juyongjiang commited on
Commit
82ee3ab
1 Parent(s): 1d6dfa5

upload model checkpoint

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 CHANGED
@@ -1,3 +1,76 @@
1
- ---
2
- license: gemma
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ tags:
4
+ - alignment-handbook
5
+ - generated_from_trainer
6
+ datasets:
7
+ - llama-duo/synth_summarize_dataset_dedup
8
+ base_model: google/gemma-7b
9
+ model-index:
10
+ - name: gemma7b-summarize-gemini1_5flash-64k
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
+ # gemma7b-summarize-gemini1_5flash-64k
18
+
19
+ This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the llama-duo/synth_summarize_dataset_dedup dataset.
20
+ It achieves the following results on the evaluation set:
21
+ - Loss: 2.5156
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: 4
42
+ - eval_batch_size: 2
43
+ - seed: 42
44
+ - distributed_type: multi-GPU
45
+ - num_devices: 8
46
+ - gradient_accumulation_steps: 2
47
+ - total_train_batch_size: 64
48
+ - total_eval_batch_size: 16
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: 10
53
+
54
+ ### Training results
55
+
56
+ | Training Loss | Epoch | Step | Validation Loss |
57
+ |:-------------:|:------:|:----:|:---------------:|
58
+ | 1.2495 | 0.9952 | 104 | 2.6543 |
59
+ | 1.0258 | 2.0 | 209 | 2.5232 |
60
+ | 0.9351 | 2.9952 | 313 | 2.4668 |
61
+ | 0.8914 | 4.0 | 418 | 2.4789 |
62
+ | 0.8487 | 4.9952 | 522 | 2.4752 |
63
+ | 0.8222 | 6.0 | 627 | 2.4928 |
64
+ | 0.7746 | 6.9952 | 731 | 2.4925 |
65
+ | 0.7644 | 8.0 | 836 | 2.5051 |
66
+ | 0.7578 | 8.9952 | 940 | 2.5211 |
67
+ | 0.7589 | 9.9522 | 1040 | 2.5156 |
68
+
69
+
70
+ ### Framework versions
71
+
72
+ - PEFT 0.10.0
73
+ - Transformers 4.40.0
74
+ - Pytorch 2.1.2+cu121
75
+ - Datasets 2.18.0
76
+ - Tokenizers 0.19.1
adapter_config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "layer_replication": null,
10
+ "layers_pattern": null,
11
+ "layers_to_transform": null,
12
+ "loftq_config": {},
13
+ "lora_alpha": 64,
14
+ "lora_dropout": 0.05,
15
+ "megatron_config": null,
16
+ "megatron_core": "megatron.core",
17
+ "modules_to_save": null,
18
+ "peft_type": "LORA",
19
+ "r": 32,
20
+ "rank_pattern": {},
21
+ "revision": null,
22
+ "target_modules": [
23
+ "v_proj",
24
+ "q_proj"
25
+ ],
26
+ "task_type": "CAUSAL_LM",
27
+ "use_dora": false,
28
+ "use_rslora": false
29
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:913e9ec1f2ef29aad9d27f217d6e7854cf1f04ea4765b41d857649bef3ec243d
3
+ size 25705248
all_results.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 9.952153110047847,
3
+ "eval_loss": 2.515638828277588,
4
+ "eval_runtime": 0.236,
5
+ "eval_samples": 25,
6
+ "eval_samples_per_second": 42.369,
7
+ "eval_steps_per_second": 4.237,
8
+ "total_flos": 3.1750846888049377e+18,
9
+ "train_loss": 2.1342261617000284,
10
+ "train_runtime": 2535.4321,
11
+ "train_samples": 63353,
12
+ "train_samples_per_second": 26.287,
13
+ "train_steps_per_second": 0.41
14
+ }
config.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "google/gemma-7b",
3
+ "architectures": [
4
+ "GemmaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 2,
9
+ "eos_token_id": 1,
10
+ "head_dim": 256,
11
+ "hidden_act": "gelu",
12
+ "hidden_activation": null,
13
+ "hidden_size": 3072,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 24576,
16
+ "max_position_embeddings": 8192,
17
+ "model_type": "gemma",
18
+ "num_attention_heads": 16,
19
+ "num_hidden_layers": 28,
20
+ "num_key_value_heads": 16,
21
+ "pad_token_id": 0,
22
+ "quantization_config": {
23
+ "_load_in_4bit": true,
24
+ "_load_in_8bit": false,
25
+ "bnb_4bit_compute_dtype": "bfloat16",
26
+ "bnb_4bit_quant_storage": "uint8",
27
+ "bnb_4bit_quant_type": "nf4",
28
+ "bnb_4bit_use_double_quant": false,
29
+ "llm_int8_enable_fp32_cpu_offload": false,
30
+ "llm_int8_has_fp16_weight": false,
31
+ "llm_int8_skip_modules": null,
32
+ "llm_int8_threshold": 6.0,
33
+ "load_in_4bit": true,
34
+ "load_in_8bit": false,
35
+ "quant_method": "bitsandbytes"
36
+ },
37
+ "rms_norm_eps": 1e-06,
38
+ "rope_scaling": null,
39
+ "rope_theta": 10000.0,
40
+ "torch_dtype": "bfloat16",
41
+ "transformers_version": "4.40.0",
42
+ "use_cache": true,
43
+ "vocab_size": 256000
44
+ }
eval_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 9.952153110047847,
3
+ "eval_loss": 2.515638828277588,
4
+ "eval_runtime": 0.236,
5
+ "eval_samples": 25,
6
+ "eval_samples_per_second": 42.369,
7
+ "eval_steps_per_second": 4.237
8
+ }
runs/Jun13_17-30-31_gpu1-1/events.out.tfevents.1718271477.gpu1-1.651682.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6af2eca768d3c646e1dd6dc73b6ec51741ab4a3bd2a1fc8bfc8b6a32fbf8a75d
3
+ size 52541
runs/Jun13_17-30-31_gpu1-1/events.out.tfevents.1718274013.gpu1-1.651682.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b7052d5bfde8c4dad65848338549e05a092809f168acbaae48eca79d82eefeee
3
+ size 359
special_tokens_map.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>"
5
+ ],
6
+ "bos_token": {
7
+ "content": "<bos>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false
12
+ },
13
+ "eos_token": {
14
+ "content": "<eos>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false
19
+ },
20
+ "pad_token": {
21
+ "content": "<pad>",
22
+ "lstrip": false,
23
+ "normalized": false,
24
+ "rstrip": false,
25
+ "single_word": false
26
+ },
27
+ "unk_token": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false
33
+ }
34
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:322a5f52ab5cab196761ab397a022d6fa3a2e1418585e532bb6efb2fedd2ae94
3
+ size 17477501
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": "<bos>",
