Qin Liu commited on
Commit
770c1db
1 Parent(s): 5479bc2

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-10p
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-10p
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: 1.1202
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
+ | 0.9388 | 1.0 | 328 | 1.1538 |
59
+ | 0.9232 | 2.0 | 656 | 1.1174 |
60
+ | 0.8492 | 3.0 | 984 | 1.1202 |
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
+ "q_proj",
23
+ "gate_proj",
24
+ "up_proj",
25
+ "o_proj",
26
+ "v_proj",
27
+ "k_proj",
28
+ "down_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 sha256:1d50cc7d80f1da36d5149527ae5169744a2ffd41ab77893b70d255a2ea1f4459
3
+ size 37555048
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "train_loss": 1.2712977070633957,
4
+ "train_runtime": 11725.5058,
5
+ "train_samples": 20971,
6
+ "train_samples_per_second": 5.365,
7
+ "train_steps_per_second": 0.084
8
+ }
runs/Apr21_05-32-12_COE-CS-sv003/events.out.tfevents.1713677554.COE-CS-sv003.350572.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:22d82db1accae130156d1d74159bb88dada80547a38f166309f00f34f7e10b9b
3
+ size 4790
runs/Apr21_16-09-42_COE-CS-sv003/events.out.tfevents.1713715802.COE-CS-sv003.361358.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8b4c98b78d2e630d48472b0fe64f3fdf0d8102cc2e51eb85e09b38ed1040feb9
3
+ size 4790
runs/Apr21_19-50-06_COE-CS-sv003/events.out.tfevents.1713729024.COE-CS-sv003.363955.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ea2d9571c66a083cbd4e0fda289827bbde1b1f86d35b2ef4efc844f935656d5d
3
+ size 47420
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": 1.2712977070633957,
4
+ "train_runtime": 11725.5058,
5
+ "train_samples": 20971,
6
+ "train_samples_per_second": 5.365,
7
+ "train_steps_per_second": 0.084
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1433 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.0,
5
+ "eval_steps": 500,
6
+ "global_step": 984,
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": 70.84053355134651,
14
+ "learning_rate": 2.0202020202020206e-06,
15
+ "loss": 16.5976,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.02,
20
+ "grad_norm": 70.40484875837613,
21
+ "learning_rate": 1.0101010101010101e-05,
22
+ "loss": 16.2878,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.03,
27
+ "grad_norm": 64.13592776682735,
28
+ "learning_rate": 2.0202020202020203e-05,
29
+ "loss": 15.6234,
30
+ "step": 10
31
+ },
32
+ {
33
+ "epoch": 0.05,
34
+ "grad_norm": 48.17657464142415,
35
+ "learning_rate": 3.0303030303030306e-05,
36
+ "loss": 13.1692,
37
+ "step": 15
38
+ },
39
+ {
40
+ "epoch": 0.06,
41
+ "grad_norm": 42.09774383233372,
42
+ "learning_rate": 4.0404040404040405e-05,
43
+ "loss": 9.9677,
44
+ "step": 20
45
+ },
46
+ {
47
+ "epoch": 0.08,
48
+ "grad_norm": 21.40017494750998,
49
+ "learning_rate": 5.050505050505051e-05,
50
+ "loss": 6.438,
51
+ "step": 25
52
+ },
53
+ {
54
+ "epoch": 0.09,
55
+ "grad_norm": 13.47184864955966,
56
+ "learning_rate": 6.060606060606061e-05,
57
+ "loss": 4.5534,
58
+ "step": 30
59
+ },
60
+ {
61
+ "epoch": 0.11,
62
+ "grad_norm": 9.0175134340465,
63
+ "learning_rate": 7.07070707070707e-05,
64
+ "loss": 3.3841,
65
+ "step": 35
66
+ },
67
+ {
68
+ "epoch": 0.12,
69
+ "grad_norm": 7.9069963595872474,
70
+ "learning_rate": 8.080808080808081e-05,
71
+ "loss": 2.4486,
72
+ "step": 40
73
+ },
74
+ {
75
+ "epoch": 0.14,
76
+ "grad_norm": 4.497563859166001,
77
+ "learning_rate": 9.090909090909092e-05,
78
+ "loss": 1.9157,
79
+ "step": 45
80
+ },
81
+ {
82
+ "epoch": 0.15,
83
+ "grad_norm": 3.8365103057277237,
84
+ "learning_rate": 0.00010101010101010102,
85
+ "loss": 1.6253,
86
+ "step": 50
87
+ },
88
+ {
89
+ "epoch": 0.17,
90
+ "grad_norm": 2.6711072114914822,
91
+ "learning_rate": 0.00011111111111111112,
92
+ "loss": 1.4531,
93
+ "step": 55
94
+ },
95
+ {
96
+ "epoch": 0.18,
97
+ "grad_norm": 2.8242292867987038,
98
+ "learning_rate": 0.00012121212121212122,
99
+ "loss": 1.355,
100
+ "step": 60
101
+ },
102
+ {
103
+ "epoch": 0.2,
104
+ "grad_norm": 1.5627627321505486,
105
+ "learning_rate": 0.00013131313131313133,
106
+ "loss": 1.3144,
107
+ "step": 65
108
+ },
109
+ {
110
+ "epoch": 0.21,
111
+ "grad_norm": 1.9610550212854403,
112
+ "learning_rate": 0.0001414141414141414,
113
+ "loss": 1.2403,
114
+ "step": 70
115
+ },
116
+ {
117
+ "epoch": 0.23,
118
+ "grad_norm": 1.4826190020479229,
119
+ "learning_rate": 0.00015151515151515152,
120
+ "loss": 1.2116,
121
