End of training
Browse files- README.md +72 -0
- adapter_config.json +34 -0
- adapter_model.safetensors +3 -0
- all_results.json +16 -0
- config.json +33 -0
- eval_results.json +9 -0
- special_tokens_map.json +23 -0
- tokenizer.json +0 -0
- tokenizer_config.json +194 -0
- train_results.json +10 -0
- trainer_state.json +3575 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,72 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: deepseek-ai/deepseek-coder-1.3b-base
|
| 3 |
+
datasets:
|
| 4 |
+
- Sohaib9920/code_tests
|
| 5 |
+
library_name: peft
|
| 6 |
+
license: other
|
| 7 |
+
tags:
|
| 8 |
+
- trl
|
| 9 |
+
- sft
|
| 10 |
+
- generated_from_trainer
|
| 11 |
+
model-index:
|
| 12 |
+
- name: sft_00
|
| 13 |
+
results: []
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 17 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 18 |
+
|
| 19 |
+
# sft_00
|
| 20 |
+
|
| 21 |
+
This model is a fine-tuned version of [deepseek-ai/deepseek-coder-1.3b-base](https://huggingface.co/deepseek-ai/deepseek-coder-1.3b-base) on the Sohaib9920/code_tests dataset.
|
| 22 |
+
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 0.2291
|
| 24 |
+
|
| 25 |
+
## Model description
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Intended uses & limitations
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training and evaluation data
|
| 34 |
+
|
| 35 |
+
More information needed
|
| 36 |
+
|
| 37 |
+
## Training procedure
|
| 38 |
+
|
| 39 |
+
### Training hyperparameters
|
| 40 |
+
|
| 41 |
+
The following hyperparameters were used during training:
|
| 42 |
+
- learning_rate: 0.0005
|
| 43 |
+
- train_batch_size: 2
|
| 44 |
+
- eval_batch_size: 4
|
| 45 |
+
- seed: 42
|
| 46 |
+
- distributed_type: multi-GPU
|
| 47 |
+
- num_devices: 2
|
| 48 |
+
- gradient_accumulation_steps: 8
|
| 49 |
+
- total_train_batch_size: 32
|
| 50 |
+
- total_eval_batch_size: 8
|
| 51 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 52 |
+
- lr_scheduler_type: cosine
|
| 53 |
+
- num_epochs: 2
|
| 54 |
+
- mixed_precision_training: Native AMP
|
| 55 |
+
|
| 56 |
+
### Training results
|
| 57 |
+
|
| 58 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 59 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 60 |
+
| 0.4228 | 0.4996 | 623 | 0.3337 |
|
| 61 |
+
| 0.3632 | 0.9993 | 1246 | 0.2687 |
|
| 62 |
+
| 0.264 | 1.4989 | 1869 | 0.2418 |
|
| 63 |
+
| 0.2535 | 1.9986 | 2492 | 0.2291 |
|
| 64 |
+
|
| 65 |
+
|
| 66 |
+
### Framework versions
|
| 67 |
+
|
| 68 |
+
- PEFT 0.12.0
|
| 69 |
+
- Transformers 4.44.2
|
| 70 |
+
- Pytorch 2.4.0
|
| 71 |
+
- Datasets 2.21.0
|
| 72 |
+
- Tokenizers 0.19.1
|
adapter_config.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "deepseek-ai/deepseek-coder-1.3b-base",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layer_replication": null,
|
| 10 |
+
"layers_pattern": null,
|
| 11 |
+
"layers_to_transform": null,
|
| 12 |
+
"loftq_config": {},
|
| 13 |
+
"lora_alpha": 64,
|
| 14 |
+
"lora_dropout": 0.05,
|
| 15 |
+
"megatron_config": null,
|
| 16 |
+
"megatron_core": "megatron.core",
|
| 17 |
+
"modules_to_save": null,
|
| 18 |
+
"peft_type": "LORA",
|
| 19 |
+
"r": 32,
|
| 20 |
+
"rank_pattern": {},
|
| 21 |
+
"revision": null,
|
| 22 |
+
"target_modules": [
|
| 23 |
+
"k_proj",
|
| 24 |
+
"v_proj",
|
| 25 |
+
"gate_proj",
|
| 26 |
+
"q_proj",
|
| 27 |
+
"o_proj",
|
| 28 |
+
"down_proj",
|
| 29 |
+
"up_proj"
|
| 30 |
+
],
|
| 31 |
+
"task_type": "CAUSAL_LM",
|
| 32 |
+
"use_dora": false,
|
| 33 |
+
"use_rslora": false
|
| 34 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:99b1f6e2f3a3249bd7eb172e5b3179923ebbd7570cd82121a5128f0ec68c9524
|
| 3 |
+
size 119975656
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.9985964912280703,
|
| 3 |
+
"eval_loss": 0.22910664975643158,
|
| 4 |
+
"eval_runtime": 8.378,
|
| 5 |
+
"eval_samples": 100,
|
| 6 |
+
"eval_samples_per_second": 6.565,
|
| 7 |
+
"eval_steps_per_second": 0.836,
|
| 8 |
+
"perplexity": 1.2574761413829163,
|
| 9 |
+
"raw_train_examples": 78434,
|
| 10 |
+
"total_flos": 8.025299762167153e+17,
|
| 11 |
+
"train_examples": 39900,
|
| 12 |
+
"train_loss": 0.37913749282088366,
|
| 13 |
+
"train_runtime": 28381.1385,
|
| 14 |
+
"train_samples_per_second": 2.812,
|
| 15 |
+
"train_steps_per_second": 0.088
|
| 16 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "deepseek-ai/deepseek-coder-1.3b-base",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 32013,
|
| 9 |
+
"eos_token_id": 32014,
|
| 10 |
+
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 2048,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 5504,
|
| 14 |
+
"max_position_embeddings": 16384,
|
| 15 |
+
"mlp_bias": false,
|
| 16 |
+
"model_type": "llama",
|
| 17 |
+
"num_attention_heads": 16,
|
| 18 |
+
"num_hidden_layers": 24,
|
| 19 |
+
"num_key_value_heads": 16,
|
| 20 |
+
"pretraining_tp": 1,
|
| 21 |
+
"rms_norm_eps": 1e-06,
|
| 22 |
+
"rope_scaling": {
|
| 23 |
+
"factor": 4.0,
|
| 24 |
+
"rope_type": "linear",
|
| 25 |
+
"type": "linear"
|
| 26 |
+
},
|
| 27 |
+
"rope_theta": 100000,
|
| 28 |
+
"tie_word_embeddings": false,
|
| 29 |
+
"torch_dtype": "float16",
|
| 30 |
+
"transformers_version": "4.44.2",
|
| 31 |
+
"use_cache": true,
|
| 32 |
+
"vocab_size": 32256
|
| 33 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.9985964912280703,
|
| 3 |
+
"eval_loss": 0.22910664975643158,
|
| 4 |
+
"eval_runtime": 8.378,
|
| 5 |
+
"eval_samples": 100,
|
| 6 |
+
"eval_samples_per_second": 6.565,
|
| 7 |
+
"eval_steps_per_second": 0.836,
|
| 8 |
+
"perplexity": 1.2574761413829163
|
| 9 |
+
}
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<｜begin▁of▁sentence｜>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<｜end▁of▁sentence｜>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "<｜end▁of▁sentence｜>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": true,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
}
|
| 23 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,194 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": null,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"32000": {
|
| 7 |
+
"content": "õ",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": true,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": false
|
| 13 |
+
},
|
| 14 |
+
"32001": {
|
| 15 |
+
"content": "÷",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": true,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": false
|
| 21 |
+
},
|
| 22 |
+
"32002": {
|
| 23 |
+
"content": "Á",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": true,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": false
|
| 29 |
+
},
|
| 30 |
+
"32003": {
|
| 31 |
+
"content": "ý",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": true,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false,
|
| 36 |
+
"special": false
|
| 37 |
+
},
|
| 38 |
+
"32004": {
|
| 39 |
+
"content": "À",
|
| 40 |
+
"lstrip": false,
|
| 41 |
+
"normalized": true,
|
| 42 |
+
"rstrip": false,
|
| 43 |
+
"single_word": false,
|
| 44 |
+
"special": false
|
| 45 |
+
},
|
| 46 |
+
"32005": {
|
| 47 |
+
"content": "ÿ",
|
| 48 |
+
"lstrip": false,
|
| 49 |
+
"normalized": true,
|
| 50 |
+
"rstrip": false,
|
| 51 |
+
"single_word": false,
|
| 52 |
+
"special": false
|
| 53 |
+
},
|
| 54 |
+
"32006": {
|
| 55 |
+
"content": "ø",
|
| 56 |
+
"lstrip": false,
|
| 57 |
+
"normalized": true,
|
| 58 |
+
"rstrip": false,
|
| 59 |
+
"single_word": false,
|
| 60 |
+
"special": false
|
| 61 |
+
},
|
| 62 |
+
"32007": {
|
| 63 |
+
"content": "ú",
|
| 64 |
+
"lstrip": false,
|
| 65 |
+
"normalized": true,
|
| 66 |
+
"rstrip": false,
|
| 67 |
+
"single_word": false,
|
| 68 |
+
"special": false
|
| 69 |
+
},
|
| 70 |
+
"32008": {
|
| 71 |
+
"content": "þ",
|
| 72 |
+
"lstrip": false,
|
| 73 |
+
"normalized": true,
|
| 74 |
+
"rstrip": false,
|
| 75 |
+
"single_word": false,
|
| 76 |
+
"special": false
|
| 77 |
+
},
|
| 78 |
+
"32009": {
|
| 79 |
+
"content": "ü",
|
| 80 |
+
"lstrip": false,
|
| 81 |
+
"normalized": true,
|
| 82 |
+
"rstrip": false,
|
| 83 |
+
"single_word": false,
|
| 84 |
+
"special": false
|
| 85 |
+
},
|
| 86 |
+
"32010": {
|
| 87 |
+
"content": "ù",
|
| 88 |
+
"lstrip": false,
|
| 89 |
+
"normalized": true,
|
| 90 |
+
"rstrip": false,
|
| 91 |
+
"single_word": false,
|
| 92 |
+
"special": false
|
| 93 |
+
},
|
| 94 |
+
"32011": {
|
| 95 |
+
"content": "ö",
|
| 96 |
+
"lstrip": false,
|
| 97 |
+
"normalized": true,
|
| 98 |
+
"rstrip": false,
|
| 99 |
+
"single_word": false,
|
| 100 |
+
"special": false
|
| 101 |
+
},
|
| 102 |
+
"32012": {
|
| 103 |
+
"content": "û",
|
| 104 |
+
"lstrip": false,
|
| 105 |
+
"normalized": true,
|
| 106 |
+
"rstrip": false,
|
| 107 |
+
"single_word": false,
|
| 108 |
+
"special": false
|
| 109 |
+
},
|
| 110 |
+
"32013": {
|
| 111 |
+
"content": "<｜begin▁of▁sentence｜>",
|
| 112 |
+
"lstrip": false,
|
| 113 |
+
"normalized": true,
|
| 114 |
+
"rstrip": false,
|
| 115 |
+
"single_word": false,
|
| 116 |
+
"special": true
|
| 117 |
+
},
|
| 118 |
+
"32014": {
|
| 119 |
+
"content": "<｜end▁of▁sentence｜>",
|
| 120 |
+
"lstrip": false,
|
| 121 |
+
"normalized": true,
|
| 122 |
+
"rstrip": false,
|
| 123 |
+
"single_word": false,
|
| 124 |
+
"special": true
|
| 125 |
+
},
|
| 126 |
+
"32015": {
|
| 127 |
+
"content": "<｜fim▁hole｜>",
|
| 128 |
+
"lstrip": false,
|
| 129 |
+
"normalized": true,
|
| 130 |
+
"rstrip": false,
|
| 131 |
+
"single_word": false,
|
| 132 |
+
"special": false
|
| 133 |
+
},
|
| 134 |
+
"32016": {
|
| 135 |
+
"content": "<｜fim▁begin｜>",
|
| 136 |
+
"lstrip": false,
|
| 137 |
+
"normalized": true,
|
| 138 |
+
"rstrip": false,
|
| 139 |
+
"single_word": false,
|
| 140 |
+
"special": false
|
| 141 |
+
},
|
| 142 |
+
"32017": {
|
| 143 |
+
"content": "<｜fim▁end｜>",
|
| 144 |
+
"lstrip": false,
|
| 145 |
+
"normalized": true,
|
| 146 |
+
"rstrip": false,
|
| 147 |
+
"single_word": false,
|
| 148 |
+
"special": false
|
| 149 |
+
},
|
| 150 |
+
"32018": {
|
| 151 |
+
"content": "<pad>",
|
| 152 |
+
"lstrip": false,
|
| 153 |
+
"normalized": true,
|
| 154 |
+
"rstrip": false,
|
| 155 |
+
"single_word": false,
|
| 156 |
+
"special": false
|
| 157 |
+
},
|
| 158 |
+
"32019": {
|
| 159 |
+
"content": "<|User|>",
|
| 160 |
+
"lstrip": false,
|
| 161 |
+
"normalized": true,
|
| 162 |
+
"rstrip": false,
|
| 163 |
+
"single_word": false,
|
| 164 |
+
"special": false
|
| 165 |
+
},
|
| 166 |
+
"32020": {
|
| 167 |
+
"content": "<|Assistant|>",
|
| 168 |
+
"lstrip": false,
|
| 169 |
+
"normalized": true,
|
| 170 |
+
"rstrip": false,
|
| 171 |
+
"single_word": false,
|
| 172 |
+
"special": false
|
| 173 |
+
},
|
| 174 |
+
"32021": {
|
| 175 |
+
"content": "<|EOT|>",
|
| 176 |
+
"lstrip": false,
|
| 177 |
+
"normalized": true,
|
| 178 |
+
"rstrip": false,
|
| 179 |
+
"single_word": false,
|
| 180 |
+
"special": false
|
| 181 |
+
}
|
| 182 |
+
},
|
| 183 |
+
"bos_token": "<｜begin▁of▁sentence｜>",
|
| 184 |
+