59
+ "chat_template": "{% if messages[0]['role'] == 'user' or messages[0]['role'] == 'system' %}{{ bos_token }}{% endif %}{% for message in messages %}{{ '<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n' }}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% elif messages[-1]['role'] == 'assistant' %}{{ eos_token }}{% endif %}",
60
+ "clean_up_tokenization_spaces": false,
61
+ "eos_token": "<eos>",
62
+ "legacy": null,
63
+ "model_max_length": 2048,
64
+ "pad_token": "<pad>",
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 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 9.952153110047847,
3
+ "total_flos": 3.1750846888049377e+18,
4
+ "train_loss": 2.1342261617000284,
5
+ "train_runtime": 2535.4321,
6
+ "train_samples": 63353,
7
+ "train_samples_per_second": 26.287,
8
+ "train_steps_per_second": 0.41
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,1573 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 9.952153110047847,
5
+ "eval_steps": 500,
6
+ "global_step": 1040,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.009569377990430622,
13
+ "grad_norm": 452.0,
14
+ "learning_rate": 1.9230769230769234e-06,
15
+ "loss": 52.8047,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.04784688995215311,
20
+ "grad_norm": 348.0,
21
+ "learning_rate": 9.615384615384616e-06,
22
+ "loss": 50.9309,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.09569377990430622,
27
+ "grad_norm": 151.0,
28
+ "learning_rate": 1.923076923076923e-05,
29
+ "loss": 41.6311,
30
+ "step": 10
31
+ },
32
+ {
33
+ "epoch": 0.14354066985645933,
34
+ "grad_norm": 38.5,
35
+ "learning_rate": 2.8846153846153845e-05,
36
+ "loss": 30.0661,
37
+ "step": 15
38
+ },
39
+ {
40
+ "epoch": 0.19138755980861244,
41
+ "grad_norm": 21.625,
42
+ "learning_rate": 3.846153846153846e-05,
43
+ "loss": 25.0223,
44
+ "step": 20
45
+ },
46
+ {
47
+ "epoch": 0.23923444976076555,
48
+ "grad_norm": 9.5,
49
+ "learning_rate": 4.8076923076923084e-05,
50
+ "loss": 22.4207,
51
+ "step": 25
52
+ },
53
+ {
54
+ "epoch": 0.28708133971291866,
55
+ "grad_norm": 5.21875,
56
+ "learning_rate": 5.769230769230769e-05,
57
+ "loss": 20.4665,
58
+ "step": 30
59
+ },
60
+ {
61
+ "epoch": 0.3349282296650718,
62
+ "grad_norm": 5.25,
63
+ "learning_rate": 6.730769230769232e-05,
64
+ "loss": 19.3811,
65
+ "step": 35
66
+ },
67
+ {
68
+ "epoch": 0.3827751196172249,
69
+ "grad_norm": 7.28125,
70
+ "learning_rate": 7.692307692307693e-05,
71
+ "loss": 18.3511,
72
+ "step": 40
73
+ },
74
+ {
75
+ "epoch": 0.430622009569378,
76
+ "grad_norm": 14.6875,
77
+ "learning_rate": 8.653846153846155e-05,
78
+ "loss": 16.8341,
79
+ "step": 45
80
+ },
81
+ {
82
+ "epoch": 0.4784688995215311,
83
+ "grad_norm": 28.25,
84
+ "learning_rate": 9.615384615384617e-05,
85
+ "loss": 13.7818,
86
+ "step": 50
87
+ },
88
+ {
89
+ "epoch": 0.5263157894736842,
90
+ "grad_norm": 28.375,
91
+ "learning_rate": 0.00010576923076923077,
92
+ "loss": 7.7465,
93
+ "step": 55
94
+ },
95
+ {
96
+ "epoch": 0.5741626794258373,
97
+ "grad_norm": 8.25,
98
+ "learning_rate": 0.00011538461538461538,
99
+ "loss": 3.0182,
100
+ "step": 60
101
+ },
102
+ {
103
+ "epoch": 0.6220095693779905,
104
+ "grad_norm": 6.90625,
105
+ "learning_rate": 0.000125,
106
+ "loss": 2.2631,
107
+ "step": 65
108
+ },
109
+ {
110
+ "epoch": 0.6698564593301436,
111
+ "grad_norm": 16.125,
112
+ "learning_rate": 0.00013461538461538464,
113
+ "loss": 1.8651,
114
+ "step": 70
115
+ },
116
+ {
117
+ "epoch": 0.7177033492822966,
118
+ "grad_norm": 0.83203125,
119
+ "learning_rate": 0.00014423076923076924,
120
+ "loss": 1.6482,
121
+ "step": 75
122
+ },
123
+ {
124
+ "epoch": 0.7655502392344498,
125
+ "grad_norm": 1.3125,
126
+ "learning_rate": 0.00015384615384615385,
127
+ "loss": 1.4966,
128
+ "step": 80
129
+ },
130
+ {
131
+ "epoch": 0.8133971291866029,
132
+ "grad_norm": 1.09375,
133
+ "learning_rate": 0.00016346153846153846,
134
+ "loss": 1.4308,
135
+ "step": 85
136
+ },
137
+ {
138
+ "epoch": 0.861244019138756,
139
+ "grad_norm": 1.0625,
140
+ "learning_rate": 0.0001730769230769231,
141
+ "loss": 1.3404,
142
+ "step": 90
143
+ },
144
+ {
145
+ "epoch": 0.9090909090909091,
146
+ "grad_norm": 1.828125,
147
+ "learning_rate": 0.0001826923076923077,
148
+ "loss": 1.2829,
149
+ "step": 95
150
+ },
151
+ {
152
+ "epoch": 0.9569377990430622,
153
+ "grad_norm": 0.72265625,
154
+ "learning_rate": 0.00019230769230769233,
155
+ "loss": 1.2495,
156
+ "step": 100
157
+ },
158
+ {
159
+ "epoch": 0.9952153110047847,
160
+ "eval_loss": 2.6542556285858154,
161
+ "eval_runtime": 0.2797,
162
+ "eval_samples_per_second": 35.754,
163
+ "eval_steps_per_second": 3.575,
164
+ "step": 104
165
+ },
166
+ {
167
+ "epoch": 1.0047846889952152,
168
+ "grad_norm": 0.90234375,
169
+ "learning_rate": 0.0001999994367286727,
170
+ "loss": 1.2336,
171
+ "step": 105
172
+ },
173
+ {
174
+ "epoch": 1.0526315789473684,
175
+ "grad_norm": 2.046875,
176
+ "learning_rate": 0.00019997972289848503,
177
+ "loss": 1.1928,
178
+ "step": 110
179
+ },
180
+ {
181
+ "epoch": 1.1004784688995215,
182
+ "grad_norm": 9.25,
183
+ "learning_rate": 0.00019993185184710165,
184
+ "loss": 1.1702,
185
+ "step": 115
186
+ },
187
+ {
188
+ "epoch": 1.1483253588516746,
189
+ "grad_norm": 1.1015625,
190
+ "learning_rate": 0.00019985583705641418,
191
+ "loss": 1.1684,
192
+ "step": 120
193
+ },
194
+ {
195
+ "epoch": 1.1961722488038278,
196
+ "grad_norm": 2.859375,
197
+ "learning_rate": 0.00019975169993441627,
198
+ "loss": 1.125,
199
+ "step": 125
200
+ },
201
+ {
202
+ "epoch": 1.244019138755981,
203
+ "grad_norm": 0.76171875,
204
+ "learning_rate": 0.00019961946980917456,
