Ritvik19 commited on
Commit
879fd07
1 Parent(s): 52db434

Upload 13 files

Browse files
README.md CHANGED
@@ -1,3 +1,52 @@
1
- ---
2
- license: apache-2.0
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ library_name: peft
4
+ tags:
5
+ - alignment-handbook
6
+ - generated_from_trainer
7
+ base_model: h2oai/h2o-danube2-1.8b-base
8
+ datasets:
9
+ - HuggingFaceH4/ultrachat_200k
10
+ model-index:
11
+ - name: zephyr-danube-sft-qlora
12
+ results: []
13
+ ---
14
+
15
+ **Note**: This model card has been generated automatically according to the information the Trainer had access to.
16
+ Visit the [model card](https://ritvik19.github.io/small-llms/) to see the full description
17
+
18
+ # zephyr-danube-sft-qlora
19
+
20
+ This model is a fine-tuned version of [h2oai/h2o-danube2-1.8b-base](https://huggingface.co/h2oai/h2o-danube2-1.8b-base) on the HuggingFaceH4/ultrachat_200k dataset.
21
+ It achieves the following results on the evaluation set:
22
+ - Loss: 1.0677
23
+
24
+ ### Training hyperparameters
25
+
26
+ The following hyperparameters were used during training:
27
+ - learning_rate: 0.0002
28
+ - train_batch_size: 1
29
+ - eval_batch_size: 1
30
+ - seed: 42
31
+ - distributed_type: multi-GPU
32
+ - gradient_accumulation_steps: 128
33
+ - total_train_batch_size: 128
34
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
35
+ - lr_scheduler_type: cosine
36
+ - lr_scheduler_warmup_ratio: 0.1
37
+ - num_epochs: 1
38
+
39
+ ### Training results
40
+
41
+ | Training Loss | Epoch | Step | Validation Loss |
42
+ |:-------------:|:------:|:----:|:---------------:|
43
+ | 1.0546 | 0.9996 | 1088 | 1.0677 |
44
+
45
+
46
+ ### Framework versions
47
+
48
+ - PEFT 0.7.1
49
+ - Transformers 4.40.1
50
+ - Pytorch 2.1.2+cu121
51
+ - Datasets 2.19.0
52
+ - Tokenizers 0.19.1
adapter_config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "h2oai/h2o-danube2-1.8b-base",
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": 16,
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": 16,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "v_proj",
23
+ "o_proj",
24
+ "gate_proj",
25
+ "q_proj",
26
+ "k_proj",
27
+ "down_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 sha256:0f755b6def912b7a84d2f14a95eae2c2062b17b7194b507b83dc0be01e3559a6
3
+ size 34647816
all_results.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 0.9995621747712183,
3
+ "eval_loss": 1.0676658153533936,
4
+ "eval_runtime": 1919.0013,
5
+ "eval_samples": 23109,
6
+ "eval_samples_per_second": 8.036,
7
+ "eval_steps_per_second": 8.036,
8
+ "total_flos": 3.023110834516656e+18,
9
+ "train_loss": 0.22257287449696483,
10
+ "train_runtime": 15104.5697,
11
+ "train_samples": 207864,
12
+ "train_samples_per_second": 9.224,
13
+ "train_steps_per_second": 0.072
14
+ }
config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "h2oai/h2o-danube2-1.8b-base",
3
+ "architectures": [
4
+ "MistralForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 1,
8
+ "eos_token_id": 2,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 2560,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 6912,
13
+ "max_position_embeddings": 8192,
14
+ "model_type": "mistral",
15
+ "num_attention_heads": 32,
16
+ "num_hidden_layers": 24,
17
+ "num_key_value_heads": 8,
18
+ "rms_norm_eps": 1e-05,
19
+ "rope_theta": 10000,
20
+ "sliding_window": null,
21
+ "tie_word_embeddings": false,
22
+ "torch_dtype": "bfloat16",
23
+ "transformers_version": "4.40.1",
24
+ "use_cache": true,
25
+ "vocab_size": 32000
26
+ }
eval_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 0.9995621747712183,
3
+ "eval_loss": 1.0676658153533936,
4
+ "eval_runtime": 1919.0013,
5
+ "eval_samples": 23109,
6
+ "eval_samples_per_second": 8.036,
7
+ "eval_steps_per_second": 8.036
8
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "</s>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
3
+ size 493443
tokenizer_config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_eos_token": false,
4
+ "added_tokens_decoder": {
5
+ "0": {
6
+ "content": "<unk>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "1": {
14
+ "content": "<s>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "2": {
22
+ "content": "</s>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ }
29
+ },
30
+ "additional_special_tokens": [],
31
+ "bos_token": "<s>",
32
+ "chat_template": "{% for message in messages %}\n{% if message['role'] == 'user' %}\n{{ '<|user|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'system' %}\n{{ '<|system|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'assistant' %}\n{{ '<|assistant|>\n' + message['content'] + eos_token }}\n{% endif %}\n{% if loop.last and add_generation_prompt %}\n{{ '<|assistant|>' }}\n{% endif %}\n{% endfor %}",
33
+ "clean_up_tokenization_spaces": false,
34
+ "eos_token": "</s>",
35
+ "legacy": true,
36
+ "model_max_length": 2048,
37
+ "pad_token": "</s>",
38
+ "sp_model_kwargs": {},
39
+ "spaces_between_special_tokens": false,
40
+ "tokenizer_class": "LlamaTokenizer",
41
+ "unk_token": "<unk>",
42
+ "use_default_system_prompt": false
43
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 0.9995621747712183,
3
+ "total_flos": 3.023110834516656e+18,
4
+ "train_loss": 0.22257287449696483,
5
+ "train_runtime": 15104.5697,
6
+ "train_samples": 207864,
7
+ "train_samples_per_second": 9.224,
8
+ "train_steps_per_second": 0.072
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,1564 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.9995621747712183,
5
+ "eval_steps": 500,
6
+ "global_step": 1088,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0009187152341647227,
