515574276,
597
+ "loss": 0.9189,
598
+ "step": 410
599
+ },
600
+ {
601
+ "epoch": 1.23,
602
+ "grad_norm": 0.9981587820990419,
603
+ "learning_rate": 0.00014697766900409074,
604
+ "loss": 0.9761,
605
+ "step": 415
606
+ },
607
+ {
608
+ "epoch": 1.24,
609
+ "grad_norm": 1.1240954604826308,
610
+ "learning_rate": 0.00014544523495299842,
611
+ "loss": 0.9174,
612
+ "step": 420
613
+ },
614
+ {
615
+ "epoch": 1.26,
616
+ "grad_norm": 1.4387424633384827,
617
+ "learning_rate": 0.00014389923059926062,
618
+ "loss": 0.905,
619
+ "step": 425
620
+ },
621
+ {
622
+ "epoch": 1.27,
623
+ "grad_norm": 1.3190393196712298,
624
+ "learning_rate": 0.00014234011759187083,
625
+ "loss": 0.9378,
626
+ "step": 430
627
+ },
628
+ {
629
+ "epoch": 1.29,
630
+ "grad_norm": 0.9858027932238713,
631
+ "learning_rate": 0.00014076836149416887,
632
+ "loss": 0.9164,
633
+ "step": 435
634
+ },
635
+ {
636
+ "epoch": 1.3,
637
+ "grad_norm": 0.8937095468783902,
638
+ "learning_rate": 0.00013918443164482046,
639
+ "loss": 0.9296,
640
+ "step": 440
641
+ },
642
+ {
643
+ "epoch": 1.32,
644
+ "grad_norm": 1.08163207719712,
645
+ "learning_rate": 0.0001375888010176686,
646
+ "loss": 0.914,
647
+ "step": 445
648
+ },
649
+ {
650
+ "epoch": 1.33,
651
+ "grad_norm": 1.0364463188268707,
652
+ "learning_rate": 0.0001359819460805001,
653
+ "loss": 0.8906,
654
+ "step": 450
655
+ },
656
+ {
657
+ "epoch": 1.35,
658
+ "grad_norm": 0.9266587461121177,
659
+ "learning_rate": 0.00013436434665276865,
660
+ "loss": 0.9395,
661
+ "step": 455
662
+ },
663
+ {
664
+ "epoch": 1.36,
665
+ "grad_norm": 1.0746148135081544,
666
+ "learning_rate": 0.0001327364857623168,
667
+ "loss": 0.9365,
668
+ "step": 460
669
+ },
670
+ {
671
+ "epoch": 1.38,
672
+ "grad_norm": 1.077761213751867,
673
+ "learning_rate": 0.00013109884950114007,
674
+ "loss": 0.9157,
675
+ "step": 465
676
+ },
677
+ {
678
+ "epoch": 1.39,
679
+ "grad_norm": 1.4149324327770756,
680
+ "learning_rate": 0.00012945192688023624,
681
+ "loss": 0.8864,
682
+ "step": 470
683
+ },
684
+ {
685
+ "epoch": 1.41,
686
+ "grad_norm": 1.0160083472695705,
687
+ "learning_rate": 0.00012779620968358273,
688
+ "loss": 0.8757,
689
+ "step": 475
690
+ },
691
+ {
692
+ "epoch": 1.42,
693
+ "grad_norm": 1.1590619362809342,
694
+ "learning_rate": 0.00012613219232128608,
695
+ "loss": 0.8985,
696
+ "step": 480
697
+ },
698
+ {
699
+ "epoch": 1.44,
700
+ "grad_norm": 1.1120230320518538,
701
+ "learning_rate": 0.00012446037168194714,
702
+ "loss": 0.9227,
703
+ "step": 485
704
+ },
705
+ {
706
+ "epoch": 1.45,
707
+ "grad_norm": 1.1834009436730764,
708
+ "learning_rate": 0.0001227812469842864,
709
+ "loss": 0.9204,
