Upload folder using huggingface_hub
Browse files- 46_128_e3_3e-5/README.md +63 -0
- 46_128_e3_3e-5/adapter_config.json +39 -0
- 46_128_e3_3e-5/adapter_model.safetensors +3 -0
- 46_128_e3_3e-5/all_results.json +9 -0
- 46_128_e3_3e-5/config.json +32 -0
- 46_128_e3_3e-5/merges.txt +0 -0
- 46_128_e3_3e-5/special_tokens_map.json +45 -0
- 46_128_e3_3e-5/tokenizer.json +0 -0
- 46_128_e3_3e-5/tokenizer_config.json +188 -0
- 46_128_e3_3e-5/train_results.json +9 -0
- 46_128_e3_3e-5/trainer_state.json +1198 -0
- 46_128_e3_3e-5/training_args.bin +3 -0
- 46_128_e3_3e-5/vocab.json +0 -0
46_128_e3_3e-5/README.md
ADDED
|
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
base_model: ibm-granite/granite-3.3-8b-base
|
| 5 |
+
tags:
|
| 6 |
+
- alignment-handbook
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
datasets:
|
| 9 |
+
- data/knowledge_lora_training_data_2000
|
| 10 |
+
model-index:
|
| 11 |
+
- name: 46_128_e3_3e-5
|
| 12 |
+
results: []
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 16 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 17 |
+
|
| 18 |
+
# 46_128_e3_3e-5
|
| 19 |
+
|
| 20 |
+
This model is a fine-tuned version of [ibm-granite/granite-3.3-8b-base](https://huggingface.co/ibm-granite/granite-3.3-8b-base) on the data/knowledge_lora_training_data_2000 dataset.
|
| 21 |
+
|
| 22 |
+
## Model description
|
| 23 |
+
|
| 24 |
+
More information needed
|
| 25 |
+
|
| 26 |
+
## Intended uses & limitations
|
| 27 |
+
|
| 28 |
+
More information needed
|
| 29 |
+
|
| 30 |
+
## Training and evaluation data
|
| 31 |
+
|
| 32 |
+
More information needed
|
| 33 |
+
|
| 34 |
+
## Training procedure
|
| 35 |
+
|
| 36 |
+
### Training hyperparameters
|
| 37 |
+
|
| 38 |
+
The following hyperparameters were used during training:
|
| 39 |
+
- learning_rate: 3e-05
|
| 40 |
+
- train_batch_size: 2
|
| 41 |
+
- eval_batch_size: 8
|
| 42 |
+
- seed: 42
|
| 43 |
+
- distributed_type: multi-GPU
|
| 44 |
+
- num_devices: 8
|
| 45 |
+
- gradient_accumulation_steps: 2
|
| 46 |
+
- total_train_batch_size: 32
|
| 47 |
+
- total_eval_batch_size: 64
|
| 48 |
+
- optimizer: Use adamw_torch with betas=(0.9,0.95) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 49 |
+
- lr_scheduler_type: cosine
|
| 50 |
+
- lr_scheduler_warmup_ratio: 0.05
|
| 51 |
+
- num_epochs: 3.0
|
| 52 |
+
|
| 53 |
+
### Training results
|
| 54 |
+
|
| 55 |
+
|
| 56 |
+
|
| 57 |
+
### Framework versions
|
| 58 |
+
|
| 59 |
+
- PEFT 0.15.2
|
| 60 |
+
- Transformers 4.52.4
|
| 61 |
+
- Pytorch 2.7.0+cu126
|
| 62 |
+
- Datasets 3.6.0
|
| 63 |
+
- Tokenizers 0.21.2
|
46_128_e3_3e-5/adapter_config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "ibm-granite/granite-3.3-8b-base",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"corda_config": null,
|
| 7 |
+
"eva_config": null,
|
| 8 |
+
"exclude_modules": null,
|
| 9 |
+
"fan_in_fan_out": false,
|
| 10 |
+
"inference_mode": true,
|
| 11 |
+
"init_lora_weights": true,
|
| 12 |
+
"layer_replication": null,
|
| 13 |
+
"layers_pattern": null,
|
| 14 |
+
"layers_to_transform": null,
|
| 15 |
+
"loftq_config": {},
|
| 16 |
+
"lora_alpha": 256,
|
| 17 |
+
"lora_bias": false,
|
| 18 |
+
"lora_dropout": 0.05,
|
| 19 |
+
"megatron_config": null,
|
| 20 |
+
"megatron_core": "megatron.core",
|
| 21 |
+
"modules_to_save": null,
|
| 22 |
+
"peft_type": "LORA",
|
| 23 |
+
"r": 128,
|
| 24 |
+
"rank_pattern": {},
|
| 25 |
+
"revision": null,
|
| 26 |
+
"target_modules": [
|
| 27 |
+
"down_proj",
|
| 28 |
+
"v_proj",
|
| 29 |
+
"o_proj",
|
| 30 |
+
"gate_proj",
|
| 31 |
+
"up_proj",
|
| 32 |
+
"q_proj",
|
| 33 |
+
"k_proj"
|
| 34 |
+
],
|
| 35 |
+
"task_type": "CAUSAL_LM",
|
| 36 |
+
"trainable_token_indices": null,
|
| 37 |
+
"use_dora": false,
|
| 38 |
+
"use_rslora": false
|
| 39 |
+
}
|
46_128_e3_3e-5/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aa4169371302c92064eca3bdd065ce370141084ca5d7c9484a5f616bfdb873f0
|
| 3 |
+
size 791751704
|
46_128_e3_3e-5/all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 1.1532968769638892e+18,
|
| 4 |
+
"train_loss": 0.5211183394930788,
|
| 5 |
+
"train_runtime": 540.0579,
|
| 6 |
+
"train_samples": 8814,
|
| 7 |
+
"train_samples_per_second": 48.961,
|
| 8 |
+
"train_steps_per_second": 1.533
|
| 9 |
+
}
|
46_128_e3_3e-5/config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"GraniteForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0,
|
| 7 |
+
"attention_multiplier": 0.0078125,
|
| 8 |
+
"bos_token_id": 0,
|
| 9 |
+
"embedding_multiplier": 12.0,
|
| 10 |
+
"eos_token_id": 0,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 4096,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 12800,
|
| 15 |
+
"logits_scaling": 16.0,
|
| 16 |
+
"max_position_embeddings": 131072,
|
| 17 |
+
"mlp_bias": false,
|
| 18 |
+
"model_type": "granite",
|
| 19 |
+
"num_attention_heads": 32,
|
| 20 |
+
"num_hidden_layers": 40,
|
| 21 |
+
"num_key_value_heads": 8,
|
| 22 |
+
"pad_token_id": 0,
|
| 23 |
+
"residual_multiplier": 0.22,
|
| 24 |
+
"rms_norm_eps": 1e-05,
|
| 25 |
+
"rope_scaling": null,
|
| 26 |
+
"rope_theta": 10000000.0,
|
| 27 |
+
"tie_word_embeddings": true,
|
| 28 |
+
"torch_dtype": "bfloat16",
|
| 29 |
+
"transformers_version": "4.52.4",
|
| 30 |
+
"use_cache": true,
|
| 31 |
+
"vocab_size": 49152
|
| 32 |
+
}
|
46_128_e3_3e-5/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
46_128_e3_3e-5/special_tokens_map.json
ADDED
|
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|endoftext|>",
|
| 4 |
+
"<fim_prefix>",
|
| 5 |
+
"<fim_middle>",
|
| 6 |
+
"<fim_suffix>",
|
| 7 |
+
"<fim_pad>",
|
| 8 |
+
"<filename>",
|
| 9 |
+
"<gh_stars>",
|
| 10 |
+
"<issue_start>",
|
| 11 |
+
"<issue_comment>",
|
| 12 |
+
"<issue_closed>",
|
| 13 |
+
"<jupyter_start>",
|
| 14 |
+
"<jupyter_text>",
|
| 15 |
+
"<jupyter_code>",
|
| 16 |
+
"<jupyter_output>",
|
| 17 |
+
"<empty_output>",
|
| 18 |
+
"<commit_before>",
|
| 19 |
+
"<commit_msg>",
|
| 20 |
+
"<commit_after>",
|
| 21 |
+
"<reponame>"
|
| 22 |
+
],
|
| 23 |
+
"bos_token": {
|
| 24 |
+
"content": "<|endoftext|>",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"eos_token": {
|
| 31 |
+
"content": "<|endoftext|>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
},
|
| 37 |
+
"pad_token": "<reponame>",
|
| 38 |
+
"unk_token": {