13
+ "grad_norm": 0.12890625,
14
+ "learning_rate": 1.8348623853211011e-06,
15
+ "loss": 1.1775,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.004593576170823614,
20
+ "grad_norm": 0.1240234375,
21
+ "learning_rate": 9.174311926605506e-06,
22
+ "loss": 1.1877,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.009187152341647228,
27
+ "grad_norm": 0.1298828125,
28
+ "learning_rate": 1.834862385321101e-05,
29
+ "loss": 1.1744,
30
+ "step": 10
31
+ },
32
+ {
33
+ "epoch": 0.01378072851247084,
34
+ "grad_norm": 0.11865234375,
35
+ "learning_rate": 2.7522935779816515e-05,
36
+ "loss": 1.1666,
37
+ "step": 15
38
+ },
39
+ {
40
+ "epoch": 0.018374304683294457,
41
+ "grad_norm": 0.0947265625,
42
+ "learning_rate": 3.669724770642202e-05,
43
+ "loss": 1.1771,
44
+ "step": 20
45
+ },
46
+ {
47
+ "epoch": 0.02296788085411807,
48
+ "grad_norm": 0.083984375,
49
+ "learning_rate": 4.587155963302753e-05,
50
+ "loss": 1.1437,
51
+ "step": 25
52
+ },
53
+ {
54
+ "epoch": 0.02756145702494168,
55
+ "grad_norm": 0.07080078125,
56
+ "learning_rate": 5.504587155963303e-05,
57
+ "loss": 1.1373,
58
+ "step": 30
59
+ },
60
+ {
61
+ "epoch": 0.0321550331957653,
62
+ "grad_norm": 0.0517578125,
63
+ "learning_rate": 6.422018348623854e-05,
64
+ "loss": 1.0934,
65
+ "step": 35
66
+ },
67
+ {
68
+ "epoch": 0.036748609366588914,
69
+ "grad_norm": 0.0439453125,
70
+ "learning_rate": 7.339449541284404e-05,
71
+ "loss": 1.0912,
72
+ "step": 40
73
+ },
74
+ {
75
+ "epoch": 0.04134218553741253,
76
+ "grad_norm": 0.041748046875,
77
+ "learning_rate": 8.256880733944955e-05,
78
+ "loss": 1.1026,
79
+ "step": 45
80
+ },
81
+ {
82
+ "epoch": 0.04593576170823614,
83
+ "grad_norm": 0.0380859375,
84
+ "learning_rate": 9.174311926605506e-05,
85
+ "loss": 1.0824,
86
+ "step": 50
87
+ },
88
+ {
89
+ "epoch": 0.05052933787905975,
90
+ "grad_norm": 0.03369140625,
91
+ "learning_rate": 0.00010091743119266055,
92
+ "loss": 1.1075,
93
+ "step": 55
94
+ },
95
+ {
96
+ "epoch": 0.05512291404988336,
97
+ "grad_norm": 0.034423828125,
98
+ "learning_rate": 0.00011009174311926606,
99
+ "loss": 1.1073,
100
+ "step": 60
101
+ },
102
+ {
103
+ "epoch": 0.05971649022070698,
104
+ "grad_norm": 0.0322265625,
105
+ "learning_rate": 0.00011926605504587157,
106
+ "loss": 1.0848,
107
+ "step": 65
108
+ },
109
+ {
110
+ "epoch": 0.0643100663915306,
111
+ "grad_norm": 0.0294189453125,
112
+ "learning_rate": 0.00012844036697247707,
113
+ "loss": 1.072,
114
+ "step": 70
115
+ },
116
+ {
117
+ "epoch": 0.0689036425623542,
118
+ "grad_norm": 0.0311279296875,
119
+ "learning_rate": 0.00013761467889908258,
120
+ "loss": 1.0763,
121
+ "step": 75
122
+ },
123
+ {
124
+ "epoch": 0.07349721873317783,
125
+ "grad_norm": 0.0294189453125,
126
+ "learning_rate": 0.0001467889908256881,
127
+ "loss": 1.0737,
128
+ "step": 80
129
+ },
130
+ {
131
+ "epoch": 0.07809079490400143,
132
+ "grad_norm": 0.0308837890625,
133
+ "learning_rate": 0.0001559633027522936,
134
+ "loss": 1.0564,
135
+ "step": 85
136
+ },
137
+ {
138
+ "epoch": 0.08268437107482506,
139
+ "grad_norm": 0.030517578125,
140
+ "learning_rate": 0.0001651376146788991,
141
+ "loss": 1.0754,
142
+ "step": 90
143
+ },
144
+ {
145
+ "epoch": 0.08727794724564866,
146
+ "grad_norm": 0.0296630859375,
147
+ "learning_rate": 0.00017431192660550458,
148
+ "loss": 1.0635,
149
+ "step": 95
150
+ },
151
+ {
152
+ "epoch": 0.09187152341647228,
153
+ "grad_norm": 0.02978515625,
154
+ "learning_rate": 0.00018348623853211012,
155
+ "loss": 1.0757,
156
+ "step": 100
157
+ },
158
+ {
159
+ "epoch": 0.09646509958729589,
160
+ "grad_norm": 0.0299072265625,
161
+ "learning_rate": 0.0001926605504587156,
162
+ "loss": 1.0715,
163
+ "step": 105
164
+ },
165
+ {
166
+ "epoch": 0.1010586757581195,
167
+ "grad_norm": 0.0301513671875,
168
+ "learning_rate": 0.00019999948512240548,
169
+ "loss": 1.0733,
170
+ "step": 110
171
+ },
172
+ {
173
+ "epoch": 0.10565225192894312,
174
+ "grad_norm": 0.0289306640625,
175
+ "learning_rate": 0.00019998146496329837,
176
+ "loss": 1.0891,
177
+ "step": 115
178
+ },
179
+ {
180
+ "epoch": 0.11024582809976673,
181
+ "grad_norm": 0.03125,
182
+ "learning_rate": 0.00019993770622619782,
183
+ "loss": 1.0702,
184
+ "step": 120
185
+ },
186
+ {
187
+ "epoch": 0.11483940427059035,
188
+ "grad_norm": 0.033447265625,
189
+ "learning_rate": 0.00019986822017606848,
190
+ "loss": 1.0861,
191
+ "step": 125
192
+ },
193
+ {
194
+ "epoch": 0.11943298044141396,
195
+ "grad_norm": 0.0294189453125,
196
+ "learning_rate": 0.00019977302470094708,
197
+ "loss": 1.0615,
198
+ "step": 130
199
+ },
200
+ {
201
+ "epoch": 0.12402655661223758,
202
+ "grad_norm": 0.0296630859375,
203
+ "learning_rate": 0.00019965214430733754,
204
+ "loss": 1.0673,
205
+ "step": 135
206
+ },
207
+ {
208
+ "epoch": 0.1286201327830612,
209
+ "grad_norm": 0.031982421875,
210
+ "learning_rate": 0.00019950561011390213,
211
+ "loss": 1.0648,
212
+ "step": 140
213
+ },
214
+ {
215
+ "epoch": 0.1332137089538848,
216
+ "grad_norm": 0.031494140625,
217
+ "learning_rate": 0.00019933345984345037,
218
+ "loss": 1.0685,
219
+ "step": 145
220
+ },
221
+ {
222
+ "epoch": 0.1378072851247084,
223
+ "grad_norm": 0.0301513671875,
224
+ "learning_rate": 0.00019913573781322818,
225
+ "loss": 1.0663,
226
+ "step": 150
227
+ },
228
+ {
229
+ "epoch": 0.14240086129553203,
230
+ "grad_norm": 0.0291748046875,
231
+ "learning_rate": 0.00019891249492350887,
232
+ "loss": 1.079,
233
+ "step": 155
234
+ },
235
+ {
236
+ "epoch": 0.14699443746635565,
237
+ "grad_norm": 0.03076171875,
238
+ "learning_rate": 0.00019866378864448985,