710
+ "step": 490
711
+ },
712
+ {
713
+ "epoch": 1.47,
714
+ "grad_norm": 1.0355132778907057,
715
+ "learning_rate": 0.00012109531962807332,
716
+ "loss": 0.889,
717
+ "step": 495
718
+ },
719
+ {
720
+ "epoch": 1.48,
721
+ "grad_norm": 1.0214392183969367,
722
+ "learning_rate": 0.00011940309304440433,
723
+ "loss": 0.8882,
724
+ "step": 500
725
+ },
726
+ {
727
+ "epoch": 1.5,
728
+ "grad_norm": 1.0496957066021488,
729
+ "learning_rate": 0.00011770507254537453,
730
+ "loss": 0.9334,
731
+ "step": 505
732
+ },
733
+ {
734
+ "epoch": 1.51,
735
+ "grad_norm": 1.0238376160697826,
736
+ "learning_rate": 0.00011600176517318741,
737
+ "loss": 0.9181,
738
+ "step": 510
739
+ },
740
+ {
741
+ "epoch": 1.53,
742
+ "grad_norm": 1.2008352860962,
743
+ "learning_rate": 0.00011429367954874819,
744
+ "loss": 0.9335,
745
+ "step": 515
746
+ },
747
+ {
748
+ "epoch": 1.54,
749
+ "grad_norm": 1.037759449688711,
750
+ "learning_rate": 0.00011258132571978555,
751
+ "loss": 0.9394,
752
+ "step": 520
753
+ },
754
+ {
755
+ "epoch": 1.56,
756
+ "grad_norm": 0.98060881335434,
757
+ "learning_rate": 0.00011086521500854745,
758
+ "loss": 0.9237,
759
+ "step": 525
760
+ },
761
+ {
762
+ "epoch": 1.57,
763
+ "grad_norm": 1.137978604803647,
764
+ "learning_rate": 0.00010914585985911632,
765
+ "loss": 0.873,
766
+ "step": 530
767
+ },
768
+ {
769
+ "epoch": 1.59,
770
+ "grad_norm": 1.0209942184078924,
771
+ "learning_rate": 0.00010742377368438914,
772
+ "loss": 0.989,
773
+ "step": 535
774
+ },
775
+ {
776
+ "epoch": 1.6,
777
+ "grad_norm": 1.0005848887814885,
778
+ "learning_rate": 0.00010569947071276847,
779
+ "loss": 0.9054,
780
+ "step": 540
781
+ },
782
+ {
783
+ "epoch": 1.61,
784
+ "grad_norm": 1.2509234265685367,
785
+ "learning_rate": 0.00010397346583460971,
786
+ "loss": 0.9068,
787
+ "step": 545
788
+ },
789
+ {
790
+ "epoch": 1.63,
791
+ "grad_norm": 1.118860273669544,
792
+ "learning_rate": 0.0001022462744484709,
793
+ "loss": 0.9185,
794
+ "step": 550
795
+ },
796
+ {
797
+ "epoch": 1.64,
798
+ "grad_norm": 0.9619304215105235,
799
+ "learning_rate": 0.00010051841230721065,
800
+ "loss": 0.9152,
801
+ "step": 555
802
+ },
803
+ {
804
+ "epoch": 1.66,
805
+ "grad_norm": 0.9312722485058544,
806
+ "learning_rate": 9.879039536398024e-05,
807
+ "loss": 0.9451,
808
+ "step": 560
809
+ },
810
+ {
811
+ "epoch": 1.67,
812
+ "grad_norm": 0.9758156277152181,
813
+ "learning_rate": 9.70627396181561e-05,
814
+ "loss": 0.9016,
815
+ "step": 565
816
+ },
817
+ {
818
+ "epoch": 1.69,
819
+ "grad_norm": 0.8860842880872017,
820
+ "learning_rate": 9.533596096125825e-05,
821
+ "loss": 0.9069,
822
+ "step": 570
823
+ },
824
+ {