+ "step": 75
122
+ },
123
+ {
124
+ "epoch": 0.24,
125
+ "grad_norm": 1.4179410680444433,
126
+ "learning_rate": 0.00016161616161616162,
127
+ "loss": 1.1361,
128
+ "step": 80
129
+ },
130
+ {
131
+ "epoch": 0.26,
132
+ "grad_norm": 1.6440373948698321,
133
+ "learning_rate": 0.00017171717171717173,
134
+ "loss": 1.0898,
135
+ "step": 85
136
+ },
137
+ {
138
+ "epoch": 0.27,
139
+ "grad_norm": 1.4741470072787244,
140
+ "learning_rate": 0.00018181818181818183,
141
+ "loss": 1.1321,
142
+ "step": 90
143
+ },
144
+ {
145
+ "epoch": 0.29,
146
+ "grad_norm": 1.4918119127052252,
147
+ "learning_rate": 0.00019191919191919191,
148
+ "loss": 1.1502,
149
+ "step": 95
150
+ },
151
+ {
152
+ "epoch": 0.3,
153
+ "grad_norm": 1.780737333023736,
154
+ "learning_rate": 0.0001999993699387764,
155
+ "loss": 1.0674,
156
+ "step": 100
157
+ },
158
+ {
159
+ "epoch": 0.32,
160
+ "grad_norm": 1.5713634753942138,
161
+ "learning_rate": 0.0001999773186295914,
162
+ "loss": 1.0976,
163
+ "step": 105
164
+ },
165
+ {
166
+ "epoch": 0.34,
167
+ "grad_norm": 1.4792461465668252,
168
+ "learning_rate": 0.0001999237721983199,
169
+ "loss": 1.0793,
170
+ "step": 110
171
+ },
172
+ {
173
+ "epoch": 0.35,
174
+ "grad_norm": 1.414080214214796,
175
+ "learning_rate": 0.0001998387475133018,
176
+ "loss": 1.1008,
177
+ "step": 115
178
+ },
179
+ {
180
+ "epoch": 0.37,
181
+ "grad_norm": 1.4612237811023239,
182
+ "learning_rate": 0.00019972227135924052,
183
+ "loss": 1.0824,
184
+ "step": 120
185
+ },
186
+ {
187
+ "epoch": 0.38,
188
+ "grad_norm": 1.5484440969715458,
189
+ "learning_rate": 0.00019957438042876542,
190
+ "loss": 1.0178,
191
+ "step": 125
192
+ },
193
+ {
194
+ "epoch": 0.4,
195
+ "grad_norm": 2.0098511861836643,
196
+ "learning_rate": 0.0001993951213108726,
197
+ "loss": 1.0397,
198
+ "step": 130
199
+ },
200
+ {
201
+ "epoch": 0.41,
202
+ "grad_norm": 1.461268889218997,
203
+ "learning_rate": 0.00019918455047624847,
204
+ "loss": 1.0558,
205
+ "step": 135
206
+ },
207
+ {
208
+ "epoch": 0.43,
209
+ "grad_norm": 1.816402545663251,
210
+ "learning_rate": 0.00019894273425948003,
211
+ "loss": 0.9983,
212
+ "step": 140
213
+ },
214
+ {
215
+ "epoch": 0.44,
216
+ "grad_norm": 1.5043083827737078,
217
+ "learning_rate": 0.0001986697488381581,
218
+ "loss": 0.9565,
219
+ "step": 145
220
+ },
221
+ {
222
+ "epoch": 0.46,
223
+ "grad_norm": 1.5595660341738484,
224
+ "learning_rate": 0.00019836568020887964,
225
+ "loss": 0.9811,
226
+ "step": 150
227
+ },
228
+ {
229
+ "epoch": 0.47,
230
+ "grad_norm": 1.3010086010553632,
231
+ "learning_rate": 0.00019803062416015675,
232
+ "loss": 1.05,
233
+ "step": 155
234
+ },
235
+ {
236
+ "epoch": 0.49,
237
+ "grad_norm": 1.936074537977103,
238
+ "learning_rate": 0.0001976646862422413,
239
+ "loss": 1.0033,
240
+ "step": 160
241
+ },
242
+ {
243
+ "epoch": 0.5,
244
+ "grad_norm": 2.43894566672306,
245
+ "learning_rate": 0.00019726798173387415,
246
+ "loss": 0.9904,
247
+ "step": 165
248
+ },
249
+ {
250
+ "epoch": 0.52,
251
+ "grad_norm": 2.0226681751936604,
252
+ "learning_rate": 0.0001968406356059696,
253
+ "loss": 1.0087,
254
+ "step": 170
255
+ },
256
+ {
257
+ "epoch": 0.53,
258
+ "grad_norm": 1.5765984883843074,
259
+ "learning_rate": 0.00019638278248224688,
260
+ "loss": 1.0074,
261
+ "step": 175
262
+ },
263
+ {
264
+ "epoch": 0.55,
265
+ "grad_norm": 1.4953203184933894,
266
+ "learning_rate": 0.00019589456659682057,
267
+ "loss": 0.9663,
268
+ "step": 180
269
+ },
270
+ {
271
+ "epoch": 0.56,
272
+ "grad_norm": 1.3905440186304812,
273
+ "learning_rate": 0.00019537614174876356,
274
+ "loss": 0.9621,
275
+ "step": 185
276
+ },
277
+ {
278
+ "epoch": 0.58,
279
+ "grad_norm": 1.7108204646804686,
280
+ "learning_rate": 0.0001948276712536569,
281
+ "loss": 0.958,
282
+ "step": 190
283
+ },
284
+ {
285
+ "epoch": 0.59,
286
+ "grad_norm": 1.146705235438316,
287
+ "learning_rate": 0.00019424932789214157,
288
+ "loss": 1.0311,
289
+ "step": 195
290
+ },
291
+ {
292
+ "epoch": 0.61,
293
+ "grad_norm": 1.1584586792685783,
294
+ "learning_rate": 0.00019364129385548892,
295
+ "loss": 1.0171,
296
+ "step": 200
297
+ },
298
+ {
299
+ "epoch": 0.62,
300
+ "grad_norm": 1.629476089763749,
301
+ "learning_rate": 0.00019300376068820602,
302
+ "loss": 0.9775,
303
+ "step": 205
304
+ },
305
+ {
306
+ "epoch": 0.64,
307
+ "grad_norm": 1.2581783118467773,
308
+ "learning_rate": 0.00019233692922769496,
309
+ "loss": 1.0471,
310
+ "step": 210
311
+ },
312
+ {
313
+ "epoch": 0.66,
314