"chat_template": "{% for message in messages %}{% if (message['role'] == 'system')%}{{ message['content'] + '\n\n' }}{% elif (message['role'] == 'user')%}{{ 'Generate a unit test case for the following Java method:\n' + message['content'] + '\n\n' }}{% elif (message['role'] == 'assistant')%}{{ 'Unit test case:\n' + message['content'] + '\n\n' }}{% endif %}{% if loop.last and message['role'] == 'user' and add_generation_prompt %}{{ 'Unit test case:\n' }}{% endif %}{% endfor %}",
|
| 185 |
+
"clean_up_tokenization_spaces": false,
|
| 186 |
+
"eos_token": "<｜end▁of▁sentence｜>",
|
| 187 |
+
"legacy": true,
|
| 188 |
+
"model_max_length": 16384,
|
| 189 |
+
"pad_token": "<｜end▁of▁sentence｜>",
|
| 190 |
+
"sp_model_kwargs": {},
|
| 191 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 192 |
+
"unk_token": null,
|
| 193 |
+
"use_default_system_prompt": false
|
| 194 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.9985964912280703,
|
| 3 |
+
"raw_train_examples": 78434,
|
| 4 |
+
"total_flos": 8.025299762167153e+17,
|
| 5 |
+
"train_examples": 39900,
|
| 6 |
+
"train_loss": 0.37913749282088366,
|
| 7 |
+
"train_runtime": 28381.1385,
|
| 8 |
+
"train_samples_per_second": 2.812,
|
| 9 |
+
"train_steps_per_second": 0.088
|
| 10 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,3575 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.9985964912280703,
|
| 5 |
+
"eval_steps": 623,
|
| 6 |
+
"global_step": 2492,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0008020050125313284,
|
| 13 |
+
"grad_norm": 0.09629630297422409,
|
| 14 |
+
"learning_rate": 0.000499999801338539,
|
| 15 |
+
"loss": 0.8462,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0040100250626566416,
|
| 20 |
+
"grad_norm": 0.09335128962993622,
|
| 21 |
+
"learning_rate": 0.0004999950334792631,
|
| 22 |
+
"loss": 0.9167,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.008020050125313283,
|
| 27 |
+
"grad_norm": 0.0814199447631836,
|
| 28 |
+
"learning_rate": 0.0004999801341143832,
|
| 29 |
+
"loss": 0.8982,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.012030075187969926,
|
| 34 |
+
"grad_norm": 0.0830962136387825,
|
| 35 |
+
"learning_rate": 0.0004999553024973442,
|
| 36 |
+
"loss": 0.8568,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.016040100250626566,
|
| 41 |
+
"grad_norm": 0.07535357028245926,
|
| 42 |
+
"learning_rate": 0.00049992053961476,
|
| 43 |
+
"loss": 0.8423,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.020050125313283207,
|
| 48 |
+
"grad_norm": 0.08100854605436325,
|
| 49 |
+
"learning_rate": 0.0004998758468478354,
|
| 50 |
+
"loss": 0.822,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.02406015037593985,
|
| 55 |
+
"grad_norm": 0.0720616802573204,
|
| 56 |
+
"learning_rate": 0.0004998212259723108,
|
| 57 |
+
"loss": 0.801,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.028070175438596492,
|
| 62 |
+
"grad_norm": 0.07352820038795471,
|
| 63 |
+
"learning_rate": 0.0004997566791583916,
|
| 64 |
+
"loss": 0.819,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.03208020050125313,
|
| 69 |
+
"grad_norm": 0.07747522741556168,
|
| 70 |
+
"learning_rate": 0.0004996822089706628,
|
| 71 |
+
"loss": 0.7917,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.03609022556390978,
|
| 76 |
+
"grad_norm": 0.0775732770562172,
|
| 77 |
+
"learning_rate": 0.0004995978183679864,
|
| 78 |
+
"loss": 0.8007,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.040100250626566414,
|
| 83 |
+
"grad_norm": 0.08353012800216675,
|
| 84 |
+
"learning_rate": 0.0004995035107033833,
|
| 85 |
+
"loss": 0.7939,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.04411027568922306,
|
| 90 |
+
"grad_norm": 0.09015482664108276,
|
| 91 |
+
"learning_rate": 0.0004993992897239018,
|
| 92 |
+
"loss": 0.7844,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.0481203007518797,
|
| 97 |
+
"grad_norm": 0.08346273750066757,
|
| 98 |
+
"learning_rate": 0.0004992851595704668,
|
| 99 |
+
"loss": 0.7805,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.05213032581453634,
|
| 104 |
+
"grad_norm": 0.08374553918838501,
|
| 105 |
+
"learning_rate": 0.0004991611247777166,
|
| 106 |
+
"loss": 0.7758,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.056140350877192984,
|
| 111 |
+
"grad_norm": 0.0855209082365036,
|
| 112 |
+
"learning_rate": 0.0004990271902738223,
|
| 113 |
+
"loss": 0.7799,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.06015037593984962,
|
| 118 |
+
"grad_norm": 0.08624681085348129,
|
| 119 |
+
"learning_rate": 0.0004988833613802917,
|
| 120 |
+
"loss": 0.7461,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.06416040100250626,
|
| 125 |
+
"grad_norm": 0.08796326816082001,
|
| 126 |
+
"learning_rate": 0.0004987296438117581,
|
| 127 |
+
"loss": 0.7181,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.06817042606516291,
|
| 132 |
+
"grad_norm": 0.08574078977108002,
|
| 133 |
+
"learning_rate": 0.0004985660436757538,
|
| 134 |
+
"loss": 0.7209,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.07218045112781955,
|
| 139 |
+
"grad_norm": 0.09403970092535019,
|
| 140 |
+
"learning_rate": 0.0004983925674724662,
|
| 141 |
+
"loss": 0.7297,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.0761904761904762,
|
| 146 |
+
"grad_norm": 0.1051216572523117,
|
| 147 |
+
"learning_rate": 0.0004982092220944808,
|
| 148 |
+
"loss": 0.7077,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.08020050125313283,
|
| 153 |
+
"grad_norm": 0.10198145359754562,
|
| 154 |
+
"learning_rate": 0.0004980160148265066,
|
| 155 |
+
"loss": 0.7369,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.08421052631578947,
|
| 160 |
+
"grad_norm": 0.10444864630699158,
|
| 161 |
+
"learning_rate": 0.0004978129533450866,
|
| 162 |
+
"loss": 0.7044,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.08822055137844612,
|
| 167 |
+
"grad_norm": 0.09810823947191238,
|
| 168 |
+
"learning_rate": 0.0004976000457182935,
|
| 169 |
+
"loss": 0.7055,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.09223057644110276,
|
| 174 |
+
"grad_norm": 0.10619106888771057,
|
| 175 |
+
"learning_rate": 0.0004973773004054081,
|
| 176 |
+
"loss": 0.6951,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.0962406015037594,
|
| 181 |
+
"grad_norm": 0.11827559024095535,
|
| 182 |
+
"learning_rate": 0.0004971447262565846,
|
| 183 |
+
"loss": 0.7285,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.10025062656641603,
|
| 188 |
+
"grad_norm": 0.09872109442949295,
|
| 189 |
+
"learning_rate": 0.0004969023325124974,
|
| 190 |
+
"loss": 0.7065,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.10426065162907268,
|
| 195 |
+
"grad_norm": 0.11186440289020538,
|
| 196 |
+
"learning_rate": 0.0004966501288039749,
|
| 197 |
+
"loss": 0.6663,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.10827067669172932,
|
| 202 |
+
"grad_norm": 0.10218355059623718,
|
| 203 |
+
"learning_rate": 0.0004963881251516169,
|
| 204 |
+
"loss": 0.6763,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.11228070175438597,
|
| 209 |
+
"grad_norm": 0.11396008729934692,
|
| 210 |
+
"learning_rate": 0.0004961163319653958,
|
| 211 |
+
"loss": 0.6704,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.11629072681704261,
|
| 216 |
+
"grad_norm": 0.10969991981983185,
|
| 217 |
+
"learning_rate": 0.0004958347600442437,
|
| 218 |
+
"loss": 0.6779,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.12030075187969924,
|
| 223 |
+
"grad_norm": 0.11111714690923691,
|
| 224 |
+
"learning_rate": 0.0004955434205756226,
|
| 225 |
+
"loss": 0.6835,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.12431077694235589,
|
| 230 |
+
"grad_norm": 0.11582209914922714,
|
| 231 |
+
"learning_rate": 0.000495242325135081,
|
| 232 |
+
"loss": 0.6966,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.12832080200501253,
|
| 237 |
+
"grad_norm": 0.11664141714572906,
|
| 238 |
+
"learning_rate": 0.0004949314856857926,
|
| 239 |
+
"loss": 0.641,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.13233082706766916,
|
| 244 |
+
"grad_norm": 0.11299869418144226,
|
| 245 |
+
"learning_rate": 0.000494610914578082,
|
| 246 |
+
"loss": 0.6349,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.13634085213032582,
|
| 251 |
+
"grad_norm": 0.10604275017976761,
|
| 252 |
+
"learning_rate": 0.0004942806245489338,
|
| 253 |
+
"loss": 0.6498,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.14035087719298245,
|
| 258 |
+
"grad_norm": 0.11253874003887177,
|
| 259 |
+
"learning_rate": 0.000493940628721486,
|
| 260 |
+
"loss": 0.6536,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.1443609022556391,
|
| 265 |
+
"grad_norm": 0.12596091628074646,
|
| 266 |
+
"learning_rate": 0.0004935909406045095,
|
| 267 |
+
"loss": 0.6348,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.14837092731829574,
|
| 272 |
+
"grad_norm": 0.12171223014593124,
|
| 273 |
+
"learning_rate": 0.0004932315740918703,
|
| 274 |
+
"loss": 0.64,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.1523809523809524,
|
| 279 |
+
"grad_norm": 0.12628182768821716,
|
| 280 |
+
"learning_rate": 0.0004928625434619785,
|
| 281 |
+
"loss": 0.6225,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.15639097744360902,
|
| 286 |
+
"grad_norm": 0.12631121277809143,
|
| 287 |
+
"learning_rate": 0.0004924838633772201,
|
| 288 |
+
"loss": 0.6197,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.16040100250626566,
|
| 293 |
+
"grad_norm": 0.12037768959999084,
|
| 294 |
+
"learning_rate": 0.0004920955488833753,
|
| 295 |
+
"loss": 0.6201,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.1644110275689223,
|
| 300 |
+
"grad_norm": 0.11302505433559418,
|
| 301 |
+
"learning_rate": 0.0004916976154090198,
|
| 302 |
+
"loss": 0.5972,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.16842105263157894,
|
| 307 |
+
"grad_norm": 0.15172865986824036,
|
| 308 |
+
"learning_rate": 0.0004912900787649124,
|
| 309 |
+
"loss": 0.6322,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.1724310776942356,
|
| 314 |
+
"grad_norm": 0.12461714446544647,
|
| 315 |
+
"learning_rate": 0.0004908729551433668,
|
| 316 |
+
"loss": 0.5681,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.17644110275689223,
|
| 321 |
+
"grad_norm": 0.11686723679304123,
|
| 322 |
+
"learning_rate": 0.0004904462611176079,
|
| 323 |
+
"loss": 0.6112,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.18045112781954886,
|
| 328 |
+
"grad_norm": 0.12082964181900024,
|
| 329 |
+
"learning_rate": 0.0004900100136411134,
|
| 330 |
+
"loss": 0.59,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.18446115288220552,
|
| 335 |
+
"grad_norm": 0.14786957204341888,
|
| 336 |
+
"learning_rate": 0.0004895642300469405,
|
| 337 |
+
"loss": 0.6008,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.18847117794486215,
|
| 342 |
+
"grad_norm": 0.11940759420394897,
|
| 343 |
+
"learning_rate": 0.0004891089280470369,
|
| 344 |
+
"loss": 0.5781,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.1924812030075188,
|
| 349 |
+
"grad_norm": 0.12067493796348572,
|
| 350 |
+
"learning_rate": 0.0004886441257315373,
|
| 351 |
+
"loss": 0.5963,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.19649122807017544,
|
| 356 |
+
"grad_norm": 0.12630167603492737,
|