825
+ "epoch": 1.7,
826
+ "grad_norm": 1.0727044167911228,
827
+ "learning_rate": 9.361057502290113e-05,
828
+ "loss": 0.8748,
829
+ "step": 575
830
+ },
831
+ {
832
+ "epoch": 1.72,
833
+ "grad_norm": 0.8634186584908737,
834
+ "learning_rate": 9.188709701682247e-05,
835
+ "loss": 0.893,
836
+ "step": 580
837
+ },
838
+ {
839
+ "epoch": 1.73,
840
+ "grad_norm": 1.1477573965693937,
841
+ "learning_rate": 9.016604158703654e-05,
842
+ "loss": 0.8893,
843
+ "step": 585
844
+ },
845
+ {
846
+ "epoch": 1.75,
847
+ "grad_norm": 0.9383547560329538,
848
+ "learning_rate": 8.844792265415738e-05,
849
+ "loss": 0.8651,
850
+ "step": 590
851
+ },
852
+ {
853
+ "epoch": 1.76,
854
+ "grad_norm": 1.1121008461123547,
855
+ "learning_rate": 8.673325326193806e-05,
856
+ "loss": 0.9379,
857
+ "step": 595
858
+ },
859
+ {
860
+ "epoch": 1.78,
861
+ "grad_norm": 1.0975479479362362,
862
+ "learning_rate": 8.502254542407186e-05,
863
+ "loss": 0.8481,
864
+ "step": 600
865
+ },
866
+ {
867
+ "epoch": 1.79,
868
+ "grad_norm": 1.113767179456745,
869
+ "learning_rate": 8.33163099713009e-05,
870
+ "loss": 0.8327,
871
+ "step": 605
872
+ },
873
+ {
874
+ "epoch": 1.81,
875
+ "grad_norm": 0.89921971591069,
876
+ "learning_rate": 8.161505639887817e-05,
877
+ "loss": 0.9102,
878
+ "step": 610
879
+ },
880
+ {
881
+ "epoch": 1.82,
882
+ "grad_norm": 1.0388536628309168,
883
+ "learning_rate": 7.991929271442817e-05,
884
+ "loss": 0.859,
885
+ "step": 615
886
+ },
887
+ {
888
+ "epoch": 1.84,
889
+ "grad_norm": 1.0521026013607777,
890
+ "learning_rate": 7.822952528625191e-05,
891
+ "loss": 0.8731,
892
+ "step": 620
893
+ },
894
+ {
895
+ "epoch": 1.85,
896
+ "grad_norm": 0.9421918430486114,
897
+ "learning_rate": 7.654625869212146e-05,
898
+ "loss": 0.9012,
899
+ "step": 625
900
+ },
901
+ {
902
+ "epoch": 1.87,
903
+ "grad_norm": 0.9484722686972686,
904
+ "learning_rate": 7.48699955686089e-05,
905
+ "loss": 0.9432,
906
+ "step": 630
907
+ },
908
+ {
909
+ "epoch": 1.88,
910
+ "grad_norm": 0.9768965209508316,
911
+ "learning_rate": 7.320123646099519e-05,
912
+ "loss": 0.8999,
913
+ "step": 635
914
+ },
915
+ {
916
+ "epoch": 1.9,
917
+ "grad_norm": 0.9453535215942106,
918
+ "learning_rate": 7.154047967380354e-05,
919
+ "loss": 0.9724,
920
+ "step": 640
921
+ },
922
+ {
923
+ "epoch": 1.91,
924
+ "grad_norm": 1.0424636870577482,
925
+ "learning_rate": 6.988822112200156e-05,
926
+ "loss": 0.8789,
927
+ "step": 645
928
+ },
929
+ {
930
+ "epoch": 1.93,
931
+ "grad_norm": 0.9224956346414365,
932
+ "learning_rate": 6.82449541829174e-05,
933
+ "loss": 0.8569,
934
+ "step": 650
935
+ },
936
+ {
937
+ "epoch": 1.94,
938
+ "grad_norm": 1.1516517831276118,