|
| 39 |
+
"content": "<|endoftext|>",
|
| 40 |
+
"lstrip": false,
|
| 41 |
+
"normalized": false,
|
| 42 |
+
"rstrip": false,
|
| 43 |
+
"single_word": false
|
| 44 |
+
}
|
| 45 |
+
}
|
46_128_e3_3e-5/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
46_128_e3_3e-5/tokenizer_config.json
ADDED
|
@@ -0,0 +1,188 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<|endoftext|>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": false,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<fim_prefix>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": false,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"2": {
|
| 21 |
+
"content": "<fim_middle>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": false,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"3": {
|
| 29 |
+
"content": "<fim_suffix>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": false,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"4": {
|
| 37 |
+
"content": "<fim_pad>",
|
| 38 |
+
"lstrip": false,
|
| 39 |
+
"normalized": false,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
},
|
| 44 |
+
"5": {
|
| 45 |
+
"content": "<filename>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": false,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false,
|
| 50 |
+
"special": true
|
| 51 |
+
},
|
| 52 |
+
"6": {
|
| 53 |
+
"content": "<gh_stars>",
|
| 54 |
+
"lstrip": false,
|
| 55 |
+
"normalized": false,
|
| 56 |
+
"rstrip": false,
|
| 57 |
+
"single_word": false,
|
| 58 |
+
"special": true
|
| 59 |
+
},
|
| 60 |
+
"7": {
|
| 61 |
+
"content": "<issue_start>",
|
| 62 |
+
"lstrip": false,
|
| 63 |
+
"normalized": false,
|
| 64 |
+
"rstrip": false,
|
| 65 |
+
"single_word": false,
|
| 66 |
+
"special": true
|
| 67 |
+
},
|
| 68 |
+
"8": {
|
| 69 |
+
"content": "<issue_comment>",
|
| 70 |
+
"lstrip": false,
|
| 71 |
+
"normalized": false,
|
| 72 |
+
"rstrip": false,
|
| 73 |
+
"single_word": false,
|
| 74 |
+
"special": true
|
| 75 |
+
},
|
| 76 |
+
"9": {
|
| 77 |
+
"content": "<issue_closed>",
|
| 78 |
+
"lstrip": false,
|
| 79 |
+
"normalized": false,
|
| 80 |
+
"rstrip": false,
|
| 81 |
+
"single_word": false,
|
| 82 |
+
"special": true
|
| 83 |
+
},
|
| 84 |
+
"10": {
|
| 85 |
+
"content": "<jupyter_start>",
|
| 86 |
+
"lstrip": false,
|
| 87 |
+
"normalized": false,
|
| 88 |
+
"rstrip": false,
|
| 89 |
+
"single_word": false,
|
| 90 |
+
"special": true
|
| 91 |
+
},
|
| 92 |
+
"11": {
|
| 93 |
+
"content": "<jupyter_text>",
|
| 94 |
+
"lstrip": false,
|
| 95 |
+
"normalized": false,
|
| 96 |
+
"rstrip": false,
|
| 97 |
+
"single_word": false,
|
| 98 |
+
"special": true
|
| 99 |
+
},
|
| 100 |
+
"12": {
|
| 101 |
+
"content": "<jupyter_code>",
|
| 102 |
+
"lstrip": false,
|
| 103 |
+
"normalized": false,
|
| 104 |
+
"rstrip": false,
|
| 105 |
+
"single_word": false,
|
| 106 |
+
"special": true
|
| 107 |
+
},
|
| 108 |
+
"13": {
|
| 109 |
+
"content": "<jupyter_output>",
|
| 110 |
+
"lstrip": false,
|
| 111 |
+
"normalized": false,
|
| 112 |
+
"rstrip": false,
|
| 113 |
+
"single_word": false,
|
| 114 |
+
"special": true
|
| 115 |
+
},
|
| 116 |
+
"14": {
|
| 117 |
+
"content": "<empty_output>",
|
| 118 |
+
"lstrip": false,
|
| 119 |
+
"normalized": false,
|
| 120 |
+
"rstrip": false,
|
| 121 |
+
"single_word": false,
|
| 122 |
+
"special": true
|
| 123 |
+
},
|
| 124 |
+
"15": {
|
| 125 |
+
"content": "<commit_before>",
|
| 126 |
+
"lstrip": false,
|
| 127 |
+
"normalized": false,
|
| 128 |
+
"rstrip": false,
|
| 129 |
+
"single_word": false,
|
| 130 |
+
"special": true
|
| 131 |
+
},
|
| 132 |
+
"16": {
|
| 133 |
+
"content": "<commit_msg>",
|
| 134 |
+
"lstrip": false,
|
| 135 |
+
"normalized": false,
|
| 136 |
+
"rstrip": false,
|
| 137 |
+
"single_word": false,
|
| 138 |
+
"special": true
|
| 139 |
+
},
|
| 140 |
+
"17": {
|
| 141 |
+
"content": "<commit_after>",
|
| 142 |
+
"lstrip": false,
|
| 143 |
+
"normalized": false,
|
| 144 |
+
"rstrip": false,
|
| 145 |
+
"single_word": false,
|
| 146 |
+
"special": true
|
| 147 |
+
},
|
| 148 |
+
"18": {
|
| 149 |
+
"content": "<reponame>",
|
| 150 |
+
"lstrip": false,
|
| 151 |
+
"normalized": false,
|
| 152 |
+
"rstrip": false,
|
| 153 |
+
"single_word": false,
|
| 154 |
+
"special": true
|
| 155 |
+
}
|
| 156 |
+
},
|
| 157 |
+
"additional_special_tokens": [
|
| 158 |
+
"<|endoftext|>",
|
| 159 |
+
"<fim_prefix>",
|
| 160 |
+
"<fim_middle>",
|
| 161 |
+
"<fim_suffix>",
|
| 162 |
+
"<fim_pad>",
|
| 163 |
+
"<filename>",
|
| 164 |
+
"<gh_stars>",
|
| 165 |
+
"<issue_start>",
|
| 166 |
+
"<issue_comment>",
|
| 167 |
+
"<issue_closed>",
|
| 168 |
+
"<jupyter_start>",
|
| 169 |
+
"<jupyter_text>",
|
| 170 |
+
"<jupyter_code>",
|
| 171 |
+
"<jupyter_output>",
|
| 172 |
+
"<empty_output>",
|
| 173 |
+
"<commit_before>",
|
| 174 |
+
"<commit_msg>",
|
| 175 |
+
"<commit_after>",
|
| 176 |
+
"<reponame>"
|
| 177 |
+
],
|
| 178 |
+
"bos_token": "<|endoftext|>",
|
| 179 |
+
"clean_up_tokenization_spaces": true,
|
| 180 |
+
"eos_token": "<|endoftext|>",
|
| 181 |
+
"extra_special_tokens": {},
|
| 182 |
+
"model_max_length": 8192,
|
| 183 |
+
"pad_token": "<reponame>",
|
| 184 |
+
"padding_side": "left",
|
| 185 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 186 |
+
"unk_token": "<|endoftext|>",
|
| 187 |
+
"vocab_size": 49152
|
| 188 |
+
}
|
46_128_e3_3e-5/train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 1.1532968769638892e+18,
|
| 4 |
+
"train_loss": 0.5211183394930788,
|
| 5 |
+
"train_runtime": 540.0579,
|
| 6 |
+
"train_samples": 8814,
|
| 7 |
+
"train_samples_per_second": 48.961,
|
| 8 |
+
"train_steps_per_second": 1.533
|
| 9 |
+
}
|
46_128_e3_3e-5/trainer_state.json
ADDED
|
@@ -0,0 +1,1198 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 3.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 828,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.018148820326678767,
|
| 14 |
+
"grad_norm": 1.053978681564331,
|
| 15 |
+
"learning_rate": 2.8571428571428573e-06,
|
| 16 |
+
"loss": 1.2026,
|
| 17 |
+
"step": 5
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.036297640653357534,
|
| 21 |
+
"grad_norm": 0.8344705104827881,
|
| 22 |
+
"learning_rate": 6.428571428571429e-06,
|
| 23 |
+
"loss": 1.2595,
|
| 24 |
+
"step": 10
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.0544464609800363,