239
+ "loss": 1.0959,
240
+ "step": 160
241
+ },
242
+ {
243
+ "epoch": 0.15158801363717925,
244
+ "grad_norm": 0.030029296875,
245
+ "learning_rate": 0.00019838968300149782,
246
+ "loss": 1.0805,
247
+ "step": 165
248
+ },
249
+ {
250
+ "epoch": 0.15618158980800287,
251
+ "grad_norm": 0.0302734375,
252
+ "learning_rate": 0.00019809024855850662,
253
+ "loss": 1.0644,
254
+ "step": 170
255
+ },
256
+ {
257
+ "epoch": 0.1607751659788265,
258
+ "grad_norm": 0.02978515625,
259
+ "learning_rate": 0.00019776556239997146,
260
+ "loss": 1.0657,
261
+ "step": 175
262
+ },
263
+ {
264
+ "epoch": 0.1653687421496501,
265
+ "grad_norm": 0.0289306640625,
266
+ "learning_rate": 0.000197415708110985,
267
+ "loss": 1.0559,
268
+ "step": 180
269
+ },
270
+ {
271
+ "epoch": 0.1699623183204737,
272
+ "grad_norm": 0.02978515625,
273
+ "learning_rate": 0.00019704077575575978,
274
+ "loss": 1.0507,
275
+ "step": 185
276
+ },
277
+ {
278
+ "epoch": 0.17455589449129733,
279
+ "grad_norm": 0.031005859375,
280
+ "learning_rate": 0.00019664086185444246,
281
+ "loss": 1.1005,
282
+ "step": 190
283
+ },
284
+ {
285
+ "epoch": 0.17914947066212095,
286
+ "grad_norm": 0.02978515625,
287
+ "learning_rate": 0.0001962160693582665,
288
+ "loss": 1.0795,
289
+ "step": 195
290
+ },
291
+ {
292
+ "epoch": 0.18374304683294457,
293
+ "grad_norm": 0.0291748046875,
294
+ "learning_rate": 0.00019576650762304903,
295
+ "loss": 1.0786,
296
+ "step": 200
297
+ },
298
+ {
299
+ "epoch": 0.18833662300376816,
300
+ "grad_norm": 0.0284423828125,
301
+ "learning_rate": 0.00019529229238103883,
302
+ "loss": 1.0602,
303
+ "step": 205
304
+ },
305
+ {
306
+ "epoch": 0.19293019917459178,
307
+ "grad_norm": 0.0301513671875,
308
+ "learning_rate": 0.00019479354571112323,
309
+ "loss": 1.072,
310
+ "step": 210
311
+ },
312
+ {
313
+ "epoch": 0.1975237753454154,
314
+ "grad_norm": 0.0286865234375,
315
+ "learning_rate": 0.00019427039600740072,
316
+ "loss": 1.0903,
317
+ "step": 215
318
+ },
319
+ {
320
+ "epoch": 0.202117351516239,
321
+ "grad_norm": 0.03076171875,
322
+ "learning_rate": 0.00019372297794612817,
323
+ "loss": 1.0734,
324
+ "step": 220
325
+ },
326
+ {
327
+ "epoch": 0.20671092768706262,
328
+ "grad_norm": 0.029541015625,
329
+ "learning_rate": 0.00019315143245105047,
330
+ "loss": 1.0638,
331
+ "step": 225
332
+ },
333
+ {
334
+ "epoch": 0.21130450385788624,
335
+ "grad_norm": 0.0302734375,
336
+ "learning_rate": 0.00019255590665712214,
337
+ "loss": 1.0663,
338
+ "step": 230
339
+ },
340
+ {
341
+ "epoch": 0.21589808002870986,
342
+ "grad_norm": 0.029052734375,
343
+ "learning_rate": 0.00019193655387262984,
344
+ "loss": 1.0778,
345
+ "step": 235
346
+ },
347
+ {
348
+ "epoch": 0.22049165619953345,
349
+ "grad_norm": 0.02978515625,
350
+ "learning_rate": 0.00019129353353972581,
351
+ "loss": 1.0732,
352
+ "step": 240
353
+ },
354
+ {
355
+ "epoch": 0.22508523237035707,
356
+ "grad_norm": 0.029052734375,
357
+ "learning_rate": 0.00019062701119338185,
358
+ "loss": 1.0463,
359
+ "step": 245
360
+ },
361
+ {
362
+ "epoch": 0.2296788085411807,
363
+ "grad_norm": 0.030517578125,
364
+ "learning_rate": 0.0001899371584187753,
365
+ "loss": 1.0674,
366
+ "step": 250
367
+ },
368
+ {
369
+ "epoch": 0.23427238471200432,
370
+ "grad_norm": 0.0294189453125,
371
+ "learning_rate": 0.00018922415280711716,
372
+ "loss": 1.0611,
373
+ "step": 255
374
+ },
375
+ {
376
+ "epoch": 0.2388659608828279,
377
+ "grad_norm": 0.0291748046875,
378
+ "learning_rate": 0.00018848817790993432,
379
+ "loss": 1.0603,
380
+ "step": 260
381
+ },
382
+ {
383
+ "epoch": 0.24345953705365153,
384
+ "grad_norm": 0.0299072265625,
385
+ "learning_rate": 0.00018772942319181696,
386
+ "loss": 1.0497,
387
+ "step": 265
388
+ },
389
+ {
390
+ "epoch": 0.24805311322447515,
391
+ "grad_norm": 0.0311279296875,
392
+ "learning_rate": 0.0001869480839816443,
393
+ "loss": 1.056,
394
+ "step": 270
395
+ },
396
+ {
397
+ "epoch": 0.2526466893952988,
398
+ "grad_norm": 0.029541015625,
399
+ "learning_rate": 0.0001861443614223002,
400
+ "loss": 1.0677,
401
+ "step": 275
402
+ },
403
+ {
404
+ "epoch": 0.2572402655661224,
405
+ "grad_norm": 0.0301513671875,
406
+ "learning_rate": 0.00018531846241889245,
407
+ "loss": 1.0646,
408
+ "step": 280
409
+ },
410
+ {
411
+ "epoch": 0.261833841736946,
412
+ "grad_norm": 0.030029296875,
413
+ "learning_rate": 0.0001844705995854882,
414
+ "loss": 1.0962,
415
+ "step": 285
416
+ },
417
+ {
418
+ "epoch": 0.2664274179077696,
419
+ "grad_norm": 0.0296630859375,
420
+ "learning_rate": 0.0001836009911903803,
421
+ "loss": 1.0587,
422
+ "step": 290
423
+ },
424
+ {
425
+ "epoch": 0.2710209940785932,
426
+ "grad_norm": 0.0299072265625,
427
+ "learning_rate": 0.00018270986109989744,
428
+ "loss": 1.0724,
429
+ "step": 295
430
+ },
431
+ {
432
+ "epoch": 0.2756145702494168,
433
+ "grad_norm": 0.0299072265625,
434
+ "learning_rate": 0.00018179743872077359,
435
+ "loss": 1.0816,
436
+ "step": 300
437
+ },
438
+ {
439
+ "epoch": 0.28020814642024044,
440
+ "grad_norm": 0.030029296875,
441
+ "learning_rate": 0.000180863958941091,
442
+ "loss": 1.0654,
443
+ "step": 305
444
+ },
445
+ {
446
+ "epoch": 0.28480172259106407,
447
+ "grad_norm": 0.032470703125,
448
+ "learning_rate": 0.00017990966206981224,
449
+ "loss": 1.0828,
450
+ "step": 310
451
+ },
452
+ {
453
+ "epoch": 0.2893952987618877,
454
+ "grad_norm": 0.02978515625,
455
+ "learning_rate": 0.0001789347937749164,
456
+ "loss": 1.0492,
457
+ "step": 315
458
+ },
459
+ {
460
+ "epoch": 0.2939888749327113,
461
+ "grad_norm": 0.0294189453125,
462
+ "learning_rate": 0.00017793960502015613,