939
+ "learning_rate": 6.661116954891328e-05,
940
+ "loss": 0.9306,
941
+ "step": 655
942
+ },
943
+ {
944
+ "epoch": 1.96,
945
+ "grad_norm": 0.9883581608392334,
946
+ "learning_rate": 6.498735508086093e-05,
947
+ "loss": 0.871,
948
+ "step": 660
949
+ },
950
+ {
951
+ "epoch": 1.97,
952
+ "grad_norm": 0.9438307796156034,
953
+ "learning_rate": 6.337399566246257e-05,
954
+ "loss": 0.9302,
955
+ "step": 665
956
+ },
957
+ {
958
+ "epoch": 1.99,
959
+ "grad_norm": 0.8930439870325309,
960
+ "learning_rate": 6.177157305546078e-05,
961
+ "loss": 0.9477,
962
+ "step": 670
963
+ },
964
+ {
965
+ "epoch": 2.0,
966
+ "grad_norm": 0.9118609372127487,
967
+ "learning_rate": 6.018056575578075e-05,
968
+ "loss": 0.8771,
969
+ "step": 675
970
+ },
971
+ {
972
+ "epoch": 2.0,
973
+ "eval_loss": 1.1150912046432495,
974
+ "eval_runtime": 156.2319,
975
+ "eval_samples_per_second": 14.786,
976
+ "eval_steps_per_second": 0.467,
977
+ "step": 675
978
+ },
979
+ {
980
+ "epoch": 2.01,
981
+ "grad_norm": 0.828432991647057,
982
+ "learning_rate": 5.860144885064751e-05,
983
+ "loss": 0.8378,
984
+ "step": 680
985
+ },
986
+ {
987
+ "epoch": 2.03,
988
+ "grad_norm": 1.0448906737327117,
989
+ "learning_rate": 5.7034693876721376e-05,
990
+ "loss": 0.8421,
991
+ "step": 685
992
+ },
993
+ {
994
+ "epoch": 2.04,
995
+ "grad_norm": 1.0620773533713173,
996
+ "learning_rate": 5.54807686792933e-05,
997
+ "loss": 0.8794,
998
+ "step": 690
999
+ },
1000
+ {
1001
+ "epoch": 2.06,
1002
+ "grad_norm": 1.0869539484563016,
1003
+ "learning_rate": 5.394013727258254e-05,
1004
+ "loss": 0.8679,
1005
+ "step": 695
1006
+ },
1007
+ {
1008
+ "epoch": 2.07,
1009
+ "grad_norm": 0.9798911637537467,
1010
+ "learning_rate": 5.2413259701178505e-05,
1011
+ "loss": 0.8898,
1012
+ "step": 700
1013
+ },
1014
+ {
1015
+ "epoch": 2.09,
1016
+ "grad_norm": 0.9047442392594395,
1017
+ "learning_rate": 5.090059190266779e-05,
1018
+ "loss": 0.8724,
1019
+ "step": 705
1020
+ },
1021
+ {
1022
+ "epoch": 2.1,
1023
+ "grad_norm": 1.0781903357123614,
1024
+ "learning_rate": 4.940258557148765e-05,
1025
+ "loss": 0.8573,
1026
+ "step": 710
1027
+ },
1028
+ {
1029
+ "epoch": 2.12,
1030
+ "grad_norm": 1.0342896185783423,
1031
+ "learning_rate": 4.791968802404648e-05,
1032
+ "loss": 0.8281,
1033
+ "step": 715
1034
+ },
1035
+ {
1036
+ "epoch": 2.13,
1037
+ "grad_norm": 0.9191153541590816,
1038
+ "learning_rate": 4.645234206515171e-05,
1039
+ "loss": 0.9009,
1040
+ "step": 720
1041
+ },
1042
+ {
1043
+ "epoch": 2.15,
1044
+ "grad_norm": 1.1845629044153207,
1045
+ "learning_rate": 4.5000985855784746e-05,
1046
+ "loss": 0.857,
1047
+ "step": 725
1048
+ },