+ "grad_norm": 1.4906993491973393,
315
+ "learning_rate": 0.0001916410095409843,
316
+ "loss": 0.9967,
317
+ "step": 215
318
+ },
319
+ {
320
+ "epoch": 0.67,
321
+ "grad_norm": 1.1893776969242746,
322
+ "learning_rate": 0.00019091622085855363,
323
+ "loss": 0.9619,
324
+ "step": 220
325
+ },
326
+ {
327
+ "epoch": 0.69,
328
+ "grad_norm": 1.4151710347672513,
329
+ "learning_rate": 0.00019016279150527044,
330
+ "loss": 1.0227,
331
+ "step": 225
332
+ },
333
+ {
334
+ "epoch": 0.7,
335
+ "grad_norm": 1.2558733422369428,
336
+ "learning_rate": 0.00018938095882846306,
337
+ "loss": 0.9832,
338
+ "step": 230
339
+ },
340
+ {
341
+ "epoch": 0.72,
342
+ "grad_norm": 1.3320637162932236,
343
+ "learning_rate": 0.00018857096912315062,
344
+ "loss": 0.9802,
345
+ "step": 235
346
+ },
347
+ {
348
+ "epoch": 0.73,
349
+ "grad_norm": 1.2878623504565412,
350
+ "learning_rate": 0.00018773307755445465,
351
+ "loss": 0.9251,
352
+ "step": 240
353
+ },
354
+ {
355
+ "epoch": 0.75,
356
+ "grad_norm": 1.213232935367416,
357
+ "learning_rate": 0.0001868675480772163,
358
+ "loss": 0.98,
359
+ "step": 245
360
+ },
361
+ {
362
+ "epoch": 0.76,
363
+ "grad_norm": 1.3405116881469274,
364
+ "learning_rate": 0.00018597465335284437,
365
+ "loss": 0.9473,
366
+ "step": 250
367
+ },
368
+ {
369
+ "epoch": 0.78,
370
+ "grad_norm": 1.543540810073074,
371
+ "learning_rate": 0.0001850546746634211,
372
+ "loss": 0.9902,
373
+ "step": 255
374
+ },
375
+ {
376
+ "epoch": 0.79,
377
+ "grad_norm": 1.3710808148280944,
378
+ "learning_rate": 0.00018410790182309169,
379
+ "loss": 0.975,
380
+ "step": 260
381
+ },
382
+ {
383
+ "epoch": 0.81,
384
+ "grad_norm": 1.21692171592665,
385
+ "learning_rate": 0.00018313463308676636,
386
+ "loss": 0.9613,
387
+ "step": 265
388
+ },
389
+ {
390
+ "epoch": 0.82,
391
+ "grad_norm": 1.1934007014753554,
392
+ "learning_rate": 0.00018213517505616338,
393
+ "loss": 0.9618,
394
+ "step": 270
395
+ },
396
+ {
397
+ "epoch": 0.84,
398
+ "grad_norm": 1.60819637926966,
399
+ "learning_rate": 0.00018110984258322237,
400
+ "loss": 0.9939,
401
+ "step": 275
402
+ },
403
+ {
404
+ "epoch": 0.85,
405
+ "grad_norm": 1.2454326978067687,
406
+ "learning_rate": 0.00018005895867091896,
407
+ "loss": 0.9191,
408
+ "step": 280
409
+ },
410
+ {
411
+ "epoch": 0.87,
412
+ "grad_norm": 0.9690417217411257,
413
+ "learning_rate": 0.0001789828543715116,
414
+ "loss": 0.9421,
415
+ "step": 285
416
+ },
417
+ {
418
+ "epoch": 0.88,
419
+ "grad_norm": 1.3978495893847322,
420
+ "learning_rate": 0.0001778818686822523,
421
+ "loss": 0.9757,
422
+ "step": 290
423
+ },
424
+ {
425
+ "epoch": 0.9,
426
+ "grad_norm": 1.2683425234668648,
427
+ "learning_rate": 0.00017675634843859514,
428
+ "loss": 0.9902,
429
+ "step": 295
430
+ },
431
+ {
432
+ "epoch": 0.91,
433
+ "grad_norm": 1.2072894977789046,
434
+ "learning_rate": 0.00017560664820493498,
435
+ "loss": 0.9517,
436
+ "step": 300
437
+ },
438
+ {
439
+ "epoch": 0.93,
440
+ "grad_norm": 1.179285109564365,
441
+ "learning_rate": 0.00017443313016291185,
442
+ "loss": 0.9657,
443
+ "step": 305
444
+ },
445
+ {
446
+ "epoch": 0.95,
447
+ "grad_norm": 1.1314300153647592,
448
+ "learning_rate": 0.00017323616399731533,
449
+ "loss": 0.9438,
450
+ "step": 310
451
+ },
452
+ {
453
+ "epoch": 0.96,
454
+ "grad_norm": 1.0638903666221389,
455
+ "learning_rate": 0.0001720161267796256,
456
+ "loss": 1.0097,
457
+ "step": 315
458
+ },
459
+ {
460
+ "epoch": 0.98,
461
+ "grad_norm": 1.0687942439764777,
462
+ "learning_rate": 0.00017077340284922732,
463
+ "loss": 0.9453,
464
+ "step": 320
465
+ },
466
+ {
467
+ "epoch": 0.99,
468
+ "grad_norm": 1.2396289777933702,
469
+ "learning_rate": 0.000169508383692334,
470
+ "loss": 0.9388,
471
+ "step": 325
472
+ },
473
+ {
474
+ "epoch": 1.0,
475
+ "eval_loss": 1.1537501811981201,
476
+ "eval_runtime": 156.3402,
477
+ "eval_samples_per_second": 14.775,
478
+ "eval_steps_per_second": 0.467,
479
+ "step": 328
480
+ },
481
+ {
482
+ "epoch": 1.01,
483
+ "grad_norm": 1.178139096940401,
484
+ "learning_rate": 0.00016822146781866098,
485
+ "loss": 0.9355,
486
+ "step": 330
487
+ },
488
+ {
489
+ "epoch": 1.02,
490
+ "grad_norm": 0.9864464240664622,
491
+ "learning_rate": 0.00016691306063588583,
492
+ "loss": 0.9135,
493
+ "step": 335
494
+ },
495
+ {
496
+ "epoch": 1.04,
497
+ "grad_norm": 1.0449361148940848,
498
+ "learning_rate": 0.00016558357432193578,
499
+ "loss": 0.9088,
500
+ "step": 340
501
+ },
502
+ {
503
+ "epoch": 1.05,
504
+ "grad_norm": 1.02719313804267,