205
+ "loss": 1.1078,
206
+ "step": 130
207
+ },
208
+ {
209
+ "epoch": 1.291866028708134,
210
+ "grad_norm": 1.546875,
211
+ "learning_rate": 0.0001994591839205691,
212
+ "loss": 1.1369,
213
+ "step": 135
214
+ },
215
+ {
216
+ "epoch": 1.339712918660287,
217
+ "grad_norm": 3.328125,
218
+ "learning_rate": 0.0001992708874098054,
219
+ "loss": 1.144,
220
+ "step": 140
221
+ },
222
+ {
223
+ "epoch": 1.38755980861244,
224
+ "grad_norm": 2.53125,
225
+ "learning_rate": 0.00019905463330670143,
226
+ "loss": 1.1301,
227
+ "step": 145
228
+ },
229
+ {
230
+ "epoch": 1.4354066985645932,
231
+ "grad_norm": 2.890625,
232
+ "learning_rate": 0.0001988104825147528,
233
+ "loss": 1.146,
234
+ "step": 150
235
+ },
236
+ {
237
+ "epoch": 1.4832535885167464,
238
+ "grad_norm": 2.453125,
239
+ "learning_rate": 0.0001985385037939806,
240
+ "loss": 1.0953,
241
+ "step": 155
242
+ },
243
+ {
244
+ "epoch": 1.5311004784688995,
245
+ "grad_norm": 2.1875,
246
+ "learning_rate": 0.00019823877374156647,
247
+ "loss": 1.0699,
248
+ "step": 160
249
+ },
250
+ {
251
+ "epoch": 1.5789473684210527,
252
+ "grad_norm": 1.984375,
253
+ "learning_rate": 0.00019791137677028082,
254
+ "loss": 1.0699,
255
+ "step": 165
256
+ },
257
+ {
258
+ "epoch": 1.6267942583732058,
259
+ "grad_norm": 2.296875,
260
+ "learning_rate": 0.00019755640508470942,
261
+ "loss": 1.0545,
262
+ "step": 170
263
+ },
264
+ {
265
+ "epoch": 1.674641148325359,
266
+ "grad_norm": 1.8828125,
267
+ "learning_rate": 0.00019717395865528602,
268
+ "loss": 1.0404,
269
+ "step": 175
270
+ },
271
+ {
272
+ "epoch": 1.722488038277512,
273
+ "grad_norm": 2.015625,
274
+ "learning_rate": 0.00019676414519013781,
275
+ "loss": 1.0512,
276
+ "step": 180
277
+ },
278
+ {
279
+ "epoch": 1.7703349282296652,
280
+ "grad_norm": 6.75,
281
+ "learning_rate": 0.00019632708010475165,
282
+ "loss": 1.0382,
283
+ "step": 185
284
+ },
285
+ {
286
+ "epoch": 1.8181818181818183,
287
+ "grad_norm": 0.8046875,
288
+ "learning_rate": 0.00019586288648946947,
289
+ "loss": 1.0383,
290
+ "step": 190
291
+ },
292
+ {
293
+ "epoch": 1.8660287081339713,
294
+ "grad_norm": 0.875,
295
+ "learning_rate": 0.0001953716950748227,
296
+ "loss": 1.0356,
297
+ "step": 195
298
+ },
299
+ {
300
+ "epoch": 1.9138755980861244,
301
+ "grad_norm": 6.03125,
302
+ "learning_rate": 0.00019485364419471454,
303
+ "loss": 1.0354,
304
+ "step": 200
305
+ },
306
+ {
307
+ "epoch": 1.9617224880382775,
308
+ "grad_norm": 1.640625,
309
+ "learning_rate": 0.0001943088797474612,
310
+ "loss": 1.0258,
311
+ "step": 205
312
+ },
313
+ {
314
+ "epoch": 2.0,
315
+ "eval_loss": 2.5231521129608154,
316
+ "eval_runtime": 0.2369,
317
+ "eval_samples_per_second": 42.205,
318
+ "eval_steps_per_second": 4.22,
319
+ "step": 209
320
+ },
321
+ {
322
+ "epoch": 2.0095693779904304,
323
+ "grad_norm": 1.578125,
324
+ "learning_rate": 0.00019373755515470254,
325
+ "loss": 1.0034,
326
+ "step": 210
327
+ },
328
+ {
329
+ "epoch": 2.0574162679425836,
330
+ "grad_norm": 1.5703125,
331
+ "learning_rate": 0.00019313983131819407,
332
+ "loss": 0.9855,
333
+ "step": 215
334
+ },
335
+ {
336
+ "epoch": 2.1052631578947367,
337
+ "grad_norm": 4.4375,
338
+ "learning_rate": 0.00019251587657449236,
339
+ "loss": 0.9769,
340
+ "step": 220
341
+ },
342
+ {
343
+ "epoch": 2.15311004784689,
344
+ "grad_norm": 1.8359375,
345
+ "learning_rate": 0.0001918658666475465,
346
+ "loss": 0.9829,
347
+ "step": 225
348
+ },
349
+ {
350
+ "epoch": 2.200956937799043,
351
+ "grad_norm": 0.7421875,
352
+ "learning_rate": 0.00019118998459920902,
353
+ "loss": 0.9843,
354
+ "step": 230
355
+ },
356
+ {
357
+ "epoch": 2.248803827751196,
358
+ "grad_norm": 2.578125,
359
+ "learning_rate": 0.0001904884207776804,
360
+ "loss": 0.9876,
361
+ "step": 235
362
+ },
363
+ {
364
+ "epoch": 2.2966507177033493,
365
+ "grad_norm": 1.5390625,
366
+ "learning_rate": 0.0001897613727639014,
367
+ "loss": 0.9713,
368
+ "step": 240
369
+ },
370
+ {
371
+ "epoch": 2.3444976076555024,
372
+ "grad_norm": 1.9609375,
373
+ "learning_rate": 0.00018900904531590846,
374
+ "loss": 0.9706,
375
+ "step": 245
376
+ },
377
+ {
378
+ "epoch": 2.3923444976076556,
379
+ "grad_norm": 2.453125,
380
+ "learning_rate": 0.0001882316503111678,
381
+ "loss": 0.9635,
382
+ "step": 250
383
+ },
384
+ {
385
+ "epoch": 2.4401913875598087,
386
+ "grad_norm": 1.6484375,
387
+ "learning_rate": 0.00018742940668690464,
388
+ "loss": 0.9608,
389
+ "step": 255
390
+ },
391
+ {
392
+ "epoch": 2.488038277511962,
393
+ "grad_norm": 1.46875,
394
+ "learning_rate": 0.00018660254037844388,
395
+ "loss": 0.9498,
396
+ "step": 260
397
+ },
398
+ {
399
+ "epoch": 2.535885167464115,
400
+ "grad_norm": 4.34375,
401
+ "learning_rate": 0.00018575128425558023,
402
+ "loss": 0.9539,
403
+ "step": 265
404
+ },
405
+ {
406
+ "epoch": 2.583732057416268,
407
+ "grad_norm": 2.90625,
408
+ "learning_rate": 0.00018487587805699526,
409
+ "loss": 0.96,
410
+ "step": 270
411
+ },
412
+ {
413
+ "epoch": 2.6315789473684212,
414
+ "grad_norm": 0.92578125,
415
+ "learning_rate": 0.0001839765683227398,
416
+ "loss": 0.9431,
417
+ "step": 275
418
+ },
419
+ {
420
+ "epoch": 2.679425837320574,
421
+ "grad_norm": 4.375,
422
+ "learning_rate": 0.00018305360832480117,
423
+ "loss": 0.9457,
424
+ "step": 280
425
+ },
426
+ {
427
+ "epoch": 2.7272727272727275,
428
+ "grad_norm": 11.1875,
429
+ "learning_rate": 0.00018210725799577439,
430
+ "loss": 0.9594,
431
+ "step": 285
432
+ },
433
+ {
434
+ "epoch": 2.77511961722488,
435
+ "grad_norm": 1.078125,
436
+ "learning_rate": 0.00018113778385565733,
437
+ "loss": 0.9542,
438
+ "step": 290
439
+ },
440