|
| 28 |
+
"grad_norm": 0.6406563520431519,
|
| 29 |
+
"learning_rate": 9.999999999999999e-06,
|
| 30 |
+
"loss": 1.2506,
|
| 31 |
+
"step": 15
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.07259528130671507,
|
| 35 |
+
"grad_norm": 0.6422258615493774,
|
| 36 |
+
"learning_rate": 1.3571428571428572e-05,
|
| 37 |
+
"loss": 1.2046,
|
| 38 |
+
"step": 20
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.09074410163339383,
|
| 42 |
+
"grad_norm": 0.5808016657829285,
|
| 43 |
+
"learning_rate": 1.7142857142857142e-05,
|
| 44 |
+
"loss": 1.1901,
|
| 45 |
+
"step": 25
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.1088929219600726,
|
| 49 |
+
"grad_norm": 0.5957618951797485,
|
| 50 |
+
"learning_rate": 2.0714285714285715e-05,
|
| 51 |
+
"loss": 1.1198,
|
| 52 |
+
"step": 30
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.12704174228675136,
|
| 56 |
+
"grad_norm": 0.5114598274230957,
|
| 57 |
+
"learning_rate": 2.4285714285714288e-05,
|
| 58 |
+
"loss": 1.1087,
|
| 59 |
+
"step": 35
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.14519056261343014,
|
| 63 |
+
"grad_norm": 0.5465165376663208,
|
| 64 |
+
"learning_rate": 2.7857142857142858e-05,
|
| 65 |
+
"loss": 1.1311,
|
| 66 |
+
"step": 40
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.16333938294010888,
|
| 70 |
+
"grad_norm": 0.6859464049339294,
|
| 71 |
+
"learning_rate": 2.999952073733835e-05,
|
| 72 |
+
"loss": 1.091,
|
| 73 |
+
"step": 45
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.18148820326678766,
|
| 77 |
+
"grad_norm": 0.5999999046325684,
|
| 78 |
+
"learning_rate": 2.9994129384105553e-05,
|
| 79 |
+
"loss": 1.117,
|
| 80 |
+
"step": 50
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.1996370235934664,
|
| 84 |
+
"grad_norm": 0.5300623178482056,
|
| 85 |
+
"learning_rate": 2.9982749759659206e-05,
|
| 86 |
+
"loss": 1.1006,
|
| 87 |
+
"step": 55
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.2177858439201452,
|
| 91 |
+
"grad_norm": 0.5001697540283203,
|
| 92 |
+
"learning_rate": 2.996538640872983e-05,
|
| 93 |
+
"loss": 1.0505,
|
| 94 |
+
"step": 60
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.23593466424682397,
|
| 98 |
+
"grad_norm": 0.5570539832115173,
|
| 99 |
+
"learning_rate": 2.9942046265795044e-05,
|
| 100 |
+
"loss": 1.0493,
|
| 101 |
+
"step": 65
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.2540834845735027,
|
| 105 |
+
"grad_norm": 0.5113873481750488,
|
| 106 |
+
"learning_rate": 2.9912738652310094e-05,
|
| 107 |
+
"loss": 1.0307,
|
| 108 |
+
"step": 70
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.27223230490018147,
|
| 112 |
+
"grad_norm": 0.549514651298523,
|
| 113 |
+
"learning_rate": 2.9877475272985114e-05,
|
| 114 |
+
"loss": 1.0725,
|
| 115 |
+
"step": 75
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.29038112522686027,
|
| 119 |
+
"grad_norm": 0.5019710659980774,
|
| 120 |
+
"learning_rate": 2.9836270211110567e-05,
|
| 121 |
+
"loss": 0.9993,
|
| 122 |
+
"step": 80
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.308529945553539,
|
| 126 |
+
"grad_norm": 0.6285251379013062,
|
| 127 |
+
"learning_rate": 2.978913992293273e-05,
|
| 128 |
+
"loss": 0.9709,
|
| 129 |
+
"step": 85
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.32667876588021777,
|
| 133 |
+
"grad_norm": 0.5799192786216736,
|
| 134 |
+
"learning_rate": 2.9736103231081494e-05,
|
| 135 |
+
"loss": 1.0036,
|
| 136 |
+
"step": 90
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.3448275862068966,
|
| 140 |
+
"grad_norm": 0.8796660304069519,
|
| 141 |
+
"learning_rate": 2.9677181317053095e-05,
|
| 142 |
+
"loss": 0.9914,
|
| 143 |
+
"step": 95
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.3629764065335753,
|
| 147 |
+
"grad_norm": 0.6010348796844482,
|
| 148 |
+
"learning_rate": 2.961239771275076e-05,
|
| 149 |
+
"loss": 0.9594,
|
| 150 |
+
"step": 100
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.3811252268602541,
|
| 154 |
+
"grad_norm": 0.6567463278770447,
|
| 155 |
+
"learning_rate": 2.9541778291086666e-05,
|
| 156 |
+
"loss": 0.9461,
|
| 157 |
+
"step": 105
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.3992740471869328,
|
| 161 |
+
"grad_norm": 0.6974070072174072,
|
| 162 |
+
"learning_rate": 2.9465351255648967e-05,
|
| 163 |
+
"loss": 0.9341,
|
| 164 |
+
"step": 110
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.41742286751361163,
|
| 168 |
+
"grad_norm": 0.6048945784568787,
|
| 169 |
+
"learning_rate": 2.9383147129438e-05,
|
| 170 |
+
"loss": 0.9432,
|
| 171 |
+
"step": 115
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.4355716878402904,
|
| 175 |
+
"grad_norm": 0.623528778553009,
|
| 176 |
+
"learning_rate": 2.929519874267619e-05,
|
| 177 |
+
"loss": 0.9048,
|
| 178 |
+
"step": 120
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.4537205081669691,
|
| 182 |
+
"grad_norm": 0.6546903848648071,
|
| 183 |
+
"learning_rate": 2.9201541219696464e-05,
|
| 184 |
+
"loss": 0.8977,
|
| 185 |
+
"step": 125
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.47186932849364793,
|
| 189 |
+
"grad_norm": 0.6157246828079224,
|
| 190 |
+
"learning_rate": 2.9102211964914554e-05,
|
| 191 |
+
"loss": 0.8994,
|
| 192 |
+
"step": 130
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.4900181488203267,
|
| 196 |
+
"grad_norm": 0.6559402942657471,
|
| 197 |
+
"learning_rate": 2.899725064789058e-05,
|
| 198 |
+
"loss": 0.8717,
|
| 199 |
+
"step": 135
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.5081669691470054,
|
| 203 |
+
"grad_norm": 0.6873505115509033,
|
| 204 |
+
"learning_rate": 2.888669918748606e-05,
|
| 205 |
+
"loss": 0.9159,
|
| 206 |
+
"step": 140
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.5263157894736842,
|
| 210 |
+
"grad_norm": 0.7308347225189209,
|
| 211 |
+
"learning_rate": 2.8770601735122593e-05,
|
| 212 |
+
"loss": 0.8488,
|
| 213 |
+
"step": 145
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.5444646098003629,
|
| 217 |
+
"grad_norm": 0.950423002243042,
|
| 218 |
+
"learning_rate": 2.864900465714891e-05,
|
| 219 |
+
"loss": 0.885,
|
| 220 |
+
"step": 150
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.5626134301270418,
|
| 224 |
+
"grad_norm": 0.7558553814888,
|
| 225 |
+