505
+ "learning_rate": 0.00016423342769514228,
506
+ "loss": 0.9361,
507
+ "step": 345
508
+ },
509
+ {
510
+ "epoch": 1.07,
511
+ "grad_norm": 1.1796531987937382,
512
+ "learning_rate": 0.00016286304608230368,
513
+ "loss": 0.9167,
514
+ "step": 350
515
+ },
516
+ {
517
+ "epoch": 1.08,
518
+ "grad_norm": 1.2165440327749395,
519
+ "learning_rate": 0.0001614728611846978,
520
+ "loss": 0.9071,
521
+ "step": 355
522
+ },
523
+ {
524
+ "epoch": 1.1,
525
+ "grad_norm": 1.0680390308725847,
526
+ "learning_rate": 0.0001600633109420861,
527
+ "loss": 0.9423,
528
+ "step": 360
529
+ },
530
+ {
531
+ "epoch": 1.11,
532
+ "grad_norm": 1.1460931817295472,
533
+ "learning_rate": 0.00015863483939475281,
534
+ "loss": 0.9414,
535
+ "step": 365
536
+ },
537
+ {
538
+ "epoch": 1.13,
539
+ "grad_norm": 1.2232830262553624,
540
+ "learning_rate": 0.00015718789654362204,
541
+ "loss": 0.9501,
542
+ "step": 370
543
+ },
544
+ {
545
+ "epoch": 1.14,
546
+ "grad_norm": 1.1171762766168023,
547
+ "learning_rate": 0.00015572293820849753,
548
+ "loss": 0.9278,
549
+ "step": 375
550
+ },
551
+ {
552
+ "epoch": 1.16,
553
+ "grad_norm": 1.134580557741046,
554
+ "learning_rate": 0.00015424042588446882,
555
+ "loss": 0.8877,
556
+ "step": 380
557
+ },
558
+ {
559
+ "epoch": 1.17,
560
+ "grad_norm": 1.1467717113467715,
561
+ "learning_rate": 0.00015274082659653,
562
+ "loss": 0.8968,
563
+ "step": 385
564
+ },
565
+ {
566
+ "epoch": 1.19,
567
+ "grad_norm": 1.0245879848355788,
568
+ "learning_rate": 0.0001512246127524561,
569
+ "loss": 0.933,
570
+ "step": 390
571
+ },
572
+ {
573
+ "epoch": 1.2,
574
+ "grad_norm": 0.9149640789208816,
575
+ "learning_rate": 0.0001496922619939842,
576
+ "loss": 0.9562,
577
+ "step": 395
578
+ },
579
+ {
580
+ "epoch": 1.22,
581
+ "grad_norm": 1.5241591126822251,
582
+ "learning_rate": 0.00014814425704634508,
583
+ "loss": 0.9906,
584
+ "step": 400
585
+ },
586
+ {
587
+ "epoch": 1.23,
588
+ "grad_norm": 1.0478360565843916,
589
+ "learning_rate": 0.00014658108556619417,
590
+ "loss": 0.9131,
591
+ "step": 405
592
+ },
593
+ {
594
+ "epoch": 1.25,
595
+ "grad_norm": 1.1580130995552973,
596
+ "learning_rate": 0.00014500323998798843,
597
+ "loss": 0.9385,
598
+ "step": 410
599
+ },
600
+ {
601
+ "epoch": 1.27,
602
+ "grad_norm": 1.247960363943888,
603
+ "learning_rate": 0.00014341121736885843,
604
+ "loss": 0.8527,
605
+ "step": 415
606
+ },
607
+ {
608
+ "epoch": 1.28,
609
+ "grad_norm": 1.1576786962267551,
610
+ "learning_rate": 0.00014180551923202405,
611
+ "loss": 0.9051,
612
+ "step": 420
613
+ },
614
+ {
615
+ "epoch": 1.3,
616
+ "grad_norm": 1.0608499549231114,
617
+ "learning_rate": 0.00014018665140880332,
618
+ "loss": 0.908,
619
+ "step": 425
620
+ },
621
+ {
622
+ "epoch": 1.31,
623
+ "grad_norm": 1.1012203287352098,
624
+ "learning_rate": 0.000138555123879264,
625
+ "loss": 0.9342,
626
+ "step": 430
627
+ },
628
+ {
629
+ "epoch": 1.33,
630
+ "grad_norm": 1.0806519524832878,
631
+ "learning_rate": 0.00013691145061156844,
632
+ "loss": 0.9136,
633
+ "step": 435
634
+ },
635
+ {
636
+ "epoch": 1.34,
637
+ "grad_norm": 1.0359931963939946,
638
+ "learning_rate": 0.00013525614940006184,
639
+ "loss": 0.8949,
640
+ "step": 440
641
+ },
642
+ {
643
+ "epoch": 1.36,
644
+ "grad_norm": 1.3063581428657738,
645
+ "learning_rate": 0.00013358974170215538,
646
+ "loss": 0.9174,
647
+ "step": 445
648
+ },
649
+ {
650
+ "epoch": 1.37,
651
+ "grad_norm": 1.0718302654560807,
652
+ "learning_rate": 0.00013191275247405527,
653
+ "loss": 0.8404,
654
+ "step": 450
655
+ },
656
+ {
657
+ "epoch": 1.39,
658
+ "grad_norm": 0.9302638253667144,
659
+ "learning_rate": 0.00013022571000538953,
660
+ "loss": 0.9538,
661
+ "step": 455
662
+ },
663
+ {
664
+ "epoch": 1.4,
665
+ "grad_norm": 1.002831930422535,
666
+ "learning_rate": 0.00012852914575278498,
667
+ "loss": 0.9497,
668
+ "step": 460
669
+ },
670
+ {
671
+ "epoch": 1.42,
672
+ "grad_norm": 1.0473392521685814,
673
+ "learning_rate": 0.0001268235941724463,
674
+ "loss": 0.9065,
675
+ "step": 465
676
+ },
677
+ {
678
+ "epoch": 1.43,
679
+ "grad_norm": 1.077577179849463,
680
+ "learning_rate": 0.00012510959255179006,
681
+ "loss": 0.9666,
682
+ "step": 470
683
+ },
684
+ {
685
+ "epoch": 1.45,
686
+ "grad_norm": 1.0906580828243861,
687
+ "learning_rate": 0.00012338768084018718,
688
+ "loss": 0.9149,
689
+ "step": 475
690
+ },
691
+ {
692
+ "epoch": 1.46,
693
+ "grad_norm": 1.1044210032346624,
694
+ "learning_rate": 0.00012165840147886656,