+ {
441
+ "epoch": 2.8229665071770333,
442
+ "grad_norm": 3.734375,
443
+ "learning_rate": 0.00018014545893679115,
444
+ "loss": 0.9407,
445
+ "step": 295
446
+ },
447
+ {
448
+ "epoch": 2.8708133971291865,
449
+ "grad_norm": 1.9765625,
450
+ "learning_rate": 0.0001791305627069662,
451
+ "loss": 0.9418,
452
+ "step": 300
453
+ },
454
+ {
455
+ "epoch": 2.9186602870813396,
456
+ "grad_norm": 0.88671875,
457
+ "learning_rate": 0.00017809338099071577,
458
+ "loss": 0.9357,
459
+ "step": 305
460
+ },
461
+ {
462
+ "epoch": 2.9665071770334928,
463
+ "grad_norm": 1.046875,
464
+ "learning_rate": 0.00017703420588881946,
465
+ "loss": 0.9351,
466
+ "step": 310
467
+ },
468
+ {
469
+ "epoch": 2.9952153110047846,
470
+ "eval_loss": 2.466839551925659,
471
+ "eval_runtime": 0.257,
472
+ "eval_samples_per_second": 38.912,
473
+ "eval_steps_per_second": 3.891,
474
+ "step": 313
475
+ },
476
+ {
477
+ "epoch": 3.014354066985646,
478
+ "grad_norm": 1.7109375,
479
+ "learning_rate": 0.0001759533356960391,
480
+ "loss": 0.9283,
481
+ "step": 315
482
+ },
483
+ {
484
+ "epoch": 3.062200956937799,
485
+ "grad_norm": 2.0625,
486
+ "learning_rate": 0.00017485107481711012,
487
+ "loss": 0.9187,
488
+ "step": 320
489
+ },
490
+ {
491
+ "epoch": 3.110047846889952,
492
+ "grad_norm": 1.8125,
493
+ "learning_rate": 0.0001737277336810124,
494
+ "loss": 0.9121,
495
+ "step": 325
496
+ },
497
+ {
498
+ "epoch": 3.1578947368421053,
499
+ "grad_norm": 3.6875,
500
+ "learning_rate": 0.00017258362865354426,
501
+ "loss": 0.9481,
502
+ "step": 330
503
+ },
504
+ {
505
+ "epoch": 3.2057416267942584,
506
+ "grad_norm": 3.84375,
507
+ "learning_rate": 0.00017141908194822446,
508
+ "loss": 0.9215,
509
+ "step": 335
510
+ },
511
+ {
512
+ "epoch": 3.2535885167464116,
513
+ "grad_norm": 8.625,
514
+ "learning_rate": 0.00017023442153554777,
515
+ "loss": 0.9082,
516
+ "step": 340
517
+ },
518
+ {
519
+ "epoch": 3.3014354066985647,
520
+ "grad_norm": 0.94140625,
521
+ "learning_rate": 0.00016902998105061844,
522
+ "loss": 0.9035,
523
+ "step": 345
524
+ },
525
+ {
526
+ "epoch": 3.349282296650718,
527
+ "grad_norm": 0.9921875,
528
+ "learning_rate": 0.0001678060996991891,
529
+ "loss": 0.9084,
530
+ "step": 350
531
+ },
532
+ {
533
+ "epoch": 3.397129186602871,
534
+ "grad_norm": 1.1484375,
535
+ "learning_rate": 0.00016656312216213034,
536
+ "loss": 0.8981,
537
+ "step": 355
538
+ },
539
+ {
540
+ "epoch": 3.444976076555024,
541
+ "grad_norm": 1.0859375,
542
+ "learning_rate": 0.0001653013984983585,
543
+ "loss": 0.9024,
544
+ "step": 360
545
+ },
546
+ {
547
+ "epoch": 3.492822966507177,
548
+ "grad_norm": 0.80078125,
549
+ "learning_rate": 0.00016402128404624882,
550
+ "loss": 0.9027,
551
+ "step": 365
552
+ },
553
+ {
554
+ "epoch": 3.5406698564593304,
555
+ "grad_norm": 0.8125,
556
+ "learning_rate": 0.00016272313932356162,
557
+ "loss": 0.8924,
558
+ "step": 370
559
+ },
560
+ {
561
+ "epoch": 3.588516746411483,
562
+ "grad_norm": 0.99609375,
563
+ "learning_rate": 0.0001614073299259101,
564
+ "loss": 0.8943,
565
+ "step": 375
566
+ },
567
+ {
568
+ "epoch": 3.6363636363636362,
569
+ "grad_norm": 1.6796875,
570
+ "learning_rate": 0.0001600742264237979,
571
+ "loss": 0.9152,
572
+ "step": 380
573
+ },
574
+ {
575
+ "epoch": 3.6842105263157894,
576
+ "grad_norm": 1.9375,
577
+ "learning_rate": 0.0001587242042582554,
578
+ "loss": 0.9076,
579
+ "step": 385
580
+ },
581
+ {
582
+ "epoch": 3.7320574162679425,
583
+ "grad_norm": 1.859375,
584
+ "learning_rate": 0.0001573576436351046,
585
+ "loss": 0.8935,
586
+ "step": 390
587
+ },
588
+ {
589
+ "epoch": 3.7799043062200957,
590
+ "grad_norm": 1.1328125,
591
+ "learning_rate": 0.00015597492941788222,
592
+ "loss": 0.8895,
593
+ "step": 395
594
+ },
595
+ {
596
+ "epoch": 3.827751196172249,
597
+ "grad_norm": 0.89453125,
598
+ "learning_rate": 0.00015457645101945046,
599
+ "loss": 0.8767,
600
+ "step": 400
601
+ },
602
+ {
603
+ "epoch": 3.875598086124402,
604
+ "grad_norm": 3.609375,
605
+ "learning_rate": 0.00015316260229232727,
606
+ "loss": 0.8795,
607
+ "step": 405
608
+ },
609
+ {
610
+ "epoch": 3.923444976076555,
611
+ "grad_norm": 2.40625,
612
+ "learning_rate": 0.00015173378141776568,
613
+ "loss": 0.8847,
614
+ "step": 410
615
+ },
616
+ {
617
+ "epoch": 3.971291866028708,
618
+ "grad_norm": 4.0,
619
+ "learning_rate": 0.00015029039079361448,
620
+ "loss": 0.8914,
621
+ "step": 415
622
+ },
623
+ {
624
+ "epoch": 4.0,
625
+ "eval_loss": 2.4789059162139893,
626
+ "eval_runtime": 0.2403,
627
+ "eval_samples_per_second": 41.62,
628
+ "eval_steps_per_second": 4.162,
629
+ "step": 418
630
+ },
631
+ {
632
+ "epoch": 4.019138755980861,
633
+ "grad_norm": 2.28125,
634
+ "learning_rate": 0.00014883283692099112,
635
+ "loss": 0.8795,
636
+ "step": 420
637
+ },
638
+ {
639
+ "epoch": 4.0669856459330145,
640
+ "grad_norm": 1.890625,
641
+ "learning_rate": 0.00014736153028979893,
642
+ "loss": 0.8598,
643
+ "step": 425
644
+ },
645
+ {
646
+ "epoch": 4.114832535885167,
647
+ "grad_norm": 2.984375,
648
+ "learning_rate": 0.00014587688526312143,
649
+ "loss": 0.8533,
650
+ "step": 430
651
+ },
652
+ {
653
+ "epoch": 4.162679425837321,
654
+ "grad_norm": 2.453125,
655
+ "learning_rate": 0.00014437931996052518,
656
+ "loss": 0.8544,
657
+ "step": 435
658
+ },
659
+ {
660
+ "epoch": 4.2105263157894735,
661
+ "grad_norm": 8.4375,
662
+ "learning_rate": 0.00014286925614030542,
663
+ "loss": 0.8635,
664
+ "step": 440
665
+ },
666
+ {
667
+ "epoch": 4.258373205741627,
668
+ "grad_norm": 4.34375,
669
+ "learning_rate": 0.00014134711908070631,
670
+ "loss": 0.8664,
671
+ "step": 445
672
+ },
673