| 357 |
+
"learning_rate": 0.0004881698415680442,
|
| 358 |
+
"loss": 0.5767,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.20050125313283207,
|
| 363 |
+
"grad_norm": 0.12695525586605072,
|
| 364 |
+
"learning_rate": 0.00048768609440089474,
|
| 365 |
+
"loss": 0.6039,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.20451127819548873,
|
| 370 |
+
"grad_norm": 0.13445709645748138,
|
| 371 |
+
"learning_rate": 0.00048719290345041166,
|
| 372 |
+
"loss": 0.6087,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.20852130325814536,
|
| 377 |
+
"grad_norm": 0.13432475924491882,
|
| 378 |
+
"learning_rate": 0.0004866902883121397,
|
| 379 |
+
"loss": 0.5491,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.21253132832080202,
|
| 384 |
+
"grad_norm": 0.15156829357147217,
|
| 385 |
+
"learning_rate": 0.0004861782689560667,
|
| 386 |
+
"loss": 0.5492,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.21654135338345865,
|
| 391 |
+
"grad_norm": 0.13093040883541107,
|
| 392 |
+
"learning_rate": 0.0004856568657258308,
|
| 393 |
+
"loss": 0.6045,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.22055137844611528,
|
| 398 |
+
"grad_norm": 0.1290189027786255,
|
| 399 |
+
"learning_rate": 0.00048512609933791165,
|
| 400 |
+
"loss": 0.5433,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.22456140350877193,
|
| 405 |
+
"grad_norm": 0.13374298810958862,
|
| 406 |
+
"learning_rate": 0.00048458599088080736,
|
| 407 |
+
"loss": 0.5575,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.22857142857142856,
|
| 412 |
+
"grad_norm": 0.1266585886478424,
|
| 413 |
+
"learning_rate": 0.00048403656181419686,
|
| 414 |
+
"loss": 0.5474,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.23258145363408522,
|
| 419 |
+
"grad_norm": 0.13824708759784698,
|
| 420 |
+
"learning_rate": 0.00048347783396808687,
|
| 421 |
+
"loss": 0.5465,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.23659147869674185,
|
| 426 |
+
"grad_norm": 0.12988659739494324,
|
| 427 |
+
"learning_rate": 0.00048290982954194485,
|
| 428 |
+
"loss": 0.5605,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.24060150375939848,
|
| 433 |
+
"grad_norm": 0.13609863817691803,
|
| 434 |
+
"learning_rate": 0.0004823325711038169,
|
| 435 |
+
"loss": 0.5812,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.24461152882205514,
|
| 440 |
+
"grad_norm": 0.14271323382854462,
|
| 441 |
+
"learning_rate": 0.00048174608158943114,
|
| 442 |
+
"loss": 0.5485,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.24862155388471177,
|
| 447 |
+
"grad_norm": 0.13438940048217773,
|
| 448 |
+
"learning_rate": 0.0004811503843012861,
|
| 449 |
+
"loss": 0.5325,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.25263157894736843,
|
| 454 |
+
"grad_norm": 0.138273686170578,
|
| 455 |
+
"learning_rate": 0.0004805455029077255,
|
| 456 |
+
"loss": 0.5283,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.25664160401002506,
|
| 461 |
+
"grad_norm": 0.15901249647140503,
|
| 462 |
+
"learning_rate": 0.0004799314614419969,
|
| 463 |
+
"loss": 0.5717,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.2606516290726817,
|
| 468 |
+
"grad_norm": 0.14647658169269562,
|
| 469 |
+
"learning_rate": 0.0004793082843012979,
|
| 470 |
+
"loss": 0.5378,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.2646616541353383,
|
| 475 |
+
"grad_norm": 0.1300746202468872,
|
| 476 |
+
"learning_rate": 0.0004786759962458059,
|
| 477 |
+
"loss": 0.5183,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.268671679197995,
|
| 482 |
+
"grad_norm": 0.13711751997470856,
|
| 483 |
+
"learning_rate": 0.000478034622397695,
|
| 484 |
+
"loss": 0.5442,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.27268170426065164,
|
| 489 |
+
"grad_norm": 0.150499626994133,
|
| 490 |
+
"learning_rate": 0.0004773841882401372,
|
| 491 |
+
"loss": 0.524,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.27669172932330827,
|
| 496 |
+
"grad_norm": 0.13912473618984222,
|
| 497 |
+
"learning_rate": 0.00047672471961629037,
|
| 498 |
+
"loss": 0.5412,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.2807017543859649,
|
| 503 |
+
"grad_norm": 0.1455031782388687,
|
| 504 |
+
"learning_rate": 0.0004760562427282712,
|
| 505 |
+
"loss": 0.5321,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.2847117794486215,
|
| 510 |
+
"grad_norm": 0.12485536932945251,
|
| 511 |
+
"learning_rate": 0.0004753787841361145,
|
| 512 |
+
"loss": 0.5113,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.2887218045112782,
|
| 517 |
+
"grad_norm": 0.13250643014907837,
|
| 518 |
+
"learning_rate": 0.00047469237075671724,
|
| 519 |
+
"loss": 0.5402,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.29273182957393484,
|
| 524 |
+
"grad_norm": 0.13680434226989746,
|
| 525 |
+
"learning_rate": 0.00047399702986276984,
|
| 526 |
+
"loss": 0.5751,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.2967418546365915,
|
| 531 |
+
"grad_norm": 0.14655643701553345,
|
| 532 |
+
"learning_rate": 0.0004732927890816719,
|
| 533 |
+
"loss": 0.5371,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.3007518796992481,
|
| 538 |
+
"grad_norm": 0.14064128696918488,
|
| 539 |
+
"learning_rate": 0.00047257967639443513,
|
| 540 |
+
"loss": 0.531,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.3047619047619048,
|
| 545 |
+
"grad_norm": 0.14199021458625793,
|
| 546 |
+
"learning_rate": 0.00047185772013457096,
|
| 547 |
+
"loss": 0.4934,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.3087719298245614,
|
| 552 |
+
"grad_norm": 0.14177988469600677,
|
| 553 |
+
"learning_rate": 0.00047112694898696533,
|
| 554 |
+
"loss": 0.5249,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.31278195488721805,
|
| 559 |
+
"grad_norm": 0.14135052263736725,
|
| 560 |
+
"learning_rate": 0.00047038739198673876,
|
| 561 |
+
"loss": 0.5272,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.3167919799498747,
|
| 566 |
+
"grad_norm": 0.12976044416427612,
|
| 567 |
+
"learning_rate": 0.00046963907851809263,
|
| 568 |
+
"loss": 0.5263,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.3208020050125313,
|
| 573 |
+
"grad_norm": 0.15720346570014954,
|
| 574 |
+
"learning_rate": 0.0004688820383131418,
|
| 575 |
+
"loss": 0.4858,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.324812030075188,
|
| 580 |
+
"grad_norm": 0.14797399938106537,
|
| 581 |
+
"learning_rate": 0.0004681163014507334,
|
| 582 |
+
"loss": 0.5201,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.3288220551378446,
|
| 587 |
+
"grad_norm": 0.15095673501491547,
|
| 588 |
+
"learning_rate": 0.0004673418983552513,
|
| 589 |
+
"loss": 0.5366,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.33283208020050126,
|
| 594 |
+
"grad_norm": 0.16217216849327087,
|
| 595 |
+
"learning_rate": 0.00046655885979540783,
|
| 596 |
+
"loss": 0.4836,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.3368421052631579,
|
| 601 |
+
"grad_norm": 0.14743471145629883,
|
| 602 |
+
"learning_rate": 0.00046576721688302105,
|
| 603 |
+
"loss": 0.5067,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.3408521303258145,
|
| 608 |
+
"grad_norm": 0.15485595166683197,
|
| 609 |
+
"learning_rate": 0.00046496700107177835,
|
| 610 |
+
"loss": 0.4614,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.3448621553884712,
|
| 615 |
+
"grad_norm": 0.15458199381828308,
|
| 616 |
+
"learning_rate": 0.0004641582441559872,
|
| 617 |
+
"loss": 0.4897,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.34887218045112783,
|
| 622 |
+
"grad_norm": 0.13934871554374695,
|
| 623 |
+
"learning_rate": 0.00046334097826931144,
|
| 624 |
+
"loss": 0.522,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.35288220551378446,
|
| 629 |
+
"grad_norm": 0.1473606377840042,
|
| 630 |
+
"learning_rate": 0.00046251523588349485,
|
| 631 |
+
"loss": 0.4851,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.3568922305764411,
|
| 636 |
+
"grad_norm": 0.14684675633907318,
|
| 637 |
+
"learning_rate": 0.00046168104980707104,
|
| 638 |
+
"loss": 0.4888,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.3609022556390977,
|
| 643 |
+
"grad_norm": 0.15129096806049347,
|
| 644 |
+
"learning_rate": 0.0004608384531840595,
|
| 645 |
+
"loss": 0.5236,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.3649122807017544,
|
| 650 |
+
"grad_norm": 0.15553347766399384,
|
| 651 |
+
"learning_rate": 0.00045998747949264895,
|
| 652 |
+
"loss": 0.4791,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.36892230576441104,
|
| 657 |
+
"grad_norm": 0.16991403698921204,
|
| 658 |
+
"learning_rate": 0.00045912816254386747,
|
| 659 |
+
"loss": 0.4672,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.37293233082706767,
|
| 664 |
+
"grad_norm": 0.15001340210437775,
|
| 665 |
+
"learning_rate": 0.0004582605364802385,
|
| 666 |
+
"loss": 0.4989,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.3769423558897243,
|
| 671 |
+
"grad_norm": 0.15301191806793213,
|
| 672 |
+
"learning_rate": 0.00045738463577442467,
|
| 673 |
+
"loss": 0.4885,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.38095238095238093,
|
| 678 |
+
"grad_norm": 0.1510782241821289,
|
| 679 |
+
"learning_rate": 0.00045650049522785827,
|
| 680 |
+
"loss": 0.4827,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.3849624060150376,
|
| 685 |
+
"grad_norm": 0.14636896550655365,
|
| 686 |
+
"learning_rate": 0.0004556081499693581,
|
| 687 |
+
"loss": 0.4791,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.38897243107769425,
|
| 692 |
+
"grad_norm": 0.1476137489080429,
|
| 693 |
+
"learning_rate": 0.00045470763545373395,
|
| 694 |
+
"loss": 0.4709,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.3929824561403509,
|
| 699 |
+
"grad_norm": 0.15410684049129486,
|
| 700 |
+
"learning_rate": 0.000453798987460378,
|
| 701 |
+
"loss": 0.4794,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.3969924812030075,
|
| 706 |
+
"grad_norm": 0.14843297004699707,
|
| 707 |
+
"learning_rate": 0.00045288224209184315,
|
| 708 |
+
"loss": 0.5077,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.40100250626566414,
|
| 713 |
+
"grad_norm": 0.1490759700536728,
|
| 714 |
+
"learning_rate": 0.00045195743577240835,
|
| 715 |
+
"loss": 0.464,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.4050125313283208,
|
| 720 |
+
"grad_norm": 0.17478294670581818,
|
| 721 |
+
"learning_rate": 0.0004510246052466317,
|
| 722 |
+
"loss": 0.4754,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.40902255639097745,
|
| 727 |
+
"grad_norm": 0.15610414743423462,
|
| 728 |
+
"learning_rate": 0.0004500837875778905,
|
| 729 |
+
"loss": 0.488,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.4130325814536341,
|
| 734 |
+
"grad_norm": 0.15166425704956055,
|
| 735 |
+
"learning_rate": 0.00044913502014690837,
|
| 736 |
+
"loss": 0.4772,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.4170426065162907,
|
| 741 |
+
"grad_norm": 0.14028780162334442,
|
| 742 |
+
"learning_rate": 0.0004481783406502703,
|
| 743 |
+
"loss": 0.5013,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.42105263157894735,
|
| 748 |
+