695
+ "loss": 0.9367,
696
+ "step": 480
697
+ },
698
+ {
699
+ "epoch": 1.48,
700
+ "grad_norm": 1.20379457933143,
701
+ "learning_rate": 0.00011992229923003377,
702
+ "loss": 0.9065,
703
+ "step": 485
704
+ },
705
+ {
706
+ "epoch": 1.49,
707
+ "grad_norm": 1.2198024955592683,
708
+ "learning_rate": 0.00011817992100525872,
709
+ "loss": 0.9161,
710
+ "step": 490
711
+ },
712
+ {
713
+ "epoch": 1.51,
714
+ "grad_norm": 0.9478289622764081,
715
+ "learning_rate": 0.00011643181569318595,
716
+ "loss": 0.9285,
717
+ "step": 495
718
+ },
719
+ {
720
+ "epoch": 1.52,
721
+ "grad_norm": 1.3556113362942024,
722
+ "learning_rate": 0.00011467853398662236,
723
+ "loss": 0.9149,
724
+ "step": 500
725
+ },
726
+ {
727
+ "epoch": 1.54,
728
+ "grad_norm": 1.0405819610764953,
729
+ "learning_rate": 0.00011292062820905651,
730
+ "loss": 0.8987,
731
+ "step": 505
732
+ },
733
+ {
734
+ "epoch": 1.55,
735
+ "grad_norm": 1.0272942715335152,
736
+ "learning_rate": 0.00011115865214066414,
737
+ "loss": 0.9296,
738
+ "step": 510
739
+ },
740
+ {
741
+ "epoch": 1.57,
742
+ "grad_norm": 1.235836010680135,
743
+ "learning_rate": 0.00010939316084385489,
744
+ "loss": 0.907,
745
+ "step": 515
746
+ },
747
+ {
748
+ "epoch": 1.59,
749
+ "grad_norm": 1.141603178243694,
750
+ "learning_rate": 0.00010762471048841501,
751
+ "loss": 0.9424,
752
+ "step": 520
753
+ },
754
+ {
755
+ "epoch": 1.6,
756
+ "grad_norm": 0.9315641865978828,
757
+ "learning_rate": 0.00010585385817630137,
758
+ "loss": 0.8801,
759
+ "step": 525
760
+ },
761
+ {
762
+ "epoch": 1.62,
763
+ "grad_norm": 1.2155191029089636,
764
+ "learning_rate": 0.00010408116176614167,
765
+ "loss": 0.9832,
766
+ "step": 530
767
+ },
768
+ {
769
+ "epoch": 1.63,
770
+ "grad_norm": 1.03073958762087,
771
+ "learning_rate": 0.00010230717969749616,
772
+ "loss": 0.9226,
773
+ "step": 535
774
+ },
775
+ {
776
+ "epoch": 1.65,
777
+ "grad_norm": 0.9081025102798024,
778
+ "learning_rate": 0.00010053247081493685,
779
+ "loss": 0.9383,
780
+ "step": 540
781
+ },
782
+ {
783
+ "epoch": 1.66,
784
+ "grad_norm": 0.9719608179885268,
785
+ "learning_rate": 9.875759419199848e-05,
786
+ "loss": 0.9139,
787
+ "step": 545
788
+ },
789
+ {
790
+ "epoch": 1.68,
791
+ "grad_norm": 0.9765085879374533,
792
+ "learning_rate": 9.698310895505784e-05,
793
+ "loss": 0.9546,
794
+ "step": 550
795
+ },
796
+ {
797
+ "epoch": 1.69,
798
+ "grad_norm": 0.9934690602517904,
799
+ "learning_rate": 9.520957410719632e-05,
800
+ "loss": 0.9085,
801
+ "step": 555
802
+ },
803
+ {
804
+ "epoch": 1.71,
805
+ "grad_norm": 1.0535937418956107,
806
+ "learning_rate": 9.343754835210106e-05,
807
+ "loss": 0.9058,
808
+ "step": 560
809
+ },
810
+ {
811
+ "epoch": 1.72,
812
+ "grad_norm": 1.023947025769762,
813
+ "learning_rate": 9.166758991806098e-05,
814
+ "loss": 0.9229,
815
+ "step": 565
816
+ },
817
+ {
818
+ "epoch": 1.74,
819
+ "grad_norm": 0.9292604570424309,
820
+ "learning_rate": 8.990025638211179e-05,
821
+ "loss": 0.9222,
822
+ "step": 570
823
+ },
824
+ {
825
+ "epoch": 1.75,
826
+ "grad_norm": 1.114153037641386,
827
+ "learning_rate": 8.813610449438692e-05,
828
+ "loss": 0.8626,
829
+ "step": 575
830
+ },
831
+ {
832
+ "epoch": 1.77,
833
+ "grad_norm": 1.040325901090811,
834
+ "learning_rate": 8.637569000272835e-05,
835
+ "loss": 0.9408,
836
+ "step": 580
837
+ },
838
+ {
839
+ "epoch": 1.78,
840
+ "grad_norm": 0.8817859269222795,
841
+ "learning_rate": 8.461956747761374e-05,
842
+ "loss": 0.8877,
843
+ "step": 585
844
+ },
845
+ {
846
+ "epoch": 1.8,
847
+ "grad_norm": 1.1284180630525456,
848
+ "learning_rate": 8.286829013745382e-05,
849
+ "loss": 0.9207,
850
+ "step": 590
851
+ },
852
+ {
853
+ "epoch": 1.81,
854
+ "grad_norm": 0.9343742775094561,
855
+ "learning_rate": 8.11224096743163e-05,
856
+ "loss": 0.8933,
857
+ "step": 595
858
+ },
859
+ {
860
+ "epoch": 1.83,
861
+ "grad_norm": 1.0538577448642068,
862
+ "learning_rate": 7.938247608013021e-05,
863
+ "loss": 0.9298,
864
+ "step": 600
865
+ },
866
+ {
867
+ "epoch": 1.84,
868
+ "grad_norm": 1.0018046757269787,
869
+ "learning_rate": 7.764903747342603e-05,
870
+ "loss": 0.9026,
871
+ "step": 605
872
+ },
873
+ {
874
+ "epoch": 1.86,
875
+ "grad_norm": 0.9082963605695086,
876
+ "learning_rate": 7.592263992666604e-05,
877
+ "loss": 0.8603,
878
+ "step": 610
879
+ },
880
+ {
881
+ "epoch": 1.88,
882
+ "grad_norm": 0.9123839165404105,
883
+ "learning_rate": 7.420382729421883e-05,
884
+ "loss": 0.9466,
885
+ "step": 615
886
+ },
887