+ {
674
+ "epoch": 4.30622009569378,
675
+ "grad_norm": 0.83984375,
676
+ "learning_rate": 0.0001398133374601501,
677
+ "loss": 0.8567,
678
+ "step": 450
679
+ },
680
+ {
681
+ "epoch": 4.354066985645933,
682
+ "grad_norm": 0.73828125,
683
+ "learning_rate": 0.000138268343236509,
684
+ "loss": 0.8579,
685
+ "step": 455
686
+ },
687
+ {
688
+ "epoch": 4.401913875598086,
689
+ "grad_norm": 1.3203125,
690
+ "learning_rate": 0.00013671257152545277,
691
+ "loss": 0.8562,
692
+ "step": 460
693
+ },
694
+ {
695
+ "epoch": 4.44976076555024,
696
+ "grad_norm": 1.140625,
697
+ "learning_rate": 0.00013514646047790775,
698
+ "loss": 0.8527,
699
+ "step": 465
700
+ },
701
+ {
702
+ "epoch": 4.497607655502392,
703
+ "grad_norm": 0.80078125,
704
+ "learning_rate": 0.0001335704511566605,
705
+ "loss": 0.8474,
706
+ "step": 470
707
+ },
708
+ {
709
+ "epoch": 4.545454545454545,
710
+ "grad_norm": 2.625,
711
+ "learning_rate": 0.00013198498741214166,
712
+ "loss": 0.856,
713
+ "step": 475
714
+ },
715
+ {
716
+ "epoch": 4.5933014354066986,
717
+ "grad_norm": 0.91015625,
718
+ "learning_rate": 0.0001303905157574247,
719
+ "loss": 0.8586,
720
+ "step": 480
721
+ },
722
+ {
723
+ "epoch": 4.641148325358852,
724
+ "grad_norm": 1.28125,
725
+ "learning_rate": 0.00012878748524247462,
726
+ "loss": 0.8519,
727
+ "step": 485
728
+ },
729
+ {
730
+ "epoch": 4.688995215311005,
731
+ "grad_norm": 1.609375,
732
+ "learning_rate": 0.00012717634732768243,
733
+ "loss": 0.8483,
734
+ "step": 490
735
+ },
736
+ {
737
+ "epoch": 4.7368421052631575,
738
+ "grad_norm": 1.6171875,
739
+ "learning_rate": 0.0001255575557567207,
740
+ "loss": 0.8663,
741
+ "step": 495
742
+ },
743
+ {
744
+ "epoch": 4.784688995215311,
745
+ "grad_norm": 0.890625,
746
+ "learning_rate": 0.0001239315664287558,
747
+ "loss": 0.8684,
748
+ "step": 500
749
+ },
750
+ {
751
+ "epoch": 4.832535885167464,
752
+ "grad_norm": 1.3828125,
753
+ "learning_rate": 0.00012229883727005365,
754
+ "loss": 0.8391,
755
+ "step": 505
756
+ },
757
+ {
758
+ "epoch": 4.880382775119617,
759
+ "grad_norm": 1.21875,
760
+ "learning_rate": 0.00012065982810501404,
761
+ "loss": 0.8491,
762
+ "step": 510
763
+ },
764
+ {
765
+ "epoch": 4.92822966507177,
766
+ "grad_norm": 3.765625,
767
+ "learning_rate": 0.00011901500052667068,
768
+ "loss": 0.8435,
769
+ "step": 515
770
+ },
771
+ {
772
+ "epoch": 4.976076555023924,
773
+ "grad_norm": 6.75,
774
+ "learning_rate": 0.00011736481776669306,
775
+ "loss": 0.8487,
776
+ "step": 520
777
+ },
778
+ {
779
+ "epoch": 4.9952153110047846,
780
+ "eval_loss": 2.475181818008423,
781
+ "eval_runtime": 0.2635,
782
+ "eval_samples_per_second": 37.955,
783
+ "eval_steps_per_second": 3.796,
784
+ "step": 522
785
+ },
786
+ {
787
+ "epoch": 5.023923444976076,
788
+ "grad_norm": 6.28125,
789
+ "learning_rate": 0.00011570974456492678,
790
+ "loss": 0.845,
791
+ "step": 525
792
+ },
793
+ {
794
+ "epoch": 5.07177033492823,
795
+ "grad_norm": 12.1875,
796
+ "learning_rate": 0.00011405024703850929,
797
+ "loss": 0.8426,
798
+ "step": 530
799
+ },
800
+ {
801
+ "epoch": 5.119617224880383,
802
+ "grad_norm": 7.71875,
803
+ "learning_rate": 0.00011238679255059752,
804
+ "loss": 0.8348,
805
+ "step": 535
806
+ },
807
+ {
808
+ "epoch": 5.167464114832536,
809
+ "grad_norm": 6.3125,
810
+ "learning_rate": 0.00011071984957874479,
811
+ "loss": 0.829,
812
+ "step": 540
813
+ },
814
+ {
815
+ "epoch": 5.215311004784689,
816
+ "grad_norm": 0.578125,
817
+ "learning_rate": 0.0001090498875829638,
818
+ "loss": 0.8233,
819
+ "step": 545
820
+ },
821
+ {
822
+ "epoch": 5.2631578947368425,
823
+ "grad_norm": 2.03125,
824
+ "learning_rate": 0.00010737737687351284,
825
+ "loss": 0.8409,
826
+ "step": 550
827
+ },
828
+ {
829
+ "epoch": 5.311004784688995,
830
+ "grad_norm": 0.984375,
831
+ "learning_rate": 0.00010570278847844275,
832
+ "loss": 0.8217,
833
+ "step": 555
834
+ },
835
+ {
836
+ "epoch": 5.358851674641148,
837
+ "grad_norm": 1.4609375,
838
+ "learning_rate": 0.00010402659401094152,
839
+ "loss": 0.8108,
840
+ "step": 560
841
+ },
842
+ {
843
+ "epoch": 5.4066985645933014,
844
+ "grad_norm": 1.2421875,
845
+ "learning_rate": 0.00010234926553651422,
846
+ "loss": 0.8351,
847
+ "step": 565
848
+ },
849
+ {
850
+ "epoch": 5.454545454545454,
851
+ "grad_norm": 1.140625,
852
+ "learning_rate": 0.00010067127544003563,
853
+ "loss": 0.8194,
854
+ "step": 570
855
+ },
856
+ {
857
+ "epoch": 5.502392344497608,
858
+ "grad_norm": 0.98046875,
859
+ "learning_rate": 9.899309629271246e-05,
860
+ "loss": 0.8161,
861
+ "step": 575
862
+ },
863
+ {
864
+ "epoch": 5.55023923444976,
865
+ "grad_norm": 1.1875,
866
+ "learning_rate": 9.73152007189939e-05,
867
+ "loss": 0.8156,
868
+ "step": 580
869
+ },
870
+ {
871
+ "epoch": 5.598086124401914,
872
+ "grad_norm": 1.4375,
873
+ "learning_rate": 9.563806126346642e-05,
874
+ "loss": 0.8183,
875
+ "step": 585
876
+ },
877
+ {
878
+ "epoch": 5.645933014354067,
879
+ "grad_norm": 0.8359375,
880
+ "learning_rate": 9.396215025777139e-05,
881
+ "loss": 0.8079,
882
+ "step": 590
883
+ },
884
+ {
885
+ "epoch": 5.69377990430622,
886
+ "grad_norm": 1.234375,
887
+ "learning_rate": 9.22879396875828e-05,
888
+ "loss": 0.821,
889
+ "step": 595
890
+ },
891
+ {
892
+ "epoch": 5.741626794258373,
893
+ "grad_norm": 0.85546875,
894
+ "learning_rate": 9.061590105968208e-05,
895
+ "loss": 0.8078,
896
+ "step": 600
897
+ },
898
+ {
899
+ "epoch": 5.7894736842105265,
900
+ "grad_norm": 1.25,
901
+ "learning_rate": 8.894650526916803e-05,
902
+ "loss": 0.8042,
903
+ "step": 605
904
+ },
905
+ {
906
+ "epoch": 5.837320574162679,
907
+ "grad_norm": 1.046875,