"grad_norm": 0.17099718749523163,
|
| 749 |
+
"learning_rate": 0.0004472137870989247,
|
| 750 |
+
"loss": 0.4811,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.42506265664160403,
|
| 755 |
+
"grad_norm": 0.1719314604997635,
|
| 756 |
+
"learning_rate": 0.0004462413978166734,
|
| 757 |
+
"loss": 0.4735,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.42907268170426066,
|
| 762 |
+
"grad_norm": 0.17192725837230682,
|
| 763 |
+
"learning_rate": 0.0004452612114386485,
|
| 764 |
+
"loss": 0.4945,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.4330827067669173,
|
| 769 |
+
"grad_norm": 0.1921142190694809,
|
| 770 |
+
"learning_rate": 0.000444273266909778,
|
| 771 |
+
"loss": 0.4975,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.4370927318295739,
|
| 776 |
+
"grad_norm": 0.16015449166297913,
|
| 777 |
+
"learning_rate": 0.00044327760348323767,
|
| 778 |
+
"loss": 0.4566,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.44110275689223055,
|
| 783 |
+
"grad_norm": 0.15686671435832977,
|
| 784 |
+
"learning_rate": 0.000442274260718892,
|
| 785 |
+
"loss": 0.449,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.44511278195488724,
|
| 790 |
+
"grad_norm": 0.1583927422761917,
|
| 791 |
+
"learning_rate": 0.0004412632784817221,
|
| 792 |
+
"loss": 0.4738,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.44912280701754387,
|
| 797 |
+
"grad_norm": 0.1498887687921524,
|
| 798 |
+
"learning_rate": 0.00044024469694024196,
|
| 799 |
+
"loss": 0.4416,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.4531328320802005,
|
| 804 |
+
"grad_norm": 0.16318589448928833,
|
| 805 |
+
"learning_rate": 0.0004392185565649024,
|
| 806 |
+
"loss": 0.4527,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.45714285714285713,
|
| 811 |
+
"grad_norm": 0.14862380921840668,
|
| 812 |
+
"learning_rate": 0.000438184898126483,
|
| 813 |
+
"loss": 0.4613,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.46115288220551376,
|
| 818 |
+
"grad_norm": 0.17321060597896576,
|
| 819 |
+
"learning_rate": 0.00043714376269447233,
|
| 820 |
+
"loss": 0.4587,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.46516290726817044,
|
| 825 |
+
"grad_norm": 0.15772369503974915,
|
| 826 |
+
"learning_rate": 0.0004360951916354361,
|
| 827 |
+
"loss": 0.4546,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.4691729323308271,
|
| 832 |
+
"grad_norm": 0.1607331931591034,
|
| 833 |
+
"learning_rate": 0.0004350392266113736,
|
| 834 |
+
"loss": 0.4836,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.4731829573934837,
|
| 839 |
+
"grad_norm": 0.16422398388385773,
|
| 840 |
+
"learning_rate": 0.0004339759095780623,
|
| 841 |
+
"loss": 0.446,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.47719298245614034,
|
| 846 |
+
"grad_norm": 0.15711261332035065,
|
| 847 |
+
"learning_rate": 0.00043290528278339097,
|
| 848 |
+
"loss": 0.4308,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.48120300751879697,
|
| 853 |
+
"grad_norm": 0.15732958912849426,
|
| 854 |
+
"learning_rate": 0.0004318273887656811,
|
| 855 |
+
"loss": 0.4687,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.48521303258145365,
|
| 860 |
+
"grad_norm": 0.15295931696891785,
|
| 861 |
+
"learning_rate": 0.00043074227035199656,
|
| 862 |
+
"loss": 0.4474,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.4892230576441103,
|
| 867 |
+
"grad_norm": 0.16534899175167084,
|
| 868 |
+
"learning_rate": 0.00042964997065644204,
|
| 869 |
+
"loss": 0.4552,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.4932330827067669,
|
| 874 |
+
"grad_norm": 0.15630511939525604,
|
| 875 |
+
"learning_rate": 0.00042855053307845056,
|
| 876 |
+
"loss": 0.4842,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.49724310776942354,
|
| 881 |
+
"grad_norm": 0.1624716818332672,
|
| 882 |
+
"learning_rate": 0.0004274440013010581,
|
| 883 |
+
"loss": 0.4228,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.49964912280701756,
|
| 888 |
+
"eval_loss": 0.33371835947036743,
|
| 889 |
+
"eval_runtime": 8.4047,
|
| 890 |
+
"eval_samples_per_second": 6.544,
|
| 891 |
+
"eval_steps_per_second": 0.833,
|
| 892 |
+
"step": 623
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.5012531328320802,
|
| 896 |
+
"grad_norm": 0.16876687109470367,
|
| 897 |
+
"learning_rate": 0.0004263304192891689,
|
| 898 |
+
"loss": 0.4628,
|
| 899 |
+
"step": 625
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.5052631578947369,
|
| 903 |
+
"grad_norm": 0.16050459444522858,
|
| 904 |
+
"learning_rate": 0.00042520983128780824,
|
| 905 |
+
"loss": 0.422,
|
| 906 |
+
"step": 630
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.5092731829573934,
|
| 910 |
+
"grad_norm": 0.1509382724761963,
|
| 911 |
+
"learning_rate": 0.00042408228182036446,
|
| 912 |
+
"loss": 0.4473,
|
| 913 |
+
"step": 635
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.5132832080200501,
|
| 917 |
+
"grad_norm": 0.16160839796066284,
|
| 918 |
+
"learning_rate": 0.00042294781568682,
|
| 919 |
+
"loss": 0.4264,
|
| 920 |
+
"step": 640
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.5172932330827068,
|
| 924 |
+
"grad_norm": 0.1524878293275833,
|
| 925 |
+
"learning_rate": 0.0004218064779619715,
|
| 926 |
+
"loss": 0.4332,
|
| 927 |
+
"step": 645
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.5213032581453634,
|
| 931 |
+
"grad_norm": 0.18008528649806976,
|
| 932 |
+
"learning_rate": 0.00042065831399363884,
|
| 933 |
+
"loss": 0.4365,
|
| 934 |
+
"step": 650
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.5253132832080201,
|
| 938 |
+
"grad_norm": 0.16481877863407135,
|
| 939 |
+
"learning_rate": 0.0004195033694008632,
|
| 940 |
+
"loss": 0.4417,
|
| 941 |
+
"step": 655
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.5293233082706766,
|
| 945 |
+
"grad_norm": 0.16098377108573914,
|
| 946 |
+
"learning_rate": 0.00041834169007209487,
|
| 947 |
+
"loss": 0.4459,
|
| 948 |
+
"step": 660
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.5333333333333333,
|
| 952 |
+
"grad_norm": 0.15441514551639557,
|
| 953 |
+
"learning_rate": 0.0004171733221633695,
|
| 954 |
+
"loss": 0.4178,
|
| 955 |
+
"step": 665
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.53734335839599,
|
| 959 |
+
"grad_norm": 0.15172994136810303,
|
| 960 |
+
"learning_rate": 0.00041599831209647475,
|
| 961 |
+
"loss": 0.4261,
|
| 962 |
+
"step": 670
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.5413533834586466,
|
| 966 |
+
"grad_norm": 0.16424188017845154,
|
| 967 |
+
"learning_rate": 0.0004148167065571055,
|
| 968 |
+
"loss": 0.4375,
|
| 969 |
+
"step": 675
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.5453634085213033,
|
| 973 |
+
"grad_norm": 0.1634334772825241,
|
| 974 |
+
"learning_rate": 0.0004136285524930091,
|
| 975 |
+
"loss": 0.4453,
|
| 976 |
+
"step": 680
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.5493734335839598,
|
| 980 |
+
"grad_norm": 0.1613541692495346,
|
| 981 |
+
"learning_rate": 0.00041243389711211997,
|
| 982 |
+
"loss": 0.4835,
|
| 983 |
+
"step": 685
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.5533834586466165,
|
| 987 |
+
"grad_norm": 0.16017889976501465,
|
| 988 |
+
"learning_rate": 0.00041123278788068375,
|
| 989 |
+
"loss": 0.4245,
|
| 990 |
+
"step": 690
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.5573934837092732,
|
| 994 |
+
"grad_norm": 0.1616198718547821,
|
| 995 |
+
"learning_rate": 0.0004100252725213718,
|
| 996 |
+
"loss": 0.4757,
|
| 997 |
+
"step": 695
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.5614035087719298,
|
| 1001 |
+
"grad_norm": 0.15503592789173126,
|
| 1002 |
+
"learning_rate": 0.00040881139901138467,
|
| 1003 |
+
"loss": 0.438,
|
| 1004 |
+
"step": 700
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.5654135338345865,
|
| 1008 |
+
"grad_norm": 0.15209273993968964,
|
| 1009 |
+
"learning_rate": 0.000407591215580546,
|
| 1010 |
+
"loss": 0.4258,
|
| 1011 |
+
"step": 705
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.569423558897243,
|
| 1015 |
+
"grad_norm": 0.16087529063224792,
|
| 1016 |
+
"learning_rate": 0.0004063647707093864,
|
| 1017 |
+
"loss": 0.4319,
|
| 1018 |
+
"step": 710
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.5734335839598997,
|
| 1022 |
+
"grad_norm": 0.16061115264892578,
|
| 1023 |
+
"learning_rate": 0.0004051321131272167,
|
| 1024 |
+
"loss": 0.4562,
|
| 1025 |
+
"step": 715
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.5774436090225564,
|
| 1029 |
+
"grad_norm": 0.15838468074798584,
|
| 1030 |
+
"learning_rate": 0.0004038932918101927,
|
| 1031 |
+
"loss": 0.4171,
|
| 1032 |
+
"step": 720
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.581453634085213,
|
| 1036 |
+
"grad_norm": 0.16260045766830444,
|
| 1037 |
+
"learning_rate": 0.00040264835597936846,
|
| 1038 |
+
"loss": 0.4197,
|
| 1039 |
+
"step": 725
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.5854636591478697,
|
| 1043 |
+
"grad_norm": 0.20412659645080566,
|
| 1044 |
+
"learning_rate": 0.0004013973550987408,
|
| 1045 |
+
"loss": 0.4399,
|
| 1046 |
+
"step": 730
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.5894736842105263,
|
| 1050 |
+
"grad_norm": 0.16249172389507294,
|
| 1051 |
+
"learning_rate": 0.0004001403388732842,
|
| 1052 |
+
"loss": 0.4141,
|
| 1053 |
+
"step": 735
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.593483709273183,
|
| 1057 |
+
"grad_norm": 0.16174912452697754,
|
| 1058 |
+
"learning_rate": 0.0003988773572469761,
|
| 1059 |
+
"loss": 0.4235,
|
| 1060 |
+
"step": 740
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.5974937343358396,
|
| 1064 |
+
"grad_norm": 0.16679826378822327,
|
| 1065 |
+
"learning_rate": 0.0003976084604008119,
|
| 1066 |
+
"loss": 0.4078,
|
| 1067 |
+
"step": 745
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.6015037593984962,
|
| 1071 |
+
"grad_norm": 0.1594962477684021,
|
| 1072 |
+
"learning_rate": 0.0003963336987508115,
|
| 1073 |
+
"loss": 0.4229,
|
| 1074 |
+
"step": 750
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.6055137844611529,
|
| 1078 |
+
"grad_norm": 0.1605675369501114,
|
| 1079 |
+
"learning_rate": 0.00039505312294601635,
|
| 1080 |
+
"loss": 0.4084,
|
| 1081 |
+
"step": 755
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.6095238095238096,
|
| 1085 |
+
"grad_norm": 0.16422423720359802,
|
| 1086 |
+
"learning_rate": 0.0003937667838664768,
|
| 1087 |
+
"loss": 0.3926,
|
| 1088 |
+
"step": 760
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.6135338345864662,
|
| 1092 |
+
"grad_norm": 0.15119189023971558,
|
| 1093 |
+
"learning_rate": 0.00039247473262123036,
|
| 1094 |
+
"loss": 0.3998,
|
| 1095 |
+
"step": 765
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.6175438596491228,
|
| 1099 |
+
"grad_norm": 0.1493472158908844,
|
| 1100 |
+
"learning_rate": 0.0003911770205462717,
|
| 1101 |
+
"loss": 0.3965,
|
| 1102 |
+
"step": 770
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.6215538847117794,
|
| 1106 |
+
"grad_norm": 0.16699016094207764,
|
| 1107 |
+
"learning_rate": 0.00038987369920251213,
|
| 1108 |
+
"loss": 0.4089,