463
+ "loss": 1.0845,
464
+ "step": 320
465
+ },
466
+ {
467
+ "epoch": 0.2985824511035349,
468
+ "grad_norm": 0.0294189453125,
469
+ "learning_rate": 0.0001769243520004511,
470
+ "loss": 1.0536,
471
+ "step": 325
472
+ },
473
+ {
474
+ "epoch": 0.3031760272743585,
475
+ "grad_norm": 0.030029296875,
476
+ "learning_rate": 0.0001758892960759348,
477
+ "loss": 1.0649,
478
+ "step": 330
479
+ },
480
+ {
481
+ "epoch": 0.3077696034451821,
482
+ "grad_norm": 0.031005859375,
483
+ "learning_rate": 0.00017483470370467178,
484
+ "loss": 1.064,
485
+ "step": 335
486
+ },
487
+ {
488
+ "epoch": 0.31236317961600574,
489
+ "grad_norm": 0.0296630859375,
490
+ "learning_rate": 0.00017376084637406222,
491
+ "loss": 1.0584,
492
+ "step": 340
493
+ },
494
+ {
495
+ "epoch": 0.31695675578682936,
496
+ "grad_norm": 0.02978515625,
497
+ "learning_rate": 0.00017266800053095232,
498
+ "loss": 1.0616,
499
+ "step": 345
500
+ },
501
+ {
502
+ "epoch": 0.321550331957653,
503
+ "grad_norm": 0.030029296875,
504
+ "learning_rate": 0.0001715564475104673,
505
+ "loss": 1.0868,
506
+ "step": 350
507
+ },
508
+ {
509
+ "epoch": 0.3261439081284766,
510
+ "grad_norm": 0.030517578125,
511
+ "learning_rate": 0.00017042647346358645,
512
+ "loss": 1.0808,
513
+ "step": 355
514
+ },
515
+ {
516
+ "epoch": 0.3307374842993002,
517
+ "grad_norm": 0.0294189453125,
518
+ "learning_rate": 0.00016927836928347826,
519
+ "loss": 1.0739,
520
+ "step": 360
521
+ },
522
+ {
523
+ "epoch": 0.3353310604701238,
524
+ "grad_norm": 0.030517578125,
525
+ "learning_rate": 0.00016811243053061487,
526
+ "loss": 1.078,
527
+ "step": 365
528
+ },
529
+ {
530
+ "epoch": 0.3399246366409474,
531
+ "grad_norm": 0.030517578125,
532
+ "learning_rate": 0.00016692895735668476,
533
+ "loss": 1.0578,
534
+ "step": 370
535
+ },
536
+ {
537
+ "epoch": 0.34451821281177103,
538
+ "grad_norm": 0.0308837890625,
539
+ "learning_rate": 0.00016572825442732366,
540
+ "loss": 1.0535,
541
+ "step": 375
542
+ },
543
+ {
544
+ "epoch": 0.34911178898259465,
545
+ "grad_norm": 0.03125,
546
+ "learning_rate": 0.0001645106308436836,
547
+ "loss": 1.0668,
548
+ "step": 380
549
+ },
550
+ {
551
+ "epoch": 0.35370536515341827,
552
+ "grad_norm": 0.031494140625,
553
+ "learning_rate": 0.00016327640006285967,
554
+ "loss": 1.05,
555
+ "step": 385
556
+ },
557
+ {
558
+ "epoch": 0.3582989413242419,
559
+ "grad_norm": 0.031494140625,
560
+ "learning_rate": 0.00016202587981719584,
561
+ "loss": 1.066,
562
+ "step": 390
563
+ },
564
+ {
565
+ "epoch": 0.3628925174950655,
566
+ "grad_norm": 0.030517578125,
567
+ "learning_rate": 0.0001607593920324899,
568
+ "loss": 1.0622,
569
+ "step": 395
570
+ },
571
+ {
572
+ "epoch": 0.36748609366588914,
573
+ "grad_norm": 0.031982421875,
574
+ "learning_rate": 0.00015947726274511908,
575
+ "loss": 1.0666,
576
+ "step": 400
577
+ },
578
+ {
579
+ "epoch": 0.3720796698367127,
580
+ "grad_norm": 0.031494140625,
581
+ "learning_rate": 0.0001581798220181073,
582
+ "loss": 1.0531,
583
+ "step": 405
584
+ },
585
+ {
586
+ "epoch": 0.3766732460075363,
587
+ "grad_norm": 0.0311279296875,
588
+ "learning_rate": 0.00015686740385615586,
589
+ "loss": 1.0829,
590
+ "step": 410
591
+ },
592
+ {
593
+ "epoch": 0.38126682217835994,
594
+ "grad_norm": 0.030517578125,
595
+ "learning_rate": 0.00015554034611965954,
596
+ "loss": 1.0592,
597
+ "step": 415
598
+ },
599
+ {
600
+ "epoch": 0.38586039834918356,
601
+ "grad_norm": 0.031494140625,
602
+ "learning_rate": 0.00015419899043772994,
603
+ "loss": 1.05,
604
+ "step": 420
605
+ },
606
+ {
607
+ "epoch": 0.3904539745200072,
608
+ "grad_norm": 0.030029296875,
609
+ "learning_rate": 0.00015284368212024877,
610
+ "loss": 1.0732,
611
+ "step": 425
612
+ },
613
+ {
614
+ "epoch": 0.3950475506908308,
615
+ "grad_norm": 0.031494140625,
616
+ "learning_rate": 0.00015147477006897337,
617
+ "loss": 1.0728,
618
+ "step": 430
619
+ },
620
+ {
621
+ "epoch": 0.3996411268616544,
622
+ "grad_norm": 0.030029296875,
623
+ "learning_rate": 0.000150092606687718,
624
+ "loss": 1.0481,
625
+ "step": 435
626
+ },
627
+ {
628
+ "epoch": 0.404234703032478,
629
+ "grad_norm": 0.03173828125,
630
+ "learning_rate": 0.0001486975477916329,
631
+ "loss": 1.0781,
632
+ "step": 440
633
+ },
634
+ {
635
+ "epoch": 0.4088282792033016,
636
+ "grad_norm": 0.03125,
637
+ "learning_rate": 0.00014728995251560596,
638
+ "loss": 1.0703,
639
+ "step": 445
640
+ },
641
+ {
642
+ "epoch": 0.41342185537412524,
643
+ "grad_norm": 0.0322265625,
644
+ "learning_rate": 0.00014587018322180905,
645
+ "loss": 1.0554,
646
+ "step": 450
647
+ },
648
+ {
649
+ "epoch": 0.41801543154494886,
650
+ "grad_norm": 0.03076171875,
651
+ "learning_rate": 0.00014443860540641406,
652
+ "loss": 1.0626,
653
+ "step": 455
654
+ },
655
+ {
656
+ "epoch": 0.4226090077157725,
657
+ "grad_norm": 0.033203125,
658
+ "learning_rate": 0.00014299558760550184,
659
+ "loss": 1.0918,
660
+ "step": 460
661
+ },
662
+ {
663
+ "epoch": 0.4272025838865961,
664
+ "grad_norm": 0.0302734375,
665
+ "learning_rate": 0.00014154150130018866,
666
+ "loss": 1.0488,
667
+ "step": 465
668
+ },
669
+ {
670
+ "epoch": 0.4317961600574197,
671
+ "grad_norm": 0.031494140625,
672
+ "learning_rate": 0.0001400767208209946,
673
+ "loss": 1.0599,
674
+ "step": 470
675
+ },
676
+ {
677
+ "epoch": 0.43638973622824334,
678
+ "grad_norm": 0.0308837890625,
679
+ "learning_rate": 0.0001386016232514784,
680
+ "loss": 1.0669,
681
+ "step": 475
682
+ },
683
+ {
684
+ "epoch": 0.4409833123990669,
685
+ "grad_norm": 0.0311279296875,
686
+ "learning_rate": 0.00013711658833116323,
687
+ "loss": 1.0721,
688
+ "step": 480