"learning_rate": 2.852195651632334e-05,
|
| 226 |
+
"loss": 0.9028,
|
| 227 |
+
"step": 155
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.5807622504537205,
|
| 231 |
+
"grad_norm": 0.824928879737854,
|
| 232 |
+
"learning_rate": 2.8389508052419107e-05,
|
| 233 |
+
"loss": 0.8566,
|
| 234 |
+
"step": 160
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.5989110707803993,
|
| 238 |
+
"grad_norm": 0.8163105845451355,
|
| 239 |
+
"learning_rate": 2.8251712161960118e-05,
|
| 240 |
+
"loss": 0.8061,
|
| 241 |
+
"step": 165
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.617059891107078,
|
| 245 |
+
"grad_norm": 0.808022677898407,
|
| 246 |
+
"learning_rate": 2.810862387709551e-05,
|
| 247 |
+
"loss": 0.8067,
|
| 248 |
+
"step": 170
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.6352087114337568,
|
| 252 |
+
"grad_norm": 0.8545414805412292,
|
| 253 |
+
"learning_rate": 2.7960300343621145e-05,
|
| 254 |
+
"loss": 0.7638,
|
| 255 |
+
"step": 175
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.6533575317604355,
|
| 259 |
+
"grad_norm": 0.7906185388565063,
|
| 260 |
+
"learning_rate": 2.7806800798157126e-05,
|
| 261 |
+
"loss": 0.7844,
|
| 262 |
+
"step": 180
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.6715063520871143,
|
| 266 |
+
"grad_norm": 0.9761213064193726,
|
| 267 |
+
"learning_rate": 2.7648186544490135e-05,
|
| 268 |
+
"loss": 0.7686,
|
| 269 |
+
"step": 185
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.6896551724137931,
|
| 273 |
+
"grad_norm": 0.8222201466560364,
|
| 274 |
+
"learning_rate": 2.7484520929090313e-05,
|
| 275 |
+
"loss": 0.7774,
|
| 276 |
+
"step": 190
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.7078039927404719,
|
| 280 |
+
"grad_norm": 0.909803032875061,
|
| 281 |
+
"learning_rate": 2.731586931581229e-05,
|
| 282 |
+
"loss": 0.7106,
|
| 283 |
+
"step": 195
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.7259528130671506,
|
| 287 |
+
"grad_norm": 0.916222095489502,
|
| 288 |
+
"learning_rate": 2.7142299059790534e-05,
|
| 289 |
+
"loss": 0.7134,
|
| 290 |
+
"step": 200
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.7441016333938294,
|
| 294 |
+
"grad_norm": 0.8470730185508728,
|
| 295 |
+
"learning_rate": 2.6963879480539437e-05,
|
| 296 |
+
"loss": 0.7378,
|
| 297 |
+
"step": 205
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.7622504537205081,
|
| 301 |
+
"grad_norm": 0.8945624232292175,
|
| 302 |
+
"learning_rate": 2.678068183426888e-05,
|
| 303 |
+
"loss": 0.7132,
|
| 304 |
+
"step": 210
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.7803992740471869,
|
| 308 |
+
"grad_norm": 0.9166932702064514,
|
| 309 |
+
"learning_rate": 2.6592779285426332e-05,
|
| 310 |
+
"loss": 0.7452,
|
| 311 |
+
"step": 215
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.7985480943738656,
|
| 315 |
+
"grad_norm": 0.981235682964325,
|
| 316 |
+
"learning_rate": 2.640024687747682e-05,
|
| 317 |
+
"loss": 0.6757,
|
| 318 |
+
"step": 220
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.8166969147005445,
|
| 322 |
+
"grad_norm": 0.9101331830024719,
|
| 323 |
+
"learning_rate": 2.6203161502932514e-05,
|
| 324 |
+
"loss": 0.6653,
|
| 325 |
+
"step": 225
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.8348457350272233,
|
| 329 |
+
"grad_norm": 0.9170770645141602,
|
| 330 |
+
"learning_rate": 2.600160187264384e-05,
|
| 331 |
+
"loss": 0.6444,
|
| 332 |
+
"step": 230
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.852994555353902,
|
| 336 |
+
"grad_norm": 0.917047381401062,
|
| 337 |
+
"learning_rate": 2.579564848436433e-05,
|
| 338 |
+
"loss": 0.6777,
|
| 339 |
+
"step": 235
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.8711433756805808,
|
| 343 |
+
"grad_norm": 1.009281873703003,
|
| 344 |
+
"learning_rate": 2.5585383590601953e-05,
|
| 345 |
+
"loss": 0.6193,
|
| 346 |
+
"step": 240
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.8892921960072595,
|
| 350 |
+
"grad_norm": 0.9169691205024719,
|
| 351 |
+
"learning_rate": 2.5370891165769498e-05,
|
| 352 |
+
"loss": 0.6573,
|
| 353 |
+
"step": 245
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.9074410163339383,
|
| 357 |
+
"grad_norm": 1.0548322200775146,
|
| 358 |
+
"learning_rate": 2.5152256872647403e-05,
|
| 359 |
+
"loss": 0.6386,
|
| 360 |
+
"step": 250
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.925589836660617,
|
| 364 |
+
"grad_norm": 1.0737444162368774,
|
| 365 |
+
"learning_rate": 2.4929568028172196e-05,
|
| 366 |
+
"loss": 0.6519,
|
| 367 |
+
"step": 255
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.9437386569872959,
|
| 371 |
+
"grad_norm": 0.9646818041801453,
|
| 372 |
+
"learning_rate": 2.470291356856437e-05,
|
| 373 |
+
"loss": 0.6805,
|
| 374 |
+
"step": 260
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.9618874773139746,
|
| 378 |
+
"grad_norm": 0.9463504552841187,
|
| 379 |
+
"learning_rate": 2.4472384013809516e-05,
|
| 380 |
+
"loss": 0.6229,
|
| 381 |
+
"step": 265
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.9800362976406534,
|
| 385 |
+
"grad_norm": 1.247211217880249,
|
| 386 |
+
"learning_rate": 2.4238071431506977e-05,
|
| 387 |
+
"loss": 0.5594,
|
| 388 |
+
"step": 270
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.9981851179673321,
|
| 392 |
+
"grad_norm": 1.0467617511749268,
|
| 393 |
+
"learning_rate": 2.400006940010041e-05,
|
| 394 |
+
"loss": 0.6216,
|
| 395 |
+
"step": 275
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.014519056261343,
|
| 399 |
+
"grad_norm": 0.9873857498168945,
|
| 400 |
+
"learning_rate": 2.375847297150491e-05,
|
| 401 |
+
"loss": 0.5546,
|
| 402 |
+
"step": 280
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.0326678765880217,
|
| 406 |
+
"grad_norm": 1.0524786710739136,
|
| 407 |
+
"learning_rate": 2.351337863314578e-05,
|
| 408 |
+
"loss": 0.5433,
|
| 409 |
+
"step": 285
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 1.0508166969147006,
|
| 413 |
+
"grad_norm": 1.0610527992248535,
|
| 414 |
+
"learning_rate": 2.3264884269423855e-05,
|
| 415 |
+
"loss": 0.5626,
|
| 416 |
+
"step": 290
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.0689655172413792,
|
| 420 |
+
"grad_norm": 1.3043826818466187,
|
| 421 |
+
"learning_rate": 2.301308912262303e-05,
|
| 422 |
+