908
+ "learning_rate": 8.728022246683894e-05,
909
+ "loss": 0.8298,
910
+ "step": 610
911
+ },
912
+ {
913
+ "epoch": 5.885167464114833,
914
+ "grad_norm": 2.34375,
915
+ "learning_rate": 8.561752192678443e-05,
916
+ "loss": 0.8122,
917
+ "step": 615
918
+ },
919
+ {
920
+ "epoch": 5.9330143540669855,
921
+ "grad_norm": 12.75,
922
+ "learning_rate": 8.395887191422397e-05,
923
+ "loss": 0.819,
924
+ "step": 620
925
+ },
926
+ {
927
+ "epoch": 5.980861244019139,
928
+ "grad_norm": 1.5703125,
929
+ "learning_rate": 8.23047395536298e-05,
930
+ "loss": 0.8222,
931
+ "step": 625
932
+ },
933
+ {
934
+ "epoch": 6.0,
935
+ "eval_loss": 2.4928200244903564,
936
+ "eval_runtime": 0.2377,
937
+ "eval_samples_per_second": 42.076,
938
+ "eval_steps_per_second": 4.208,
939
+ "step": 627
940
+ },
941
+ {
942
+ "epoch": 6.028708133971292,
943
+ "grad_norm": 2.8125,
944
+ "learning_rate": 8.065559069717088e-05,
945
+ "loss": 0.8044,
946
+ "step": 630
947
+ },
948
+ {
949
+ "epoch": 6.076555023923445,
950
+ "grad_norm": 12.0,
951
+ "learning_rate": 7.901188979351526e-05,
952
+ "loss": 0.8055,
953
+ "step": 635
954
+ },
955
+ {
956
+ "epoch": 6.124401913875598,
957
+ "grad_norm": 1.46875,
958
+ "learning_rate": 7.73740997570278e-05,
959
+ "loss": 0.7972,
960
+ "step": 640
961
+ },
962
+ {
963
+ "epoch": 6.172248803827751,
964
+ "grad_norm": 5.34375,
965
+ "learning_rate": 7.574268183739989e-05,
966
+ "loss": 0.8078,
967
+ "step": 645
968
+ },
969
+ {
970
+ "epoch": 6.220095693779904,
971
+ "grad_norm": 4.5,
972
+ "learning_rate": 7.411809548974792e-05,
973
+ "loss": 0.7981,
974
+ "step": 650
975
+ },
976
+ {
977
+ "epoch": 6.267942583732057,
978
+ "grad_norm": 1.1328125,
979
+ "learning_rate": 7.250079824521743e-05,
980
+ "loss": 0.8073,
981
+ "step": 655
982
+ },
983
+ {
984
+ "epoch": 6.315789473684211,
985
+ "grad_norm": 5.21875,
986
+ "learning_rate": 7.089124558212871e-05,
987
+ "loss": 0.8028,
988
+ "step": 660
989
+ },
990
+ {
991
+ "epoch": 6.363636363636363,
992
+ "grad_norm": 6.84375,
993
+ "learning_rate": 6.928989079770094e-05,
994
+ "loss": 0.7954,
995
+ "step": 665
996
+ },
997
+ {
998
+ "epoch": 6.411483253588517,
999
+ "grad_norm": 1.1640625,
1000
+ "learning_rate": 6.769718488039023e-05,
1001
+ "loss": 0.7899,
1002
+ "step": 670
1003
+ },
1004
+ {
1005
+ "epoch": 6.45933014354067,
1006
+ "grad_norm": 7.21875,
1007
+ "learning_rate": 6.611357638287823e-05,
1008
+ "loss": 0.8025,
1009
+ "step": 675
1010
+ },
1011
+ {
1012
+ "epoch": 6.507177033492823,
1013
+ "grad_norm": 5.1875,
1014
+ "learning_rate": 6.453951129574644e-05,
1015
+ "loss": 0.7917,
1016
+ "step": 680
1017
+ },
1018
+ {
1019
+ "epoch": 6.555023923444976,
1020
+ "grad_norm": 0.7421875,
1021
+ "learning_rate": 6.297543292187215e-05,
1022
+ "loss": 0.7903,
1023
+ "step": 685
1024
+ },
1025
+ {
1026
+ "epoch": 6.6028708133971294,
1027
+ "grad_norm": 2.390625,
1028
+ "learning_rate": 6.142178175158149e-05,
1029
+ "loss": 0.7894,
1030
+ "step": 690
1031
+ },
1032
+ {
1033
+ "epoch": 6.650717703349282,
1034
+ "grad_norm": 2.46875,
1035
+ "learning_rate": 5.9878995338594224e-05,
1036
+ "loss": 0.7887,
1037
+ "step": 695
1038
+ },
1039
+ {
1040
+ "epoch": 6.698564593301436,
1041
+ "grad_norm": 0.52734375,
1042
+ "learning_rate": 5.834750817679606e-05,
1043
+ "loss": 0.7866,
1044
+ "step": 700
1045
+ },
1046
+ {
1047
+ "epoch": 6.746411483253588,
1048
+ "grad_norm": 1.140625,
1049
+ "learning_rate": 5.682775157787213e-05,
1050
+ "loss": 0.7845,
1051
+ "step": 705
1052
+ },
1053
+ {
1054
+ "epoch": 6.794258373205742,
1055
+ "grad_norm": 0.9296875,
1056
+ "learning_rate": 5.5320153549837415e-05,
1057
+ "loss": 0.7874,
1058
+ "step": 710
1059
+ },
1060
+ {
1061
+ "epoch": 6.842105263157895,
1062
+ "grad_norm": 0.98046875,
1063
+ "learning_rate": 5.382513867649663e-05,
1064
+ "loss": 0.7865,
1065
+ "step": 715
1066
+ },
1067
+ {
1068
+ "epoch": 6.889952153110048,
1069
+ "grad_norm": 0.8515625,
1070
+ "learning_rate": 5.234312799786921e-05,
1071
+ "loss": 0.7866,
1072
+ "step": 720
1073
+ },
1074
+ {
1075
+ "epoch": 6.937799043062201,
1076
+ "grad_norm": 3.40625,
1077
+ "learning_rate": 5.087453889161229e-05,
1078
+ "loss": 0.7742,
1079
+ "step": 725
1080
+ },
1081
+ {
1082
+ "epoch": 6.985645933014354,
1083
+ "grad_norm": 0.546875,
1084
+ "learning_rate": 4.9419784955474524e-05,
1085
+ "loss": 0.7746,
1086
+ "step": 730
1087
+ },
1088
+ {
1089
+ "epoch": 6.9952153110047846,
1090
+ "eval_loss": 2.4925007820129395,
1091
+ "eval_runtime": 0.2584,
1092
+ "eval_samples_per_second": 38.704,
1093
+ "eval_steps_per_second": 3.87,
1094
+ "step": 731
1095
+ },
1096
+ {
1097
+ "epoch": 7.033492822966507,
1098
+ "grad_norm": 0.95703125,
1099
+ "learning_rate": 4.797927589081509e-05,
1100
+ "loss": 0.7674,
1101
+ "step": 735
1102
+ },
1103
+ {
1104
+ "epoch": 7.08133971291866,
1105
+ "grad_norm": 0.5,
1106
+ "learning_rate": 4.6553417387219886e-05,
1107
+ "loss": 0.7691,
1108
+ "step": 740
1109
+ },
1110
+ {
1111
+ "epoch": 7.1291866028708135,
1112
+ "grad_norm": 0.5234375,
1113
+ "learning_rate": 4.514261100824709e-05,
1114
+ "loss": 0.7602,
1115
+ "step": 745
1116
+ },
1117
+ {
1118
+ "epoch": 7.177033492822966,
1119
+ "grad_norm": 0.41015625,
1120
+ "learning_rate": 4.374725407833532e-05,
1121
+ "loss": 0.7656,
1122
+ "step": 750
1123
+ },
1124
+ {
1125
+ "epoch": 7.22488038277512,
1126
+ "grad_norm": 0.6171875,
1127
+ "learning_rate": 4.236773957090548e-05,
1128
+ "loss": 0.7673,
1129
+ "step": 755
1130
+ },
1131
+ {
1132
+ "epoch": 7.2727272727272725,
1133
+ "grad_norm": 0.53515625,
1134
+ "learning_rate": 4.100445599768774e-05,
1135
+ "loss": 0.7712,
