Ritvik19 commited on
Commit
bcec71a
1 Parent(s): 0a0c8fb

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
+ - Ritvik19/open-hermes-2_5-reformatted
10
+ model-index:
11
+ - name: openhermes-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
+ # openhermes-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 Ritvik19/open-hermes-2_5-reformatted dataset.
21
+ It achieves the following results on the evaluation set:
22
+ - Loss: 1.1197
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
+ | 1.0838 | 0.9999 | 1704 | 1.1197 |
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
+ "k_proj",
23
+ "up_proj",
24
+ "gate_proj",
25
+ "down_proj",
26
+ "o_proj",
27
+ "v_proj",
28
+ "q_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 sha256:292595ab91dae1211730882f85941f43c026d2f74a346c76e35e0402ebd3a276
3
+ size 34647816
all_results.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 0.9999209511896846,
3
+ "eval_loss": 1.1197459697723389,
4
+ "eval_runtime": 1941.0266,
5
+ "eval_samples": 23109,
6
+ "eval_samples_per_second": 7.945,
7
+ "eval_steps_per_second": 7.945,
8
+ "total_flos": 4.668038789401149e+18,
9
+ "train_loss": 0.0,
10
+ "train_runtime": 0.023,
11
+ "train_samples": 1001538,
12
+ "train_samples_per_second": 9337043.614,
13
+ "train_steps_per_second": 72939.887
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": 0.9999209511896846,
3
+ "eval_loss": 1.1197459697723389,
4
+ "eval_runtime": 1941.0266,
5
+ "eval_samples": 23109,
6
+ "eval_samples_per_second": 7.945,
7
+ "eval_steps_per_second": 7.945
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": 0.9999209511896846,
3
+ "total_flos": 4.668038789401149e+18,
4
+ "train_loss": 0.0,
5
+ "train_runtime": 0.023,
6
+ "train_samples": 1001538,
7
+ "train_samples_per_second": 9337043.614,
8
+ "train_steps_per_second": 72939.887
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,2389 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.9999209511896846,
5
+ "eval_steps": 500,
6
+ "global_step": 1680,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0005951910423748122,
13
+ "grad_norm": 0.32421875,
14
+ "learning_rate": 1.1904761904761904e-06,
15
+ "loss": 1.0942,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.0029759552118740614,
20
+ "grad_norm": 0.306640625,
21
+ "learning_rate": 5.9523809523809525e-06,
22
+ "loss": 1.1063,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.005951910423748123,
27
+ "grad_norm": 0.29296875,
28
+ "learning_rate": 1.1904761904761905e-05,
29
+ "loss": 1.0913,
30
+ "step": 10
31
+ },
32
+ {
33
+ "epoch": 0.008927865635622183,
34
+ "grad_norm": 0.294921875,
35
+ "learning_rate": 1.785714285714286e-05,
36
+ "loss": 1.0858,
37
+ "step": 15
38
+ },
39
+ {
40
+ "epoch": 0.011903820847496246,
41
+ "grad_norm": 0.2734375,
42
+ "learning_rate": 2.380952380952381e-05,
43
+ "loss": 1.0707,
44
+ "step": 20
45
+ },
46
+ {
47
+ "epoch": 0.014879776059370306,
48
+ "grad_norm": 0.2451171875,
49
+ "learning_rate": 2.9761904761904762e-05,
50
+ "loss": 1.0274,
51
+ "step": 25
52
+ },
53
+ {
54
+ "epoch": 0.017855731271244367,
55
+ "grad_norm": 0.171875,
56
+ "learning_rate": 3.571428571428572e-05,
57
+ "loss": 0.9783,
58
+ "step": 30
59
+ },
60
+ {
61
+ "epoch": 0.02083168648311843,
62
+ "grad_norm": 0.16796875,
63
+ "learning_rate": 4.166666666666667e-05,
64
+ "loss": 0.9564,
65
+ "step": 35
66
+ },
67
+ {
68
+ "epoch": 0.02380764169499249,
69
+ "grad_norm": 0.1328125,
70
+ "learning_rate": 4.761904761904762e-05,
71
+ "loss": 0.9325,
72
+ "step": 40
73
+ },
74
+ {
75
+ "epoch": 0.02678359690686655,
76
+ "grad_norm": 0.11767578125,
77
+ "learning_rate": 5.3571428571428575e-05,
78
+ "loss": 0.9242,
79
+ "step": 45
80
+ },
81
+ {
82
+ "epoch": 0.029759552118740613,
83
+ "grad_norm": 0.0859375,
84
+ "learning_rate": 5.9523809523809524e-05,
85
+ "loss": 0.8951,
86
+ "step": 50
87
+ },
88
+ {
89
+ "epoch": 0.03273550733061467,
90
+ "grad_norm": 0.0849609375,
91
+ "learning_rate": 6.547619047619048e-05,
92
+ "loss": 0.8703,
93
+ "step": 55
94
+ },
95
+ {
96
+ "epoch": 0.035711462542488734,
97
+ "grad_norm": 0.076171875,
98
+ "learning_rate": 7.142857142857143e-05,
99
+ "loss": 0.8513,
100
+ "step": 60
101
+ },
102
+ {
103
+ "epoch": 0.038687417754362796,
104
+ "grad_norm": 0.06494140625,
105
+ "learning_rate": 7.738095238095239e-05,
106
+ "loss": 0.8488,
107
+ "step": 65
108
+ },
109
+ {
110
+ "epoch": 0.04166337296623686,
111
+ "grad_norm": 0.06494140625,
112
+ "learning_rate": 8.333333333333334e-05,
113
+ "loss": 0.8218,
114
+ "step": 70
115
+ },
116
+ {
117
+ "epoch": 0.04463932817811092,
118
+ "grad_norm": 0.068359375,
119
+ "learning_rate": 8.92857142857143e-05,
120
+ "loss": 0.8321,
121
+ "step": 75
122
+ },
123
+ {
124
+ "epoch": 0.04761528338998498,
125
+ "grad_norm": 0.06640625,
126
+ "learning_rate": 9.523809523809524e-05,
127
+ "loss": 0.8271,
128
+ "step": 80
129
+ },
130
+ {
131
+ "epoch": 0.050591238601859045,
132
+ "grad_norm": 0.0537109375,
133
+ "learning_rate": 0.0001011904761904762,
134
+ "loss": 0.8097,
135
+ "step": 85
136
+ },
137
+ {
138
+ "epoch": 0.0535671938137331,
139
+ "grad_norm": 0.0546875,
140
+ "learning_rate": 0.00010714285714285715,
141
+ "loss": 0.7871,
142
+ "step": 90
143
+ },
144
+ {
145
+ "epoch": 0.05654314902560716,
146
+ "grad_norm": 0.050537109375,
147
+ "learning_rate": 0.00011309523809523809,
148
+ "loss": 0.8205,
149
+ "step": 95
150
+ },
151
+ {
152
+ "epoch": 0.059519104237481225,
153
+ "grad_norm": 0.051513671875,
154
+ "learning_rate": 0.00011904761904761905,
155
+ "loss": 0.8013,
156
+ "step": 100
157
+ },
158
+ {
159
+ "epoch": 0.06249505944935529,
160
+ "grad_norm": 0.059814453125,
161
+ "learning_rate": 0.000125,
162
+ "loss": 0.7996,
163
+ "step": 105
164
+ },
165
+ {
166
+ "epoch": 0.06547101466122934,
167
+ "grad_norm": 0.056396484375,
168
+ "learning_rate": 0.00013095238095238096,
169
+ "loss": 0.8194,
170
+ "step": 110
171
+ },
172
+ {
173
+ "epoch": 0.0684469698731034,
174
+ "grad_norm": 0.052734375,
175
+ "learning_rate": 0.0001369047619047619,
176
+ "loss": 0.7937,
177
+ "step": 115
178
+ },
179
+ {
180
+ "epoch": 0.07142292508497747,
181
+ "grad_norm": 0.0517578125,
182
+ "learning_rate": 0.00014285714285714287,
183
+ "loss": 0.7972,
184
+ "step": 120
185
+ },
186
+ {
187
+ "epoch": 0.07439888029685153,
188
+ "grad_norm": 0.056640625,
189
+ "learning_rate": 0.00014880952380952382,
190
+ "loss": 0.8139,
191
+ "step": 125
192
+ },
193
+ {
194
+ "epoch": 0.07737483550872559,
195
+ "grad_norm": 0.056640625,
196
+ "learning_rate": 0.00015476190476190478,
197
+ "loss": 0.7933,
198
+ "step": 130
199
+ },
200
+ {
201
+ "epoch": 0.08035079072059965,
202
+ "grad_norm": 0.05126953125,
203
+ "learning_rate": 0.00016071428571428573,
204
+ "loss": 0.7852,
205
+ "step": 135
206
+ },
207
+ {
208
+ "epoch": 0.08332674593247372,
209
+ "grad_norm": 0.048095703125,
210
+ "learning_rate": 0.0001666666666666667,
211
+ "loss": 0.8041,
212
+ "step": 140
213
+ },
214
+ {
215
+ "epoch": 0.08630270114434778,
216
+ "grad_norm": 0.055419921875,
217
+ "learning_rate": 0.00017261904761904764,
218
+ "loss": 0.8021,
219
+ "step": 145
220
+ },
221
+ {
222
+ "epoch": 0.08927865635622184,
223
+ "grad_norm": 0.0498046875,
224
+ "learning_rate": 0.0001785714285714286,
225
+ "loss": 0.7883,
226
+ "step": 150
227
+ },
228
+ {
229
+ "epoch": 0.0922546115680959,
230
+ "grad_norm": 0.0537109375,
231
+ "learning_rate": 0.00018452380952380955,
232
+ "loss": 0.8034,
233
+ "step": 155
234
+ },
235
+ {
236
+ "epoch": 0.09523056677996997,
237
+ "grad_norm": 0.05322265625,
238
+ "learning_rate": 0.00019047619047619048,
239
+ "loss": 0.7865,
240
+ "step": 160
241
+ },
242
+ {
243
+ "epoch": 0.09820652199184403,
244
+ "grad_norm": 0.0546875,
245
+ "learning_rate": 0.00019642857142857144,
246
+ "loss": 0.7945,
247
+ "step": 165