"loss": 0.4991,
|
| 423 |
+
"step": 295
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.087114337568058,
|
| 427 |
+
"grad_norm": 1.0931960344314575,
|
| 428 |
+
"learning_rate": 2.27580937532754e-05,
|
| 429 |
+
"loss": 0.5176,
|
| 430 |
+
"step": 300
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 1.1052631578947367,
|
| 434 |
+
"grad_norm": 1.0401699542999268,
|
| 435 |
+
"learning_rate": 2.25e-05,
|
| 436 |
+
"loss": 0.5199,
|
| 437 |
+
"step": 305
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 1.1234119782214156,
|
| 441 |
+
"grad_norm": 1.1369975805282593,
|
| 442 |
+
"learning_rate": 2.2238910938831012e-05,
|
| 443 |
+
"loss": 0.4874,
|
| 444 |
+
"step": 310
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.1415607985480944,
|
| 448 |
+
"grad_norm": 0.9706031084060669,
|
| 449 |
+
"learning_rate": 2.197493084205189e-05,
|
| 450 |
+
"loss": 0.4802,
|
| 451 |
+
"step": 315
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 1.159709618874773,
|
| 455 |
+
"grad_norm": 1.2659469842910767,
|
| 456 |
+
"learning_rate": 2.1708165136551643e-05,
|
| 457 |
+
"loss": 0.5044,
|
| 458 |
+
"step": 320
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 1.177858439201452,
|
| 462 |
+
"grad_norm": 1.0018097162246704,
|
| 463 |
+
"learning_rate": 2.1438720361720074e-05,
|
| 464 |
+
"loss": 0.4702,
|
| 465 |
+
"step": 325
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 1.1960072595281306,
|
| 469 |
+
"grad_norm": 1.3021812438964844,
|
| 470 |
+
"learning_rate": 2.1166704126898627e-05,
|
| 471 |
+
"loss": 0.493,
|
| 472 |
+
"step": 330
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 1.2141560798548094,
|
| 476 |
+
"grad_norm": 1.0300713777542114,
|
| 477 |
+
"learning_rate": 2.089222506840402e-05,
|
| 478 |
+
"loss": 0.5138,
|
| 479 |
+
"step": 335
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 1.232304900181488,
|
| 483 |
+
"grad_norm": 1.052535057067871,
|
| 484 |
+
"learning_rate": 2.0615392806141663e-05,
|
| 485 |
+
"loss": 0.5095,
|
| 486 |
+
"step": 340
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 1.250453720508167,
|
| 490 |
+
"grad_norm": 0.9547827243804932,
|
| 491 |
+
"learning_rate": 2.033631789982626e-05,
|
| 492 |
+
"loss": 0.4992,
|
| 493 |
+
"step": 345
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 1.2686025408348458,
|
| 497 |
+
"grad_norm": 1.0180929899215698,
|
| 498 |
+
"learning_rate": 2.005511180482712e-05,
|
| 499 |
+
"loss": 0.4418,
|
| 500 |
+
"step": 350
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 1.2867513611615244,
|
| 504 |
+
"grad_norm": 1.2542915344238281,
|
| 505 |
+
"learning_rate": 1.9771886827655725e-05,
|
| 506 |
+
"loss": 0.4595,
|
| 507 |
+
"step": 355
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 1.3049001814882033,
|
| 511 |
+
"grad_norm": 1.25581955909729,
|
| 512 |
+
"learning_rate": 1.9486756081113373e-05,
|
| 513 |
+
"loss": 0.4912,
|
| 514 |
+
"step": 360
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 1.3230490018148822,
|
| 518 |
+
"grad_norm": 1.1318669319152832,
|
| 519 |
+
"learning_rate": 1.9199833439116844e-05,
|
| 520 |
+
"loss": 0.4498,
|
| 521 |
+
"step": 365
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 1.3411978221415608,
|
| 525 |
+
"grad_norm": 1.1164065599441528,
|
| 526 |
+
"learning_rate": 1.8911233491220054e-05,
|
| 527 |
+
"loss": 0.4431,
|
| 528 |
+
"step": 370
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 1.3593466424682394,
|
| 532 |
+
"grad_norm": 1.2040191888809204,
|
| 533 |
+
"learning_rate": 1.8621071496849948e-05,
|
| 534 |
+
"loss": 0.4677,
|
| 535 |
+
"step": 375
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 1.3774954627949183,
|
| 539 |
+
"grad_norm": 1.002925992012024,
|
| 540 |
+
"learning_rate": 1.8329463339274812e-05,
|
| 541 |
+
"loss": 0.4368,
|
| 542 |
+
"step": 380
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.3956442831215972,
|
| 546 |
+
"grad_norm": 1.4232462644577026,
|
| 547 |
+
"learning_rate": 1.8036525479323464e-05,
|
| 548 |
+
"loss": 0.4729,
|
| 549 |
+
"step": 385
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.4137931034482758,
|
| 553 |
+
"grad_norm": 1.0863258838653564,
|
| 554 |
+
"learning_rate": 1.7742374908873817e-05,
|
| 555 |
+
"loss": 0.4353,
|
| 556 |
+
"step": 390
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 1.4319419237749547,
|
| 560 |
+
"grad_norm": 1.1894892454147339,
|
| 561 |
+
"learning_rate": 1.744712910412927e-05,
|
| 562 |
+
"loss": 0.4151,
|
| 563 |
+
"step": 395
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 1.4500907441016335,
|
| 567 |
+
"grad_norm": 1.1045399904251099,
|
| 568 |
+
"learning_rate": 1.7150905978701765e-05,
|
| 569 |
+
"loss": 0.4311,
|
| 570 |
+
"step": 400
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 1.4682395644283122,
|
| 574 |
+
"grad_norm": 1.2653708457946777,
|
| 575 |
+
"learning_rate": 1.6853823836520076e-05,
|
| 576 |
+
"loss": 0.4086,
|
| 577 |
+
"step": 405
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 1.4863883847549908,
|
| 581 |
+
"grad_norm": 1.2789379358291626,
|
| 582 |
+
"learning_rate": 1.6556001324582275e-05,
|
| 583 |
+
"loss": 0.4614,
|
| 584 |
+
"step": 410
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 1.5045372050816697,
|
| 588 |
+
"grad_norm": 1.3415888547897339,
|
| 589 |
+
"learning_rate": 1.6257557385571147e-05,
|
| 590 |
+
"loss": 0.4009,
|
| 591 |
+
"step": 415
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 1.5226860254083485,
|
| 595 |
+
"grad_norm": 1.0844992399215698,
|
| 596 |
+
"learning_rate": 1.595861121035154e-05,
|
| 597 |
+
"loss": 0.4049,
|
| 598 |
+
"step": 420
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 1.5408348457350272,
|
| 602 |
+
"grad_norm": 1.1509233713150024,
|
| 603 |
+
"learning_rate": 1.565928219036856e-05,
|
| 604 |
+
"loss": 0.4281,
|
| 605 |
+
"step": 425
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 1.558983666061706,
|
| 609 |
+
"grad_norm": 1.1053664684295654,
|
| 610 |
+
"learning_rate": 1.535968986996573e-05,
|
| 611 |
+
"loss": 0.3987,
|
| 612 |
+
"step": 430
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 1.5771324863883849,
|
| 616 |
+
"grad_norm": 1.1549456119537354,
|
| 617 |
+
"learning_rate": 1.5059953898641993e-05,
|
| 618 |
+
"loss": 0.4173,
|
| 619 |