689
+ },
690
+ {
691
+ "epoch": 0.44557688856989053,
692
+ "grad_norm": 0.03173828125,
693
+ "learning_rate": 0.00013562199835777934,
694
+ "loss": 1.0829,
695
+ "step": 485
696
+ },
697
+ {
698
+ "epoch": 0.45017046474071415,
699
+ "grad_norm": 0.03125,
700
+ "learning_rate": 0.00013411823808884765,
701
+ "loss": 1.0611,
702
+ "step": 490
703
+ },
704
+ {
705
+ "epoch": 0.45476404091153777,
706
+ "grad_norm": 0.031494140625,
707
+ "learning_rate": 0.00013260569464263036,
708
+ "loss": 1.0547,
709
+ "step": 495
710
+ },
711
+ {
712
+ "epoch": 0.4593576170823614,
713
+ "grad_norm": 0.032958984375,
714
+ "learning_rate": 0.00013108475739847362,
715
+ "loss": 1.0635,
716
+ "step": 500
717
+ },
718
+ {
719
+ "epoch": 0.463951193253185,
720
+ "grad_norm": 0.0302734375,
721
+ "learning_rate": 0.00012955581789656843,
722
+ "loss": 1.0746,
723
+ "step": 505
724
+ },
725
+ {
726
+ "epoch": 0.46854476942400863,
727
+ "grad_norm": 0.03125,
728
+ "learning_rate": 0.00012801926973715483,
729
+ "loss": 1.0665,
730
+ "step": 510
731
+ },
732
+ {
733
+ "epoch": 0.4731383455948322,
734
+ "grad_norm": 0.03125,
735
+ "learning_rate": 0.0001264755084791963,
736
+ "loss": 1.0547,
737
+ "step": 515
738
+ },
739
+ {
740
+ "epoch": 0.4777319217656558,
741
+ "grad_norm": 0.03271484375,
742
+ "learning_rate": 0.00012492493153854937,
743
+ "loss": 1.0508,
744
+ "step": 520
745
+ },
746
+ {
747
+ "epoch": 0.48232549793647944,
748
+ "grad_norm": 0.03173828125,
749
+ "learning_rate": 0.0001233679380856557,
750
+ "loss": 1.082,
751
+ "step": 525
752
+ },
753
+ {
754
+ "epoch": 0.48691907410730306,
755
+ "grad_norm": 0.03173828125,
756
+ "learning_rate": 0.00012180492894278206,
757
+ "loss": 1.0656,
758
+ "step": 530
759
+ },
760
+ {
761
+ "epoch": 0.4915126502781267,
762
+ "grad_norm": 0.032958984375,
763
+ "learning_rate": 0.00012023630648083528,
764
+ "loss": 1.0703,
765
+ "step": 535
766
+ },
767
+ {
768
+ "epoch": 0.4961062264489503,
769
+ "grad_norm": 0.031982421875,
770
+ "learning_rate": 0.00011866247451577864,
771
+ "loss": 1.0685,
772
+ "step": 540
773
+ },
774
+ {
775
+ "epoch": 0.5006998026197739,
776
+ "grad_norm": 0.0306396484375,
777
+ "learning_rate": 0.00011708383820467595,
778
+ "loss": 1.0568,
779
+ "step": 545
780
+ },
781
+ {
782
+ "epoch": 0.5052933787905975,
783
+ "grad_norm": 0.0322265625,
784
+ "learning_rate": 0.00011550080394139062,
785
+ "loss": 1.0405,
786
+ "step": 550
787
+ },
788
+ {
789
+ "epoch": 0.5098869549614211,
790
+ "grad_norm": 0.033203125,
791
+ "learning_rate": 0.00011391377925196626,
792
+ "loss": 1.0654,
793
+ "step": 555
794
+ },
795
+ {
796
+ "epoch": 0.5144805311322448,
797
+ "grad_norm": 0.031005859375,
798
+ "learning_rate": 0.00011232317268971585,
799
+ "loss": 1.0578,
800
+ "step": 560
801
+ },
802
+ {
803
+ "epoch": 0.5190741073030684,
804
+ "grad_norm": 0.03125,
805
+ "learning_rate": 0.00011072939373004647,
806
+ "loss": 1.0581,
807
+ "step": 565
808
+ },
809
+ {
810
+ "epoch": 0.523667683473892,
811
+ "grad_norm": 0.031005859375,
812
+ "learning_rate": 0.00010913285266504636,
813
+ "loss": 1.0464,
814
+ "step": 570
815
+ },
816
+ {
817
+ "epoch": 0.5282612596447156,
818
+ "grad_norm": 0.0308837890625,
819
+ "learning_rate": 0.0001075339604978624,
820
+ "loss": 1.0744,
821
+ "step": 575
822
+ },
823
+ {
824
+ "epoch": 0.5328548358155392,
825
+ "grad_norm": 0.0308837890625,
826
+ "learning_rate": 0.0001059331288368938,
827
+ "loss": 1.0675,
828
+ "step": 580
829
+ },
830
+ {
831
+ "epoch": 0.5374484119863628,
832
+ "grad_norm": 0.031982421875,
833
+ "learning_rate": 0.0001043307697898305,
834
+ "loss": 1.0682,
835
+ "step": 585
836
+ },
837
+ {
838
+ "epoch": 0.5420419881571864,
839
+ "grad_norm": 0.032470703125,
840
+ "learning_rate": 0.00010272729585756275,
841
+ "loss": 1.0577,
842
+ "step": 590
843
+ },
844
+ {
845
+ "epoch": 0.5466355643280101,
846
+ "grad_norm": 0.0311279296875,
847
+ "learning_rate": 0.00010112311982798959,
848
+ "loss": 1.0535,
849
+ "step": 595
850
+ },
851
+ {
852
+ "epoch": 0.5512291404988336,
853
+ "grad_norm": 0.0306396484375,
854
+ "learning_rate": 9.951865466975344e-05,
855
+ "loss": 1.0713,
856
+ "step": 600
857
+ },
858
+ {
859
+ "epoch": 0.5558227166696573,
860
+ "grad_norm": 0.0311279296875,
861
+ "learning_rate": 9.791431342592811e-05,
862
+ "loss": 1.0532,
863
+ "step": 605
864
+ },
865
+ {
866
+ "epoch": 0.5604162928404809,
867
+ "grad_norm": 0.0311279296875,
868
+ "learning_rate": 9.631050910768773e-05,
869
+ "loss": 1.0807,
870
+ "step": 610
871
+ },
872
+ {
873
+ "epoch": 0.5650098690113045,
874
+ "grad_norm": 0.031005859375,
875
+ "learning_rate": 9.470765458798368e-05,
876
+ "loss": 1.0504,
877
+ "step": 615
878
+ },
879
+ {
880
+ "epoch": 0.5696034451821281,
881
+ "grad_norm": 0.032470703125,
882
+ "learning_rate": 9.310616249525759e-05,
883
+ "loss": 1.0507,
884
+ "step": 620
885
+ },
886
+ {
887
+ "epoch": 0.5741970213529517,
888
+ "grad_norm": 0.0308837890625,
889
+ "learning_rate": 9.15064451072165e-05,
890
+ "loss": 1.0434,
891
+ "step": 625
892
+ },
893
+ {
894
+ "epoch": 0.5787905975237754,
895
+ "grad_norm": 0.031005859375,
896
+ "learning_rate": 8.990891424469927e-05,
897
+ "loss": 1.0744,
898
+ "step": 630
899
+ },
900
+ {
901
+ "epoch": 0.5833841736945989,
902
+ "grad_norm": 0.03173828125,
903
+ "learning_rate": 8.831398116565964e-05,
904
+ "loss": 1.0792,
905
+ "step": 635
906
+ },
907
+ {
908
+ "epoch": 0.5879777498654226,
909
+ "grad_norm": 0.03173828125,
910
+ "learning_rate": 8.672205645929503e-05,
911
+ "loss": 1.0723,
912
+ "step": 640
913
+ },
914
+ {
915
+ "epoch": 0.5925713260362462,