248
+ },
249
+ {
250
+ "epoch": 0.10118247720371809,
251
+ "grad_norm": 0.04833984375,
252
+ "learning_rate": 0.0001999991365731819,
253
+ "loss": 0.7968,
254
+ "step": 170
255
+ },
256
+ {
257
+ "epoch": 0.10415843241559214,
258
+ "grad_norm": 0.058349609375,
259
+ "learning_rate": 0.00019998942319271077,
260
+ "loss": 0.7942,
261
+ "step": 175
262
+ },
263
+ {
264
+ "epoch": 0.1071343876274662,
265
+ "grad_norm": 0.0517578125,
266
+ "learning_rate": 0.00019996891820008164,
267
+ "loss": 0.8003,
268
+ "step": 180
269
+ },
270
+ {
271
+ "epoch": 0.11011034283934026,
272
+ "grad_norm": 0.052734375,
273
+ "learning_rate": 0.00019993762380834785,
274
+ "loss": 0.7792,
275
+ "step": 185
276
+ },
277
+ {
278
+ "epoch": 0.11308629805121433,
279
+ "grad_norm": 0.05029296875,
280
+ "learning_rate": 0.00019989554339503612,
281
+ "loss": 0.7829,
282
+ "step": 190
283
+ },
284
+ {
285
+ "epoch": 0.11606225326308839,
286
+ "grad_norm": 0.05322265625,
287
+ "learning_rate": 0.00019984268150178167,
288
+ "loss": 0.7946,
289
+ "step": 195
290
+ },
291
+ {
292
+ "epoch": 0.11903820847496245,
293
+ "grad_norm": 0.04931640625,
294
+ "learning_rate": 0.0001997790438338385,
295
+ "loss": 0.7961,
296
+ "step": 200
297
+ },
298
+ {
299
+ "epoch": 0.12201416368683651,
300
+ "grad_norm": 0.053466796875,
301
+ "learning_rate": 0.00019970463725946336,
302
+ "loss": 0.7896,
303
+ "step": 205
304
+ },
305
+ {
306
+ "epoch": 0.12499011889871058,
307
+ "grad_norm": 0.05224609375,
308
+ "learning_rate": 0.00019961946980917456,
309
+ "loss": 0.782,
310
+ "step": 210
311
+ },
312
+ {
313
+ "epoch": 0.12796607411058464,
314
+ "grad_norm": 0.049072265625,
315
+ "learning_rate": 0.00019952355067488523,
316
+ "loss": 0.8018,
317
+ "step": 215
318
+ },
319
+ {
320
+ "epoch": 0.1309420293224587,
321
+ "grad_norm": 0.051513671875,
322
+ "learning_rate": 0.0001994168902089112,
323
+ "loss": 0.7943,
324
+ "step": 220
325
+ },
326
+ {
327
+ "epoch": 0.13391798453433276,
328
+ "grad_norm": 0.052001953125,
329
+ "learning_rate": 0.00019929949992285396,
330
+ "loss": 0.8005,
331
+ "step": 225
332
+ },
333
+ {
334
+ "epoch": 0.1368939397462068,
335
+ "grad_norm": 0.055908203125,
336
+ "learning_rate": 0.00019917139248635786,
337
+ "loss": 0.7776,
338
+ "step": 230
339
+ },
340
+ {
341
+ "epoch": 0.1398698949580809,
342
+ "grad_norm": 0.052734375,
343
+ "learning_rate": 0.00019903258172574302,
344
+ "loss": 0.7905,
345
+ "step": 235
346
+ },
347
+ {
348
+ "epoch": 0.14284585016995494,
349
+ "grad_norm": 0.051513671875,
350
+ "learning_rate": 0.00019888308262251285,
351
+ "loss": 0.7782,
352
+ "step": 240
353
+ },
354
+ {
355
+ "epoch": 0.145821805381829,
356
+ "grad_norm": 0.07958984375,
357
+ "learning_rate": 0.00019872291131173742,
358
+ "loss": 0.7798,
359
+ "step": 245
360
+ },
361
+ {
362
+ "epoch": 0.14879776059370306,
363
+ "grad_norm": 0.0615234375,
364
+ "learning_rate": 0.0001985520850803117,
365
+ "loss": 0.7842,
366
+ "step": 250
367
+ },
368
+ {
369
+ "epoch": 0.15177371580557714,
370
+ "grad_norm": 0.054443359375,
371
+ "learning_rate": 0.00019837062236509014,
372
+ "loss": 0.7778,
373
+ "step": 255
374
+ },
375
+ {
376
+ "epoch": 0.15474967101745118,
377
+ "grad_norm": 0.04833984375,
378
+ "learning_rate": 0.0001981785427508966,
379
+ "loss": 0.7813,
380
+ "step": 260
381
+ },
382
+ {
383
+ "epoch": 0.15772562622932526,
384
+ "grad_norm": 0.0546875,
385
+ "learning_rate": 0.00019797586696841072,
386
+ "loss": 0.7956,
387
+ "step": 265
388
+ },
389
+ {
390
+ "epoch": 0.1607015814411993,
391
+ "grad_norm": 0.052978515625,
392
+ "learning_rate": 0.00019776261689193048,
393
+ "loss": 0.8056,
394
+ "step": 270
395
+ },
396
+ {
397
+ "epoch": 0.16367753665307336,
398
+ "grad_norm": 0.0498046875,
399
+ "learning_rate": 0.00019753881553701138,
400
+ "loss": 0.7886,
401
+ "step": 275
402
+ },
403
+ {
404
+ "epoch": 0.16665349186494743,
405
+ "grad_norm": 0.0576171875,
406
+ "learning_rate": 0.00019730448705798239,
407
+ "loss": 0.8025,
408
+ "step": 280
409
+ },
410
+ {
411
+ "epoch": 0.16962944707682148,
412
+ "grad_norm": 0.05712890625,
413
+ "learning_rate": 0.0001970596567453391,
414
+ "loss": 0.7803,
415
+ "step": 285
416
+ },
417
+ {
418
+ "epoch": 0.17260540228869556,
419
+ "grad_norm": 0.049560546875,
420
+ "learning_rate": 0.00019680435102301412,
421
+ "loss": 0.7844,
422
+ "step": 290
423
+ },
424
+ {
425
+ "epoch": 0.1755813575005696,
426
+ "grad_norm": 0.052978515625,
427
+ "learning_rate": 0.0001965385974455251,
428
+ "loss": 0.797,
429
+ "step": 295
430
+ },
431
+ {
432
+ "epoch": 0.17855731271244368,
433
+ "grad_norm": 0.051513671875,
434
+ "learning_rate": 0.0001962624246950012,
435
+ "loss": 0.7774,
436
+ "step": 300
437
+ },
438
+ {
439
+ "epoch": 0.18153326792431773,
440
+ "grad_norm": 0.05029296875,
441
+ "learning_rate": 0.00019597586257808712,
442
+ "loss": 0.7756,
443
+ "step": 305
444
+ },
445
+ {
446
+ "epoch": 0.1845092231361918,
447
+ "grad_norm": 0.052734375,
448
+ "learning_rate": 0.0001956789420227262,
449
+ "loss": 0.7703,
450
+ "step": 310
451
+ },
452
+ {
453
+ "epoch": 0.18748517834806586,
454
+ "grad_norm": 0.275390625,
455
+ "learning_rate": 0.0001953716950748227,
456
+ "loss": 0.7901,
457
+ "step": 315
458
+ },
459
+ {
460
+ "epoch": 0.19046113355993993,
461
+ "grad_norm": 0.05029296875,
462
+ "learning_rate": 0.0001950541548947829,
463
+ "loss": 0.78,
464
+ "step": 320
465
+ },
466
+ {
467
+ "epoch": 0.19343708877181398,
468
+ "grad_norm": 0.052978515625,
469
+ "learning_rate": 0.0001947263557539363,
470
+ "loss": 0.767,
471
+ "step": 325
472
+ },
473
+ {
474
+ "epoch": 0.19641304398368806,
475
+ "grad_norm": 0.050048828125,
476
+ "learning_rate": 0.00019438833303083678,
477
+ "loss": 0.7963,
478
+ "step": 330
479
+ },
480
+ {
481
+ "epoch": 0.1993889991955621,
482
+ "grad_norm": 0.051025390625,
483
+ "learning_rate": 0.00019404012320744417,
484
+ "loss": 0.783,
485
+ "step": 335
486
+ },
487
+ {
488
+ "epoch": 0.20236495440743618,
489
+ "grad_norm": 0.049560546875,
490
+ "learning_rate": 0.0001936817638651871,
491
+ "loss": 0.7814,
492
+ "step": 340
493
+ },
494
+ {
495
+ "epoch": 0.20534090961931023,
496
+ "grad_norm": 0.05322265625,
497
+ "learning_rate": 0.00019331329368090666,
498
+ "loss": 0.7791,
499
+ "step": 345
500
+ },
501
+ {
502
+ "epoch": 0.20831686483118428,
503
+ "grad_norm": 0.04833984375,
504
+ "learning_rate": 0.00019293475242268223,
505
+ "loss": 0.7782,
506
+ "step": 350
507
+ },
508
+ {
509
+ "epoch": 0.21129282004305835,
510
+ "grad_norm": 0.049560546875,
511
+ "learning_rate": 0.00019254618094553949,
512
+ "loss": 0.7835,
513
+ "step": 355
514
+ },
515
+ {
516
+ "epoch": 0.2142687752549324,
517
+ "grad_norm": 0.048095703125,
518
+ "learning_rate": 0.00019214762118704076,
519
+ "loss": 0.7886,
520
+ "step": 360
521
+ },
522
+ {
523
+ "epoch": 0.21724473046680648,
524
+ "grad_norm": 0.048583984375,
525
+ "learning_rate": 0.00019173911616275917,
526
+ "loss": 0.7766,
527
+ "step": 365
528
+ },
529
+ {
530
+ "epoch": 0.22022068567868053,
531
+ "grad_norm": 0.0478515625,
532
+ "learning_rate": 0.00019132070996163568,
533
+ "loss": 0.7838,
534
+ "step": 370
535
+ },
536
+ {
537
+ "epoch": 0.2231966408905546,
538
+ "grad_norm": 0.0478515625,
539
+ "learning_rate": 0.0001908924477412211,
540
+ "loss": 0.7837,
541
+ "step": 375
542
+ },
543
+ {
544
+ "epoch": 0.22617259610242865,
545
+ "grad_norm": 0.044921875,
546
+ "learning_rate": 0.00019045437572280194,
547
+ "loss": 0.7961,
548
+ "step": 380
549
+ },
550
+ {
551
+ "epoch": 0.22914855131430273,
552
+ "grad_norm": 0.04638671875,
553
+ "learning_rate": 0.00019000654118641211,
554
+ "loss": 0.7903,
555
+ "step": 385
556
+ },
557
+ {
558
+ "epoch": 0.23212450652617678,
559
+ "grad_norm": 0.048828125,
560
+ "learning_rate": 0.0001895489924657301,
561
+ "loss": 0.7776,
562