+
"step": 435
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 1.5952813067150635,
|
| 623 |
+
"grad_norm": 1.2164968252182007,
|
| 624 |
+
"learning_rate": 1.4760193983266813e-05,
|
| 625 |
+
"loss": 0.4142,
|
| 626 |
+
"step": 440
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 1.6134301270417422,
|
| 630 |
+
"grad_norm": 1.2697888612747192,
|
| 631 |
+
"learning_rate": 1.4460529840272285e-05,
|
| 632 |
+
"loss": 0.3463,
|
| 633 |
+
"step": 445
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 1.631578947368421,
|
| 637 |
+
"grad_norm": 1.072198748588562,
|
| 638 |
+
"learning_rate": 1.4161081147841456e-05,
|
| 639 |
+
"loss": 0.4271,
|
| 640 |
+
"step": 450
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 1.6497277676950999,
|
| 644 |
+
"grad_norm": 1.0467942953109741,
|
| 645 |
+
"learning_rate": 1.3861967498111952e-05,
|
| 646 |
+
"loss": 0.4064,
|
| 647 |
+
"step": 455
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.6678765880217785,
|
| 651 |
+
"grad_norm": 1.1074597835540771,
|
| 652 |
+
"learning_rate": 1.3563308349413926e-05,
|
| 653 |
+
"loss": 0.3801,
|
| 654 |
+
"step": 460
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.6860254083484574,
|
| 658 |
+
"grad_norm": 1.3617875576019287,
|
| 659 |
+
"learning_rate": 1.3265222978561453e-05,
|
| 660 |
+
"loss": 0.3967,
|
| 661 |
+
"step": 465
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.7041742286751362,
|
| 665 |
+
"grad_norm": 1.3225388526916504,
|
| 666 |
+
"learning_rate": 1.2967830433216436e-05,
|
| 667 |
+
"loss": 0.3527,
|
| 668 |
+
"step": 470
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.7223230490018149,
|
| 672 |
+
"grad_norm": 1.3548988103866577,
|
| 673 |
+
"learning_rate": 1.2671249484344016e-05,
|
| 674 |
+
"loss": 0.4025,
|
| 675 |
+
"step": 475
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 1.7404718693284935,
|
| 679 |
+
"grad_norm": 1.2789945602416992,
|
| 680 |
+
"learning_rate": 1.2375598578778496e-05,
|
| 681 |
+
"loss": 0.4124,
|
| 682 |
+
"step": 480
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 1.7586206896551724,
|
| 686 |
+
"grad_norm": 1.0328474044799805,
|
| 687 |
+
"learning_rate": 1.2080995791918697e-05,
|
| 688 |
+
"loss": 0.3887,
|
| 689 |
+
"step": 485
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.7767695099818512,
|
| 693 |
+
"grad_norm": 1.1773806810379028,
|
| 694 |
+
"learning_rate": 1.1787558780571678e-05,
|
| 695 |
+
"loss": 0.3677,
|
| 696 |
+
"step": 490
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 1.7949183303085299,
|
| 700 |
+
"grad_norm": 1.7615448236465454,
|
| 701 |
+
"learning_rate": 1.1495404735963622e-05,
|
| 702 |
+
"loss": 0.3594,
|
| 703 |
+
"step": 495
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 1.8130671506352087,
|
| 707 |
+
"grad_norm": 1.188841462135315,
|
| 708 |
+
"learning_rate": 1.1204650336936665e-05,
|
| 709 |
+
"loss": 0.3816,
|
| 710 |
+
"step": 500
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 1.8312159709618876,
|
| 714 |
+
"grad_norm": 1.3961434364318848,
|
| 715 |
+
"learning_rate": 1.0915411703350326e-05,
|
| 716 |
+
"loss": 0.3678,
|
| 717 |
+
"step": 505
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 1.8493647912885662,
|
| 721 |
+
"grad_norm": 1.4136226177215576,
|
| 722 |
+
"learning_rate": 1.0627804349706252e-05,
|
| 723 |
+
"loss": 0.345,
|
| 724 |
+
"step": 510
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 1.8675136116152449,
|
| 728 |
+
"grad_norm": 1.217195749282837,
|
| 729 |
+
"learning_rate": 1.034194313901466e-05,
|
| 730 |
+
"loss": 0.3923,
|
| 731 |
+
"step": 515
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 1.8856624319419237,
|
| 735 |
+
"grad_norm": 1.1499128341674805,
|
| 736 |
+
"learning_rate": 1.0057942236921005e-05,
|
| 737 |
+
"loss": 0.3437,
|
| 738 |
+
"step": 520
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 1.9038112522686026,
|
| 742 |
+
"grad_norm": 1.1922115087509155,
|
| 743 |
+
"learning_rate": 9.775915066111116e-06,
|
| 744 |
+
"loss": 0.3315,
|
| 745 |
+
"step": 525
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 1.9219600725952812,
|
| 749 |
+
"grad_norm": 1.088967204093933,
|
| 750 |
+
"learning_rate": 9.495974261013126e-06,
|
| 751 |
+
"loss": 0.3399,
|
| 752 |
+
"step": 530
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.94010889292196,
|
| 756 |
+
"grad_norm": 1.1436982154846191,
|
| 757 |
+
"learning_rate": 9.21823162281413e-06,
|
| 758 |
+
"loss": 0.3424,
|
| 759 |
+
"step": 535
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 1.958257713248639,
|
| 763 |
+
"grad_norm": 1.0752840042114258,
|
| 764 |
+
"learning_rate": 8.94279807480968e-06,
|
| 765 |
+
"loss": 0.2802,
|
| 766 |
+
"step": 540
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 1.9764065335753176,
|
| 770 |
+
"grad_norm": 1.1687134504318237,
|
| 771 |
+
"learning_rate": 8.669783618103834e-06,
|
| 772 |
+
"loss": 0.3245,
|
| 773 |
+
"step": 545
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 1.9945553539019962,
|
| 777 |
+
"grad_norm": 1.2609084844589233,
|
| 778 |
+
"learning_rate": 8.39929728767756e-06,
|
| 779 |
+
"loss": 0.3023,
|
| 780 |
+
"step": 550
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 2.010889292196007,
|
| 784 |
+
"grad_norm": 1.1618186235427856,
|
| 785 |
+
"learning_rate": 8.13144710884294e-06,
|
| 786 |
+
"loss": 0.3145,
|
| 787 |
+
"step": 555
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 2.029038112522686,
|
| 791 |
+
"grad_norm": 1.4103319644927979,
|
| 792 |
+
"learning_rate": 7.866340054100642e-06,
|
| 793 |
+
"loss": 0.2724,
|
| 794 |
+
"step": 560
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 2.047186932849365,
|
| 798 |
+
"grad_norm": 1.2527198791503906,
|
| 799 |
+
"learning_rate": 7.604082000417818e-06,
|
| 800 |
+
"loss": 0.2782,
|
| 801 |
+
"step": 565
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 2.0653357531760435,
|
| 805 |
+
"grad_norm": 1.284936785697937,
|
| 806 |
+
"learning_rate": 7.344777686943588e-06,
|
| 807 |
+
"loss": 0.2525,
|
| 808 |
+
"step": 570
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 2.0834845735027225,
|
| 812 |
+
"grad_norm": 1.242108941078186,
|
| 813 |
+
"learning_rate": 7.088530673178884e-06,
|
| 814 |
+
"loss": 0.2925,
|
| 815 |
+
"step": 575