|
| 1109 |
+
"step": 775
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.6255639097744361,
|
| 1113 |
+
"grad_norm": 0.15896373987197876,
|
| 1114 |
+
"learning_rate": 0.00038856482037373155,
|
| 1115 |
+
"loss": 0.3785,
|
| 1116 |
+
"step": 780
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.6295739348370928,
|
| 1120 |
+
"grad_norm": 0.16796669363975525,
|
| 1121 |
+
"learning_rate": 0.0003872504360645206,
|
| 1122 |
+
"loss": 0.3973,
|
| 1123 |
+
"step": 785
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.6335839598997494,
|
| 1127 |
+
"grad_norm": 0.14833569526672363,
|
| 1128 |
+
"learning_rate": 0.0003859305984982149,
|
| 1129 |
+
"loss": 0.4149,
|
| 1130 |
+
"step": 790
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.637593984962406,
|
| 1134 |
+
"grad_norm": 0.1650949865579605,
|
| 1135 |
+
"learning_rate": 0.0003846053601148194,
|
| 1136 |
+
"loss": 0.4253,
|
| 1137 |
+
"step": 795
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.6416040100250626,
|
| 1141 |
+
"grad_norm": 0.15514247119426727,
|
| 1142 |
+
"learning_rate": 0.00038327477356892546,
|
| 1143 |
+
"loss": 0.3654,
|
| 1144 |
+
"step": 800
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.6456140350877193,
|
| 1148 |
+
"grad_norm": 0.16105593740940094,
|
| 1149 |
+
"learning_rate": 0.0003819388917276185,
|
| 1150 |
+
"loss": 0.4399,
|
| 1151 |
+
"step": 805
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.649624060150376,
|
| 1155 |
+
"grad_norm": 0.15280301868915558,
|
| 1156 |
+
"learning_rate": 0.00038059776766837753,
|
| 1157 |
+
"loss": 0.4166,
|
| 1158 |
+
"step": 810
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.6536340852130326,
|
| 1162 |
+
"grad_norm": 0.16363288462162018,
|
| 1163 |
+
"learning_rate": 0.000379251454676966,
|
| 1164 |
+
"loss": 0.3977,
|
| 1165 |
+
"step": 815
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.6576441102756893,
|
| 1169 |
+
"grad_norm": 0.16015923023223877,
|
| 1170 |
+
"learning_rate": 0.00037790000624531505,
|
| 1171 |
+
"loss": 0.3918,
|
| 1172 |
+
"step": 820
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.6616541353383458,
|
| 1176 |
+
"grad_norm": 0.14514611661434174,
|
| 1177 |
+
"learning_rate": 0.0003765434760693981,
|
| 1178 |
+
"loss": 0.3801,
|
| 1179 |
+
"step": 825
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.6656641604010025,
|
| 1183 |
+
"grad_norm": 0.16084541380405426,
|
| 1184 |
+
"learning_rate": 0.0003751819180470969,
|
| 1185 |
+
"loss": 0.4123,
|
| 1186 |
+
"step": 830
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.6696741854636592,
|
| 1190 |
+
"grad_norm": 0.14666476845741272,
|
| 1191 |
+
"learning_rate": 0.000373815386276061,
|
| 1192 |
+
"loss": 0.3712,
|
| 1193 |
+
"step": 835
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.6736842105263158,
|
| 1197 |
+
"grad_norm": 0.16997689008712769,
|
| 1198 |
+
"learning_rate": 0.0003724439350515571,
|
| 1199 |
+
"loss": 0.3999,
|
| 1200 |
+
"step": 840
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.6776942355889725,
|
| 1204 |
+
"grad_norm": 0.15905480086803436,
|
| 1205 |
+
"learning_rate": 0.000371067618864313,
|
| 1206 |
+
"loss": 0.4068,
|
| 1207 |
+
"step": 845
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.681704260651629,
|
| 1211 |
+
"grad_norm": 0.1562589704990387,
|
| 1212 |
+
"learning_rate": 0.0003696864923983516,
|
| 1213 |
+
"loss": 0.3843,
|
| 1214 |
+
"step": 850
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.6857142857142857,
|
| 1218 |
+
"grad_norm": 0.16755416989326477,
|
| 1219 |
+
"learning_rate": 0.00036830061052881897,
|
| 1220 |
+
"loss": 0.4426,
|
| 1221 |
+
"step": 855
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.6897243107769424,
|
| 1225 |
+
"grad_norm": 0.14882604777812958,
|
| 1226 |
+
"learning_rate": 0.00036691002831980323,
|
| 1227 |
+
"loss": 0.3939,
|
| 1228 |
+
"step": 860
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.693734335839599,
|
| 1232 |
+
"grad_norm": 0.16580046713352203,
|
| 1233 |
+
"learning_rate": 0.0003655148010221476,
|
| 1234 |
+
"loss": 0.3823,
|
| 1235 |
+
"step": 865
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.6977443609022557,
|
| 1239 |
+
"grad_norm": 0.15855063498020172,
|
| 1240 |
+
"learning_rate": 0.00036411498407125435,
|
| 1241 |
+
"loss": 0.3907,
|
| 1242 |
+
"step": 870
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.7017543859649122,
|
| 1246 |
+
"grad_norm": 0.1665424108505249,
|
| 1247 |
+
"learning_rate": 0.00036271063308488294,
|
| 1248 |
+
"loss": 0.4133,
|
| 1249 |
+
"step": 875
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.7057644110275689,
|
| 1253 |
+
"grad_norm": 0.1685834378004074,
|
| 1254 |
+
"learning_rate": 0.0003613018038609398,
|
| 1255 |
+
"loss": 0.4359,
|
| 1256 |
+
"step": 880
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.7097744360902256,
|
| 1260 |
+
"grad_norm": 0.15352065861225128,
|
| 1261 |
+
"learning_rate": 0.0003598885523752611,
|
| 1262 |
+
"loss": 0.4037,
|
| 1263 |
+
"step": 885
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.7137844611528822,
|
| 1267 |
+
"grad_norm": 0.17349396646022797,
|
| 1268 |
+
"learning_rate": 0.00035847093477938953,
|
| 1269 |
+
"loss": 0.411,
|
| 1270 |
+
"step": 890
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.7177944862155389,
|
| 1274 |
+
"grad_norm": 0.15559545159339905,
|
| 1275 |
+
"learning_rate": 0.0003570490073983425,
|
| 1276 |
+
"loss": 0.3817,
|
| 1277 |
+
"step": 895
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.7218045112781954,
|
| 1281 |
+
"grad_norm": 0.16843955218791962,
|
| 1282 |
+
"learning_rate": 0.00035562282672837477,
|
| 1283 |
+
"loss": 0.4162,
|
| 1284 |
+
"step": 900
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.7258145363408521,
|
| 1288 |
+
"grad_norm": 0.17087852954864502,
|
| 1289 |
+
"learning_rate": 0.00035419244943473305,
|
| 1290 |
+
"loss": 0.412,
|
| 1291 |
+
"step": 905
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.7298245614035088,
|
| 1295 |
+
"grad_norm": 0.16542008519172668,
|
| 1296 |
+
"learning_rate": 0.00035275793234940545,
|
| 1297 |
+
"loss": 0.421,
|
| 1298 |
+
"step": 910
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.7338345864661654,
|
| 1302 |
+
"grad_norm": 0.17914004623889923,
|
| 1303 |
+
"learning_rate": 0.0003513193324688627,
|
| 1304 |
+
"loss": 0.3759,
|
| 1305 |
+
"step": 915
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.7378446115288221,
|
| 1309 |
+
"grad_norm": 0.16926446557044983,
|
| 1310 |
+
"learning_rate": 0.00034987670695179403,
|
| 1311 |
+
"loss": 0.3988,
|
| 1312 |
+
"step": 920
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.7418546365914787,
|
| 1316 |
+
"grad_norm": 0.15922464430332184,
|
| 1317 |
+
"learning_rate": 0.00034843011311683566,
|
| 1318 |
+
"loss": 0.3641,
|
| 1319 |
+
"step": 925
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.7458646616541353,
|
| 1323 |
+
"grad_norm": 0.16068434715270996,
|
| 1324 |
+
"learning_rate": 0.00034697960844029383,
|
| 1325 |
+
"loss": 0.3746,
|
| 1326 |
+
"step": 930
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.749874686716792,
|
| 1330 |
+
"grad_norm": 0.18251362442970276,
|
| 1331 |
+
"learning_rate": 0.0003455252505538611,
|
| 1332 |
+
"loss": 0.4308,
|
| 1333 |
+
"step": 935
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.7538847117794486,
|
| 1337 |
+
"grad_norm": 0.1559171974658966,
|
| 1338 |
+
"learning_rate": 0.0003440670972423262,
|
| 1339 |
+
"loss": 0.4045,
|
| 1340 |
+
"step": 940
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.7578947368421053,
|
| 1344 |
+
"grad_norm": 0.17260219156742096,
|
| 1345 |
+
"learning_rate": 0.00034260520644127846,
|
| 1346 |
+
"loss": 0.3642,
|
| 1347 |
+
"step": 945
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.7619047619047619,
|
| 1351 |
+
"grad_norm": 0.1719062477350235,
|
| 1352 |
+
"learning_rate": 0.0003411396362348056,
|
| 1353 |
+
"loss": 0.3973,
|
| 1354 |
+
"step": 950
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.7659147869674185,
|
| 1358 |
+
"grad_norm": 0.16649049520492554,
|
| 1359 |
+
"learning_rate": 0.0003396704448531863,
|
| 1360 |
+
"loss": 0.4031,
|
| 1361 |
+
"step": 955
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.7699248120300752,
|
| 1365 |
+
"grad_norm": 0.15968754887580872,
|
| 1366 |
+
"learning_rate": 0.00033819769067057617,
|
| 1367 |
+
"loss": 0.3648,
|
| 1368 |
+
"step": 960
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.7739348370927318,
|
| 1372 |
+
"grad_norm": 0.18287472426891327,
|
| 1373 |
+
"learning_rate": 0.0003367214322026888,
|
| 1374 |
+
"loss": 0.3921,
|
| 1375 |
+
"step": 965
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.7779448621553885,
|
| 1379 |
+
"grad_norm": 0.1621864289045334,
|
| 1380 |
+
"learning_rate": 0.00033524172810447055,
|
| 1381 |
+
"loss": 0.3956,
|
| 1382 |
+
"step": 970
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.7819548872180451,
|
| 1386 |
+
"grad_norm": 0.1772567480802536,
|
| 1387 |
+
"learning_rate": 0.00033375863716777007,
|
| 1388 |
+
"loss": 0.3745,
|
| 1389 |
+
"step": 975
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.7859649122807018,
|
| 1393 |
+
"grad_norm": 0.17318162322044373,
|
| 1394 |
+
"learning_rate": 0.0003322722183190025,
|
| 1395 |
+
"loss": 0.3924,
|
| 1396 |
+
"step": 980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.7899749373433584,
|
| 1400 |
+
"grad_norm": 0.17136339843273163,
|
| 1401 |
+
"learning_rate": 0.0003307825306168082,
|
| 1402 |
+
"loss": 0.3897,
|
| 1403 |
+
"step": 985
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.793984962406015,
|
| 1407 |
+
"grad_norm": 0.15933868288993835,
|
| 1408 |
+
"learning_rate": 0.00032928963324970595,
|
| 1409 |
+
"loss": 0.3723,
|
| 1410 |
+
"step": 990
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.7979949874686717,
|
| 1414 |
+
"grad_norm": 0.1778361201286316,
|
| 1415 |
+
"learning_rate": 0.0003277935855337417,
|
| 1416 |
+
"loss": 0.3822,
|
| 1417 |
+
"step": 995
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.8020050125313283,
|
| 1421 |
+
"grad_norm": 0.15671303868293762,
|
| 1422 |
+
"learning_rate": 0.00032629444691013144,
|
| 1423 |
+
"loss": 0.3805,
|
| 1424 |
+
"step": 1000
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.806015037593985,
|
| 1428 |
+
"grad_norm": 0.16916193068027496,
|
| 1429 |
+
"learning_rate": 0.0003247922769428998,
|
| 1430 |
+
"loss": 0.3931,
|
| 1431 |
+
"step": 1005
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.8100250626566416,
|
| 1435 |
+
"grad_norm": 0.15111452341079712,
|
| 1436 |
+
"learning_rate": 0.0003232871353165129,
|
| 1437 |
+
"loss": 0.3398,
|
| 1438 |
+
"step": 1010
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.8140350877192982,
|
| 1442 |
+
"grad_norm": 0.14872141182422638,
|
| 1443 |
+
"learning_rate": 0.0003217790818335077,
|
| 1444 |
+
"loss": 0.3541,
|
| 1445 |
+
"step": 1015
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.8180451127819549,
|
| 1449 |
+
"grad_norm": 0.1576554924249649,
|
| 1450 |
+
"learning_rate": 0.00032026817641211524,