+ "step": 390
563
+ },
564
+ {
565
+ "epoch": 0.23510046173805085,
566
+ "grad_norm": 0.05029296875,
567
+ "learning_rate": 0.00018908177894286232,
568
+ "loss": 0.7768,
569
+ "step": 395
570
+ },
571
+ {
572
+ "epoch": 0.2380764169499249,
573
+ "grad_norm": 0.0478515625,
574
+ "learning_rate": 0.00018860495104301345,
575
+ "loss": 0.7842,
576
+ "step": 400
577
+ },
578
+ {
579
+ "epoch": 0.24105237216179898,
580
+ "grad_norm": 0.0517578125,
581
+ "learning_rate": 0.00018811856022904423,
582
+ "loss": 0.7728,
583
+ "step": 405
584
+ },
585
+ {
586
+ "epoch": 0.24402832737367303,
587
+ "grad_norm": 0.050537109375,
588
+ "learning_rate": 0.00018762265899591722,
589
+ "loss": 0.7787,
590
+ "step": 410
591
+ },
592
+ {
593
+ "epoch": 0.24700428258554707,
594
+ "grad_norm": 0.05078125,
595
+ "learning_rate": 0.000187117300865031,
596
+ "loss": 0.7893,
597
+ "step": 415
598
+ },
599
+ {
600
+ "epoch": 0.24998023779742115,
601
+ "grad_norm": 0.05078125,
602
+ "learning_rate": 0.00018660254037844388,
603
+ "loss": 0.7846,
604
+ "step": 420
605
+ },
606
+ {
607
+ "epoch": 0.2529561930092952,
608
+ "grad_norm": 0.055908203125,
609
+ "learning_rate": 0.00018607843309298723,
610
+ "loss": 0.785,
611
+ "step": 425
612
+ },
613
+ {
614
+ "epoch": 0.2559321482211693,
615
+ "grad_norm": 0.052001953125,
616
+ "learning_rate": 0.00018554503557426948,
617
+ "loss": 0.7861,
618
+ "step": 430
619
+ },
620
+ {
621
+ "epoch": 0.2589081034330433,
622
+ "grad_norm": 0.052490234375,
623
+ "learning_rate": 0.0001850024053905709,
624
+ "loss": 0.7605,
625
+ "step": 435
626
+ },
627
+ {
628
+ "epoch": 0.2618840586449174,
629
+ "grad_norm": 0.051025390625,
630
+ "learning_rate": 0.0001844506011066308,
631
+ "loss": 0.7691,
632
+ "step": 440
633
+ },
634
+ {
635
+ "epoch": 0.2648600138567915,
636
+ "grad_norm": 0.05322265625,
637
+ "learning_rate": 0.00018388968227732626,
638
+ "loss": 0.7743,
639
+ "step": 445
640
+ },
641
+ {
642
+ "epoch": 0.2678359690686655,
643
+ "grad_norm": 0.047607421875,
644
+ "learning_rate": 0.0001833197094412449,
645
+ "loss": 0.7834,
646
+ "step": 450
647
+ },
648
+ {
649
+ "epoch": 0.2708119242805396,
650
+ "grad_norm": 0.050048828125,
651
+ "learning_rate": 0.00018274074411415105,
652
+ "loss": 0.7804,
653
+ "step": 455
654
+ },
655
+ {
656
+ "epoch": 0.2737878794924136,
657
+ "grad_norm": 0.051025390625,
658
+ "learning_rate": 0.00018215284878234642,
659
+ "loss": 0.7985,
660
+ "step": 460
661
+ },
662
+ {
663
+ "epoch": 0.2767638347042877,
664
+ "grad_norm": 0.050048828125,
665
+ "learning_rate": 0.00018155608689592604,
666
+ "loss": 0.7852,
667
+ "step": 465
668
+ },
669
+ {
670
+ "epoch": 0.2797397899161618,
671
+ "grad_norm": 0.048828125,
672
+ "learning_rate": 0.0001809505228619304,
673
+ "loss": 0.7697,
674
+ "step": 470
675
+ },
676
+ {
677
+ "epoch": 0.2827157451280358,
678
+ "grad_norm": 0.04736328125,
679
+ "learning_rate": 0.0001803362220373942,
680
+ "loss": 0.772,
681
+ "step": 475
682
+ },
683
+ {
684
+ "epoch": 0.28569170033990987,
685
+ "grad_norm": 0.05126953125,
686
+ "learning_rate": 0.00017971325072229226,
687
+ "loss": 0.7613,
688
+ "step": 480
689
+ },
690
+ {
691
+ "epoch": 0.2886676555517839,
692
+ "grad_norm": 0.04833984375,
693
+ "learning_rate": 0.00017908167615238415,
694
+ "loss": 0.7737,
695
+ "step": 485
696
+ },
697
+ {
698
+ "epoch": 0.291643610763658,
699
+ "grad_norm": 0.047607421875,
700
+ "learning_rate": 0.00017844156649195759,
701
+ "loss": 0.7813,
702
+ "step": 490
703
+ },
704
+ {
705
+ "epoch": 0.29461956597553207,
706
+ "grad_norm": 0.04931640625,
707
+ "learning_rate": 0.00017779299082647148,
708
+ "loss": 0.7709,
709
+ "step": 495
710
+ },
711
+ {
712
+ "epoch": 0.2975955211874061,
713
+ "grad_norm": 0.046875,
714
+ "learning_rate": 0.0001771360191551,
715
+ "loss": 0.7838,
716
+ "step": 500
717
+ },
718
+ {
719
+ "epoch": 0.30057147639928017,
720
+ "grad_norm": 0.050537109375,
721
+ "learning_rate": 0.00017647072238317728,
722
+ "loss": 0.7638,
723
+ "step": 505
724
+ },
725
+ {
726
+ "epoch": 0.3035474316111543,
727
+ "grad_norm": 0.04833984375,
728
+ "learning_rate": 0.0001757971723145453,
729
+ "loss": 0.7838,
730
+ "step": 510
731
+ },
732
+ {
733
+ "epoch": 0.3065233868230283,
734
+ "grad_norm": 0.05126953125,
735
+ "learning_rate": 0.00017511544164380388,
736
+ "loss": 0.7852,
737
+ "step": 515
738
+ },
739
+ {
740
+ "epoch": 0.30949934203490237,
741
+ "grad_norm": 0.0478515625,
742
+ "learning_rate": 0.00017442560394846516,
743
+ "loss": 0.7664,
744
+ "step": 520
745
+ },
746
+ {
747
+ "epoch": 0.3124752972467764,
748
+ "grad_norm": 0.05029296875,
749
+ "learning_rate": 0.0001737277336810124,
750
+ "loss": 0.7715,
751
+ "step": 525
752
+ },
753
+ {
754
+ "epoch": 0.3154512524586505,
755
+ "grad_norm": 0.05029296875,
756
+ "learning_rate": 0.00017302190616086464,
757
+ "loss": 0.7474,
758
+ "step": 530
759
+ },
760
+ {
761
+ "epoch": 0.31842720767052457,
762
+ "grad_norm": 0.046630859375,
763
+ "learning_rate": 0.0001723081975662476,
764
+ "loss": 0.7906,
765
+ "step": 535
766
+ },
767
+ {
768
+ "epoch": 0.3214031628823986,
769
+ "grad_norm": 0.05029296875,
770
+ "learning_rate": 0.00017158668492597186,
771
+ "loss": 0.7694,
772
+ "step": 540
773
+ },
774
+ {
775
+ "epoch": 0.32437911809427267,
776
+ "grad_norm": 0.0478515625,
777
+ "learning_rate": 0.00017085744611111957,
778
+ "loss": 0.7855,
779
+ "step": 545
780
+ },
781
+ {
782
+ "epoch": 0.3273550733061467,
783
+ "grad_norm": 0.049072265625,
784
+ "learning_rate": 0.0001701205598266398,
785
+ "loss": 0.7674,
786
+ "step": 550
787
+ },
788
+ {
789
+ "epoch": 0.3303310285180208,
790
+ "grad_norm": 0.048095703125,
791
+ "learning_rate": 0.00016937610560285418,
792
+ "loss": 0.7619,
793
+ "step": 555
794
+ },
795
+ {
796
+ "epoch": 0.33330698372989487,
797
+ "grad_norm": 0.0458984375,
798
+ "learning_rate": 0.0001686241637868734,
799
+ "loss": 0.7777,
800
+ "step": 560
801
+ },
802
+ {
803
+ "epoch": 0.3362829389417689,
804
+ "grad_norm": 0.052490234375,
805
+ "learning_rate": 0.00016786481553392548,
806
+ "loss": 0.7792,
807
+ "step": 565
808
+ },
809
+ {
810
+ "epoch": 0.33925889415364296,
811
+ "grad_norm": 0.048095703125,
812
+ "learning_rate": 0.00016709814279859702,
813
+ "loss": 0.7601,
814
+ "step": 570
815
+ },
816
+ {
817
+ "epoch": 0.34223484936551707,
818
+ "grad_norm": 0.050048828125,
819
+ "learning_rate": 0.00016632422832598795,
820
+ "loss": 0.7817,
821
+ "step": 575
822
+ },
823
+ {
824
+ "epoch": 0.3452108045773911,
825
+ "grad_norm": 0.04931640625,
826
+ "learning_rate": 0.000165543155642781,
827
+ "loss": 0.7905,
828
+ "step": 580
829
+ },
830
+ {
831
+ "epoch": 0.34818675978926517,
832
+ "grad_norm": 0.04736328125,
833
+ "learning_rate": 0.00016475500904822706,
834
+ "loss": 0.7581,
835
+ "step": 585
836
+ },
837
+ {
838
+ "epoch": 0.3511627150011392,
839
+ "grad_norm": 0.049072265625,
840
+ "learning_rate": 0.00016395987360504668,
841
+ "loss": 0.7693,
842
+ "step": 590
843
+ },
844
+ {
845
+ "epoch": 0.3541386702130133,
846
+ "grad_norm": 0.051513671875,
847
+ "learning_rate": 0.00016315783513024977,
848
+ "loss": 0.7861,
849
+ "step": 595
850
+ },
851
+ {
852
+ "epoch": 0.35711462542488737,
853
+ "grad_norm": 0.049072265625,
854
+ "learning_rate": 0.00016234898018587337,
855
+ "loss": 0.7765,
856
+ "step": 600
857
+ },
858
+ {
859
+ "epoch": 0.3600905806367614,
860
+ "grad_norm": 0.0517578125,
861
+ "learning_rate": 0.0001615333960696393,
862
+ "loss": 0.7682,
863
+ "step": 605
864
+ },
865
+ {
866
+ "epoch": 0.36306653584863546,
867
+ "grad_norm": 0.05029296875,
868
+ "learning_rate": 0.00016071117080553236,
869
+ "loss": 0.7768,
870
+ "step": 610
871
+ },
872
+ {
873
+ "epoch": 0.3660424910605095,
874
+ "grad_norm": 0.048095703125,
875
+ "learning_rate": 0.00015988239313430004,
876
+ "loss": 0.7684,
877