+ {
888
+ "epoch": 1.89,
889
+ "grad_norm": 0.8612729466211886,
890
+ "learning_rate": 7.249314104103315e-05,
891
+ "loss": 0.9053,
892
+ "step": 620
893
+ },
894
+ {
895
+ "epoch": 1.91,
896
+ "grad_norm": 0.9068865831245562,
897
+ "learning_rate": 7.079112007206394e-05,
898
+ "loss": 0.9099,
899
+ "step": 625
900
+ },
901
+ {
902
+ "epoch": 1.92,
903
+ "grad_norm": 0.9654244669285843,
904
+ "learning_rate": 6.909830056250527e-05,
905
+ "loss": 0.922,
906
+ "step": 630
907
+ },
908
+ {
909
+ "epoch": 1.94,
910
+ "grad_norm": 0.8439836577250905,
911
+ "learning_rate": 6.741521578888272e-05,
912
+ "loss": 0.9283,
913
+ "step": 635
914
+ },
915
+ {
916
+ "epoch": 1.95,
917
+ "grad_norm": 1.0735235072492841,
918
+ "learning_rate": 6.574239596105951e-05,
919
+ "loss": 0.9292,
920
+ "step": 640
921
+ },
922
+ {
923
+ "epoch": 1.97,
924
+ "grad_norm": 1.1672700848634472,
925
+ "learning_rate": 6.408036805520801e-05,
926
+ "loss": 0.8986,
927
+ "step": 645
928
+ },
929
+ {
930
+ "epoch": 1.98,
931
+ "grad_norm": 0.952568318833893,
932
+ "learning_rate": 6.242965564780071e-05,
933
+ "loss": 0.8812,
934
+ "step": 650
935
+ },
936
+ {
937
+ "epoch": 2.0,
938
+ "grad_norm": 1.1182556181299996,
939
+ "learning_rate": 6.079077875067136e-05,
940
+ "loss": 0.9232,
941
+ "step": 655
942
+ },
943
+ {
944
+ "epoch": 2.0,
945
+ "eval_loss": 1.1174074411392212,
946
+ "eval_runtime": 155.7531,
947
+ "eval_samples_per_second": 14.831,
948
+ "eval_steps_per_second": 0.469,
949
+ "step": 656
950
+ },
951
+ {
952
+ "epoch": 2.01,
953
+ "grad_norm": 1.0745550951971172,
954
+ "learning_rate": 5.916425364719975e-05,
955
+ "loss": 0.8609,
956
+ "step": 660
957
+ },
958
+ {
959
+ "epoch": 2.03,
960
+ "grad_norm": 0.9414624320547766,
961
+ "learning_rate": 5.755059272967054e-05,
962
+ "loss": 0.8921,
963
+ "step": 665
964
+ },
965
+ {
966
+ "epoch": 2.04,
967
+ "grad_norm": 0.9382686352429901,
968
+ "learning_rate": 5.5950304337858176e-05,
969
+ "loss": 0.8712,
970
+ "step": 670
971
+ },
972
+ {
973
+ "epoch": 2.06,
974
+ "grad_norm": 1.0370060367806875,
975
+ "learning_rate": 5.436389259888841e-05,
976
+ "loss": 0.9143,
977
+ "step": 675
978
+ },
979
+ {
980
+ "epoch": 2.07,
981
+ "grad_norm": 1.0449428824200535,
982
+ "learning_rate": 5.279185726842658e-05,
983
+ "loss": 0.8609,
984
+ "step": 680
985
+ },
986
+ {
987
+ "epoch": 2.09,
988
+ "grad_norm": 1.2140796388499904,
989
+ "learning_rate": 5.1234693573243554e-05,
990
+ "loss": 0.8814,
991
+ "step": 685
992
+ },
993
+ {
994
+ "epoch": 2.1,
995
+ "grad_norm": 1.0118415907930658,
996
+ "learning_rate": 4.969289205520778e-05,
997
+ "loss": 0.8797,
998
+ "step": 690
999
+ },
1000
+ {
1001
+ "epoch": 2.12,
1002
+ "grad_norm": 1.0784940351527341,
1003
+ "learning_rate": 4.816693841675368e-05,
1004
+ "loss": 0.8898,
1005
+ "step": 695
1006
+ },
1007
+ {
1008
+ "epoch": 2.13,
1009
+ "grad_norm": 1.231043264244889,
1010
+ "learning_rate": 4.6657313367874256e-05,
1011
+ "loss": 0.8772,
1012
+ "step": 700
1013
+ },
1014
+ {
1015
+ "epoch": 2.15,
1016
+ "grad_norm": 0.93091465787962,
1017
+ "learning_rate": 4.516449247468666e-05,
1018
+ "loss": 0.9097,
1019
+ "step": 705
1020
+ },
1021
+ {
1022
+ "epoch": 2.16,
1023
+ "grad_norm": 1.1113936820092343,
1024
+ "learning_rate": 4.368894600961792e-05,
1025
+ "loss": 0.8757,
1026
+ "step": 710
1027
+ },
1028
+ {
1029
+ "epoch": 2.18,
1030
+ "grad_norm": 0.9533707648230687,
1031
+ "learning_rate": 4.223113880325865e-05,
1032
+ "loss": 0.872,
1033
+ "step": 715
1034
+ },
1035
+ {
1036
+ "epoch": 2.2,
1037
+ "grad_norm": 1.0584356613150987,
1038
+ "learning_rate": 4.0791530097930676e-05,
1039
+ "loss": 0.848,
1040
+ "step": 720
1041
+ },
1042
+ {
1043
+ "epoch": 2.21,
1044
+ "grad_norm": 1.0955157548141539,
1045
+ "learning_rate": 3.937057340301551e-05,
1046
+ "loss": 0.8831,
1047
+ "step": 725
1048
+ },
1049
+ {
1050
+ "epoch": 2.23,
1051
+ "grad_norm": 1.0974350105007893,
1052
+ "learning_rate": 3.7968716352088406e-05,
1053
+ "loss": 0.8965,
1054
+ "step": 730
1055
+ },
1056
+ {
1057
+ "epoch": 2.24,
1058
+ "grad_norm": 1.07417322379521,
1059
+ "learning_rate": 3.658640056190378e-05,
1060
+ "loss": 0.8024,
1061
+ "step": 735
1062
+ },
1063
+ {
1064
+ "epoch": 2.26,
1065
+ "grad_norm": 0.932073914233341,
1066
+ "learning_rate": 3.52240614932758e-05,
1067
+ "loss": 0.859,
1068
+ "step": 740
1069
+ },
1070
+ {
1071
+ "epoch": 2.27,
1072
+ "grad_norm": 0.960220324993126,
1073
+ "learning_rate": 3.388212831389854e-05,
1074