916
+ "grad_norm": 0.031982421875,
917
+ "learning_rate": 8.513354994034681e-05,
918
+ "loss": 1.0627,
919
+ "step": 645
920
+ },
921
+ {
922
+ "epoch": 0.5971649022070697,
923
+ "grad_norm": 0.031982421875,
924
+ "learning_rate": 8.35488705436006e-05,
925
+ "loss": 1.0844,
926
+ "step": 650
927
+ },
928
+ {
929
+ "epoch": 0.6017584783778934,
930
+ "grad_norm": 0.031982421875,
931
+ "learning_rate": 8.196842621861242e-05,
932
+ "loss": 1.0566,
933
+ "step": 655
934
+ },
935
+ {
936
+ "epoch": 0.606352054548717,
937
+ "grad_norm": 0.03173828125,
938
+ "learning_rate": 8.039262382468905e-05,
939
+ "loss": 1.0697,
940
+ "step": 660
941
+ },
942
+ {
943
+ "epoch": 0.6109456307195407,
944
+ "grad_norm": 0.03271484375,
945
+ "learning_rate": 7.882186902614865e-05,
946
+ "loss": 1.0719,
947
+ "step": 665
948
+ },
949
+ {
950
+ "epoch": 0.6155392068903642,
951
+ "grad_norm": 0.032470703125,
952
+ "learning_rate": 7.725656618788937e-05,
953
+ "loss": 1.0519,
954
+ "step": 670
955
+ },
956
+ {
957
+ "epoch": 0.6201327830611879,
958
+ "grad_norm": 0.03271484375,
959
+ "learning_rate": 7.569711827129208e-05,
960
+ "loss": 1.0508,
961
+ "step": 675
962
+ },
963
+ {
964
+ "epoch": 0.6247263592320115,
965
+ "grad_norm": 0.031982421875,
966
+ "learning_rate": 7.414392673048478e-05,
967
+ "loss": 1.0631,
968
+ "step": 680
969
+ },
970
+ {
971
+ "epoch": 0.6293199354028352,
972
+ "grad_norm": 0.031494140625,
973
+ "learning_rate": 7.259739140899462e-05,
974
+ "loss": 1.0702,
975
+ "step": 685
976
+ },
977
+ {
978
+ "epoch": 0.6339135115736587,
979
+ "grad_norm": 0.031494140625,
980
+ "learning_rate": 7.105791043681519e-05,
981
+ "loss": 1.0642,
982
+ "step": 690
983
+ },
984
+ {
985
+ "epoch": 0.6385070877444823,
986
+ "grad_norm": 0.031494140625,
987
+ "learning_rate": 6.952588012791405e-05,
988
+ "loss": 1.0585,
989
+ "step": 695
990
+ },
991
+ {
992
+ "epoch": 0.643100663915306,
993
+ "grad_norm": 0.03125,
994
+ "learning_rate": 6.80016948782086e-05,
995
+ "loss": 1.0691,
996
+ "step": 700
997
+ },
998
+ {
999
+ "epoch": 0.6476942400861295,
1000
+ "grad_norm": 0.031494140625,
1001
+ "learning_rate": 6.648574706403522e-05,
1002
+ "loss": 1.0614,
1003
+ "step": 705
1004
+ },
1005
+ {
1006
+ "epoch": 0.6522878162569532,
1007
+ "grad_norm": 0.031982421875,
1008
+ "learning_rate": 6.497842694113842e-05,
1009
+ "loss": 1.0649,
1010
+ "step": 710
1011
+ },
1012
+ {
1013
+ "epoch": 0.6568813924277768,
1014
+ "grad_norm": 0.03173828125,
1015
+ "learning_rate": 6.348012254420606e-05,
1016
+ "loss": 1.07,
1017
+ "step": 715
1018
+ },
1019
+ {
1020
+ "epoch": 0.6614749685986004,
1021
+ "grad_norm": 0.0308837890625,
1022
+ "learning_rate": 6.199121958697604e-05,
1023
+ "loss": 1.0709,
1024
+ "step": 720
1025
+ },
1026
+ {
1027
+ "epoch": 0.666068544769424,
1028
+ "grad_norm": 0.03466796875,
1029
+ "learning_rate": 6.051210136294089e-05,
1030
+ "loss": 1.063,
1031
+ "step": 725
1032
+ },
1033
+ {
1034
+ "epoch": 0.6706621209402476,
1035
+ "grad_norm": 0.03173828125,
1036
+ "learning_rate": 5.904314864667497e-05,
1037
+ "loss": 1.0622,
1038
+ "step": 730
1039
+ },
1040
+ {
1041
+ "epoch": 0.6752556971110713,
1042
+ "grad_norm": 0.031005859375,
1043
+ "learning_rate": 5.758473959581061e-05,
1044
+ "loss": 1.0558,
1045
+ "step": 735
1046
+ },
1047
+ {
1048
+ "epoch": 0.6798492732818948,
1049
+ "grad_norm": 0.03125,
1050
+ "learning_rate": 5.613724965368723e-05,
1051
+ "loss": 1.0724,
1052
+ "step": 740
1053
+ },
1054
+ {
1055
+ "epoch": 0.6844428494527185,
1056
+ "grad_norm": 0.031982421875,
1057
+ "learning_rate": 5.4701051452700245e-05,
1058
+ "loss": 1.0675,
1059
+ "step": 745
1060
+ },
1061
+ {
1062
+ "epoch": 0.6890364256235421,
1063
+ "grad_norm": 0.0322265625,
1064
+ "learning_rate": 5.327651471837242e-05,
1065
+ "loss": 1.079,
1066
+ "step": 750
1067
+ },
1068
+ {
1069
+ "epoch": 0.6936300017943657,
1070
+ "grad_norm": 0.03173828125,
1071
+ "learning_rate": 5.1864006174174504e-05,
1072
+ "loss": 1.0626,
1073
+ "step": 755
1074
+ },
1075
+ {
1076
+ "epoch": 0.6982235779651893,
1077
+ "grad_norm": 0.033447265625,
1078
+ "learning_rate": 5.046388944711824e-05,
1079
+ "loss": 1.0637,
1080
+ "step": 760
1081
+ },
1082
+ {
1083
+ "epoch": 0.7028171541360129,
1084
+ "grad_norm": 0.0322265625,
1085
+ "learning_rate": 4.9076524974146507e-05,
1086
+ "loss": 1.0565,
1087
+ "step": 765
1088
+ },
1089
+ {
1090
+ "epoch": 0.7074107303068365,
1091
+ "grad_norm": 0.031494140625,
1092
+ "learning_rate": 4.7702269909344907e-05,
1093
+ "loss": 1.068,
1094
+ "step": 770
1095
+ },
1096
+ {
1097
+ "epoch": 0.7120043064776601,
1098
+ "grad_norm": 0.031494140625,
1099
+ "learning_rate": 4.6341478031998265e-05,
1100
+ "loss": 1.0632,
1101
+ "step": 775
1102
+ },
1103
+ {
1104
+ "epoch": 0.7165978826484838,
1105
+ "grad_norm": 0.0322265625,
1106
+ "learning_rate": 4.4994499655515865e-05,
1107
+ "loss": 1.0809,
1108
+ "step": 780
1109
+ },
1110
+ {
1111
+ "epoch": 0.7211914588193074,
1112
+ "grad_norm": 0.03173828125,
1113
+ "learning_rate": 4.3661681537249455e-05,
1114
+ "loss": 1.0694,
1115
+ "step": 785
1116
+ },
1117
+ {
1118
+ "epoch": 0.725785034990131,
1119
+ "grad_norm": 0.03271484375,
1120
+ "learning_rate": 4.234336678922569e-05,
1121
+ "loss": 1.06,
1122
+ "step": 790
1123
+ },
1124
+ {
1125
+ "epoch": 0.7303786111609546,
1126
+ "grad_norm": 0.03173828125,
1127
+ "learning_rate": 4.103989478981827e-05,
1128
+ "loss": 1.0498,
1129
+ "step": 795
1130
+ },
1131
+ {
1132
+ "epoch": 0.7349721873317783,
1133
+ "grad_norm": 0.0322265625,
1134
+ "learning_rate": 3.975160109637992e-05,
1135
+ "loss": 1.0527,
1136