+ "step": 615
878
+ },
879
+ {
880
+ "epoch": 0.3690184462723836,
881
+ "grad_norm": 0.047119140625,
882
+ "learning_rate": 0.00015904715250387498,
883
+ "loss": 0.7677,
884
+ "step": 620
885
+ },
886
+ {
887
+ "epoch": 0.37199440148425766,
888
+ "grad_norm": 0.0478515625,
889
+ "learning_rate": 0.0001582055390597212,
890
+ "loss": 0.7866,
891
+ "step": 625
892
+ },
893
+ {
894
+ "epoch": 0.3749703566961317,
895
+ "grad_norm": 0.04736328125,
896
+ "learning_rate": 0.0001573576436351046,
897
+ "loss": 0.7753,
898
+ "step": 630
899
+ },
900
+ {
901
+ "epoch": 0.37794631190800576,
902
+ "grad_norm": 0.049072265625,
903
+ "learning_rate": 0.00015650355774129,
904
+ "loss": 0.7854,
905
+ "step": 635
906
+ },
907
+ {
908
+ "epoch": 0.38092226711987986,
909
+ "grad_norm": 0.05078125,
910
+ "learning_rate": 0.00015564337355766412,
911
+ "loss": 0.7639,
912
+ "step": 640
913
+ },
914
+ {
915
+ "epoch": 0.3838982223317539,
916
+ "grad_norm": 0.05029296875,
917
+ "learning_rate": 0.00015477718392178716,
918
+ "loss": 0.7574,
919
+ "step": 645
920
+ },
921
+ {
922
+ "epoch": 0.38687417754362796,
923
+ "grad_norm": 0.0458984375,
924
+ "learning_rate": 0.00015390508231937297,
925
+ "loss": 0.7663,
926
+ "step": 650
927
+ },
928
+ {
929
+ "epoch": 0.389850132755502,
930
+ "grad_norm": 0.0478515625,
931
+ "learning_rate": 0.00015302716287419945,
932
+ "loss": 0.7746,
933
+ "step": 655
934
+ },
935
+ {
936
+ "epoch": 0.3928260879673761,
937
+ "grad_norm": 0.052001953125,
938
+ "learning_rate": 0.0001521435203379498,
939
+ "loss": 0.7617,
940
+ "step": 660
941
+ },
942
+ {
943
+ "epoch": 0.39580204317925016,
944
+ "grad_norm": 0.04736328125,
945
+ "learning_rate": 0.00015125425007998653,
946
+ "loss": 0.7993,
947
+ "step": 665
948
+ },
949
+ {
950
+ "epoch": 0.3987779983911242,
951
+ "grad_norm": 0.048583984375,
952
+ "learning_rate": 0.0001503594480770581,
953
+ "loss": 0.7887,
954
+ "step": 670
955
+ },
956
+ {
957
+ "epoch": 0.40175395360299826,
958
+ "grad_norm": 0.04931640625,
959
+ "learning_rate": 0.00014945921090294076,
960
+ "loss": 0.7648,
961
+ "step": 675
962
+ },
963
+ {
964
+ "epoch": 0.40472990881487236,
965
+ "grad_norm": 0.04638671875,
966
+ "learning_rate": 0.00014855363571801523,
967
+ "loss": 0.7811,
968
+ "step": 680
969
+ },
970
+ {
971
+ "epoch": 0.4077058640267464,
972
+ "grad_norm": 0.04833984375,
973
+ "learning_rate": 0.00014764282025878068,
974
+ "loss": 0.7674,
975
+ "step": 685
976
+ },
977
+ {
978
+ "epoch": 0.41068181923862046,
979
+ "grad_norm": 0.0498046875,
980
+ "learning_rate": 0.0001467268628273062,
981
+ "loss": 0.7684,
982
+ "step": 690
983
+ },
984
+ {
985
+ "epoch": 0.4136577744504945,
986
+ "grad_norm": 0.050048828125,
987
+ "learning_rate": 0.00014580586228062122,
988
+ "loss": 0.7957,
989
+ "step": 695
990
+ },
991
+ {
992
+ "epoch": 0.41663372966236856,
993
+ "grad_norm": 0.050537109375,
994
+ "learning_rate": 0.00014487991802004623,
995
+ "loss": 0.7778,
996
+ "step": 700
997
+ },
998
+ {
999
+ "epoch": 0.41960968487424266,
1000
+ "grad_norm": 0.056396484375,
1001
+ "learning_rate": 0.0001439491299804645,
1002
+ "loss": 0.7593,
1003
+ "step": 705
1004
+ },
1005
+ {
1006
+ "epoch": 0.4225856400861167,
1007
+ "grad_norm": 0.0498046875,
1008
+ "learning_rate": 0.0001430135986195365,
1009
+ "loss": 0.7576,
1010
+ "step": 710
1011
+ },
1012
+ {
1013
+ "epoch": 0.42556159529799076,
1014
+ "grad_norm": 0.057373046875,
1015
+ "learning_rate": 0.00014207342490685774,
1016
+ "loss": 0.7772,
1017
+ "step": 715
1018
+ },
1019
+ {
1020
+ "epoch": 0.4285375505098648,
1021
+ "grad_norm": 0.05078125,
1022
+ "learning_rate": 0.00014112871031306119,
1023
+ "loss": 0.7811,
1024
+ "step": 720
1025
+ },
1026
+ {
1027
+ "epoch": 0.4315135057217389,
1028
+ "grad_norm": 0.0478515625,
1029
+ "learning_rate": 0.00014017955679886598,
1030
+ "loss": 0.7644,
1031
+ "step": 725
1032
+ },
1033
+ {
1034
+ "epoch": 0.43448946093361296,
1035
+ "grad_norm": 0.05029296875,
1036
+ "learning_rate": 0.00013922606680407307,
1037
+ "loss": 0.7757,
1038
+ "step": 730
1039
+ },
1040
+ {
1041
+ "epoch": 0.437465416145487,
1042
+ "grad_norm": 0.050048828125,
1043
+ "learning_rate": 0.000138268343236509,
1044
+ "loss": 0.7764,
1045
+ "step": 735
1046
+ },
1047
+ {
1048
+ "epoch": 0.44044137135736106,
1049
+ "grad_norm": 0.05029296875,
1050
+ "learning_rate": 0.0001373064894609194,
1051
+ "loss": 0.7784,
1052
+ "step": 740
1053
+ },
1054
+ {
1055
+ "epoch": 0.44341732656923516,
1056
+ "grad_norm": 0.04833984375,
1057
+ "learning_rate": 0.0001363406092878131,
1058
+ "loss": 0.7663,
1059
+ "step": 745
1060
+ },
1061
+ {
1062
+ "epoch": 0.4463932817811092,
1063
+ "grad_norm": 0.04736328125,
1064
+ "learning_rate": 0.00013537080696225814,
1065
+ "loss": 0.7828,
1066
+ "step": 750
1067
+ },
1068
+ {
1069
+ "epoch": 0.44936923699298326,
1070
+ "grad_norm": 0.0517578125,
1071
+ "learning_rate": 0.0001343971871526307,
1072
+ "loss": 0.7764,
1073
+ "step": 755
1074
+ },
1075
+ {
1076
+ "epoch": 0.4523451922048573,
1077
+ "grad_norm": 0.053466796875,
1078
+ "learning_rate": 0.00013341985493931877,
1079
+ "loss": 0.7753,
1080
+ "step": 760
1081
+ },
1082
+ {
1083
+ "epoch": 0.45532114741673135,
1084
+ "grad_norm": 0.0498046875,
1085
+ "learning_rate": 0.00013243891580338072,
1086
+ "loss": 0.7663,
1087
+ "step": 765
1088
+ },
1089
+ {
1090
+ "epoch": 0.45829710262860546,
1091
+ "grad_norm": 0.052001953125,
1092
+ "learning_rate": 0.00013145447561516138,
1093
+ "loss": 0.7801,
1094
+ "step": 770
1095
+ },
1096
+ {
1097
+ "epoch": 0.4612730578404795,
1098
+ "grad_norm": 0.053466796875,
1099
+ "learning_rate": 0.00013046664062286545,
1100
+ "loss": 0.7917,
1101
+ "step": 775
1102
+ },
1103
+ {
1104
+ "epoch": 0.46424901305235355,
1105
+ "grad_norm": 0.051513671875,
1106
+ "learning_rate": 0.00012947551744109043,
1107
+ "loss": 0.7751,
1108
+ "step": 780
1109
+ },
1110
+ {
1111
+ "epoch": 0.4672249682642276,
1112
+ "grad_norm": 0.049560546875,
1113
+ "learning_rate": 0.00012848121303932013,
1114
+ "loss": 0.7585,
1115
+ "step": 785
1116
+ },
1117
+ {
1118
+ "epoch": 0.4702009234761017,
1119
+ "grad_norm": 0.04736328125,
1120
+ "learning_rate": 0.00012748383473037948,
1121
+ "loss": 0.7888,
1122
+ "step": 790
1123
+ },
1124
+ {
1125
+ "epoch": 0.47317687868797575,
1126
+ "grad_norm": 0.0537109375,
1127
+ "learning_rate": 0.00012648349015885273,
1128
+ "loss": 0.7766,
1129
+ "step": 795
1130
+ },
1131
+ {
1132
+ "epoch": 0.4761528338998498,
1133
+ "grad_norm": 0.052490234375,
1134
+ "learning_rate": 0.0001254802872894655,
1135
+ "loss": 0.7707,
1136
+ "step": 800
1137
+ },
1138
+ {
1139
+ "epoch": 0.47912878911172385,
1140
+ "grad_norm": 0.04931640625,
1141
+ "learning_rate": 0.0001244743343954324,
1142
+ "loss": 0.7678,
1143
+ "step": 805
1144
+ },
1145
+ {
1146
+ "epoch": 0.48210474432359796,
1147
+ "grad_norm": 0.048828125,
1148
+ "learning_rate": 0.00012346574004677154,
1149
+ "loss": 0.7477,
1150
+ "step": 810
1151
+ },
1152
+ {
1153
+ "epoch": 0.485080699535472,
1154
+ "grad_norm": 0.0498046875,
1155
+ "learning_rate": 0.0001224546130985867,
1156
+ "loss": 0.7758,
1157
+ "step": 815
1158
+ },
1159
+ {
1160
+ "epoch": 0.48805665474734605,
1161
+ "grad_norm": 0.047119140625,
1162
+ "learning_rate": 0.00012144106267931876,
1163
+ "loss": 0.7797,
1164
+ "step": 820
1165
+ },
1166
+ {
1167
+ "epoch": 0.4910326099592201,
1168
+ "grad_norm": 0.046142578125,
1169
+ "learning_rate": 0.00012042519817896804,
1170
+ "loss": 0.764,
1171
+ "step": 825
1172
+ },
1173
+ {
1174
+ "epoch": 0.49400856517109415,
1175
+ "grad_norm": 0.048583984375,
1176
+ "learning_rate": 0.00011940712923728783,
1177
+ "loss": 0.7602,
1178
+ "step": 830
1179
+ },
1180
+ {
1181
+ "epoch": 0.49698452038296825,
1182
+ "grad_norm": 0.048583984375,
1183
+ "learning_rate": 0.00011838696573195139,
1184