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 2.101633393829401,
|
| 819 |
+
"grad_norm": 1.121969223022461,
|
| 820 |
+
"learning_rate": 6.835443297617446e-06,
|
| 821 |
+
"loss": 0.268,
|
| 822 |
+
"step": 580
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 2.11978221415608,
|
| 826 |
+
"grad_norm": 1.1498106718063354,
|
| 827 |
+
"learning_rate": 6.585616636874438e-06,
|
| 828 |
+
"loss": 0.2782,
|
| 829 |
+
"step": 585
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 2.1379310344827585,
|
| 833 |
+
"grad_norm": 1.008657455444336,
|
| 834 |
+
"learning_rate": 6.339150465319057e-06,
|
| 835 |
+
"loss": 0.2781,
|
| 836 |
+
"step": 590
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 2.1560798548094375,
|
| 840 |
+
"grad_norm": 1.1087747812271118,
|
| 841 |
+
"learning_rate": 6.096143215227199e-06,
|
| 842 |
+
"loss": 0.2714,
|
| 843 |
+
"step": 595
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 2.174228675136116,
|
| 847 |
+
"grad_norm": 1.0672491788864136,
|
| 848 |
+
"learning_rate": 5.856691937470104e-06,
|
| 849 |
+
"loss": 0.2729,
|
| 850 |
+
"step": 600
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 2.192377495462795,
|
| 854 |
+
"grad_norm": 1.2487846612930298,
|
| 855 |
+
"learning_rate": 5.620892262754764e-06,
|
| 856 |
+
"loss": 0.2485,
|
| 857 |
+
"step": 605
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 2.2105263157894735,
|
| 861 |
+
"grad_norm": 1.0378252267837524,
|
| 862 |
+
"learning_rate": 5.388838363431443e-06,
|
| 863 |
+
"loss": 0.2736,
|
| 864 |
+
"step": 610
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 2.2286751361161525,
|
| 868 |
+
"grad_norm": 1.0895459651947021,
|
| 869 |
+
"learning_rate": 5.1606229158836625e-06,
|
| 870 |
+
"loss": 0.2478,
|
| 871 |
+
"step": 615
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 2.246823956442831,
|
| 875 |
+
"grad_norm": 1.0891096591949463,
|
| 876 |
+
"learning_rate": 4.936337063515611e-06,
|
| 877 |
+
"loss": 0.2491,
|
| 878 |
+
"step": 620
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 2.26497277676951,
|
| 882 |
+
"grad_norm": 1.1727381944656372,
|
| 883 |
+
"learning_rate": 4.7160703803518306e-06,
|
| 884 |
+
"loss": 0.2835,
|
| 885 |
+
"step": 625
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 2.283121597096189,
|
| 889 |
+
"grad_norm": 1.1127444505691528,
|
| 890 |
+
"learning_rate": 4.499910835263625e-06,
|
| 891 |
+
"loss": 0.2647,
|
| 892 |
+
"step": 630
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 2.3012704174228675,
|
| 896 |
+
"grad_norm": 1.679655909538269,
|
| 897 |
+
"learning_rate": 4.2879447568365435e-06,
|
| 898 |
+
"loss": 0.2483,
|
| 899 |
+
"step": 635
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 2.319419237749546,
|
| 903 |
+
"grad_norm": 1.133689284324646,
|
| 904 |
+
"learning_rate": 4.080256798892981e-06,
|
| 905 |
+
"loss": 0.2568,
|
| 906 |
+
"step": 640
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 2.337568058076225,
|
| 910 |
+
"grad_norm": 1.2178751230239868,
|
| 911 |
+
"learning_rate": 3.87692990668357e-06,
|
| 912 |
+
"loss": 0.2633,
|
| 913 |
+
"step": 645
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 2.355716878402904,
|
| 917 |
+
"grad_norm": 1.3777053356170654,
|
| 918 |
+
"learning_rate": 3.6780452837610123e-06,
|
| 919 |
+
"loss": 0.2498,
|
| 920 |
+
"step": 650
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 2.3738656987295825,
|
| 924 |
+
"grad_norm": 1.2385168075561523,
|
| 925 |
+
"learning_rate": 3.4836823595494388e-06,
|
| 926 |
+
"loss": 0.2424,
|
| 927 |
+
"step": 655
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 2.392014519056261,
|
| 931 |
+
"grad_norm": 1.2520159482955933,
|
| 932 |
+
"learning_rate": 3.2939187576223232e-06,
|
| 933 |
+
"loss": 0.2503,
|
| 934 |
+
"step": 660
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 2.4101633393829403,
|
| 938 |
+
"grad_norm": 1.2936742305755615,
|
| 939 |
+
"learning_rate": 3.108830264701636e-06,
|
| 940 |
+
"loss": 0.2499,
|
| 941 |
+
"step": 665
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 2.428312159709619,
|
| 945 |
+
"grad_norm": 1.39249849319458,
|
| 946 |
+
"learning_rate": 2.9284908003905615e-06,
|
| 947 |
+
"loss": 0.2485,
|
| 948 |
+
"step": 670
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 2.4464609800362975,
|
| 952 |
+
"grad_norm": 1.1934243440628052,
|
| 953 |
+
"learning_rate": 2.752972387651897e-06,
|
| 954 |
+
"loss": 0.2419,
|
| 955 |
+
"step": 675
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 2.464609800362976,
|
| 959 |
+
"grad_norm": 1.207430362701416,
|
| 960 |
+
"learning_rate": 2.5823451240439238e-06,
|
| 961 |
+
"loss": 0.2575,
|
| 962 |
+
"step": 680
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 2.4827586206896552,
|
| 966 |
+
"grad_norm": 1.2185099124908447,
|
| 967 |
+
"learning_rate": 2.4166771537252496e-06,
|
| 968 |
+
"loss": 0.2972,
|
| 969 |
+
"step": 685
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 2.500907441016334,
|
| 973 |
+
"grad_norm": 1.516843318939209,
|
| 974 |
+
"learning_rate": 2.2560346402397747e-06,
|
| 975 |
+
"loss": 0.223,
|
| 976 |
+
"step": 690
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 2.519056261343013,
|
| 980 |
+
"grad_norm": 1.1260871887207031,
|
| 981 |
+
"learning_rate": 2.1004817400926475e-06,
|
| 982 |
+
"loss": 0.2606,
|
| 983 |
+
"step": 695
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 2.5372050816696916,
|
| 987 |
+
"grad_norm": 1.060727596282959,
|
| 988 |
+
"learning_rate": 1.950080577127832e-06,
|
| 989 |
+
"loss": 0.252,
|
| 990 |
+
"step": 700
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 2.5553539019963702,
|
| 994 |
+
"grad_norm": 1.0646460056304932,
|
| 995 |
+
"learning_rate": 1.8048912177174093e-06,
|
| 996 |
+
"loss": 0.2495,
|
| 997 |
+
"step": 705
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 2.573502722323049,
|
| 1001 |
+
"grad_norm": 1.1557650566101074,
|
| 1002 |
+
"learning_rate": 1.6649716467726272e-06,
|
| 1003 |
+
"loss": 0.2467,
|
| 1004 |
+
"step": 710
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 2.5916515426497275,
|
| 1008 |
+
"grad_norm": 1.4203789234161377,
|
| 1009 |
+
"learning_rate": 1.5303777445861855e-06,
|
| 1010 |
+
"loss": 0.2591,
|
| 1011 |
+
"step": 715