+ "step": 800
1137
+ },
1138
+ {
1139
+ "epoch": 0.7395657635026018,
1140
+ "grad_norm": 0.033447265625,
1141
+ "learning_rate": 3.847881735885918e-05,
1142
+ "loss": 1.0593,
1143
+ "step": 805
1144
+ },
1145
+ {
1146
+ "epoch": 0.7441593396734254,
1147
+ "grad_norm": 0.031982421875,
1148
+ "learning_rate": 3.722187123442249e-05,
1149
+ "loss": 1.0568,
1150
+ "step": 810
1151
+ },
1152
+ {
1153
+ "epoch": 0.7487529158442491,
1154
+ "grad_norm": 0.0322265625,
1155
+ "learning_rate": 3.598108630310399e-05,
1156
+ "loss": 1.0651,
1157
+ "step": 815
1158
+ },
1159
+ {
1160
+ "epoch": 0.7533464920150726,
1161
+ "grad_norm": 0.031494140625,
1162
+ "learning_rate": 3.475678198450555e-05,
1163
+ "loss": 1.0451,
1164
+ "step": 820
1165
+ },
1166
+ {
1167
+ "epoch": 0.7579400681858963,
1168
+ "grad_norm": 0.032470703125,
1169
+ "learning_rate": 3.354927345556723e-05,
1170
+ "loss": 1.0591,
1171
+ "step": 825
1172
+ },
1173
+ {
1174
+ "epoch": 0.7625336443567199,
1175
+ "grad_norm": 0.03173828125,
1176
+ "learning_rate": 3.235887156943029e-05,
1177
+ "loss": 1.0519,
1178
+ "step": 830
1179
+ },
1180
+ {
1181
+ "epoch": 0.7671272205275436,
1182
+ "grad_norm": 0.031982421875,
1183
+ "learning_rate": 3.118588277541312e-05,
1184
+ "loss": 1.0833,
1185
+ "step": 835
1186
+ },
1187
+ {
1188
+ "epoch": 0.7717207966983671,
1189
+ "grad_norm": 0.03271484375,
1190
+ "learning_rate": 3.003060904012096e-05,
1191
+ "loss": 1.0543,
1192
+ "step": 840
1193
+ },
1194
+ {
1195
+ "epoch": 0.7763143728691907,
1196
+ "grad_norm": 0.031982421875,
1197
+ "learning_rate": 2.8893347769709476e-05,
1198
+ "loss": 1.0755,
1199
+ "step": 845
1200
+ },
1201
+ {
1202
+ "epoch": 0.7809079490400144,
1203
+ "grad_norm": 0.03271484375,
1204
+ "learning_rate": 2.7774391733322713e-05,
1205
+ "loss": 1.0682,
1206
+ "step": 850
1207
+ },
1208
+ {
1209
+ "epoch": 0.7855015252108379,
1210
+ "grad_norm": 0.03173828125,
1211
+ "learning_rate": 2.6674028987724163e-05,
1212
+ "loss": 1.064,
1213
+ "step": 855
1214
+ },
1215
+ {
1216
+ "epoch": 0.7900951013816616,
1217
+ "grad_norm": 0.031494140625,
1218
+ "learning_rate": 2.559254280314156e-05,
1219
+ "loss": 1.0648,
1220
+ "step": 860
1221
+ },
1222
+ {
1223
+ "epoch": 0.7946886775524852,
1224
+ "grad_norm": 0.031982421875,
1225
+ "learning_rate": 2.4530211590343578e-05,
1226
+ "loss": 1.0699,
1227
+ "step": 865
1228
+ },
1229
+ {
1230
+ "epoch": 0.7992822537233089,
1231
+ "grad_norm": 0.03173828125,
1232
+ "learning_rate": 2.3487308828967493e-05,
1233
+ "loss": 1.0647,
1234
+ "step": 870
1235
+ },
1236
+ {
1237
+ "epoch": 0.8038758298941324,
1238
+ "grad_norm": 0.031982421875,
1239
+ "learning_rate": 2.2464102997116475e-05,
1240
+ "loss": 1.0636,
1241
+ "step": 875
1242
+ },
1243
+ {
1244
+ "epoch": 0.808469406064956,
1245
+ "grad_norm": 0.03125,
1246
+ "learning_rate": 2.1460857502244248e-05,
1247
+ "loss": 1.0793,
1248
+ "step": 880
1249
+ },
1250
+ {
1251
+ "epoch": 0.8130629822357797,
1252
+ "grad_norm": 0.03125,
1253
+ "learning_rate": 2.047783061334523e-05,
1254
+ "loss": 1.0501,
1255
+ "step": 885
1256
+ },
1257
+ {
1258
+ "epoch": 0.8176565584066032,
1259
+ "grad_norm": 0.031494140625,
1260
+ "learning_rate": 1.9515275394467446e-05,
1261
+ "loss": 1.0553,
1262
+ "step": 890
1263
+ },
1264
+ {
1265
+ "epoch": 0.8222501345774269,
1266
+ "grad_norm": 0.03173828125,
1267
+ "learning_rate": 1.8573439639565282e-05,
1268
+ "loss": 1.059,
1269
+ "step": 895
1270
+ },
1271
+ {
1272
+ "epoch": 0.8268437107482505,
1273
+ "grad_norm": 0.031494140625,
1274
+ "learning_rate": 1.765256580870924e-05,
1275
+ "loss": 1.0653,
1276
+ "step": 900
1277
+ },
1278
+ {
1279
+ "epoch": 0.8314372869190741,
1280
+ "grad_norm": 0.0322265625,
1281
+ "learning_rate": 1.6752890965668266e-05,
1282
+ "loss": 1.0616,
1283
+ "step": 905
1284
+ },
1285
+ {
1286
+ "epoch": 0.8360308630898977,
1287
+ "grad_norm": 0.031982421875,
1288
+ "learning_rate": 1.587464671688187e-05,
1289
+ "loss": 1.0612,
1290
+ "step": 910
1291
+ },
1292
+ {
1293
+ "epoch": 0.8406244392607214,
1294
+ "grad_norm": 0.031982421875,
1295
+ "learning_rate": 1.501805915183685e-05,
1296
+ "loss": 1.0723,
1297
+ "step": 915
1298
+ },
1299
+ {
1300
+ "epoch": 0.845218015431545,
1301
+ "grad_norm": 0.031982421875,
1302
+ "learning_rate": 1.4183348784864037e-05,
1303
+ "loss": 1.0664,
1304
+ "step": 920
1305
+ },
1306
+ {
1307
+ "epoch": 0.8498115916023685,
1308
+ "grad_norm": 0.03125,
1309
+ "learning_rate": 1.3370730498370831e-05,
1310
+ "loss": 1.0693,
1311
+ "step": 925
1312
+ },
1313
+ {
1314
+ "epoch": 0.8544051677731922,
1315
+ "grad_norm": 0.03173828125,
1316
+ "learning_rate": 1.258041348752308e-05,
1317
+ "loss": 1.0715,
1318
+ "step": 930
1319
+ },
1320
+ {
1321
+ "epoch": 0.8589987439440158,
1322
+ "grad_norm": 0.031982421875,
1323
+ "learning_rate": 1.1812601206391304e-05,
1324
+ "loss": 1.0755,
1325
+ "step": 935
1326
+ },
1327
+ {
1328
+ "epoch": 0.8635923201148394,
1329
+ "grad_norm": 0.03125,
1330
+ "learning_rate": 1.1067491315574797e-05,
1331
+ "loss": 1.0522,
1332
+ "step": 940
1333
+ },
1334
+ {
1335
+ "epoch": 0.868185896285663,
1336
+ "grad_norm": 0.03173828125,
1337
+ "learning_rate": 1.0345275631317163e-05,
1338
+ "loss": 1.0495,
1339
+ "step": 945
1340
+ },
1341
+ {
1342
+ "epoch": 0.8727794724564867,
1343
+ "grad_norm": 0.03173828125,
1344
+ "learning_rate": 9.64614007612633e-06,
1345
+ "loss": 1.0637,
1346
+ "step": 950
1347
+ },
1348
+ {
1349
+ "epoch": 0.8773730486273102,
1350
+ "grad_norm": 0.031494140625,
1351
+ "learning_rate": 8.970264630912061e-06,
1352
+ "loss": 1.064,
1353
+ "step": 955
1354
+ },
1355
+ {
1356