+ "loss": 0.8955,
1075
+ "step": 745
1076
+ },
1077
+ {
1078
+ "epoch": 2.29,
1079
+ "grad_norm": 0.9981005413964692,
1080
+ "learning_rate": 3.256102376314824e-05,
1081
+ "loss": 0.94,
1082
+ "step": 750
1083
+ },
1084
+ {
1085
+ "epoch": 2.3,
1086
+ "grad_norm": 1.0397017457559017,
1087
+ "learning_rate": 3.126116401891085e-05,
1088
+ "loss": 0.8587,
1089
+ "step": 755
1090
+ },
1091
+ {
1092
+ "epoch": 2.32,
1093
+ "grad_norm": 0.9448895278738276,
1094
+ "learning_rate": 2.9982958566476705e-05,
1095
+ "loss": 0.8707,
1096
+ "step": 760
1097
+ },
1098
+ {
1099
+ "epoch": 2.33,
1100
+ "grad_norm": 1.0610274165270486,
1101
+ "learning_rate": 2.872681006954315e-05,
1102
+ "loss": 0.8825,
1103
+ "step": 765
1104
+ },
1105
+ {
1106
+ "epoch": 2.35,
1107
+ "grad_norm": 1.0210002250645476,
1108
+ "learning_rate": 2.749311424336659e-05,
1109
+ "loss": 0.8915,
1110
+ "step": 770
1111
+ },
1112
+ {
1113
+ "epoch": 2.36,
1114
+ "grad_norm": 0.9667083953810022,
1115
+ "learning_rate": 2.6282259730103e-05,
1116
+ "loss": 0.8727,
1117
+ "step": 775
1118
+ },
1119
+ {
1120
+ "epoch": 2.38,
1121
+ "grad_norm": 0.9251306242277755,
1122
+ "learning_rate": 2.5094627976376927e-05,
1123
+ "loss": 0.8887,
1124
+ "step": 780
1125
+ },
1126
+ {
1127
+ "epoch": 2.39,
1128
+ "grad_norm": 0.9383384405043022,
1129
+ "learning_rate": 2.393059311311715e-05,
1130
+ "loss": 0.8894,
1131
+ "step": 785
1132
+ },
1133
+ {
1134
+ "epoch": 2.41,
1135
+ "grad_norm": 0.8717419970020719,
1136
+ "learning_rate": 2.2790521837697034e-05,
1137
+ "loss": 0.8279,
1138
+ "step": 790
1139
+ },
1140
+ {
1141
+ "epoch": 2.42,
1142
+ "grad_norm": 0.9486113739115168,
1143
+ "learning_rate": 2.167477329841633e-05,
1144
+ "loss": 0.8545,
1145
+ "step": 795
1146
+ },
1147
+ {
1148
+ "epoch": 2.44,
1149
+ "grad_norm": 0.9988727624604463,
1150
+ "learning_rate": 2.0583698981361577e-05,
1151
+ "loss": 0.8624,
1152
+ "step": 800
1153
+ },
1154
+ {
1155
+ "epoch": 2.45,
1156
+ "grad_norm": 0.9856023900634717,
1157
+ "learning_rate": 1.9517642599679807e-05,
1158
+ "loss": 0.8953,
1159
+ "step": 805
1160
+ },
1161
+ {
1162
+ "epoch": 2.47,
1163
+ "grad_norm": 1.146341438442503,
1164
+ "learning_rate": 1.8476939985301256e-05,
1165
+ "loss": 0.8583,
1166
+ "step": 810
1167
+ },
1168
+ {
1169
+ "epoch": 2.48,
1170
+ "grad_norm": 0.9934930089425651,
1171
+ "learning_rate": 1.7461918983144588e-05,
1172
+ "loss": 0.8236,
1173
+ "step": 815
1174
+ },
1175
+ {
1176
+ "epoch": 2.5,
1177
+ "grad_norm": 1.0570151368189196,
1178
+ "learning_rate": 1.6472899347838356e-05,
1179
+ "loss": 0.8521,
1180
+ "step": 820
1181
+ },
1182
+ {
1183
+ "epoch": 2.52,
1184
+ "grad_norm": 0.940846894356948,
1185
+ "learning_rate": 1.5510192642991073e-05,
1186
+ "loss": 0.8412,
1187
+ "step": 825
1188
+ },
1189
+ {
1190
+ "epoch": 2.53,
1191
+ "grad_norm": 0.9182797173905372,
1192
+ "learning_rate": 1.4574102143041512e-05,
1193
+ "loss": 0.8292,
1194
+ "step": 830
1195
+ },
1196
+ {
1197
+ "epoch": 2.55,
1198
+ "grad_norm": 1.0514937418330759,
1199
+ "learning_rate": 1.3664922737720586e-05,
1200
+ "loss": 0.8463,
1201
+ "step": 835
1202
+ },
1203
+ {
1204
+ "epoch": 2.56,
1205
+ "grad_norm": 0.9672420613383418,
1206
+ "learning_rate": 1.2782940839154111e-05,
1207
+ "loss": 0.8677,
1208
+ "step": 840
1209
+ },
1210
+ {
1211
+ "epoch": 2.58,
1212
+ "grad_norm": 1.0215480963434165,
1213
+ "learning_rate": 1.192843429163677e-05,
1214
+ "loss": 0.8555,
1215
+ "step": 845
1216
+ },
1217
+ {
1218
+ "epoch": 2.59,
1219
+ "grad_norm": 1.133368133421988,
1220
+ "learning_rate": 1.1101672284104624e-05,
1221
+ "loss": 0.8732,
1222
+ "step": 850
1223
+ },
1224
+ {
1225
+ "epoch": 2.61,
1226
+ "grad_norm": 0.9880449036656486,
1227
+ "learning_rate": 1.0302915265334723e-05,
1228
+ "loss": 0.8832,
1229
+ "step": 855
1230
+ },
1231
+ {
1232
+ "epoch": 2.62,
1233
+ "grad_norm": 0.934408626896597,
1234
+ "learning_rate": 9.532414861897632e-06,
1235
+ "loss": 0.8527,
1236
+ "step": 860
1237
+ },
1238
+ {
1239
+ "epoch": 2.64,
1240
+ "grad_norm": 0.90008489968748,
1241
+ "learning_rate": 8.790413798889452e-06,
1242
+ "loss": 0.8611,
1243
+ "step": 865
1244
+ },
1245
+ {
1246
+ "epoch": 2.65,
1247
+ "grad_norm": 0.9910034914102378,
1248
+ "learning_rate": 8.077145823467924e-06,
1249
+ "loss": 0.932,
1250
+ "step": 870
1251
+ },
1252
+ {
1253
+ "epoch": 2.67,
1254
+ "grad_norm": 1.1138535634921578,
1255
+ "learning_rate": 7.392835631216766e-06,
1256
+ "loss": 0.878,
1257
+ "step": 875
1258