1136
+ "step": 760
1137
+ },
1138
+ {
1139
+ "epoch": 7.320574162679426,
1140
+ "grad_norm": 0.546875,
1141
+ "learning_rate": 3.96577872993053e-05,
1142
+ "loss": 0.7658,
1143
+ "step": 765
1144
+ },
1145
+ {
1146
+ "epoch": 7.368421052631579,
1147
+ "grad_norm": 0.671875,
1148
+ "learning_rate": 3.832811273714569e-05,
1149
+ "loss": 0.7612,
1150
+ "step": 770
1151
+ },
1152
+ {
1153
+ "epoch": 7.416267942583732,
1154
+ "grad_norm": 0.90234375,
1155
+ "learning_rate": 3.701580678654925e-05,
1156
+ "loss": 0.7695,
1157
+ "step": 775
1158
+ },
1159
+ {
1160
+ "epoch": 7.464114832535885,
1161
+ "grad_norm": 0.54296875,
1162
+ "learning_rate": 3.5721239031346066e-05,
1163
+ "loss": 0.7702,
1164
+ "step": 780
1165
+ },
1166
+ {
1167
+ "epoch": 7.511961722488039,
1168
+ "grad_norm": 0.494140625,
1169
+ "learning_rate": 3.4444774059770536e-05,
1170
+ "loss": 0.7654,
1171
+ "step": 785
1172
+ },
1173
+ {
1174
+ "epoch": 7.559808612440191,
1175
+ "grad_norm": 0.61328125,
1176
+ "learning_rate": 3.318677136178228e-05,
1177
+ "loss": 0.763,
1178
+ "step": 790
1179
+ },
1180
+ {
1181
+ "epoch": 7.607655502392344,
1182
+ "grad_norm": 0.7265625,
1183
+ "learning_rate": 3.1947585227823394e-05,
1184
+ "loss": 0.772,
1185
+ "step": 795
1186
+ },
1187
+ {
1188
+ "epoch": 7.655502392344498,
1189
+ "grad_norm": 0.396484375,
1190
+ "learning_rate": 3.072756464904006e-05,
1191
+ "loss": 0.7763,
1192
+ "step": 800
1193
+ },
1194
+ {
1195
+ "epoch": 7.703349282296651,
1196
+ "grad_norm": 0.400390625,
1197
+ "learning_rate": 2.9527053218996037e-05,
1198
+ "loss": 0.7663,
1199
+ "step": 805
1200
+ },
1201
+ {
1202
+ "epoch": 7.751196172248804,
1203
+ "grad_norm": 0.427734375,
1204
+ "learning_rate": 2.8346389036906828e-05,
1205
+ "loss": 0.7635,
1206
+ "step": 810
1207
+ },
1208
+ {
1209
+ "epoch": 7.7990430622009566,
1210
+ "grad_norm": 0.466796875,
1211
+ "learning_rate": 2.7185904612421176e-05,
1212
+ "loss": 0.7679,
1213
+ "step": 815
1214
+ },
1215
+ {
1216
+ "epoch": 7.84688995215311,
1217
+ "grad_norm": 0.61328125,
1218
+ "learning_rate": 2.6045926771976303e-05,
1219
+ "loss": 0.7752,
1220
+ "step": 820
1221
+ },
1222
+ {
1223
+ "epoch": 7.894736842105263,
1224
+ "grad_norm": 0.462890625,
1225
+ "learning_rate": 2.492677656675414e-05,
1226
+ "loss": 0.7733,
1227
+ "step": 825
1228
+ },
1229
+ {
1230
+ "epoch": 7.942583732057416,
1231
+ "grad_norm": 0.392578125,
1232
+ "learning_rate": 2.382876918226409e-05,
1233
+ "loss": 0.7713,
1234
+ "step": 830
1235
+ },
1236
+ {
1237
+ "epoch": 7.990430622009569,
1238
+ "grad_norm": 0.5078125,
1239
+ "learning_rate": 2.2752213849577188e-05,
1240
+ "loss": 0.7644,
1241
+ "step": 835
1242
+ },
1243
+ {
1244
+ "epoch": 8.0,
1245
+ "eval_loss": 2.5051019191741943,
1246
+ "eval_runtime": 0.2382,
1247
+ "eval_samples_per_second": 41.986,
1248
+ "eval_steps_per_second": 4.199,
1249
+ "step": 836
1250
+ },
1251
+ {
1252
+ "epoch": 8.038277511961722,
1253
+ "grad_norm": 0.400390625,
1254
+ "learning_rate": 2.1697413758237784e-05,
1255
+ "loss": 0.7573,
1256
+ "step": 840
1257
+ },
1258
+ {
1259
+ "epoch": 8.086124401913876,
1260
+ "grad_norm": 0.451171875,
1261
+ "learning_rate": 2.0664665970876496e-05,
1262
+ "loss": 0.7533,
1263
+ "step": 845
1264
+ },
1265
+ {
1266
+ "epoch": 8.133971291866029,
1267
+ "grad_norm": 0.494140625,
1268
+ "learning_rate": 1.965426133954854e-05,
1269
+ "loss": 0.7638,
1270
+ "step": 850
1271
+ },
1272
+ {
1273
+ "epoch": 8.181818181818182,
1274
+ "grad_norm": 0.55078125,
1275
+ "learning_rate": 1.8666484423821373e-05,
1276
+ "loss": 0.7594,
1277
+ "step": 855
1278
+ },
1279
+ {
1280
+ "epoch": 8.229665071770334,
1281
+ "grad_norm": 0.6015625,
1282
+ "learning_rate": 1.7701613410634365e-05,
1283
+ "loss": 0.758,
1284
+ "step": 860
1285
+ },
1286
+ {
1287
+ "epoch": 8.277511961722489,
1288
+ "grad_norm": 0.55859375,
1289
+ "learning_rate": 1.6759920035953093e-05,
1290
+ "loss": 0.7619,
1291
+ "step": 865
1292
+ },
1293
+ {
1294
+ "epoch": 8.325358851674642,
1295
+ "grad_norm": 0.3828125,
1296
+ "learning_rate": 1.584166950824061e-05,
1297
+ "loss": 0.7578,
1298
+ "step": 870
1299
+ },
1300
+ {
1301
+ "epoch": 8.373205741626794,
1302
+ "grad_norm": 0.380859375,
1303
+ "learning_rate": 1.4947120433767047e-05,
1304
+ "loss": 0.7544,
1305
+ "step": 875
1306
+ },
1307
+ {
1308
+ "epoch": 8.421052631578947,
1309
+ "grad_norm": 0.388671875,
1310
+ "learning_rate": 1.4076524743778319e-05,
1311
+ "loss": 0.7606,
1312
+ "step": 880
1313
+ },
1314
+ {
1315
+ "epoch": 8.4688995215311,
1316
+ "grad_norm": 0.39453125,
1317
+ "learning_rate": 1.3230127623545064e-05,
1318
+ "loss": 0.7512,
1319
+ "step": 885
1320
+ },
1321
+ {
1322
+ "epoch": 8.516746411483254,
1323
+ "grad_norm": 0.439453125,
1324
+ "learning_rate": 1.2408167443311214e-05,
1325
+ "loss": 0.7481,
1326
+ "step": 890
1327
+ },
1328
+ {
1329
+ "epoch": 8.564593301435407,
1330
+ "grad_norm": 0.53125,
1331
+ "learning_rate": 1.1610875691161915e-05,
1332
+ "loss": 0.7625,
1333
+ "step": 895
1334
+ },
1335
+ {
1336
+ "epoch": 8.61244019138756,
1337
+ "grad_norm": 0.396484375,
1338
+ "learning_rate": 1.083847690782972e-05,
1339
+ "loss": 0.7542,
1340
+ "step": 900
1341
+ },
1342
+ {
1343
+ "epoch": 8.660287081339712,
1344
+ "grad_norm": 0.388671875,
1345
+ "learning_rate": 1.0091188623457415e-05,
1346
+ "loss": 0.7658,
1347
+ "step": 905
1348
+ },
1349
+ {
1350
+ "epoch": 8.708133971291867,
1351
+ "grad_norm": 0.451171875,
1352
+ "learning_rate": 9.369221296335006e-06,
1353
+ "loss": 0.7581,
1354
+ "step": 910
1355
+ },
1356
+ {
1357
+ "epoch": 8.75598086124402,
1358
+ "grad_norm": 0.412109375,
1359