1049
+ {
1050
+ "epoch": 2.16,
1051
+ "grad_norm": 1.0434089629049308,
1052
+ "learning_rate": 4.3566052782262735e-05,
1053
+ "loss": 0.858,
1054
+ "step": 730
1055
+ },
1056
+ {
1057
+ "epoch": 2.18,
1058
+ "grad_norm": 0.990812188911313,
1059
+ "learning_rate": 4.2147971326825966e-05,
1060
+ "loss": 0.8516,
1061
+ "step": 735
1062
+ },
1063
+ {
1064
+ "epoch": 2.19,
1065
+ "grad_norm": 0.905736044179964,
1066
+ "learning_rate": 4.074716493968975e-05,
1067
+ "loss": 0.8476,
1068
+ "step": 740
1069
+ },
1070
+ {
1071
+ "epoch": 2.21,
1072
+ "grad_norm": 1.016168128211076,
1073
+ "learning_rate": 3.936405191259891e-05,
1074
+ "loss": 0.8388,
1075
+ "step": 745
1076
+ },
1077
+ {
1078
+ "epoch": 2.22,
1079
+ "grad_norm": 0.9657889846503783,
1080
+ "learning_rate": 3.79990452539225e-05,
1081
+ "loss": 0.8796,
1082
+ "step": 750
1083
+ },
1084
+ {
1085
+ "epoch": 2.24,
1086
+ "grad_norm": 0.8488273704750955,
1087
+ "learning_rate": 3.665255256532638e-05,
1088
+ "loss": 0.8226,
1089
+ "step": 755
1090
+ },
1091
+ {
1092
+ "epoch": 2.25,
1093
+ "grad_norm": 0.8515631446002137,
1094
+ "learning_rate": 3.53249759200601e-05,
1095
+ "loss": 0.8848,
1096
+ "step": 760
1097
+ },
1098
+ {
1099
+ "epoch": 2.27,
1100
+ "grad_norm": 0.9625777666321612,
1101
+ "learning_rate": 3.401671174289469e-05,
1102
+ "loss": 0.8581,
1103
+ "step": 765
1104
+ },
1105
+ {
1106
+ "epoch": 2.28,
1107
+ "grad_norm": 1.1027956646622936,
1108
+ "learning_rate": 3.2728150691747115e-05,
1109
+ "loss": 0.8205,
1110
+ "step": 770
1111
+ },
1112
+ {
1113
+ "epoch": 2.3,
1114
+ "grad_norm": 0.959658890342703,
1115
+ "learning_rate": 3.145967754102691e-05,
1116
+ "loss": 0.8332,
1117
+ "step": 775
1118
+ },
1119
+ {
1120
+ "epoch": 2.31,
1121
+ "grad_norm": 1.014525985812299,
1122
+ "learning_rate": 3.021167106673928e-05,
1123
+ "loss": 0.8485,
1124
+ "step": 780
1125
+ },
1126
+ {
1127
+ "epoch": 2.33,
1128
+ "grad_norm": 1.0119872246269643,
1129
+ "learning_rate": 2.898450393337977e-05,
1130
+ "loss": 0.856,
1131
+ "step": 785
1132
+ },
1133
+ {
1134
+ "epoch": 2.34,
1135
+ "grad_norm": 1.321701548482505,
1136
+ "learning_rate": 2.7778542582653744e-05,
1137
+ "loss": 0.8304,
1138
+ "step": 790
1139
+ },
1140
+ {
1141
+ "epoch": 2.36,
1142
+ "grad_norm": 1.0899857889914895,
1143
+ "learning_rate": 2.659414712405398e-05,
1144
+ "loss": 0.8442,
1145
+ "step": 795
1146
+ },
1147
+ {
1148
+ "epoch": 2.37,
1149
+ "grad_norm": 1.0994834827590023,
1150
+ "learning_rate": 2.543167122732918e-05,
1151
+ "loss": 0.8645,
1152
+ "step": 800
1153
+ },
1154
+ {
1155
+ "epoch": 2.39,
1156
+ "grad_norm": 0.8030087173097793,
1157
+ "learning_rate": 2.429146201687538e-05,