+ },
1259
+ {
1260
+ "epoch": 2.68,
1261
+ "grad_norm": 1.1422833234291747,
1262
+ "learning_rate": 6.7376987953614246e-06,
1263
+ "loss": 0.8607,
1264
+ "step": 880
1265
+ },
1266
+ {
1267
+ "epoch": 2.7,
1268
+ "grad_norm": 1.1003844781363714,
1269
+ "learning_rate": 6.111941698858681e-06,
1270
+ "loss": 0.912,
1271
+ "step": 885
1272
+ },
1273
+ {
1274
+ "epoch": 2.71,
1275
+ "grad_norm": 1.137696707362488,
1276
+ "learning_rate": 5.5157614693812865e-06,
1277
+ "loss": 0.8416,
1278
+ "step": 890
1279
+ },
1280
+ {
1281
+ "epoch": 2.73,
1282
+ "grad_norm": 0.9640103484120158,
1283
+ "learning_rate": 4.9493459172183285e-06,
1284
+ "loss": 0.8452,
1285
+ "step": 895
1286
+ },
1287
+ {
1288
+ "epoch": 2.74,
1289
+ "grad_norm": 0.9223029933937373,
1290
+ "learning_rate": 4.412873476110702e-06,
1291
+ "loss": 0.8568,
1292
+ "step": 900
1293
+ },
1294
+ {
1295
+ "epoch": 2.76,
1296
+ "grad_norm": 0.9007733331158723,
1297
+ "learning_rate": 3.906513147040425e-06,
1298
+ "loss": 0.8713,
1299
+ "step": 905
1300
+ },
1301
+ {
1302
+ "epoch": 2.77,
1303
+ "grad_norm": 1.0288738184384674,
1304
+ "learning_rate": 3.43042444499152e-06,
1305
+ "loss": 0.9235,
1306
+ "step": 910
1307
+ },
1308
+ {
1309
+ "epoch": 2.79,
1310
+ "grad_norm": 1.0184524502121122,
1311
+ "learning_rate": 2.984757348699152e-06,
1312
+ "loss": 0.846,
1313
+ "step": 915
1314
+ },
1315
+ {
1316
+ "epoch": 2.8,
1317
+ "grad_norm": 1.1092808565987327,
1318
+ "learning_rate": 2.569652253402999e-06,
1319
+ "loss": 0.8701,
1320
+ "step": 920
1321
+ },
1322
+ {
1323
+ "epoch": 2.82,
1324
+ "grad_norm": 0.9711566457168485,
1325
+ "learning_rate": 2.1852399266194314e-06,
1326
+ "loss": 0.84,
1327
+ "step": 925
1328
+ },
1329
+ {
1330
+ "epoch": 2.84,
1331
+ "grad_norm": 1.0636030655856668,
1332
+ "learning_rate": 1.8316414669469539e-06,
1333
+ "loss": 0.8756,
1334
+ "step": 930
1335
+ },
1336
+ {
1337
+ "epoch": 2.85,
1338
+ "grad_norm": 1.062584489248108,
1339
+ "learning_rate": 1.5089682659172321e-06,
1340
+ "loss": 0.848,
1341
+ "step": 935
1342
+ },
1343
+ {
1344
+ "epoch": 2.87,
1345
+ "grad_norm": 1.109498548513231,
1346
+ "learning_rate": 1.2173219729043507e-06,
1347
+ "loss": 0.8488,
1348
+ "step": 940
1349
+ },
1350
+ {
1351
+ "epoch": 2.88,
1352
+ "grad_norm": 0.9140768109712519,
1353
+ "learning_rate": 9.567944631029169e-07,
1354
+ "loss": 0.8614,
1355
+ "step": 945
1356
+ },
1357
+ {
1358
+ "epoch": 2.9,
1359
+ "grad_norm": 0.9332248724423268,
1360
+ "learning_rate": 7.274678085852693e-07,
1361
+ "loss": 0.8839,
1362
+ "step": 950
1363
+ },
1364
+ {
1365
+ "epoch": 2.91,
1366
+ "grad_norm": 0.9421812832473311,
1367
+ "learning_rate": 5.294142524469359e-07,
1368
+ "loss": 0.8691,
1369
+ "step": 955
1370
+ },
1371
+ {
1372
+ "epoch": 2.93,
1373
+ "grad_norm": 1.0718790701802063,
1374
+ "learning_rate": 3.6269618604847233e-07,
1375
+ "loss": 0.8243,
1376
+ "step": 960
1377
+ },
1378
+ {
1379
+ "epoch": 2.94,
1380
+ "grad_norm": 0.996460942018122,
1381
+ "learning_rate": 2.2736612936065106e-07,
1382
+ "loss": 0.842,
1383
+ "step": 965
1384
+ },
1385
+ {
1386
+ "epoch": 2.96,
1387
+ "grad_norm": 1.1337948192404557,
1388
+ "learning_rate": 1.2346671441958447e-07,
1389
+ "loss": 0.8646,
1390
+ "step": 970
1391
+ },
1392
+ {
1393
+ "epoch": 2.97,
1394
+ "grad_norm": 0.9562746029081519,
1395
+ "learning_rate": 5.103067189662358e-08,
1396
+ "loss": 0.8854,
1397
+ "step": 975
1398
+ },
1399
+ {
1400
+ "epoch": 2.99,
1401
+ "grad_norm": 0.9192847747786613,
1402
+ "learning_rate": 1.0080820787450319e-08,
1403
+ "loss": 0.8492,
1404
+ "step": 980
1405
+ },
1406
+ {
1407
+ "epoch": 3.0,
1408
+ "eval_loss": 1.1201515197753906,
1409
+ "eval_runtime": 155.5537,
1410
+ "eval_samples_per_second": 14.85,
1411
+ "eval_steps_per_second": 0.469,
1412
+ "step": 984
1413
+ },
1414
+ {
1415
+ "epoch": 3.0,
1416
+ "step": 984,
1417
+ "total_flos": 2528700107063296.0,
1418
+ "train_loss": 1.2712977070633957,
1419
+ "train_runtime": 11725.5058,
1420
+ "train_samples_per_second": 5.365,
1421
+ "train_steps_per_second": 0.084
1422
+ }
1423
+ ],
1424
+ "logging_steps": 5,
1425
+ "max_steps": 984,
1426
+ "num_input_tokens_seen": 0,
1427
+ "num_train_epochs": 3,
1428
+ "save_steps": 500,
1429
+ "total_flos": 2528700107063296.0,
1430
+ "train_batch_size": 8,
1431
+ "trial_name": null,
1432
+ "trial_params": null
1433
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:de480259f04a252180e5041fdbb2ec4e11d8fd026c2b1bea4a6221c3a2214abd
3
+ size 6136