+ "epoch": 0.8819666247981338,
1357
+ "grad_norm": 0.031494140625,
1358
+ "learning_rate": 8.317823288652526e-06,
1359
+ "loss": 1.0457,
1360
+ "step": 960
1361
+ },
1362
+ {
1363
+ "epoch": 0.8865602009689575,
1364
+ "grad_norm": 0.03173828125,
1365
+ "learning_rate": 7.688984009603062e-06,
1366
+ "loss": 1.0633,
1367
+ "step": 965
1368
+ },
1369
+ {
1370
+ "epoch": 0.8911537771397811,
1371
+ "grad_norm": 0.0322265625,
1372
+ "learning_rate": 7.083908678057194e-06,
1373
+ "loss": 1.0677,
1374
+ "step": 970
1375
+ },
1376
+ {
1377
+ "epoch": 0.8957473533106047,
1378
+ "grad_norm": 0.03125,
1379
+ "learning_rate": 6.502753060672495e-06,
1380
+ "loss": 1.0482,
1381
+ "step": 975
1382
+ },
1383
+ {
1384
+ "epoch": 0.9003409294814283,
1385
+ "grad_norm": 0.0322265625,
1386
+ "learning_rate": 5.945666766370861e-06,
1387
+ "loss": 1.0749,
1388
+ "step": 980
1389
+ },
1390
+ {
1391
+ "epoch": 0.904934505652252,
1392
+ "grad_norm": 0.03173828125,
1393
+ "learning_rate": 5.412793207824252e-06,
1394
+ "loss": 1.0588,
1395
+ "step": 985
1396
+ },
1397
+ {
1398
+ "epoch": 0.9095280818230755,
1399
+ "grad_norm": 0.031494140625,
1400
+ "learning_rate": 4.904269564535391e-06,
1401
+ "loss": 1.0575,
1402
+ "step": 990
1403
+ },
1404
+ {
1405
+ "epoch": 0.9141216579938991,
1406
+ "grad_norm": 0.03173828125,
1407
+ "learning_rate": 4.42022674752326e-06,
1408
+ "loss": 1.0554,
1409
+ "step": 995
1410
+ },
1411
+ {
1412
+ "epoch": 0.9187152341647228,
1413
+ "grad_norm": 0.03271484375,
1414
+ "learning_rate": 3.960789365622075e-06,
1415
+ "loss": 1.0504,
1416
+ "step": 1000
1417
+ },
1418
+ {
1419
+ "epoch": 0.9233088103355463,
1420
+ "grad_norm": 0.031494140625,
1421
+ "learning_rate": 3.526075693402986e-06,
1422
+ "loss": 1.064,
1423
+ "step": 1005
1424
+ },
1425
+ {
1426
+ "epoch": 0.92790238650637,
1427
+ "grad_norm": 0.031494140625,
1428
+ "learning_rate": 3.116197640726104e-06,
1429
+ "loss": 1.0566,
1430
+ "step": 1010
1431
+ },
1432
+ {
1433
+ "epoch": 0.9324959626771936,
1434
+ "grad_norm": 0.03173828125,
1435
+ "learning_rate": 2.7312607239311505e-06,
1436
+ "loss": 1.0623,
1437
+ "step": 1015
1438
+ },
1439
+ {
1440
+ "epoch": 0.9370895388480173,
1441
+ "grad_norm": 0.032958984375,
1442
+ "learning_rate": 2.3713640386741396e-06,
1443
+ "loss": 1.0708,
1444
+ "step": 1020
1445
+ },
1446
+ {
1447
+ "epoch": 0.9416831150188408,
1448
+ "grad_norm": 0.032470703125,
1449
+ "learning_rate": 2.0366002344166745e-06,
1450
+ "loss": 1.0646,
1451
+ "step": 1025
1452
+ },
1453
+ {
1454
+ "epoch": 0.9462766911896644,
1455
+ "grad_norm": 0.03173828125,
1456
+ "learning_rate": 1.7270554905750137e-06,
1457
+ "loss": 1.0753,
1458
+ "step": 1030
1459
+ },
1460
+ {
1461
+ "epoch": 0.9508702673604881,
1462
+ "grad_norm": 0.031494140625,
1463
+ "learning_rate": 1.4428094943345294e-06,
1464
+ "loss": 1.0618,
1465
+ "step": 1035
1466
+ },
1467
+ {
1468
+ "epoch": 0.9554638435313116,
1469
+ "grad_norm": 0.03173828125,
1470
+ "learning_rate": 1.1839354201355513e-06,
1471
+ "loss": 1.0533,
1472
+ "step": 1040
1473
+ },
1474
+ {
1475
+ "epoch": 0.9600574197021353,
1476
+ "grad_norm": 0.03125,
1477
+ "learning_rate": 9.50499910835867e-07,
1478
+ "loss": 1.0471,
1479
+ "step": 1045
1480
+ },
1481
+ {
1482
+ "epoch": 0.9646509958729589,
1483
+ "grad_norm": 0.03173828125,
1484
+ "learning_rate": 7.425630605545575e-07,
1485
+ "loss": 1.0564,
1486
+ "step": 1050
1487
+ },
1488
+ {
1489
+ "epoch": 0.9692445720437826,
1490
+ "grad_norm": 0.0322265625,
1491
+ "learning_rate": 5.60178399201805e-07,
1492
+ "loss": 1.0793,
1493
+ "step": 1055
1494
+ },
1495
+ {
1496
+ "epoch": 0.9738381482146061,
1497
+ "grad_norm": 0.03173828125,
1498
+ "learning_rate": 4.0339287869847197e-07,
1499
+ "loss": 1.0729,
1500
+ "step": 1060
1501
+ },
1502
+ {
1503
+ "epoch": 0.9784317243854298,
1504
+ "grad_norm": 0.031982421875,
1505
+ "learning_rate": 2.7224686088910265e-07,
1506
+ "loss": 1.0542,
1507
+ "step": 1065
1508
+ },
1509
+ {
1510
+ "epoch": 0.9830253005562534,
1511
+ "grad_norm": 0.03173828125,
1512
+ "learning_rate": 1.6677410715149054e-07,
1513
+ "loss": 1.0532,
1514
+ "step": 1070
1515
+ },
1516
+ {
1517
+ "epoch": 0.9876188767270769,
1518
+ "grad_norm": 0.031982421875,
1519
+ "learning_rate": 8.700176970527497e-08,
1520
+ "loss": 1.0724,
1521
+ "step": 1075
1522
+ },
1523
+ {
1524
+ "epoch": 0.9922124528979006,
1525
+ "grad_norm": 0.0322265625,
1526
+ "learning_rate": 3.29503846221213e-08,
1527
+ "loss": 1.0617,
1528
+ "step": 1080
1529
+ },
1530
+ {
1531
+ "epoch": 0.9968060290687242,
1532
+ "grad_norm": 0.031494140625,
1533
+ "learning_rate": 4.633866539005549e-09,
1534
+ "loss": 1.0546,
1535
+ "step": 1085
1536
+ },
1537
+ {
1538
+ "epoch": 0.9995621747712183,
1539
+ "eval_loss": 1.0676658153533936,
1540
+ "eval_runtime": 1909.6941,
1541
+ "eval_samples_per_second": 8.075,
1542
+ "eval_steps_per_second": 8.075,
1543
+ "step": 1088
1544
+ },
1545
+ {
1546
+ "epoch": 0.9995621747712183,
1547
+ "step": 1088,
1548
+ "total_flos": 3.023110834516656e+18,
1549
+ "train_loss": 0.22257287449696483,
1550
+ "train_runtime": 15104.5697,
1551
+ "train_samples_per_second": 9.224,
1552
+ "train_steps_per_second": 0.072
1553
+ }
1554
+ ],
1555
+ "logging_steps": 5,
1556
+ "max_steps": 1088,
1557
+ "num_input_tokens_seen": 0,
1558
+ "num_train_epochs": 1,
1559
+ "save_steps": 20,
1560
+ "total_flos": 3.023110834516656e+18,
1561
+ "train_batch_size": 1,
1562
+ "trial_name": null,
1563
+ "trial_params": null
1564
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:083ba15627c3a11156123908752af71f4c4736df9a8307135c75c10977c0b45f
3
+ size 5112