1158
+ "loss": 0.8567,
1159
+ "step": 805
1160
+ },
1161
+ {
1162
+ "epoch": 2.4,
1163
+ "grad_norm": 1.01607316930185,
1164
+ "learning_rate": 2.3173859968081944e-05,
1165
+ "loss": 0.9123,
1166
+ "step": 810
1167
+ },
1168
+ {
1169
+ "epoch": 2.41,
1170
+ "grad_norm": 0.9294802520942274,
1171
+ "learning_rate": 2.2079198805662914e-05,
1172
+ "loss": 0.8374,
1173
+ "step": 815
1174
+ },
1175
+ {
1176
+ "epoch": 2.43,
1177
+ "grad_norm": 0.9523609743858644,
1178
+ "learning_rate": 2.1007805404004242e-05,
1179
+ "loss": 0.8568,
1180
+ "step": 820
1181
+ },
1182
+ {
1183
+ "epoch": 2.44,
1184
+ "grad_norm": 0.8957405654003763,
1185
+ "learning_rate": 1.995999968955641e-05,
1186
+ "loss": 0.8873,
1187
+ "step": 825
1188
+ },
1189
+ {
1190
+ "epoch": 2.46,
1191
+ "grad_norm": 1.0700819784398359,
1192
+ "learning_rate": 1.8936094545302095e-05,
1193
+ "loss": 0.8564,
1194
+ "step": 830
1195
+ },
1196
+ {
1197
+ "epoch": 2.47,
1198
+ "grad_norm": 1.0564758507529628,
1199
+ "learning_rate": 1.7936395717326704e-05,
1200
+ "loss": 0.8221,
1201
+ "step": 835
1202
+ },
1203
+ {
1204
+ "epoch": 2.49,
1205
+ "grad_norm": 1.015095925986807,
1206
+ "learning_rate": 1.696120172352025e-05,
1207
+ "loss": 0.8416,
1208
+ "step": 840
1209
+ },
1210
+ {
1211
+ "epoch": 2.5,
1212
+ "grad_norm": 1.2346772662125298,
1213
+ "learning_rate": 1.601080376443763e-05,
1214
+ "loss": 0.8742,
1215
+ "step": 845
1216
+ },
1217
+ {
1218
+ "epoch": 2.52,
1219
+ "grad_norm": 0.9234353972167334,
1220
+ "learning_rate": 1.5085485636343755e-05,
1221
+ "loss": 0.9049,
1222
+ "step": 850
1223
+ },
1224
+ {
1225
+ "epoch": 2.53,
1226
+ "grad_norm": 1.0570174679975244,
1227
+ "learning_rate": 1.4185523646469822e-05,
1228
+ "loss": 0.8367,
1229
+ "step": 855
1230
+ },
1231
+ {
1232
+ "epoch": 2.55,
1233
+ "grad_norm": 0.8795669021553979,
1234
+ "learning_rate": 1.3311186530505838e-05,
1235
+ "loss": 0.8661,
1236
+ "step": 860
1237
+ },
1238
+ {
1239
+ "epoch": 2.56,
1240
+ "grad_norm": 0.9638299921267924,
1241
+ "learning_rate": 1.2462735372353996e-05,
1242
+ "loss": 0.8363,
1243
+ "step": 865
1244
+ },
1245
+ {
1246
+ "epoch": 2.58,
1247
+ "grad_norm": 1.2030489918077059,
1248
+ "learning_rate": 1.1640423526166988e-05,
1249
+ "loss": 0.8583,
1250
+ "step": 870
1251
+ },
1252
+ {
1253
+ "epoch": 2.59,
1254
+ "grad_norm": 1.0716487215374508,
1255
+ "learning_rate": 1.0844496540694515e-05,
1256
+ "loss": 0.8588,
1257
+ "step": 875
1258
+ },
1259
+ {
1260
+ "epoch": 2.61,
1261
+ "grad_norm": 1.0669537623700072,
1262
+ "learning_rate": 1.007519208596045e-05,
1263
+ "loss": 0.9143,
1264
+ "step": 880
1265
+ },
1266
+ {
1267