+ "loss": 0.7631,
1185
+ "step": 835
1186
+ },
1187
+ {
1188
+ "epoch": 0.4999604755948423,
1189
+ "grad_norm": 0.051513671875,
1190
+ "learning_rate": 0.00011736481776669306,
1191
+ "loss": 0.7793,
1192
+ "step": 840
1193
+ },
1194
+ {
1195
+ "epoch": 0.5029364308067164,
1196
+ "grad_norm": 0.048828125,
1197
+ "learning_rate": 0.00011634079565942497,
1198
+ "loss": 0.7658,
1199
+ "step": 845
1200
+ },
1201
+ {
1202
+ "epoch": 0.5059123860185905,
1203
+ "grad_norm": 0.04931640625,
1204
+ "learning_rate": 0.00011531500993033093,
1205
+ "loss": 0.7499,
1206
+ "step": 850
1207
+ },
1208
+ {
1209
+ "epoch": 0.5088883412304644,
1210
+ "grad_norm": 0.047119140625,
1211
+ "learning_rate": 0.00011428757128993802,
1212
+ "loss": 0.7706,
1213
+ "step": 855
1214
+ },
1215
+ {
1216
+ "epoch": 0.5118642964423386,
1217
+ "grad_norm": 0.05029296875,
1218
+ "learning_rate": 0.00011325859062716795,
1219
+ "loss": 0.7757,
1220
+ "step": 860
1221
+ },
1222
+ {
1223
+ "epoch": 0.5148402516542127,
1224
+ "grad_norm": 0.0478515625,
1225
+ "learning_rate": 0.00011222817899736914,
1226
+ "loss": 0.7828,
1227
+ "step": 865
1228
+ },
1229
+ {
1230
+ "epoch": 0.5178162068660866,
1231
+ "grad_norm": 0.048095703125,
1232
+ "learning_rate": 0.00011119644761033078,
1233
+ "loss": 0.7687,
1234
+ "step": 870
1235
+ },
1236
+ {
1237
+ "epoch": 0.5207921620779608,
1238
+ "grad_norm": 0.05029296875,
1239
+ "learning_rate": 0.00011016350781828019,
1240
+ "loss": 0.775,
1241
+ "step": 875
1242
+ },
1243
+ {
1244
+ "epoch": 0.5237681172898347,
1245
+ "grad_norm": 0.05078125,
1246
+ "learning_rate": 0.00010912947110386484,
1247
+ "loss": 0.7664,
1248
+ "step": 880
1249
+ },
1250
+ {
1251
+ "epoch": 0.5267440725017088,
1252
+ "grad_norm": 0.050048828125,
1253
+ "learning_rate": 0.00010809444906812033,
1254
+ "loss": 0.7962,
1255
+ "step": 885
1256
+ },
1257
+ {
1258
+ "epoch": 0.529720027713583,
1259
+ "grad_norm": 0.048583984375,
1260
+ "learning_rate": 0.00010705855341842563,
1261
+ "loss": 0.7806,
1262
+ "step": 890
1263
+ },
1264
+ {
1265
+ "epoch": 0.532695982925457,
1266
+ "grad_norm": 0.046875,
1267
+ "learning_rate": 0.0001060218959564466,
1268
+ "loss": 0.7626,
1269
+ "step": 895
1270
+ },
1271
+ {
1272
+ "epoch": 0.535671938137331,
1273
+ "grad_norm": 0.04931640625,
1274
+ "learning_rate": 0.00010498458856606972,
1275
+ "loss": 0.7747,
1276
+ "step": 900
1277
+ },
1278
+ {
1279
+ "epoch": 0.538647893349205,
1280
+ "grad_norm": 0.049072265625,
1281
+ "learning_rate": 0.00010394674320132662,
1282
+ "loss": 0.7813,
1283
+ "step": 905
1284
+ },
1285
+ {
1286
+ "epoch": 0.5416238485610791,
1287
+ "grad_norm": 0.0498046875,
1288
+ "learning_rate": 0.00010290847187431113,
1289
+ "loss": 0.7913,
1290
+ "step": 910
1291
+ },
1292
+ {
1293
+ "epoch": 0.5445998037729533,
1294
+ "grad_norm": 0.05029296875,
1295
+ "learning_rate": 0.00010186988664309023,
1296
+ "loss": 0.7648,
1297
+ "step": 915
1298
+ },
1299
+ {
1300
+ "epoch": 0.5475757589848272,
1301
+ "grad_norm": 0.04833984375,
1302
+ "learning_rate": 0.00010083109959960973,
1303
+ "loss": 0.7837,
1304
+ "step": 920
1305
+ },
1306
+ {
1307
+ "epoch": 0.5505517141967013,
1308
+ "grad_norm": 0.04638671875,
1309
+ "learning_rate": 9.979222285759651e-05,
1310
+ "loss": 0.7756,
1311
+ "step": 925
1312
+ },
1313
+ {
1314
+ "epoch": 0.5535276694085755,
1315
+ "grad_norm": 0.045166015625,
1316
+ "learning_rate": 9.875336854045851e-05,
1317
+ "loss": 0.7512,
1318
+ "step": 930
1319
+ },
1320
+ {
1321
+ "epoch": 0.5565036246204494,
1322
+ "grad_norm": 0.04833984375,
1323
+ "learning_rate": 9.771464876918331e-05,
1324
+ "loss": 0.7696,
1325
+ "step": 935
1326
+ },
1327
+ {
1328
+ "epoch": 0.5594795798323235,
1329
+ "grad_norm": 0.82421875,
1330
+ "learning_rate": 9.667617565023735e-05,
1331
+ "loss": 0.7855,
1332
+ "step": 940
1333
+ },
1334
+ {
1335
+ "epoch": 0.5624555350441975,
1336
+ "grad_norm": 0.061279296875,
1337
+ "learning_rate": 9.563806126346642e-05,
1338
+ "loss": 0.7613,
1339
+ "step": 945
1340
+ },
1341
+ {
1342
+ "epoch": 0.5654314902560716,
1343
+ "grad_norm": 0.049072265625,
1344
+ "learning_rate": 9.460041764999928e-05,
1345
+ "loss": 0.7782,
1346
+ "step": 950
1347
+ },
1348
+ {
1349
+ "epoch": 0.5684074454679457,
1350
+ "grad_norm": 0.05029296875,
1351
+ "learning_rate": 9.356335680015534e-05,
1352
+ "loss": 0.7781,
1353
+ "step": 955
1354
+ },
1355
+ {
1356
+ "epoch": 0.5713834006798197,
1357
+ "grad_norm": 0.0498046875,
1358
+ "learning_rate": 9.252699064135758e-05,
1359
+ "loss": 0.7953,
1360
+ "step": 960
1361
+ },
1362
+ {
1363
+ "epoch": 0.5743593558916938,
1364
+ "grad_norm": 0.049560546875,
1365
+ "learning_rate": 9.149143102605294e-05,
1366
+ "loss": 0.7649,
1367
+ "step": 965
1368
+ },
1369
+ {
1370
+ "epoch": 0.5773353111035678,
1371
+ "grad_norm": 0.048583984375,
1372
+ "learning_rate": 9.045678971963988e-05,
1373
+ "loss": 0.7635,
1374
+ "step": 970
1375
+ },
1376
+ {
1377
+ "epoch": 0.5803112663154419,
1378
+ "grad_norm": 0.050048828125,
1379
+ "learning_rate": 8.942317838840623e-05,
1380
+ "loss": 0.7689,
1381
+ "step": 975
1382
+ },
1383
+ {
1384
+ "epoch": 0.583287221527316,
1385
+ "grad_norm": 0.0517578125,
1386
+ "learning_rate": 8.839070858747697e-05,
1387
+ "loss": 0.7869,
1388
+ "step": 980
1389
+ },
1390
+ {
1391
+ "epoch": 0.58626317673919,
1392
+ "grad_norm": 0.0498046875,
1393
+ "learning_rate": 8.735949174877466e-05,
1394
+ "loss": 0.7606,
1395
+ "step": 985
1396
+ },
1397
+ {
1398
+ "epoch": 0.5892391319510641,
1399
+ "grad_norm": 0.0498046875,
1400
+ "learning_rate": 8.632963916899268e-05,
1401
+ "loss": 0.7665,
1402
+ "step": 990
1403
+ },
1404
+ {
1405
+ "epoch": 0.5922150871629382,
1406
+ "grad_norm": 0.04931640625,
1407
+ "learning_rate": 8.530126199758323e-05,
1408
+ "loss": 0.7967,
1409
+ "step": 995
1410
+ },
1411
+ {
1412
+ "epoch": 0.5951910423748122,
1413
+ "grad_norm": 0.052001953125,
1414
+ "learning_rate": 8.427447122476148e-05,
1415
+ "loss": 0.7784,
1416
+ "step": 1000
1417
+ },
1418
+ {
1419
+ "epoch": 0.5981669975866863,
1420
+ "grad_norm": 0.048583984375,
1421
+ "learning_rate": 8.324937766952638e-05,
1422
+ "loss": 0.7651,
1423
+ "step": 1005
1424
+ },
1425
+ {
1426
+ "epoch": 0.6011429527985603,
1427
+ "grad_norm": 0.05810546875,
1428
+ "learning_rate": 8.222609196770036e-05,
1429
+ "loss": 0.7773,
1430
+ "step": 1010
1431
+ },
1432
+ {
1433
+ "epoch": 0.6041189080104344,
1434
+ "grad_norm": 0.05224609375,
1435
+ "learning_rate": 8.120472455998882e-05,
1436
+ "loss": 0.7509,
1437
+ "step": 1015
1438
+ },
1439
+ {
1440
+ "epoch": 0.6070948632223085,
1441
+ "grad_norm": 0.048828125,
1442
+ "learning_rate": 8.018538568006027e-05,
1443
+ "loss": 0.7623,
1444
+ "step": 1020
1445
+ },
1446
+ {
1447
+ "epoch": 0.6100708184341825,
1448
+ "grad_norm": 0.0693359375,
1449
+ "learning_rate": 7.916818534264921e-05,
1450
+ "loss": 0.7492,
1451
+ "step": 1025
1452
+ },
1453
+ {
1454
+ "epoch": 0.6130467736460566,
1455
+ "grad_norm": 0.0478515625,
1456
+ "learning_rate": 7.815323333168262e-05,
1457
+ "loss": 0.7612,
1458
+ "step": 1030
1459
+ },
1460
+ {
1461
+ "epoch": 0.6160227288579306,
1462
+ "grad_norm": 0.0556640625,
1463
+ "learning_rate": 7.714063918843106e-05,
1464
+ "loss": 0.7799,
1465
+ "step": 1035
1466
+ },
1467
+ {
1468
+ "epoch": 0.6189986840698047,
1469
+ "grad_norm": 0.049072265625,
1470
+ "learning_rate": 7.613051219968623e-05,
1471
+ "loss": 0.7611,
1472
+ "step": 1040
1473
+ },
1474
+ {
1475
+ "epoch": 0.6219746392816788,
1476
+ "grad_norm": 0.05029296875,
1477
+ "learning_rate": 7.512296138596601e-05,
1478
+ "loss": 0.7782,
1479
+ "step": 1045
1480
+ },
1481
+ {
1482
+ "epoch": 0.6249505944935528,
1483
+ "grad_norm": 0.048095703125,
1484
+ "learning_rate": 7.411809548974792e-05,
1485
+ "loss": 0.7671,
1486
+ "step": 1050
1487
+ },
1488
+ {
1489