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 2.6098003629764066,
|
| 1015 |
+
"grad_norm": 1.1007254123687744,
|
| 1016 |
+
"learning_rate": 1.4011632645151078e-06,
|
| 1017 |
+
"loss": 0.2785,
|
| 1018 |
+
"step": 720
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 2.6279491833030852,
|
| 1022 |
+
"grad_norm": 1.1692945957183838,
|
| 1023 |
+
"learning_rate": 1.2773798115130114e-06,
|
| 1024 |
+
"loss": 0.2426,
|
| 1025 |
+
"step": 725
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 2.6460980036297643,
|
| 1029 |
+
"grad_norm": 1.1155871152877808,
|
| 1030 |
+
"learning_rate": 1.159076821520414e-06,
|
| 1031 |
+
"loss": 0.2391,
|
| 1032 |
+
"step": 730
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 2.664246823956443,
|
| 1036 |
+
"grad_norm": 1.0203512907028198,
|
| 1037 |
+
"learning_rate": 1.046301541721279e-06,
|
| 1038 |
+
"loss": 0.2399,
|
| 1039 |
+
"step": 735
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 2.6823956442831216,
|
| 1043 |
+
"grad_norm": 1.134413719177246,
|
| 1044 |
+
"learning_rate": 9.39099011673713e-07,
|
| 1045 |
+
"loss": 0.2328,
|
| 1046 |
+
"step": 740
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 2.7005444646098002,
|
| 1050 |
+
"grad_norm": 1.065251350402832,
|
| 1051 |
+
"learning_rate": 8.375120453223123e-07,
|
| 1052 |
+
"loss": 0.2266,
|
| 1053 |
+
"step": 745
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 2.718693284936479,
|
| 1057 |
+
"grad_norm": 1.07843816280365,
|
| 1058 |
+
"learning_rate": 7.415812138993688e-07,
|
| 1059 |
+
"loss": 0.2185,
|
| 1060 |
+
"step": 750
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 2.736842105263158,
|
| 1064 |
+
"grad_norm": 1.0000211000442505,
|
| 1065 |
+
"learning_rate": 6.513448297217616e-07,
|
| 1066 |
+
"loss": 0.2534,
|
| 1067 |
+
"step": 755
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 2.7549909255898366,
|
| 1071 |
+
"grad_norm": 1.0788525342941284,
|
| 1072 |
+
"learning_rate": 5.668389308900012e-07,
|
| 1073 |
+
"loss": 0.2638,
|
| 1074 |
+
"step": 760
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 2.7731397459165157,
|
| 1078 |
+
"grad_norm": 1.1353278160095215,
|
| 1079 |
+
"learning_rate": 4.880972668955291e-07,
|
| 1080 |
+
"loss": 0.2486,
|
| 1081 |
+
"step": 765
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 2.7912885662431943,
|
| 1085 |
+
"grad_norm": 1.0414013862609863,
|
| 1086 |
+
"learning_rate": 4.151512851420447e-07,
|
| 1087 |
+
"loss": 0.2177,
|
| 1088 |
+
"step": 770
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 2.809437386569873,
|
| 1092 |
+
"grad_norm": 1.09066903591156,
|
| 1093 |
+
"learning_rate": 3.480301183862017e-07,
|
| 1094 |
+
"loss": 0.2556,
|
| 1095 |
+
"step": 775
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 2.8275862068965516,
|
| 1099 |
+
"grad_norm": 1.1300103664398193,
|
| 1100 |
+
"learning_rate": 2.867605731027534e-07,
|
| 1101 |
+
"loss": 0.2247,
|
| 1102 |
+
"step": 780
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 2.8457350272232302,
|
| 1106 |
+
"grad_norm": 1.1673924922943115,
|
| 1107 |
+
"learning_rate": 2.3136711877871308e-07,
|
| 1108 |
+
"loss": 0.2462,
|
| 1109 |
+
"step": 785
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 2.8638838475499093,
|
| 1113 |
+
"grad_norm": 1.2144715785980225,
|
| 1114 |
+
"learning_rate": 1.8187187814088456e-07,
|
| 1115 |
+
"loss": 0.2586,
|
| 1116 |
+
"step": 790
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 2.882032667876588,
|
| 1120 |
+
"grad_norm": 1.1757094860076904,
|
| 1121 |
+
"learning_rate": 1.382946183205913e-07,
|
| 1122 |
+
"loss": 0.2349,
|
| 1123 |
+
"step": 795
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 2.900181488203267,
|
| 1127 |
+
"grad_norm": 1.1305650472640991,
|
| 1128 |
+
"learning_rate": 1.0065274295921301e-07,
|
| 1129 |
+
"loss": 0.263,
|
| 1130 |
+
"step": 800
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 2.9183303085299457,
|
| 1134 |
+
"grad_norm": 1.1334455013275146,
|
| 1135 |
+
"learning_rate": 6.896128525760381e-08,
|
| 1136 |
+
"loss": 0.2458,
|
| 1137 |
+
"step": 805
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 2.9364791288566243,
|
| 1141 |
+
"grad_norm": 1.1247256994247437,
|
| 1142 |
+
"learning_rate": 4.3232901972216634e-08,
|
| 1143 |
+
"loss": 0.2387,
|
| 1144 |
+
"step": 810
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 2.954627949183303,
|
| 1148 |
+
"grad_norm": 1.1602647304534912,
|
| 1149 |
+
"learning_rate": 2.347786836032484e-08,
|
| 1150 |
+
"loss": 0.246,
|
| 1151 |
+
"step": 815
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 2.9727767695099816,
|
| 1155 |
+
"grad_norm": 1.250677227973938,
|
| 1156 |
+
"learning_rate": 9.704074076344882e-09,
|
| 1157 |
+
"loss": 0.2455,
|
| 1158 |
+
"step": 820
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 2.9909255898366607,
|
| 1162 |
+
"grad_norm": 1.039875864982605,
|
| 1163 |
+
"learning_rate": 1.917020020906768e-09,
|
| 1164 |
+
"loss": 0.2292,
|
| 1165 |
+
"step": 825
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 3.0,
|
| 1169 |
+
"step": 828,
|
| 1170 |
+
"total_flos": 1.1532968769638892e+18,
|
| 1171 |
+
"train_loss": 0.5211183394930788,
|
| 1172 |
+
"train_runtime": 540.0579,
|
| 1173 |
+
"train_samples_per_second": 48.961,
|
| 1174 |
+
"train_steps_per_second": 1.533
|
| 1175 |
+
}
|
| 1176 |
+
],
|
| 1177 |
+
"logging_steps": 5,
|
| 1178 |
+
"max_steps": 828,
|
| 1179 |
+
"num_input_tokens_seen": 0,
|
| 1180 |
+
"num_train_epochs": 3,
|
| 1181 |
+
"save_steps": 2000,
|
| 1182 |
+
"stateful_callbacks": {
|
| 1183 |
+
"TrainerControl": {
|
| 1184 |
+
"args": {
|
| 1185 |
+
"should_epoch_stop": false,
|
| 1186 |
+
"should_evaluate": false,
|
| 1187 |
+
"should_log": false,
|
| 1188 |
+
"should_save": false,
|
| 1189 |
+
"should_training_stop": false
|
| 1190 |
+
},
|
| 1191 |
+
"attributes": {}
|
| 1192 |
+
}
|
| 1193 |
+
},
|
| 1194 |
+
"total_flos": 1.1532968769638892e+18,
|
| 1195 |
+
"train_batch_size": 2,
|
| 1196 |
+
"trial_name": null,
|
| 1197 |
+
"trial_params": null
|
| 1198 |
+
}
|
46_128_e3_3e-5/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4f285f6d99f34f84b2b27af9fa3732523ac47f377f45087df7f9b77025bfe1f6
|
| 3 |
+
size 8145
|
46_128_e3_3e-5/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|