+ "epoch": 2.62,
1268
+ "grad_norm": 0.9415770157718919,
1269
+ "learning_rate": 9.332739882292752e-06,
1270
+ "loss": 0.906,
1271
+ "step": 885
1272
+ },
1273
+ {
1274
+ "epoch": 2.64,
1275
+ "grad_norm": 0.8466881399276756,
1276
+ "learning_rate": 8.617361631727138e-06,
1277
+ "loss": 0.8823,
1278
+ "step": 890
1279
+ },
1280
+ {
1281
+ "epoch": 2.65,
1282
+ "grad_norm": 1.173003775227567,
1283
+ "learning_rate": 7.929270951805178e-06,
1284
+ "loss": 0.8866,
1285
+ "step": 895
1286
+ },
1287
+ {
1288
+ "epoch": 2.67,
1289
+ "grad_norm": 0.9695275155903833,
1290
+ "learning_rate": 7.2686733117863784e-06,
1291
+ "loss": 0.8717,
1292
+ "step": 900
1293
+ },
1294
+ {
1295
+ "epoch": 2.68,
1296
+ "grad_norm": 0.9278647388922164,
1297
+ "learning_rate": 6.635765971293484e-06,
1298
+ "loss": 0.8999,
1299
+ "step": 905
1300
+ },
1301
+ {
1302
+ "epoch": 2.7,
1303
+ "grad_norm": 0.9905531986540063,
1304
+ "learning_rate": 6.030737921409169e-06,
1305
+ "loss": 0.8772,
1306
+ "step": 910
1307
+ },
1308
+ {
1309
+ "epoch": 2.71,
1310
+ "grad_norm": 1.068092774140916,
1311
+ "learning_rate": 5.453769828241872e-06,
1312
+ "loss": 0.8974,
1313
+ "step": 915
1314
+ },
1315
+ {
1316
+ "epoch": 2.73,
1317
+ "grad_norm": 1.1961165825237712,
1318
+ "learning_rate": 4.905033978977491e-06,
1319
+ "loss": 0.86,
1320
+ "step": 920
1321
+ },
1322
+ {
1323
+ "epoch": 2.74,
1324
+ "grad_norm": 1.0848450931313236,
1325
+ "learning_rate": 4.384694230432984e-06,
1326
+ "loss": 0.8363,
1327
+ "step": 925
1328
+ },
1329
+ {
1330
+ "epoch": 2.76,
1331
+ "grad_norm": 1.0929622662706844,
1332
+ "learning_rate": 3.892905960127546e-06,
1333
+ "loss": 0.8727,
1334
+ "step": 930
1335
+ },
1336
+ {
1337
+ "epoch": 2.77,
1338
+ "grad_norm": 1.0564158056197959,
1339
+ "learning_rate": 3.4298160198856568e-06,
1340
+ "loss": 0.8169,
1341
+ "step": 935
1342
+ },
1343
+ {
1344
+ "epoch": 2.79,
1345
+ "grad_norm": 1.0136059442310454,
1346
+ "learning_rate": 2.995562691985898e-06,
1347
+ "loss": 0.8574,
1348
+ "step": 940
1349
+ },
1350
+ {
1351
+ "epoch": 2.8,
1352
+ "grad_norm": 1.092221087850955,
1353
+ "learning_rate": 2.590275647868867e-06,
1354
+ "loss": 0.8386,
1355
+ "step": 945
1356
+ },
1357
+ {
1358
+ "epoch": 2.81,
1359
+ "grad_norm": 1.0072864796741217,
1360
+ "learning_rate": 2.2140759094162467e-06,
1361
+ "loss": 0.8685,
1362
+ "step": 950
1363
+ },
1364
+ {
1365
+ "epoch": 2.83,
1366
+ "grad_norm": 0.8433396823974946,
1367
+ "learning_rate": 1.8670758128126909e-06,
1368
+ "loss": 0.849,
1369
+ "step": 955
1370
+ },
1371
+ {
1372
+ "epoch": 2.84,
1373
+ "grad_norm": 0.9684221245764032,
1374
+ "learning_rate": 1.5493789750014031e-06,