+ "learning_rate": 8.672778253628621e-06,
1360
+ "loss": 0.7485,
1361
+ "step": 915
1362
+ },
1363
+ {
1364
+ "epoch": 8.803827751196172,
1365
+ "grad_norm": 0.400390625,
1366
+ "learning_rate": 8.002055634117578e-06,
1367
+ "loss": 0.7544,
1368
+ "step": 920
1369
+ },
1370
+ {
1371
+ "epoch": 8.851674641148325,
1372
+ "grad_norm": 0.373046875,
1373
+ "learning_rate": 7.357242332955916e-06,
1374
+ "loss": 0.76,
1375
+ "step": 925
1376
+ },
1377
+ {
1378
+ "epoch": 8.89952153110048,
1379
+ "grad_norm": 0.498046875,
1380
+ "learning_rate": 6.738519948473976e-06,
1381
+ "loss": 0.7572,
1382
+ "step": 930
1383
+ },
1384
+ {
1385
+ "epoch": 8.947368421052632,
1386
+ "grad_norm": 0.388671875,
1387
+ "learning_rate": 6.146062731035129e-06,
1388
+ "loss": 0.7559,
1389
+ "step": 935
1390
+ },
1391
+ {
1392
+ "epoch": 8.995215311004785,
1393
+ "grad_norm": 0.373046875,
1394
+ "learning_rate": 5.580037533961546e-06,
1395
+ "loss": 0.7578,
1396
+ "step": 940
1397
+ },
1398
+ {
1399
+ "epoch": 8.995215311004785,
1400
+ "eval_loss": 2.52113676071167,
1401
+ "eval_runtime": 0.2561,
1402
+ "eval_samples_per_second": 39.042,
1403
+ "eval_steps_per_second": 3.904,
1404
+ "step": 940
1405
+ },
1406
+ {
1407
+ "epoch": 9.043062200956937,
1408
+ "grad_norm": 0.416015625,
1409
+ "learning_rate": 5.040603766543594e-06,
1410
+ "loss": 0.7518,
1411
+ "step": 945
1412
+ },
1413
+ {
1414
+ "epoch": 9.090909090909092,
1415
+ "grad_norm": 0.375,
1416
+ "learning_rate": 4.527913349145441e-06,
1417
+ "loss": 0.7595,
1418
+ "step": 950
1419
+ },
1420
+ {
1421
+ "epoch": 9.138755980861244,
1422
+ "grad_norm": 0.435546875,
1423
+ "learning_rate": 4.042110670419763e-06,
1424
+ "loss": 0.7471,
1425
+ "step": 955
1426
+ },
1427
+ {
1428
+ "epoch": 9.186602870813397,
1429
+ "grad_norm": 0.3984375,
1430
+ "learning_rate": 3.5833325466437694e-06,
1431
+ "loss": 0.7538,
1432
+ "step": 960
1433
+ },
1434
+ {
1435
+ "epoch": 9.23444976076555,
1436
+ "grad_norm": 0.412109375,
1437
+ "learning_rate": 3.1517081831876737e-06,
1438
+ "loss": 0.75,
1439
+ "step": 965
1440
+ },
1441
+ {
1442
+ "epoch": 9.282296650717702,
1443
+ "grad_norm": 0.37109375,
1444
+ "learning_rate": 2.7473591381266708e-06,
1445
+ "loss": 0.7547,
1446
+ "step": 970
1447
+ },
1448
+ {
1449
+ "epoch": 9.330143540669857,
1450
+ "grad_norm": 0.384765625,
1451
+ "learning_rate": 2.3703992880066638e-06,
1452
+ "loss": 0.764,
1453
+ "step": 975
1454
+ },
1455
+ {
1456
+ "epoch": 9.37799043062201,
1457
+ "grad_norm": 0.38671875,
1458
+ "learning_rate": 2.0209347957732328e-06,
1459
+ "loss": 0.7584,
1460
+ "step": 980
1461
+ },
1462
+ {
1463
+ "epoch": 9.425837320574162,
1464
+ "grad_norm": 0.390625,
1465
+ "learning_rate": 1.6990640808730696e-06,
1466
+ "loss": 0.7524,
1467
+ "step": 985
1468
+ },
1469
+ {
1470
+ "epoch": 9.473684210526315,
1471
+ "grad_norm": 0.412109375,
1472
+ "learning_rate": 1.404877791536141e-06,
1473
+ "loss": 0.7543,
1474
+ "step": 990
1475
+ },
1476
+ {
1477
+ "epoch": 9.52153110047847,
1478
+ "grad_norm": 0.50390625,
1479
+ "learning_rate": 1.1384587792465872e-06,
1480
+ "loss": 0.7579,
1481
+ "step": 995
1482
+ },
1483
+ {
1484
+ "epoch": 9.569377990430622,
1485
+ "grad_norm": 0.380859375,
1486
+ "learning_rate": 8.998820754091531e-07,
1487
+ "loss": 0.7534,
1488
+ "step": 1000
1489
+ },
1490
+ {
1491
+ "epoch": 9.617224880382775,
1492
+ "grad_norm": 0.3828125,
1493
+ "learning_rate": 6.892148702183133e-07,
1494
+ "loss": 0.7522,
1495
+ "step": 1005
1496
+ },
1497
+ {
1498
+ "epoch": 9.665071770334928,
1499
+ "grad_norm": 0.462890625,
1500
+ "learning_rate": 5.065164937354428e-07,
1501
+ "loss": 0.7598,
1502
+ "step": 1010
1503
+ },
1504
+ {
1505
+ "epoch": 9.712918660287082,
1506
+ "grad_norm": 0.37109375,
1507
+ "learning_rate": 3.5183839917972697e-07,
1508
+ "loss": 0.7556,
1509
+ "step": 1015
1510
+ },
1511
+ {
1512
+ "epoch": 9.760765550239235,
1513
+ "grad_norm": 0.376953125,
1514
+ "learning_rate": 2.2522414843748618e-07,
1515
+ "loss": 0.7503,
1516
+ "step": 1020
1517
+ },
1518
+ {
1519
+ "epoch": 9.808612440191387,
1520
+ "grad_norm": 0.375,
1521
+ "learning_rate": 1.2670939979384512e-07,
1522
+ "loss": 0.7479,
1523
+ "step": 1025
1524
+ },
1525
+ {
1526
+ "epoch": 9.85645933014354,
1527
+ "grad_norm": 0.41015625,
1528
+ "learning_rate": 5.632189789027687e-08,
1529
+ "loss": 0.7581,
1530
+ "step": 1030
1531
+ },
1532
+ {
1533
+ "epoch": 9.904306220095695,
1534
+ "grad_norm": 0.376953125,
1535
+ "learning_rate": 1.4081465910975588e-08,
1536
+ "loss": 0.7526,
1537
+ "step": 1035
1538
+ },
1539
+ {
1540
+ "epoch": 9.952153110047847,
1541
+ "grad_norm": 0.4453125,
1542
+ "learning_rate": 0.0,
1543
+ "loss": 0.7589,
1544
+ "step": 1040
1545
+ },
1546
+ {
1547
+ "epoch": 9.952153110047847,
1548
+ "eval_loss": 2.515638828277588,
1549
+ "eval_runtime": 0.236,
1550
+ "eval_samples_per_second": 42.374,
1551
+ "eval_steps_per_second": 4.237,
1552
+ "step": 1040
1553
+ },
1554
+ {
1555
+ "epoch": 9.952153110047847,
1556
+ "step": 1040,
1557
+ "total_flos": 3.1750846888049377e+18,
1558
+ "train_loss": 2.1342261617000284,
1559
+ "train_runtime": 2535.4321,
1560
+ "train_samples_per_second": 26.287,
1561
+ "train_steps_per_second": 0.41
1562
+ }
1563
+ ],
1564
+ "logging_steps": 5,
1565
+ "max_steps": 1040,
1566
+ "num_input_tokens_seen": 0,
1567
+ "num_train_epochs": 10,
1568
+ "save_steps": 100,
1569
+ "total_flos": 3.1750846888049377e+18,
1570
+ "train_batch_size": 4,
1571
+ "trial_name": null,
1572
+ "trial_params": null
1573
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e8c49b3ec7e48206287c61f83a6f59c45977b8621711047285e3a6450c9af077
3
+ size 5176