+ "epoch": 0.6279265497054269,
1490
+ "grad_norm": 0.05078125,
1491
+ "learning_rate": 7.31160229637331e-05,
1492
+ "loss": 0.7753,
1493
+ "step": 1055
1494
+ },
1495
+ {
1496
+ "epoch": 0.630902504917301,
1497
+ "grad_norm": 0.05078125,
1498
+ "learning_rate": 7.211685195914097e-05,
1499
+ "loss": 0.7663,
1500
+ "step": 1060
1501
+ },
1502
+ {
1503
+ "epoch": 0.633878460129175,
1504
+ "grad_norm": 0.055419921875,
1505
+ "learning_rate": 7.112069031403704e-05,
1506
+ "loss": 0.7702,
1507
+ "step": 1065
1508
+ },
1509
+ {
1510
+ "epoch": 0.6368544153410491,
1511
+ "grad_norm": 0.0517578125,
1512
+ "learning_rate": 7.012764554169393e-05,
1513
+ "loss": 0.7724,
1514
+ "step": 1070
1515
+ },
1516
+ {
1517
+ "epoch": 0.6398303705529231,
1518
+ "grad_norm": 0.0478515625,
1519
+ "learning_rate": 6.913782481898789e-05,
1520
+ "loss": 0.7607,
1521
+ "step": 1075
1522
+ },
1523
+ {
1524
+ "epoch": 0.6428063257647972,
1525
+ "grad_norm": 0.049560546875,
1526
+ "learning_rate": 6.815133497483157e-05,
1527
+ "loss": 0.7879,
1528
+ "step": 1080
1529
+ },
1530
+ {
1531
+ "epoch": 0.6457822809766713,
1532
+ "grad_norm": 0.04931640625,
1533
+ "learning_rate": 6.71682824786439e-05,
1534
+ "loss": 0.7736,
1535
+ "step": 1085
1536
+ },
1537
+ {
1538
+ "epoch": 0.6487582361885453,
1539
+ "grad_norm": 0.049560546875,
1540
+ "learning_rate": 6.618877342885945e-05,
1541
+ "loss": 0.7452,
1542
+ "step": 1090
1543
+ },
1544
+ {
1545
+ "epoch": 0.6517341914004194,
1546
+ "grad_norm": 0.05126953125,
1547
+ "learning_rate": 6.521291354147727e-05,
1548
+ "loss": 0.7774,
1549
+ "step": 1095
1550
+ },
1551
+ {
1552
+ "epoch": 0.6547101466122934,
1553
+ "grad_norm": 0.0498046875,
1554
+ "learning_rate": 6.424080813865138e-05,
1555
+ "loss": 0.7825,
1556
+ "step": 1100
1557
+ },
1558
+ {
1559
+ "epoch": 0.6576861018241675,
1560
+ "grad_norm": 0.04736328125,
1561
+ "learning_rate": 6.327256213732344e-05,
1562
+ "loss": 0.7568,
1563
+ "step": 1105
1564
+ },
1565
+ {
1566
+ "epoch": 0.6606620570360416,
1567
+ "grad_norm": 0.048583984375,
1568
+ "learning_rate": 6.230828003789949e-05,
1569
+ "loss": 0.7697,
1570
+ "step": 1110
1571
+ },
1572
+ {
1573
+ "epoch": 0.6636380122479156,
1574
+ "grad_norm": 0.052490234375,
1575
+ "learning_rate": 6.134806591297133e-05,
1576
+ "loss": 0.7536,
1577
+ "step": 1115
1578
+ },
1579
+ {
1580
+ "epoch": 0.6666139674597897,
1581
+ "grad_norm": 0.04931640625,
1582
+ "learning_rate": 6.039202339608432e-05,
1583
+ "loss": 0.7799,
1584
+ "step": 1120
1585
+ },
1586
+ {
1587
+ "epoch": 0.6695899226716638,
1588
+ "grad_norm": 0.049560546875,
1589
+ "learning_rate": 5.944025567055251e-05,
1590
+ "loss": 0.7605,
1591
+ "step": 1125
1592
+ },
1593
+ {
1594
+ "epoch": 0.6725658778835378,
1595
+ "grad_norm": 0.051025390625,
1596
+ "learning_rate": 5.849286545832211e-05,
1597
+ "loss": 0.7747,
1598
+ "step": 1130
1599
+ },
1600
+ {
1601
+ "epoch": 0.6755418330954119,
1602
+ "grad_norm": 0.05078125,
1603
+ "learning_rate": 5.7549955008885294e-05,
1604
+ "loss": 0.7746,
1605
+ "step": 1135
1606
+ },
1607
+ {
1608
+ "epoch": 0.6785177883072859,
1609
+ "grad_norm": 0.04833984375,
1610
+ "learning_rate": 5.6611626088244194e-05,
1611
+ "loss": 0.7533,
1612
+ "step": 1140
1613
+ },
1614
+ {
1615
+ "epoch": 0.68149374351916,
1616
+ "grad_norm": 0.049072265625,
1617
+ "learning_rate": 5.567797996792813e-05,
1618
+ "loss": 0.7644,
1619
+ "step": 1145
1620
+ },
1621
+ {
1622
+ "epoch": 0.6844696987310341,
1623
+ "grad_norm": 0.048828125,
1624
+ "learning_rate": 5.47491174140631e-05,
1625
+ "loss": 0.7807,
1626
+ "step": 1150
1627
+ },
1628
+ {
1629
+ "epoch": 0.6874456539429081,
1630
+ "grad_norm": 0.047607421875,
1631
+ "learning_rate": 5.382513867649663e-05,
1632
+ "loss": 0.7398,
1633
+ "step": 1155
1634
+ },
1635
+ {
1636
+ "epoch": 0.6904216091547822,
1637
+ "grad_norm": 0.055908203125,
1638
+ "learning_rate": 5.290614347797802e-05,
1639
+ "loss": 0.8002,
1640
+ "step": 1160
1641
+ },
1642
+ {
1643
+ "epoch": 0.6933975643666562,
1644
+ "grad_norm": 0.047119140625,
1645
+ "learning_rate": 5.199223100339539e-05,
1646
+ "loss": 0.76,
1647
+ "step": 1165
1648
+ },
1649
+ {
1650
+ "epoch": 0.6963735195785303,
1651
+ "grad_norm": 0.050048828125,
1652
+ "learning_rate": 5.108349988907111e-05,
1653
+ "loss": 0.7712,
1654
+ "step": 1170
1655
+ },
1656
+ {
1657
+ "epoch": 0.6993494747904044,
1658
+ "grad_norm": 0.04931640625,
1659
+ "learning_rate": 5.0180048212115924e-05,
1660
+ "loss": 0.781,
1661
+ "step": 1175
1662
+ },
1663
+ {
1664
+ "epoch": 0.7023254300022784,
1665
+ "grad_norm": 0.0478515625,
1666
+ "learning_rate": 4.92819734798441e-05,
1667
+ "loss": 0.7758,
1668
+ "step": 1180
1669
+ },
1670
+ {
1671
+ "epoch": 0.7053013852141525,
1672
+ "grad_norm": 0.048828125,
1673
+ "learning_rate": 4.8389372619249326e-05,
1674
+ "loss": 0.7727,
1675
+ "step": 1185
1676
+ },
1677
+ {
1678
+ "epoch": 0.7082773404260266,
1679
+ "grad_norm": 0.048095703125,
1680
+ "learning_rate": 4.7502341966544e-05,
1681
+ "loss": 0.7605,
1682
+ "step": 1190
1683
+ },
1684
+ {
1685
+ "epoch": 0.7112532956379006,
1686
+ "grad_norm": 0.051513671875,
1687
+ "learning_rate": 4.6620977256761514e-05,
1688
+ "loss": 0.7649,
1689
+ "step": 1195
1690
+ },
1691
+ {
1692
+ "epoch": 0.7142292508497747,
1693
+ "grad_norm": 0.0478515625,
1694
+ "learning_rate": 4.574537361342407e-05,
1695
+ "loss": 0.7786,
1696
+ "step": 1200
1697
+ },
1698
+ {
1699
+ "epoch": 0.7172052060616487,
1700
+ "grad_norm": 0.04931640625,
1701
+ "learning_rate": 4.487562553827622e-05,
1702
+ "loss": 0.753,
1703
+ "step": 1205
1704
+ },
1705
+ {
1706
+ "epoch": 0.7201811612735228,
1707
+ "grad_norm": 0.048583984375,
1708
+ "learning_rate": 4.401182690108534e-05,
1709
+ "loss": 0.7849,
1710
+ "step": 1210
1711
+ },
1712
+ {
1713
+ "epoch": 0.7231571164853969,
1714
+ "grad_norm": 0.052978515625,
1715
+ "learning_rate": 4.315407092951078e-05,
1716
+ "loss": 0.7742,
1717
+ "step": 1215
1718
+ },
1719
+ {
1720
+ "epoch": 0.7261330716972709,
1721
+ "grad_norm": 0.05224609375,
1722
+ "learning_rate": 4.23024501990417e-05,
1723
+ "loss": 0.7956,
1724
+ "step": 1220
1725
+ },
1726
+ {
1727
+ "epoch": 0.729109026909145,
1728
+ "grad_norm": 0.0498046875,
1729
+ "learning_rate": 4.145705662300595e-05,
1730
+ "loss": 0.7837,
1731
+ "step": 1225
1732
+ },
1733
+ {
1734
+ "epoch": 0.732084982121019,
1735
+ "grad_norm": 0.05029296875,
1736
+ "learning_rate": 4.0617981442649855e-05,
1737
+ "loss": 0.7597,
1738
+ "step": 1230
1739
+ },
1740
+ {
1741
+ "epoch": 0.7350609373328931,
1742
+ "grad_norm": 0.048583984375,
1743
+ "learning_rate": 3.978531521729084e-05,
1744
+ "loss": 0.7705,
1745
+ "step": 1235
1746
+ },
1747
+ {
1748
+ "epoch": 0.7380368925447672,
1749
+ "grad_norm": 0.049560546875,
1750
+ "learning_rate": 3.89591478145437e-05,
1751
+ "loss": 0.7775,
1752
+ "step": 1240
1753
+ },
1754
+ {
1755
+ "epoch": 0.7410128477566412,
1756
+ "grad_norm": 0.05078125,
1757
+ "learning_rate": 3.813956840062118e-05,
1758
+ "loss": 0.7534,
1759
+ "step": 1245
1760
+ },
1761
+ {
1762
+ "epoch": 0.7439888029685153,
1763
+ "grad_norm": 0.04833984375,
1764
+ "learning_rate": 3.732666543071079e-05,
1765
+ "loss": 0.7725,
1766
+ "step": 1250
1767
+ },
1768
+ {
1769
+ "epoch": 0.7469647581803894,
1770
+ "grad_norm": 0.047119140625,
1771
+ "learning_rate": 3.652052663942769e-05,
1772
+ "loss": 0.7661,
1773
+ "step": 1255
1774
+ },
1775
+ {
1776
+ "epoch": 0.7499407133922634,
1777
+ "grad_norm": 0.05078125,
1778
+ "learning_rate": 3.5721239031346066e-05,
1779
+ "loss": 0.7778,
1780
+ "step": 1260
1781
+ },
1782
+ {
1783
+ "epoch": 0.7529166686041375,
1784
+ "grad_norm": 0.048095703125,
1785
+ "learning_rate": 3.492888887160866e-05,
1786
+ "loss": 0.7704,
1787
+ "step": 1265
1788
+ },
1789
+ {
1790
+ "epoch": 0.7558926238160115,
1791
+ "grad_norm": 0.04736328125,
1792