1375
+ "loss": 0.8525,
1376
+ "step": 960
1377
+ },
1378
+ {
1379
+ "epoch": 2.86,
1380
+ "grad_norm": 0.9141081630302206,
1381
+ "learning_rate": 1.261080262743297e-06,
1382
+ "loss": 0.8315,
1383
+ "step": 965
1384
+ },
1385
+ {
1386
+ "epoch": 2.87,
1387
+ "grad_norm": 0.9014966731205527,
1388
+ "learning_rate": 1.0022657642890231e-06,
1389
+ "loss": 0.8316,
1390
+ "step": 970
1391
+ },
1392
+ {
1393
+ "epoch": 2.89,
1394
+ "grad_norm": 0.9464472386186885,
1395
+ "learning_rate": 7.730127636723539e-07,
1396
+ "loss": 0.8605,
1397
+ "step": 975
1398
+ },
1399
+ {
1400
+ "epoch": 2.9,
1401
+ "grad_norm": 0.896721952301425,
1402
+ "learning_rate": 5.733897176325665e-07,
1403
+ "loss": 0.8137,
1404
+ "step": 980
1405
+ },
1406
+ {
1407
+ "epoch": 2.92,
1408
+ "grad_norm": 1.0368172773007387,
1409
+ "learning_rate": 4.034562351727389e-07,
1410
+ "loss": 0.8664,
1411
+ "step": 985
1412
+ },
1413
+ {
1414
+ "epoch": 2.93,
1415
+ "grad_norm": 0.9777429571054969,
1416
+ "learning_rate": 2.6326305976001055e-07,
1417
+ "loss": 0.8557,
1418
+ "step": 990
1419
+ },
1420
+ {
1421
+ "epoch": 2.95,
1422
+ "grad_norm": 0.96487435678213,
1423
+ "learning_rate": 1.5285205417319149e-07,
1424
+ "loss": 0.8691,
1425
+ "step": 995
1426
+ },
1427
+ {
1428
+ "epoch": 2.96,
1429
+ "grad_norm": 0.8599085672586297,
1430
+ "learning_rate": 7.225618800222877e-08,
1431
+ "loss": 0.8806,
1432
+ "step": 1000
1433
+ },
1434
+ {
1435
+ "epoch": 2.98,
1436
+ "grad_norm": 1.0670556057589013,
1437
+ "learning_rate": 2.1499527803214846e-08,
1438
+ "loss": 0.8669,
1439
+ "step": 1005
1440
+ },
1441
+ {
1442
+ "epoch": 2.99,
1443
+ "grad_norm": 1.0796304331427407,
1444
+ "learning_rate": 5.972299119250125e-10,
1445
+ "loss": 0.8486,
1446
+ "step": 1010
1447
+ },
1448
+ {
1449
+ "epoch": 3.0,
1450
+ "eval_loss": 1.1171541213989258,
1451
+ "eval_runtime": 156.2002,
1452
+ "eval_samples_per_second": 14.789,
1453
+ "eval_steps_per_second": 0.467,
1454
+ "step": 1011
1455
+ },
1456
+ {
1457
+ "epoch": 3.0,
1458
+ "step": 1011,
1459
+ "total_flos": 2597843400392704.0,
1460
+ "train_loss": 1.2570129054000658,
1461
+ "train_runtime": 12020.0319,
1462
+ "train_samples_per_second": 5.39,
1463
+ "train_steps_per_second": 0.084
1464
+ }
1465
+ ],
1466
+ "logging_steps": 5,
1467
+ "max_steps": 1011,
1468
+ "num_input_tokens_seen": 0,
1469
+ "num_train_epochs": 3,
1470
+ "save_steps": 500,
1471
+ "total_flos": 2597843400392704.0,
1472
+ "train_batch_size": 8,
1473
+ "trial_name": null,
1474
+ "trial_params": null
1475
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4b66f4e756294394e1b9616e65ee37344fc99b0ed37e245591abbb33dd19f6a5
3
+ size 6136