|
| 1451 |
+
"loss": 0.3612,
|
| 1452 |
+
"step": 1020
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.8220551378446115,
|
| 1456 |
+
"grad_norm": 0.16731838881969452,
|
| 1457 |
+
"learning_rate": 0.0003187544790838805,
|
| 1458 |
+
"loss": 0.3742,
|
| 1459 |
+
"step": 1025
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.8260651629072682,
|
| 1463 |
+
"grad_norm": 0.17792581021785736,
|
| 1464 |
+
"learning_rate": 0.0003172380499912768,
|
| 1465 |
+
"loss": 0.3835,
|
| 1466 |
+
"step": 1030
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.8300751879699249,
|
| 1470 |
+
"grad_norm": 0.15929646790027618,
|
| 1471 |
+
"learning_rate": 0.0003157189493853164,
|
| 1472 |
+
"loss": 0.3547,
|
| 1473 |
+
"step": 1035
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.8340852130325814,
|
| 1477 |
+
"grad_norm": 0.17030423879623413,
|
| 1478 |
+
"learning_rate": 0.00031419723762315656,
|
| 1479 |
+
"loss": 0.3833,
|
| 1480 |
+
"step": 1040
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.8380952380952381,
|
| 1484 |
+
"grad_norm": 0.16632360219955444,
|
| 1485 |
+
"learning_rate": 0.0003126729751657015,
|
| 1486 |
+
"loss": 0.3851,
|
| 1487 |
+
"step": 1045
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.8421052631578947,
|
| 1491 |
+
"grad_norm": 0.16034843027591705,
|
| 1492 |
+
"learning_rate": 0.0003111462225752,
|
| 1493 |
+
"loss": 0.3457,
|
| 1494 |
+
"step": 1050
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.8461152882205514,
|
| 1498 |
+
"grad_norm": 0.1584872305393219,
|
| 1499 |
+
"learning_rate": 0.0003096170405128393,
|
| 1500 |
+
"loss": 0.3627,
|
| 1501 |
+
"step": 1055
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.8501253132832081,
|
| 1505 |
+
"grad_norm": 0.15513049066066742,
|
| 1506 |
+
"learning_rate": 0.0003080854897363348,
|
| 1507 |
+
"loss": 0.3571,
|
| 1508 |
+
"step": 1060
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.8541353383458646,
|
| 1512 |
+
"grad_norm": 0.18553143739700317,
|
| 1513 |
+
"learning_rate": 0.000306551631097516,
|
| 1514 |
+
"loss": 0.3844,
|
| 1515 |
+
"step": 1065
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.8581453634085213,
|
| 1519 |
+
"grad_norm": 0.17072825133800507,
|
| 1520 |
+
"learning_rate": 0.00030501552553990885,
|
| 1521 |
+
"loss": 0.3968,
|
| 1522 |
+
"step": 1070
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.8621553884711779,
|
| 1526 |
+
"grad_norm": 0.17334377765655518,
|
| 1527 |
+
"learning_rate": 0.00030347723409631413,
|
| 1528 |
+
"loss": 0.3759,
|
| 1529 |
+
"step": 1075
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.8661654135338346,
|
| 1533 |
+
"grad_norm": 0.16847221553325653,
|
| 1534 |
+
"learning_rate": 0.00030193681788638274,
|
| 1535 |
+
"loss": 0.3437,
|
| 1536 |
+
"step": 1080
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.8701754385964913,
|
| 1540 |
+
"grad_norm": 0.17665381729602814,
|
| 1541 |
+
"learning_rate": 0.000300394338114187,
|
| 1542 |
+
"loss": 0.3691,
|
| 1543 |
+
"step": 1085
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.8741854636591478,
|
| 1547 |
+
"grad_norm": 0.17721857130527496,
|
| 1548 |
+
"learning_rate": 0.00029884985606578907,
|
| 1549 |
+
"loss": 0.3776,
|
| 1550 |
+
"step": 1090
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.8781954887218045,
|
| 1554 |
+
"grad_norm": 0.14506922662258148,
|
| 1555 |
+
"learning_rate": 0.0002973034331068061,
|
| 1556 |
+
"loss": 0.3329,
|
| 1557 |
+
"step": 1095
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.8822055137844611,
|
| 1561 |
+
"grad_norm": 0.14164331555366516,
|
| 1562 |
+
"learning_rate": 0.0002957551306799715,
|
| 1563 |
+
"loss": 0.3296,
|
| 1564 |
+
"step": 1100
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.8862155388471178,
|
| 1568 |
+
"grad_norm": 0.17468836903572083,
|
| 1569 |
+
"learning_rate": 0.0002942050103026942,
|
| 1570 |
+
"loss": 0.3814,
|
| 1571 |
+
"step": 1105
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.8902255639097745,
|
| 1575 |
+
"grad_norm": 0.1707754135131836,
|
| 1576 |
+
"learning_rate": 0.0002926531335646141,
|
| 1577 |
+
"loss": 0.3288,
|
| 1578 |
+
"step": 1110
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.894235588972431,
|
| 1582 |
+
"grad_norm": 0.1548461616039276,
|
| 1583 |
+
"learning_rate": 0.0002910995621251554,
|
| 1584 |
+
"loss": 0.3837,
|
| 1585 |
+
"step": 1115
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.8982456140350877,
|
| 1589 |
+
"grad_norm": 0.18606321513652802,
|
| 1590 |
+
"learning_rate": 0.000289544357711076,
|
| 1591 |
+
"loss": 0.411,
|
| 1592 |
+
"step": 1120
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.9022556390977443,
|
| 1596 |
+
"grad_norm": 0.1885378062725067,
|
| 1597 |
+
"learning_rate": 0.00028798758211401586,
|
| 1598 |
+
"loss": 0.3965,
|
| 1599 |
+
"step": 1125
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.906265664160401,
|
| 1603 |
+
"grad_norm": 0.15879781544208527,
|
| 1604 |
+
"learning_rate": 0.00028642929718804126,
|
| 1605 |
+
"loss": 0.3217,
|
| 1606 |
+
"step": 1130
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.9102756892230577,
|
| 1610 |
+
"grad_norm": 0.16221250593662262,
|
| 1611 |
+
"learning_rate": 0.00028486956484718735,
|
| 1612 |
+
"loss": 0.3431,
|
| 1613 |
+
"step": 1135
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.9142857142857143,
|
| 1617 |
+
"grad_norm": 0.15938705205917358,
|
| 1618 |
+
"learning_rate": 0.0002833084470629983,
|
| 1619 |
+
"loss": 0.3645,
|
| 1620 |
+
"step": 1140
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.9182957393483709,
|
| 1624 |
+
"grad_norm": 0.18234948813915253,
|
| 1625 |
+
"learning_rate": 0.0002817460058620647,
|
| 1626 |
+
"loss": 0.3696,
|
| 1627 |
+
"step": 1145
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.9223057644110275,
|
| 1631 |
+
"grad_norm": 0.17734579741954803,
|
| 1632 |
+
"learning_rate": 0.0002801823033235598,
|
| 1633 |
+
"loss": 0.3939,
|
| 1634 |
+
"step": 1150
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.9263157894736842,
|
| 1638 |
+
"grad_norm": 0.17581012845039368,
|
| 1639 |
+
"learning_rate": 0.00027861740157677204,
|
| 1640 |
+
"loss": 0.3857,
|
| 1641 |
+
"step": 1155
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.9303258145363409,
|
| 1645 |
+
"grad_norm": 0.16594088077545166,
|
| 1646 |
+
"learning_rate": 0.00027705136279863754,
|
| 1647 |
+
"loss": 0.3784,
|
| 1648 |
+
"step": 1160
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.9343358395989975,
|
| 1652 |
+
"grad_norm": 0.17413030564785004,
|
| 1653 |
+
"learning_rate": 0.0002754842492112685,
|
| 1654 |
+
"loss": 0.3791,
|
| 1655 |
+
"step": 1165
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.9383458646616541,
|
| 1659 |
+
"grad_norm": 0.16821810603141785,
|
| 1660 |
+
"learning_rate": 0.0002739161230794822,
|
| 1661 |
+
"loss": 0.3896,
|
| 1662 |
+
"step": 1170
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.9423558897243107,
|
| 1666 |
+
"grad_norm": 0.18337774276733398,
|
| 1667 |
+
"learning_rate": 0.000272347046708326,
|
| 1668 |
+
"loss": 0.3745,
|
| 1669 |
+
"step": 1175
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.9463659147869674,
|
| 1673 |
+
"grad_norm": 0.15810732543468475,
|
| 1674 |
+
"learning_rate": 0.00027077708244060283,
|
| 1675 |
+
"loss": 0.3576,
|
| 1676 |
+
"step": 1180
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.9503759398496241,
|
| 1680 |
+
"grad_norm": 0.16330723464488983,
|
| 1681 |
+
"learning_rate": 0.00026920629265439326,
|
| 1682 |
+
"loss": 0.3641,
|
| 1683 |
+
"step": 1185
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.9543859649122807,
|
| 1687 |
+
"grad_norm": 0.168942391872406,
|
| 1688 |
+
"learning_rate": 0.0002676347397605777,
|
| 1689 |
+
"loss": 0.36,
|
| 1690 |
+
"step": 1190
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.9583959899749374,
|
| 1694 |
+
"grad_norm": 0.1779577136039734,
|
| 1695 |
+
"learning_rate": 0.0002660624862003566,
|
| 1696 |
+
"loss": 0.3644,
|
| 1697 |
+
"step": 1195
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.9624060150375939,
|
| 1701 |
+
"grad_norm": 0.175537571310997,
|
| 1702 |
+
"learning_rate": 0.00026448959444276896,
|
| 1703 |
+
"loss": 0.3766,
|
| 1704 |
+
"step": 1200
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.9664160401002506,
|
| 1708 |
+
"grad_norm": 0.17511993646621704,
|
| 1709 |
+
"learning_rate": 0.0002629161269822113,
|
| 1710 |
+
"loss": 0.3636,
|
| 1711 |
+
"step": 1205
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.9704260651629073,
|
| 1715 |
+
"grad_norm": 0.17956364154815674,
|
| 1716 |
+
"learning_rate": 0.00026134214633595347,
|
| 1717 |
+
"loss": 0.3916,
|
| 1718 |
+
"step": 1210
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.9744360902255639,
|
| 1722 |
+
"grad_norm": 0.15933747589588165,
|
| 1723 |
+
"learning_rate": 0.000259767715041656,
|
| 1724 |
+
"loss": 0.3316,
|
| 1725 |
+
"step": 1215
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.9784461152882206,
|
| 1729 |
+
"grad_norm": 0.18015708029270172,
|
| 1730 |
+
"learning_rate": 0.000258192895654884,
|
| 1731 |
+
"loss": 0.3791,
|
| 1732 |
+
"step": 1220
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.9824561403508771,
|
| 1736 |
+
"grad_norm": 0.15591025352478027,
|
| 1737 |
+
"learning_rate": 0.00025661775074662275,
|
| 1738 |
+
"loss": 0.3222,
|
| 1739 |
+
"step": 1225
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.9864661654135338,
|
| 1743 |
+
"grad_norm": 0.1800549179315567,
|
| 1744 |
+
"learning_rate": 0.0002550423429007909,
|
| 1745 |
+
"loss": 0.3282,
|
| 1746 |
+
"step": 1230
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.9904761904761905,
|
| 1750 |
+
"grad_norm": 0.1630546748638153,
|
| 1751 |
+
"learning_rate": 0.0002534667347117544,
|
| 1752 |
+
"loss": 0.3659,
|
| 1753 |
+
"step": 1235
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.9944862155388471,
|
| 1757 |
+
"grad_norm": 0.169642373919487,
|
| 1758 |
+
"learning_rate": 0.0002518909887818393,
|
| 1759 |
+
"loss": 0.3409,
|
| 1760 |
+
"step": 1240
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.9984962406015038,
|
| 1764 |
+
"grad_norm": 0.1650785356760025,
|
| 1765 |
+
"learning_rate": 0.00025031516771884416,
|
| 1766 |
+
"loss": 0.3632,
|
| 1767 |
+
"step": 1245
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.9992982456140351,
|
| 1771 |
+
"eval_loss": 0.2687419354915619,
|
| 1772 |
+
"eval_runtime": 8.3871,
|
| 1773 |
+
"eval_samples_per_second": 6.558,
|
| 1774 |
+
"eval_steps_per_second": 0.835,
|
| 1775 |
+
"step": 1246
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 1.0025062656641603,
|
| 1779 |
+
"grad_norm": 0.1833638846874237,
|
| 1780 |
+
"learning_rate": 0.000248739334133553,
|
| 1781 |
+
"loss": 0.3377,
|
| 1782 |
+
"step": 1250
|
| 1783 |
+
},
|
| 1784 |
+
{
|
| 1785 |
+
"epoch": 1.006516290726817,
|
| 1786 |
+
"grad_norm": 0.16308297216892242,
|
| 1787 |
+
"learning_rate": 0.00024716355063724707,
|
| 1788 |
+
"loss": 0.2999,
|
| 1789 |
+
"step": 1255
|
| 1790 |
+
},
|
| 1791 |
+
{
|
| 1792 |
+