+ "learning_rate": 3.4143561676616575e-05,
1793
+ "loss": 0.7785,
1794
+ "step": 1270
1795
+ },
1796
+ {
1797
+ "epoch": 0.7588685790278856,
1798
+ "grad_norm": 0.05615234375,
1799
+ "learning_rate": 3.336534220479961e-05,
1800
+ "loss": 0.7834,
1801
+ "step": 1275
1802
+ },
1803
+ {
1804
+ "epoch": 0.7618445342397597,
1805
+ "grad_norm": 0.051513671875,
1806
+ "learning_rate": 3.259431444746846e-05,
1807
+ "loss": 0.7778,
1808
+ "step": 1280
1809
+ },
1810
+ {
1811
+ "epoch": 0.7648204894516337,
1812
+ "grad_norm": 0.053955078125,
1813
+ "learning_rate": 3.1830561619749863e-05,
1814
+ "loss": 0.7658,
1815
+ "step": 1285
1816
+ },
1817
+ {
1818
+ "epoch": 0.7677964446635078,
1819
+ "grad_norm": 0.05029296875,
1820
+ "learning_rate": 3.10741661516053e-05,
1821
+ "loss": 0.7822,
1822
+ "step": 1290
1823
+ },
1824
+ {
1825
+ "epoch": 0.7707723998753819,
1826
+ "grad_norm": 0.049560546875,
1827
+ "learning_rate": 3.032520967893453e-05,
1828
+ "loss": 0.7763,
1829
+ "step": 1295
1830
+ },
1831
+ {
1832
+ "epoch": 0.7737483550872559,
1833
+ "grad_norm": 0.0498046875,
1834
+ "learning_rate": 2.9583773034764826e-05,
1835
+ "loss": 0.7855,
1836
+ "step": 1300
1837
+ },
1838
+ {
1839
+ "epoch": 0.77672431029913,
1840
+ "grad_norm": 0.049560546875,
1841
+ "learning_rate": 2.8849936240527008e-05,
1842
+ "loss": 0.754,
1843
+ "step": 1305
1844
+ },
1845
+ {
1846
+ "epoch": 0.779700265511004,
1847
+ "grad_norm": 0.047607421875,
1848
+ "learning_rate": 2.8123778497418685e-05,
1849
+ "loss": 0.772,
1850
+ "step": 1310
1851
+ },
1852
+ {
1853
+ "epoch": 0.7826762207228781,
1854
+ "grad_norm": 0.04736328125,
1855
+ "learning_rate": 2.74053781778565e-05,
1856
+ "loss": 0.7646,
1857
+ "step": 1315
1858
+ },
1859
+ {
1860
+ "epoch": 0.7856521759347522,
1861
+ "grad_norm": 0.050048828125,
1862
+ "learning_rate": 2.669481281701739e-05,
1863
+ "loss": 0.7848,
1864
+ "step": 1320
1865
+ },
1866
+ {
1867
+ "epoch": 0.7886281311466262,
1868
+ "grad_norm": 0.048583984375,
1869
+ "learning_rate": 2.5992159104470526e-05,
1870
+ "loss": 0.7679,
1871
+ "step": 1325
1872
+ },
1873
+ {
1874
+ "epoch": 0.7916040863585003,
1875
+ "grad_norm": 0.050537109375,
1876
+ "learning_rate": 2.529749287590042e-05,
1877
+ "loss": 0.7532,
1878
+ "step": 1330
1879
+ },
1880
+ {
1881
+ "epoch": 0.7945800415703743,
1882
+ "grad_norm": 0.048095703125,
1883
+ "learning_rate": 2.461088910492202e-05,
1884
+ "loss": 0.77,
1885
+ "step": 1335
1886
+ },
1887
+ {
1888
+ "epoch": 0.7975559967822484,
1889
+ "grad_norm": 0.0478515625,
1890
+ "learning_rate": 2.3932421894989167e-05,
1891
+ "loss": 0.7768,
1892
+ "step": 1340
1893
+ },
1894
+ {
1895
+ "epoch": 0.8005319519941225,
1896
+ "grad_norm": 0.050048828125,
1897
+ "learning_rate": 2.326216447139663e-05,
1898
+ "loss": 0.7543,
1899
+ "step": 1345
1900
+ },
1901
+ {
1902
+ "epoch": 0.8035079072059965,
1903
+ "grad_norm": 0.050048828125,
1904
+ "learning_rate": 2.260018917337726e-05,
1905
+ "loss": 0.7606,
1906
+ "step": 1350
1907
+ },
1908
+ {
1909
+ "epoch": 0.8064838624178706,
1910
+ "grad_norm": 0.047607421875,
1911
+ "learning_rate": 2.194656744629442e-05,
1912
+ "loss": 0.7629,
1913
+ "step": 1355
1914
+ },
1915
+ {
1916
+ "epoch": 0.8094598176297447,
1917
+ "grad_norm": 0.049560546875,
1918
+ "learning_rate": 2.1301369833931117e-05,
1919
+ "loss": 0.7619,
1920
+ "step": 1360
1921
+ },
1922
+ {
1923
+ "epoch": 0.8124357728416187,
1924
+ "grad_norm": 0.04931640625,
1925
+ "learning_rate": 2.0664665970876496e-05,
1926
+ "loss": 0.7534,
1927
+ "step": 1365
1928
+ },
1929
+ {
1930
+ "epoch": 0.8154117280534928,
1931
+ "grad_norm": 0.048828125,
1932
+ "learning_rate": 2.0036524575010172e-05,
1933
+ "loss": 0.7765,
1934
+ "step": 1370
1935
+ },
1936
+ {
1937
+ "epoch": 0.8183876832653668,
1938
+ "grad_norm": 0.052001953125,
1939
+ "learning_rate": 1.9417013440085864e-05,
1940
+ "loss": 0.7622,
1941
+ "step": 1375
1942
+ },
1943
+ {
1944
+ "epoch": 0.8213636384772409,
1945
+ "grad_norm": 0.04833984375,
1946
+ "learning_rate": 1.880619942841435e-05,
1947
+ "loss": 0.7795,
1948
+ "step": 1380
1949
+ },
1950
+ {
1951
+ "epoch": 0.824339593689115,
1952
+ "grad_norm": 0.049072265625,
1953
+ "learning_rate": 1.8204148463647453e-05,
1954
+ "loss": 0.7599,
1955
+ "step": 1385
1956
+ },
1957
+ {
1958
+ "epoch": 0.827315548900989,
1959
+ "grad_norm": 0.052734375,
1960
+ "learning_rate": 1.7610925523662835e-05,
1961
+ "loss": 0.7617,
1962
+ "step": 1390
1963
+ },
1964
+ {
1965
+ "epoch": 0.8302915041128631,
1966
+ "grad_norm": 0.047607421875,
1967
+ "learning_rate": 1.702659463355125e-05,
1968
+ "loss": 0.7608,
1969
+ "step": 1395
1970
+ },
1971
+ {
1972
+ "epoch": 0.8332674593247371,
1973
+ "grad_norm": 0.0478515625,
1974
+ "learning_rate": 1.6451218858706374e-05,
1975
+ "loss": 0.782,
1976
+ "step": 1400
1977
+ },
1978
+ {
1979
+ "epoch": 0.8362434145366112,
1980
+ "grad_norm": 0.049072265625,
1981
+ "learning_rate": 1.5884860298018322e-05,
1982
+ "loss": 0.7622,
1983
+ "step": 1405
1984
+ },
1985
+ {
1986
+ "epoch": 0.8392193697484853,
1987
+ "grad_norm": 0.04736328125,
1988
+ "learning_rate": 1.5327580077171587e-05,
1989
+ "loss": 0.771,
1990
+ "step": 1410
1991
+ },
1992
+ {
1993
+ "epoch": 0.8421953249603593,
1994
+ "grad_norm": 0.047119140625,
1995
+ "learning_rate": 1.4779438342047713e-05,
1996
+ "loss": 0.7684,
1997
+ "step": 1415
1998
+ },
1999
+ {
2000
+ "epoch": 0.8451712801722334,
2001
+ "grad_norm": 0.048583984375,
2002
+ "learning_rate": 1.4240494252234049e-05,
2003
+ "loss": 0.7886,
2004
+ "step": 1420
2005
+ },
2006
+ {
2007
+ "epoch": 0.8481472353841075,
2008
+ "grad_norm": 0.050537109375,
2009
+ "learning_rate": 1.3710805974638696e-05,
2010
+ "loss": 0.7813,
2011
+ "step": 1425
2012
+ },
2013
+ {
2014
+ "epoch": 0.8511231905959815,
2015
+ "grad_norm": 0.052978515625,
2016
+ "learning_rate": 1.3190430677212794e-05,
2017
+ "loss": 0.7557,
2018
+ "step": 1430
2019
+ },
2020
+ {
2021
+ "epoch": 0.8540991458078556,
2022
+ "grad_norm": 0.048095703125,
2023
+ "learning_rate": 1.2679424522780426e-05,
2024
+ "loss": 0.7487,
2025
+ "step": 1435
2026
+ },
2027
+ {
2028
+ "epoch": 0.8570751010197296,
2029
+ "grad_norm": 0.049560546875,
2030
+ "learning_rate": 1.2177842662977135e-05,
2031
+ "loss": 0.777,
2032
+ "step": 1440
2033
+ },
2034
+ {
2035
+ "epoch": 0.8600510562316037,
2036
+ "grad_norm": 0.050048828125,
2037
+ "learning_rate": 1.1685739232297643e-05,
2038
+ "loss": 0.7724,
2039
+ "step": 1445
2040
+ },
2041
+ {
2042
+ "epoch": 0.8630270114434778,
2043
+ "grad_norm": 0.050537109375,
2044
+ "learning_rate": 1.1203167342253062e-05,
2045
+ "loss": 0.7689,
2046
+ "step": 1450
2047
+ },
2048
+ {
2049
+ "epoch": 0.8660029666553518,
2050
+ "grad_norm": 0.054443359375,
2051
+ "learning_rate": 1.0730179075638868e-05,
2052
+ "loss": 0.7415,
2053
+ "step": 1455
2054
+ },
2055
+ {
2056
+ "epoch": 0.8689789218672259,
2057
+ "grad_norm": 0.049072265625,
2058
+ "learning_rate": 1.0266825480913611e-05,
2059
+ "loss": 0.7772,
2060
+ "step": 1460
2061
+ },
2062
+ {
2063
+ "epoch": 0.8719548770790999,
2064
+ "grad_norm": 0.04833984375,
2065
+ "learning_rate": 9.813156566689518e-06,
2066
+ "loss": 0.7715,
2067
+ "step": 1465
2068
+ },
2069
+ {
2070
+ "epoch": 0.874930832290974,
2071
+ "grad_norm": 0.04931640625,
2072
+ "learning_rate": 9.369221296335006e-06,
2073
+ "loss": 0.7646,
2074
+ "step": 1470
2075
+ },
2076
+ {
2077
+ "epoch": 0.8779067875028481,
2078
+ "grad_norm": 0.048583984375,
2079
+ "learning_rate": 8.935067582690382e-06,
2080
+ "loss": 0.7446,
2081
+ "step": 1475
2082
+ },
2083
+ {
2084
+ "epoch": 0.8808827427147221,
2085
+ "grad_norm": 0.0478515625,
2086
+ "learning_rate": 8.510742282896544e-06,
2087
+ "loss": 0.7768,
2088
+ "step": 1480
2089
+ },
2090
+ {
2091
+ "epoch": 0.8838586979265962,
2092
+ "grad_norm": 0.047607421875,
2093
+ "learning_rate": 8.096291193337934e-06,
2094
+ "loss": 0.7535,
2095
+ "step": 1485
2096
+ },
2097
+ {
2098
+ "epoch": 0.8868346531384703,
2099
+ "grad_norm": 0.048828125,
2100
+ "learning_rate": 7.69175904469982e-06,
2101
+ "loss": 0.7691,
2102
+ "step": 1490
2103
+ },
2104
+ {
2105
+ "epoch": 0.8898106083503443,
2106
+ "grad_norm": 0.050537109375,
2107
+ "learning_rate": 7.2971894971405665e-06,
2108
+ "loss": 0.7562,
2109
+ "step": 1495
2110
+ },
2111
+ {
2112
+ "epoch": 0.8927865635622184,
2113
+ "grad_norm": 0.05224609375,
2114
+ "learning_rate": 6.9126251355795864e-06,
2115
+ "loss": 0.7574,
2116
+ "step": 1500
2117
+ },
2118
+ {
2119
+ "epoch": 0.8957625187740924,
2120
+ "grad_norm": 0.04931640625,
2121
+ "learning_rate": 6.538107465101162e-06,
2122
+ "loss": 0.7733,
2123
+ "step": 1505
2124
+ },
2125
+ {
2126
+ "epoch": 0.8987384739859665,
2127
+ "grad_norm": 0.04736328125,
2128
+ "learning_rate": 6.173676906475012e-06,
2129
+ "loss": 0.7656,
2130
+ "step": 1510
2131
+ },
2132
+ {
2133
+ "epoch": 0.9017144291978406,
2134
+ "grad_norm": 0.04931640625,
2135
+ "learning_rate": 5.8193727917936536e-06,
2136
+ "loss": 0.7604,
2137
+ "step": 1515
2138
+ },
2139
+ {
2140
+ "epoch": 0.9046903844097146,
2141
+ "grad_norm": 0.04931640625,
2142
+ "learning_rate": 5.475233360227516e-06,
2143
+ "loss": 0.7688,
2144
+ "step": 1520
2145
+ },
2146
+ {
2147
+ "epoch": 0.9076663396215887,
2148
+ "grad_norm": 0.0478515625,
2149
+ "learning_rate": 5.14129575389779e-06,
2150
+ "loss": 0.7556,
2151
+ "step": 1525
2152
+ },
2153
+ {
2154
+ "epoch": 0.9106422948334627,
2155
+ "grad_norm": 0.0537109375,
2156
+ "learning_rate": 4.817596013867764e-06,
2157
+ "loss": 0.7765,
2158
+ "step": 1530
2159
+ },
2160
+ {
2161
+ "epoch": 0.9136182500453368,
2162
+ "grad_norm": 0.0517578125,
2163
+ "learning_rate": 4.504169076253084e-06,
2164
+ "loss": 0.7788,
2165
+ "step": 1535
2166
+ },
2167
+ {
2168
+ "epoch": 0.9165942052572109,
2169
+ "grad_norm": 0.04931640625,
2170
+ "learning_rate": 4.20104876845111e-06,
2171
+ "loss": 0.7723,
2172
+ "step": 1540
2173
+ },
2174
+ {
2175
+ "epoch": 0.9195701604690849,
2176
+ "grad_norm": 0.04736328125,
2177
+ "learning_rate": 3.908267805490051e-06,
2178
+ "loss": 0.7748,
2179
+ "step": 1545
2180
+ },
2181
+ {
2182
+ "epoch": 0.922546115680959,
2183
+ "grad_norm": 0.048095703125,
2184
+ "learning_rate": 3.625857786498055e-06,
2185
+ "loss": 0.762,
2186
+ "step": 1550
2187
+ },
2188
+ {
2189
+ "epoch": 0.9255220708928331,
2190
+ "grad_norm": 0.049560546875,
2191
+ "learning_rate": 3.3538491912928792e-06,
2192
+ "loss": 0.7728,
2193
+ "step": 1555
2194
+ },
2195
+ {
2196
+ "epoch": 0.9284980261047071,
2197
+ "grad_norm": 0.047119140625,
2198
+ "learning_rate": 3.092271377092215e-06,
2199
+ "loss": 0.7595,
2200
+ "step": 1560
2201
+ },
2202
+ {
2203
+ "epoch": 0.9314739813165812,
2204
+ "grad_norm": 0.049072265625,
2205
+ "learning_rate": 2.8411525753452185e-06,
2206
+ "loss": 0.7776,
2207
+ "step": 1565
2208
+ },
2209
+ {
2210
+ "epoch": 0.9344499365284552,
2211
+ "grad_norm": 0.07470703125,
2212
+ "learning_rate": 2.6005198886856487e-06,
2213
+ "loss": 0.7814,
2214
+ "step": 1570
2215
+ },
2216
+ {
2217
+ "epoch": 0.9374258917403293,
2218
+ "grad_norm": 0.05224609375,
2219
+ "learning_rate": 2.3703992880066638e-06,
2220
+ "loss": 0.7533,
2221
+ "step": 1575
2222
+ },
2223
+ {
2224
+ "epoch": 0.9404018469522034,
2225
+ "grad_norm": 0.049072265625,
2226
+ "learning_rate": 2.150815609657875e-06,
2227
+ "loss": 0.7801,
2228
+ "step": 1580
2229
+ },
2230
+ {
2231
+ "epoch": 0.9433778021640774,
2232
+ "grad_norm": 0.05126953125,
2233
+ "learning_rate": 1.9417925527648096e-06,
2234
+ "loss": 0.766,
2235
+ "step": 1585
2236
+ },
2237
+ {
2238
+ "epoch": 0.9463537573759515,
2239
+ "grad_norm": 0.048095703125,
2240
+ "learning_rate": 1.7433526766711728e-06,
2241
+ "loss": 0.7911,
2242
+ "step": 1590
2243
+ },
2244
+ {
2245
+ "epoch": 0.9493297125878255,
2246
+ "grad_norm": 0.051513671875,
2247
+ "learning_rate": 1.5555173985039918e-06,
2248
+ "loss": 0.7798,
2249
+ "step": 1595
2250
+ },
2251
+ {
2252
+ "epoch": 0.9523056677996996,
2253
+ "grad_norm": 0.05029296875,
2254
+ "learning_rate": 1.378306990862177e-06,
2255
+ "loss": 0.7833,
2256
+ "step": 1600
2257
+ },
2258
+ {
2259
+ "epoch": 0.9552816230115737,
2260
+ "grad_norm": 0.046875,
2261
+ "learning_rate": 1.2117405796285286e-06,
2262
+ "loss": 0.7826,
2263
+ "step": 1605
2264
+ },
2265
+ {
2266
+ "epoch": 0.9582575782234477,
2267
+ "grad_norm": 0.055908203125,
2268
+ "learning_rate": 1.055836141905553e-06,
2269
+ "loss": 0.7792,
2270
+ "step": 1610
2271
+ },
2272
+ {
2273
+ "epoch": 0.9612335334353218,
2274
+ "grad_norm": 0.0517578125,
2275
+ "learning_rate": 9.106105040751822e-07,
2276
+ "loss": 0.7687,
2277
+ "step": 1615
2278
+ },
2279
+ {
2280
+ "epoch": 0.9642094886471959,
2281
+ "grad_norm": 0.048583984375,
2282
+ "learning_rate": 7.760793399827937e-07,
2283
+ "loss": 0.7452,
2284
+ "step": 1620
2285
+ },
2286
+ {
2287
+ "epoch": 0.9671854438590699,
2288
+ "grad_norm": 0.0576171875,
2289
+ "learning_rate": 6.522571692455736e-07,
2290
+ "loss": 0.7721,
2291
+ "step": 1625
2292
+ },
2293
+ {
2294
+ "epoch": 0.970161399070944,
2295
+ "grad_norm": 0.1337890625,
2296
+ "learning_rate": 5.391573556854157e-07,
2297
+ "loss": 0.7848,
2298
+ "step": 1630
2299
+ },
2300
+ {
2301
+ "epoch": 0.973137354282818,
2302
+ "grad_norm": 0.048583984375,
2303
+ "learning_rate": 4.3679210588661866e-07,
2304
+ "loss": 0.7767,
2305
+ "step": 1635
2306
+ },
2307
+ {
2308
+ "epoch": 0.9761133094946921,
2309
+ "grad_norm": 0.048095703125,
2310
+ "learning_rate": 3.451724678784518e-07,
2311
+ "loss": 0.7726,
2312
+ "step": 1640
2313
+ },
2314
+ {
2315
+ "epoch": 0.9790892647065662,
2316
+ "grad_norm": 0.04931640625,
2317
+ "learning_rate": 2.643083299427751e-07,
2318
+ "loss": 0.7683,
2319
+ "step": 1645
2320
+ },
2321
+ {
2322
+ "epoch": 0.9820652199184402,
2323
+ "grad_norm": 0.054931640625,
2324
+ "learning_rate": 1.9420841954681525e-07,
2325
+ "loss": 0.7589,
2326
+ "step": 1650
2327
+ },
2328
+ {
2329
+ "epoch": 0.9850411751303143,
2330
+ "grad_norm": 0.05078125,
2331
+ "learning_rate": 1.3488030240123017e-07,
2332
+ "loss": 0.7689,
2333
+ "step": 1655
2334
+ },
2335
+ {
2336
+ "epoch": 0.9880171303421883,
2337
+ "grad_norm": 0.052001953125,
2338
+ "learning_rate": 8.633038164358454e-08,
2339
+ "loss": 0.7738,
2340
+ "step": 1660
2341
+ },
2342
+ {
2343
+ "epoch": 0.9909930855540624,
2344
+ "grad_norm": 0.07275390625,
2345
+ "learning_rate": 4.856389714723575e-08,
2346
+ "loss": 0.7728,
2347
+ "step": 1665
2348
+ },
2349
+ {
2350
+ "epoch": 0.9939690407659365,
2351
+ "grad_norm": 0.05322265625,
2352
+ "learning_rate": 2.1584924955819764e-08,
2353
+ "loss": 0.7646,
2354
+ "step": 1670
2355
+ },
2356
+ {
2357
+ "epoch": 0.9969449959778105,
2358
+ "grad_norm": 0.04931640625,
2359
+ "learning_rate": 5.396376843369577e-09,
2360
+ "loss": 0.7784,
2361
+ "step": 1675
2362
+ },
2363
+ {
2364
+ "epoch": 0.9999209511896846,
2365
+ "grad_norm": 0.048583984375,
2366
+ "learning_rate": 0.0,
2367
+ "loss": 0.7765,
2368
+ "step": 1680
2369
+ },
2370
+ {
2371
+ "epoch": 0.9999209511896846,
2372
+ "step": 1680,
2373
+ "total_flos": 4.668038789401149e+18,
2374
+ "train_loss": 0.0,
2375
+ "train_runtime": 0.023,
2376
+ "train_samples_per_second": 9337043.614,
2377
+ "train_steps_per_second": 72939.887
2378
+ }
2379
+ ],
2380
+ "logging_steps": 5,
2381
+ "max_steps": 1680,
2382
+ "num_input_tokens_seen": 0,
2383
+ "num_train_epochs": 1,
2384
+ "save_steps": 20,
2385
+ "total_flos": 4.668038789401149e+18,
2386
+ "train_batch_size": 1,
2387
+ "trial_name": null,
2388
+ "trial_params": null
2389
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9d31bf40bd2f44498c1efbb0db922fa05e17d634e936782d148c1a93a9b316a0
3
+ size 5112