"epoch": 1.0105263157894737,
|
| 1793 |
+
"grad_norm": 0.16679681837558746,
|
| 1794 |
+
"learning_rate": 0.00024558787983921783,
|
| 1795 |
+
"loss": 0.2986,
|
| 1796 |
+
"step": 1260
|
| 1797 |
+
},
|
| 1798 |
+
{
|
| 1799 |
+
"epoch": 1.0145363408521304,
|
| 1800 |
+
"grad_norm": 0.17338265478610992,
|
| 1801 |
+
"learning_rate": 0.0002440123843442788,
|
| 1802 |
+
"loss": 0.2845,
|
| 1803 |
+
"step": 1265
|
| 1804 |
+
},
|
| 1805 |
+
{
|
| 1806 |
+
"epoch": 1.0185463659147869,
|
| 1807 |
+
"grad_norm": 0.16227693855762482,
|
| 1808 |
+
"learning_rate": 0.00024243712675027827,
|
| 1809 |
+
"loss": 0.3051,
|
| 1810 |
+
"step": 1270
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 1.0225563909774436,
|
| 1814 |
+
"grad_norm": 0.15637050569057465,
|
| 1815 |
+
"learning_rate": 0.00024086216964561236,
|
| 1816 |
+
"loss": 0.2896,
|
| 1817 |
+
"step": 1275
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 1.0265664160401002,
|
| 1821 |
+
"grad_norm": 0.1684030145406723,
|
| 1822 |
+
"learning_rate": 0.00023928757560673808,
|
| 1823 |
+
"loss": 0.272,
|
| 1824 |
+
"step": 1280
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 1.030576441102757,
|
| 1828 |
+
"grad_norm": 0.1560041755437851,
|
| 1829 |
+
"learning_rate": 0.00023771340719568688,
|
| 1830 |
+
"loss": 0.2795,
|
| 1831 |
+
"step": 1285
|
| 1832 |
+
},
|
| 1833 |
+
{
|
| 1834 |
+
"epoch": 1.0345864661654136,
|
| 1835 |
+
"grad_norm": 0.15775898098945618,
|
| 1836 |
+
"learning_rate": 0.00023613972695757919,
|
| 1837 |
+
"loss": 0.2721,
|
| 1838 |
+
"step": 1290
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 1.03859649122807,
|
| 1842 |
+
"grad_norm": 0.17277134954929352,
|
| 1843 |
+
"learning_rate": 0.00023456659741813944,
|
| 1844 |
+
"loss": 0.2726,
|
| 1845 |
+
"step": 1295
|
| 1846 |
+
},
|
| 1847 |
+
{
|
| 1848 |
+
"epoch": 1.0426065162907268,
|
| 1849 |
+
"grad_norm": 0.15520408749580383,
|
| 1850 |
+
"learning_rate": 0.00023299408108121134,
|
| 1851 |
+
"loss": 0.2681,
|
| 1852 |
+
"step": 1300
|
| 1853 |
+
},
|
| 1854 |
+
{
|
| 1855 |
+
"epoch": 1.0466165413533834,
|
| 1856 |
+
"grad_norm": 0.16784396767616272,
|
| 1857 |
+
"learning_rate": 0.00023142224042627495,
|
| 1858 |
+
"loss": 0.2817,
|
| 1859 |
+
"step": 1305
|
| 1860 |
+
},
|
| 1861 |
+
{
|
| 1862 |
+
"epoch": 1.0506265664160401,
|
| 1863 |
+
"grad_norm": 0.16616085171699524,
|
| 1864 |
+
"learning_rate": 0.00022985113790596393,
|
| 1865 |
+
"loss": 0.2863,
|
| 1866 |
+
"step": 1310
|
| 1867 |
+
},
|
| 1868 |
+
{
|
| 1869 |
+
"epoch": 1.0546365914786968,
|
| 1870 |
+
"grad_norm": 0.15373535454273224,
|
| 1871 |
+
"learning_rate": 0.00022828083594358416,
|
| 1872 |
+
"loss": 0.2837,
|
| 1873 |
+
"step": 1315
|
| 1874 |
+
},
|
| 1875 |
+
{
|
| 1876 |
+
"epoch": 1.0586466165413533,
|
| 1877 |
+
"grad_norm": 0.15861855447292328,
|
| 1878 |
+
"learning_rate": 0.00022671139693063386,
|
| 1879 |
+
"loss": 0.2878,
|
| 1880 |
+
"step": 1320
|
| 1881 |
+
},
|
| 1882 |
+
{
|
| 1883 |
+
"epoch": 1.06265664160401,
|
| 1884 |
+
"grad_norm": 0.15624883770942688,
|
| 1885 |
+
"learning_rate": 0.00022514288322432414,
|
| 1886 |
+
"loss": 0.2952,
|
| 1887 |
+
"step": 1325
|
| 1888 |
+
},
|
| 1889 |
+
{
|
| 1890 |
+
"epoch": 1.0666666666666667,
|
| 1891 |
+
"grad_norm": 0.17166998982429504,
|
| 1892 |
+
"learning_rate": 0.0002235753571451018,
|
| 1893 |
+
"loss": 0.2957,
|
| 1894 |
+
"step": 1330
|
| 1895 |
+
},
|
| 1896 |
+
{
|
| 1897 |
+
"epoch": 1.0706766917293233,
|
| 1898 |
+
"grad_norm": 0.15910623967647552,
|
| 1899 |
+
"learning_rate": 0.00022200888097417305,
|
| 1900 |
+
"loss": 0.2954,
|
| 1901 |
+
"step": 1335
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 1.07468671679198,
|
| 1905 |
+
"grad_norm": 0.154972106218338,
|
| 1906 |
+
"learning_rate": 0.00022044351695102907,
|
| 1907 |
+
"loss": 0.2624,
|
| 1908 |
+
"step": 1340
|
| 1909 |
+
},
|
| 1910 |
+
{
|
| 1911 |
+
"epoch": 1.0786967418546367,
|
| 1912 |
+
"grad_norm": 0.16736608743667603,
|
| 1913 |
+
"learning_rate": 0.00021887932727097278,
|
| 1914 |
+
"loss": 0.2677,
|
| 1915 |
+
"step": 1345
|
| 1916 |
+
},
|
| 1917 |
+
{
|
| 1918 |
+
"epoch": 1.0827067669172932,
|
| 1919 |
+
"grad_norm": 0.17113475501537323,
|
| 1920 |
+
"learning_rate": 0.00021731637408264822,
|
| 1921 |
+
"loss": 0.2613,
|
| 1922 |
+
"step": 1350
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"epoch": 1.0867167919799499,
|
| 1926 |
+
"grad_norm": 0.17451037466526031,
|
| 1927 |
+
"learning_rate": 0.00021575471948557062,
|
| 1928 |
+
"loss": 0.291,
|
| 1929 |
+
"step": 1355
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 1.0907268170426065,
|
| 1933 |
+
"grad_norm": 0.15798042714595795,
|
| 1934 |
+
"learning_rate": 0.00021419442552765948,
|
| 1935 |
+
"loss": 0.2628,
|
| 1936 |
+
"step": 1360
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 1.0947368421052632,
|
| 1940 |
+
"grad_norm": 0.17537885904312134,
|
| 1941 |
+
"learning_rate": 0.00021263555420277335,
|
| 1942 |
+
"loss": 0.2764,
|
| 1943 |
+
"step": 1365
|
| 1944 |
+
},
|
| 1945 |
+
{
|
| 1946 |
+
"epoch": 1.09874686716792,
|
| 1947 |
+
"grad_norm": 0.14623847603797913,
|
| 1948 |
+
"learning_rate": 0.00021107816744824617,
|
| 1949 |
+
"loss": 0.29,
|
| 1950 |
+
"step": 1370
|
| 1951 |
+
},
|
| 1952 |
+
{
|
| 1953 |
+
"epoch": 1.1027568922305764,
|
| 1954 |
+
"grad_norm": 0.17818230390548706,
|
| 1955 |
+
"learning_rate": 0.00020952232714242685,
|
| 1956 |
+
"loss": 0.2866,
|
| 1957 |
+
"step": 1375
|
| 1958 |
+
},
|
| 1959 |
+
{
|
| 1960 |
+
"epoch": 1.106766917293233,
|
| 1961 |
+
"grad_norm": 0.16421560943126678,
|
| 1962 |
+
"learning_rate": 0.00020796809510222048,
|
| 1963 |
+
"loss": 0.2718,
|
| 1964 |
+
"step": 1380
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 1.1107769423558898,
|
| 1968 |
+
"grad_norm": 0.16824981570243835,
|
| 1969 |
+
"learning_rate": 0.00020641553308063247,
|
| 1970 |
+
"loss": 0.2976,
|
| 1971 |
+
"step": 1385
|
| 1972 |
+
},
|
| 1973 |
+
{
|
| 1974 |
+
"epoch": 1.1147869674185464,
|
| 1975 |
+
"grad_norm": 0.15799076855182648,
|
| 1976 |
+
"learning_rate": 0.00020486470276431446,
|
| 1977 |
+
"loss": 0.2934,
|
| 1978 |
+
"step": 1390
|
| 1979 |
+
},
|
| 1980 |
+
{
|
| 1981 |
+
"epoch": 1.1187969924812031,
|
| 1982 |
+
"grad_norm": 0.18710538744926453,
|
| 1983 |
+
"learning_rate": 0.000203315665771114,
|
| 1984 |
+
"loss": 0.2792,
|
| 1985 |
+
"step": 1395
|
| 1986 |
+
},
|
| 1987 |
+
{
|
| 1988 |
+
"epoch": 1.1228070175438596,
|
| 1989 |
+
"grad_norm": 0.18664434552192688,
|
| 1990 |
+
"learning_rate": 0.00020176848364762578,
|
| 1991 |
+
"loss": 0.3125,
|
| 1992 |
+
"step": 1400
|
| 1993 |
+
},
|
| 1994 |
+
{
|
| 1995 |
+
"epoch": 1.1268170426065163,
|
| 1996 |
+
"grad_norm": 0.14843712747097015,
|
| 1997 |
+
"learning_rate": 0.00020022321786674664,
|
| 1998 |
+
"loss": 0.2636,
|
| 1999 |
+
"step": 1405
|
| 2000 |
+
},
|
| 2001 |
+
{
|
| 2002 |
+
"epoch": 1.130827067669173,
|
| 2003 |
+
"grad_norm": 0.18187515437602997,
|
| 2004 |
+
"learning_rate": 0.000198679929825233,
|
| 2005 |
+
"loss": 0.3186,
|
| 2006 |
+
"step": 1410
|
| 2007 |
+
},
|
| 2008 |
+
{
|
| 2009 |
+
"epoch": 1.1348370927318296,
|
| 2010 |
+
"grad_norm": 0.15874765813350677,
|
| 2011 |
+
"learning_rate": 0.0001971386808412612,
|
| 2012 |
+
"loss": 0.2578,
|
| 2013 |
+
"step": 1415
|
| 2014 |
+
},
|
| 2015 |
+
{
|
| 2016 |
+
"epoch": 1.1388471177944863,
|
| 2017 |
+
"grad_norm": 0.17650403082370758,
|
| 2018 |
+
"learning_rate": 0.00019559953215199168,
|
| 2019 |
+
"loss": 0.3012,
|
| 2020 |
+
"step": 1420
|
| 2021 |
+
},
|
| 2022 |
+
{
|
| 2023 |
+
"epoch": 1.1428571428571428,
|
| 2024 |
+
"grad_norm": 0.17228610813617706,
|
| 2025 |
+
"learning_rate": 0.0001940625449111354,
|
| 2026 |
+
"loss": 0.2666,
|
| 2027 |
+
"step": 1425
|
| 2028 |
+
},
|
| 2029 |
+
{
|
| 2030 |
+
"epoch": 1.1468671679197995,
|
| 2031 |
+
"grad_norm": 0.15302354097366333,
|
| 2032 |
+
"learning_rate": 0.00019252778018652447,
|
| 2033 |
+
"loss": 0.272,
|
| 2034 |
+
"step": 1430
|
| 2035 |
+
},
|
| 2036 |
+
{
|
| 2037 |
+
"epoch": 1.1508771929824562,
|
| 2038 |
+
"grad_norm": 0.17536306381225586,
|
| 2039 |
+
"learning_rate": 0.0001909952989576855,
|
| 2040 |
+
"loss": 0.3183,
|
| 2041 |
+
"step": 1435
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 1.1548872180451129,
|
| 2045 |
+
"grad_norm": 0.15911608934402466,
|
| 2046 |
+
"learning_rate": 0.00018946516211341695,
|
| 2047 |
+
"loss": 0.3022,
|
| 2048 |
+
"step": 1440
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 1.1588972431077695,
|
| 2052 |
+
"grad_norm": 0.15123997628688812,
|
| 2053 |
+
"learning_rate": 0.00018793743044936972,
|
| 2054 |
+
"loss": 0.289,
|
| 2055 |
+
"step": 1445
|
| 2056 |
+
},
|
| 2057 |
+
{
|
| 2058 |
+
"epoch": 1.162907268170426,
|
| 2059 |
+
"grad_norm": 0.16768620908260345,
|
| 2060 |
+
"learning_rate": 0.00018641216466563183,
|
| 2061 |
+
"loss": 0.2694,
|
| 2062 |
+
"step": 1450
|
| 2063 |
+
},
|
| 2064 |
+
{
|
| 2065 |
+
"epoch": 1.1669172932330827,
|
| 2066 |
+
"grad_norm": 0.16061028838157654,
|
| 2067 |
+
"learning_rate": 0.00018488942536431618,
|
| 2068 |
+
"loss": 0.2824,
|
| 2069 |
+
"step": 1455
|
| 2070 |
+
},
|
| 2071 |
+
{
|
| 2072 |
+
"epoch": 1.1709273182957394,
|
| 2073 |
+
"grad_norm": 0.15412503480911255,
|
| 2074 |
+
"learning_rate": 0.00018336927304715356,
|
| 2075 |
+
"loss": 0.3025,
|
| 2076 |
+
"step": 1460
|
| 2077 |
+
},
|
| 2078 |
+
{
|
| 2079 |
+
"epoch": 1.174937343358396,
|
| 2080 |
+
"grad_norm": 0.17830166220664978,
|
| 2081 |
+
"learning_rate": 0.0001818517681130879,
|
| 2082 |
+
"loss": 0.3046,
|
| 2083 |
+
"step": 1465
|
| 2084 |
+
},
|
| 2085 |
+
{
|
| 2086 |
+
"epoch": 1.1789473684210527,
|
| 2087 |
+
"grad_norm": 0.17978624999523163,
|
| 2088 |
+
"learning_rate": 0.00018033697085587696,
|
| 2089 |
+
"loss": 0.2966,
|
| 2090 |
+
"step": 1470
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"epoch": 1.1829573934837092,
|
| 2094 |
+
"grad_norm": 0.1743180751800537,
|
| 2095 |
+
"learning_rate": 0.00017882494146169678,
|
| 2096 |
+
"loss": 0.3072,
|
| 2097 |
+
"step": 1475
|
| 2098 |
+
},
|
| 2099 |
+
{
|
| 2100 |
+
"epoch": 1.186967418546366,
|
| 2101 |
+
"grad_norm": 0.16470114886760712,
|
| 2102 |
+
"learning_rate": 0.00017731574000674996,
|
| 2103 |
+
"loss": 0.2797,
|
| 2104 |
+
"step": 1480
|
| 2105 |
+
},
|
| 2106 |
+
{
|
| 2107 |
+
"epoch": 1.1909774436090226,
|
| 2108 |
+
"grad_norm": 0.18040384352207184,
|
| 2109 |
+
"learning_rate": 0.0001758094264548792,
|
| 2110 |
+
"loss": 0.3093,
|
| 2111 |
+
"step": 1485
|
| 2112 |
+
},
|
| 2113 |
+
{
|
| 2114 |
+
"epoch": 1.1949874686716793,
|
| 2115 |
+
"grad_norm": 0.17426797747612,
|
| 2116 |
+
"learning_rate": 0.00017430606065518435,
|
| 2117 |
+
"loss": 0.318,
|
| 2118 |
+
"step": 1490
|
| 2119 |
+
},
|
| 2120 |
+
{
|
| 2121 |
+
"epoch": 1.198997493734336,
|
| 2122 |
+
"grad_norm": 0.1632205694913864,
|
| 2123 |
+
"learning_rate": 0.00017280570233964485,
|
| 2124 |
+
"loss": 0.2691,
|
| 2125 |
+
"step": 1495
|
| 2126 |
+
},
|
| 2127 |
+
{
|
| 2128 |
+
"epoch": 1.2030075187969924,
|
| 2129 |
+
"grad_norm": 0.1813742071390152,
|
| 2130 |
+
"learning_rate": 0.00017130841112074602,
|
| 2131 |
+
"loss": 0.2769,
|
| 2132 |
+
"step": 1500
|
| 2133 |
+
},
|
| 2134 |
+
{
|
| 2135 |
+
"epoch": 1.207017543859649,
|
| 2136 |
+
"grad_norm": 0.17400313913822174,
|
| 2137 |
+
"learning_rate": 0.00016981424648911112,
|
| 2138 |
+
"loss": 0.2723,
|
| 2139 |
+
"step": 1505
|
| 2140 |
+
},
|
| 2141 |
+
{
|
| 2142 |
+
"epoch": 1.2110275689223058,
|
| 2143 |
+
"grad_norm": 0.15530717372894287,
|
| 2144 |
+
"learning_rate": 0.0001683232678111371,
|
| 2145 |
+
"loss": 0.2971,
|
| 2146 |
+
"step": 1510
|
| 2147 |
+
},
|
| 2148 |
+
{
|
| 2149 |
+
"epoch": 1.2150375939849625,
|
| 2150 |
+
"grad_norm": 0.17954862117767334,
|
| 2151 |
+
"learning_rate": 0.00016683553432663616,
|
| 2152 |
+
"loss": 0.2866,
|
| 2153 |
+
"step": 1515
|
| 2154 |
+
},
|
| 2155 |
+
{
|
| 2156 |
+
"epoch": 1.2190476190476192,
|
| 2157 |
+
"grad_norm": 0.16163484752178192,
|
| 2158 |
+
"learning_rate": 0.0001653511051464819,
|
| 2159 |
+
"loss": 0.2557,
|
| 2160 |
+
"step": 1520
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"epoch": 1.2230576441102756,
|
| 2164 |
+
"grad_norm": 0.16374574601650238,
|
| 2165 |
+
"learning_rate": 0.00016387003925026077,
|
| 2166 |
+
"loss": 0.2956,
|
| 2167 |
+
"step": 1525
|
| 2168 |
+
},
|
| 2169 |
+
{
|
| 2170 |
+
"epoch": 1.2270676691729323,
|
| 2171 |
+
"grad_norm": 0.17792537808418274,
|
| 2172 |
+
"learning_rate": 0.0001623923954839287,
|
| 2173 |
+
"loss": 0.2777,
|
| 2174 |
+
"step": 1530
|
| 2175 |
+
},
|
| 2176 |
+
{
|
| 2177 |
+
"epoch": 1.231077694235589,
|
| 2178 |
+
"grad_norm": 0.18563567101955414,
|
| 2179 |
+
"learning_rate": 0.00016091823255747292,
|
| 2180 |
+
"loss": 0.301,
|
| 2181 |
+
"step": 1535
|
| 2182 |
+
},
|
| 2183 |
+
{
|
| 2184 |
+
"epoch": 1.2350877192982457,
|
| 2185 |
+
"grad_norm": 0.15728415548801422,
|
| 2186 |
+
"learning_rate": 0.00015944760904257942,
|
| 2187 |
+
"loss": 0.2784,
|
| 2188 |
+
"step": 1540
|
| 2189 |
+
},
|
| 2190 |
+
{
|
| 2191 |
+
"epoch": 1.2390977443609024,
|
| 2192 |
+
"grad_norm": 0.15651676058769226,
|
| 2193 |
+
"learning_rate": 0.00015798058337030551,
|
| 2194 |
+
"loss": 0.2876,
|
| 2195 |
+
"step": 1545
|
| 2196 |
+
},
|
| 2197 |
+
{
|
| 2198 |
+
"epoch": 1.2431077694235588,
|
| 2199 |
+
"grad_norm": 0.1799081265926361,
|
| 2200 |
+
"learning_rate": 0.00015651721382875874,
|
| 2201 |
+
"loss": 0.2919,
|
| 2202 |
+
"step": 1550
|
| 2203 |
+
},
|
| 2204 |
+
{
|
| 2205 |
+
"epoch": 1.2471177944862155,
|
| 2206 |
+
"grad_norm": 0.18436437845230103,
|
| 2207 |
+
"learning_rate": 0.00015505755856078048,
|
| 2208 |
+
"loss": 0.2934,
|
| 2209 |
+
"step": 1555
|
| 2210 |
+
},
|
| 2211 |
+
{
|
| 2212 |
+
"epoch": 1.2511278195488722,
|
| 2213 |
+
"grad_norm": 0.16269451379776,
|
| 2214 |
+
"learning_rate": 0.00015360167556163583,
|
| 2215 |
+
"loss": 0.2719,
|
| 2216 |
+
"step": 1560
|
| 2217 |
+
},
|
| 2218 |
+
{
|
| 2219 |
+
"epoch": 1.2551378446115289,
|
| 2220 |
+
"grad_norm": 0.15247991681098938,
|
| 2221 |
+
"learning_rate": 0.0001521496226767098,
|
| 2222 |
+
"loss": 0.2678,
|
| 2223 |
+
"step": 1565
|
| 2224 |
+
},
|
| 2225 |
+
{
|
| 2226 |
+
"epoch": 1.2591478696741856,
|
| 2227 |
+
"grad_norm": 0.1753372848033905,
|
| 2228 |
+
"learning_rate": 0.0001507014575992085,
|
| 2229 |
+
"loss": 0.3065,
|
| 2230 |
+
"step": 1570
|
| 2231 |
+
},
|
| 2232 |
+
{
|
| 2233 |
+
"epoch": 1.263157894736842,
|
| 2234 |
+
"grad_norm": 0.16919946670532227,
|
| 2235 |
+
"learning_rate": 0.0001492572378678669,
|
| 2236 |
+
"loss": 0.2659,
|
| 2237 |
+
"step": 1575
|
| 2238 |
+
},
|
| 2239 |
+
{
|
| 2240 |
+
"epoch": 1.2671679197994987,
|
| 2241 |
+
"grad_norm": 0.17600032687187195,
|
| 2242 |
+
"learning_rate": 0.00014781702086466314,
|
| 2243 |
+
"loss": 0.2987,
|
| 2244 |
+
"step": 1580
|
| 2245 |
+
},
|
| 2246 |
+
{
|
| 2247 |
+
"epoch": 1.2711779448621554,
|
| 2248 |
+
"grad_norm": 0.15909601747989655,
|
| 2249 |
+
"learning_rate": 0.000146380863812538,
|
| 2250 |
+
"loss": 0.2811,
|
| 2251 |
+
"step": 1585
|
| 2252 |
+
},
|
| 2253 |
+
{
|
| 2254 |
+
"epoch": 1.275187969924812,
|
| 2255 |
+
"grad_norm": 0.165226012468338,
|
| 2256 |
+
"learning_rate": 0.00014494882377312168,
|
| 2257 |
+
"loss": 0.2682,
|
| 2258 |
+
"step": 1590
|
| 2259 |
+
},
|
| 2260 |
+
{
|
| 2261 |
+
"epoch": 1.2791979949874688,
|
| 2262 |
+
"grad_norm": 0.17594917118549347,
|
| 2263 |
+
"learning_rate": 0.00014352095764446675,
|
| 2264 |
+
"loss": 0.2602,
|
| 2265 |
+
"step": 1595
|
| 2266 |
+
},
|
| 2267 |
+
{
|
| 2268 |
+
"epoch": 1.2832080200501252,
|
| 2269 |
+
"grad_norm": 0.154309943318367,
|
| 2270 |
+
"learning_rate": 0.0001420973221587869,
|
| 2271 |
+
"loss": 0.2579,
|
| 2272 |
+
"step": 1600
|
| 2273 |
+
},
|
| 2274 |
+
{
|
| 2275 |
+
"epoch": 1.287218045112782,
|
| 2276 |
+
"grad_norm": 0.16517868638038635,
|
| 2277 |
+
"learning_rate": 0.00014067797388020353,
|
| 2278 |
+
"loss": 0.2929,
|
| 2279 |
+
"step": 1605
|
| 2280 |
+
},
|
| 2281 |
+
{
|
| 2282 |
+
"epoch": 1.2912280701754386,
|
| 2283 |
+
"grad_norm": 0.1717553734779358,
|
| 2284 |
+
"learning_rate": 0.00013926296920249796,
|
| 2285 |
+
"loss": 0.2849,
|
| 2286 |
+
"step": 1610
|
| 2287 |
+
},
|
| 2288 |
+
{
|
| 2289 |
+
"epoch": 1.2952380952380953,
|
| 2290 |
+
"grad_norm": 0.16944153606891632,
|
| 2291 |
+
"learning_rate": 0.0001378523643468708,
|
| 2292 |
+
"loss": 0.2527,
|
| 2293 |
+
"step": 1615
|
| 2294 |
+
},
|
| 2295 |
+
{
|
| 2296 |
+
"epoch": 1.299248120300752,
|
| 2297 |
+
"grad_norm": 0.16444195806980133,
|
| 2298 |
+
"learning_rate": 0.00013644621535970804,
|
| 2299 |
+
"loss": 0.2665,
|
| 2300 |
+
"step": 1620
|
| 2301 |
+
},
|
| 2302 |
+
{
|
| 2303 |
+
"epoch": 1.3032581453634084,
|
| 2304 |
+
"grad_norm": 0.17226529121398926,
|
| 2305 |
+
"learning_rate": 0.0001350445781103547,
|
| 2306 |
+
"loss": 0.2758,
|
| 2307 |
+
"step": 1625
|
| 2308 |
+
},
|
| 2309 |
+
{
|
| 2310 |
+
"epoch": 1.3072681704260651,
|
| 2311 |
+
"grad_norm": 0.18174055218696594,
|
| 2312 |
+
"learning_rate": 0.0001336475082888943,
|
| 2313 |
+
"loss": 0.2831,
|
| 2314 |
+
"step": 1630
|
| 2315 |
+
},
|
| 2316 |
+
{
|
| 2317 |
+
"epoch": 1.3112781954887218,
|
| 2318 |
+
"grad_norm": 0.16900789737701416,
|
| 2319 |
+
"learning_rate": 0.00013225506140393696,
|
| 2320 |
+
"loss": 0.2882,
|
| 2321 |
+
"step": 1635
|
| 2322 |
+
},
|
| 2323 |
+
{
|
| 2324 |
+
"epoch": 1.3152882205513785,
|
| 2325 |
+
"grad_norm": 0.16686789691448212,
|
| 2326 |
+
"learning_rate": 0.00013086729278041304,
|
| 2327 |
+
"loss": 0.2556,
|
| 2328 |
+
"step": 1640
|
| 2329 |
+
},
|
| 2330 |
+
{
|
| 2331 |
+
"epoch": 1.3192982456140352,
|
| 2332 |
+
"grad_norm": 0.16731151938438416,
|
| 2333 |
+
"learning_rate": 0.00012948425755737592,
|
| 2334 |
+
"loss": 0.2665,
|
| 2335 |
+
"step": 1645
|
| 2336 |
+
},
|
| 2337 |
+
{
|
| 2338 |
+
"epoch": 1.3233082706766917,
|
| 2339 |
+
"grad_norm": 0.17180107533931732,
|
| 2340 |
+
"learning_rate": 0.00012810601068581056,
|
| 2341 |
+
"loss": 0.3001,
|
| 2342 |
+
"step": 1650
|
| 2343 |
+
},
|
| 2344 |
+
{
|
| 2345 |
+
"epoch": 1.3273182957393483,
|
| 2346 |
+
"grad_norm": 0.1701911836862564,
|
| 2347 |
+
"learning_rate": 0.0001267326069264501,
|
| 2348 |
+
"loss": 0.2977,
|
| 2349 |
+
"step": 1655
|
| 2350 |
+
},
|
| 2351 |
+
{
|
| 2352 |
+
"epoch": 1.331328320802005,
|
| 2353 |
+
"grad_norm": 0.19208161532878876,
|
| 2354 |
+
"learning_rate": 0.0001253641008476007,
|
| 2355 |
+
"loss": 0.2752,
|
| 2356 |
+
"step": 1660
|
| 2357 |
+
},
|
| 2358 |
+
{
|
| 2359 |
+
"epoch": 1.3353383458646617,
|
| 2360 |
+
"grad_norm": 0.15200825035572052,
|
| 2361 |
+
"learning_rate": 0.00012400054682297298,
|
| 2362 |
+
"loss": 0.2875,
|
| 2363 |
+
"step": 1665
|
| 2364 |
+
},
|
| 2365 |
+
{
|
| 2366 |
+
"epoch": 1.3393483709273184,
|
| 2367 |
+
"grad_norm": 0.17044006288051605,
|
| 2368 |
+
"learning_rate": 0.00012264199902952154,
|
| 2369 |
+
"loss": 0.2501,
|
| 2370 |
+
"step": 1670
|
| 2371 |
+
},
|
| 2372 |
+
{
|
| 2373 |
+
"epoch": 1.3433583959899749,
|
| 2374 |
+
"grad_norm": 0.16576853394508362,
|
| 2375 |
+
"learning_rate": 0.00012128851144529257,
|
| 2376 |
+
"loss": 0.2758,
|
| 2377 |
+
"step": 1675
|
| 2378 |
+
},
|
| 2379 |
+
{
|
| 2380 |
+
"epoch": 1.3473684210526315,
|
| 2381 |
+
"grad_norm": 0.17262569069862366,
|
| 2382 |
+
"learning_rate": 0.00011994013784727947,
|
| 2383 |
+
"loss": 0.2661,
|
| 2384 |
+
"step": 1680
|
| 2385 |
+
},
|
| 2386 |
+
{
|
| 2387 |
+
"epoch": 1.3513784461152882,
|
| 2388 |
+
"grad_norm": 0.161615252494812,
|
| 2389 |
+
"learning_rate": 0.00011859693180928574,
|
| 2390 |
+
"loss": 0.2689,
|
| 2391 |
+
"step": 1685
|
| 2392 |
+
},
|
| 2393 |
+
{
|
| 2394 |
+
"epoch": 1.355388471177945,
|
| 2395 |
+
"grad_norm": 0.16010525822639465,
|
| 2396 |
+
"learning_rate": 0.00011725894669979639,
|
| 2397 |
+
"loss": 0.2623,
|
| 2398 |
+
"step": 1690
|
| 2399 |
+
},
|
| 2400 |
+
{
|
| 2401 |
+
"epoch": 1.3593984962406016,
|
| 2402 |
+
"grad_norm": 0.1819092184305191,
|
| 2403 |
+
"learning_rate": 0.00011592623567985783,
|
| 2404 |
+
"loss": 0.2667,
|
| 2405 |
+
"step": 1695
|
| 2406 |
+
},
|
| 2407 |
+
{
|
| 2408 |
+
"epoch": 1.363408521303258,
|
| 2409 |
+
"grad_norm": 0.16817572712898254,
|
| 2410 |
+
"learning_rate": 0.00011459885170096537,
|
| 2411 |
+
"loss": 0.2756,
|
| 2412 |
+
"step": 1700
|
| 2413 |
+
},
|
| 2414 |
+
{
|
| 2415 |
+
"epoch": 1.3674185463659148,
|
| 2416 |
+
"grad_norm": 0.17827191948890686,
|
| 2417 |
+
"learning_rate": 0.00011327684750295958,
|
| 2418 |
+
"loss": 0.3166,
|
| 2419 |
+
"step": 1705
|
| 2420 |
+
},
|
| 2421 |
+
{
|
| 2422 |
+
"epoch": 1.3714285714285714,
|
| 2423 |
+
"grad_norm": 0.15747743844985962,
|
| 2424 |
+
"learning_rate": 0.00011196027561193045,
|
| 2425 |
+
"loss": 0.2952,
|
| 2426 |
+
"step": 1710
|
| 2427 |
+
},
|
| 2428 |
+
{
|
| 2429 |
+
"epoch": 1.3754385964912281,
|
| 2430 |
+
"grad_norm": 0.17462944984436035,
|
| 2431 |
+
"learning_rate": 0.00011064918833813073,
|
| 2432 |
+
"loss": 0.2754,
|
| 2433 |
+
"step": 1715
|
| 2434 |
+
},
|
| 2435 |
+
{
|
| 2436 |
+
"epoch": 1.3794486215538848,
|
| 2437 |
+
"grad_norm": 0.16550135612487793,
|
| 2438 |
+
"learning_rate": 0.00010934363777389752,
|
| 2439 |
+
"loss": 0.2658,
|
| 2440 |
+
"step": 1720
|
| 2441 |
+
},
|
| 2442 |
+
{
|
| 2443 |
+
"epoch": 1.3834586466165413,
|
| 2444 |
+
"grad_norm": 0.19006794691085815,
|
| 2445 |
+
"learning_rate": 0.00010804367579158256,
|
| 2446 |
+
"loss": 0.2679,
|
| 2447 |
+
"step": 1725
|
| 2448 |
+
},
|
| 2449 |
+
{
|
| 2450 |
+
"epoch": 1.387468671679198,
|
| 2451 |
+
"grad_norm": 0.2023632973432541,
|
| 2452 |
+
"learning_rate": 0.00010674935404149075,
|
| 2453 |
+
"loss": 0.2739,
|
| 2454 |
+
"step": 1730
|
| 2455 |
+
},
|
| 2456 |
+
{
|
| 2457 |
+
"epoch": 1.3914786967418546,
|
| 2458 |
+
"grad_norm": 0.17994622886180878,
|
| 2459 |
+
"learning_rate": 0.00010546072394982872,
|
| 2460 |
+
"loss": 0.2585,
|
| 2461 |
+
"step": 1735
|
| 2462 |
+
},
|
| 2463 |
+
{
|
| 2464 |
+
"epoch": 1.3954887218045113,
|
| 2465 |
+
"grad_norm": 0.1834658533334732,
|
| 2466 |
+
"learning_rate": 0.00010417783671666114,
|
| 2467 |
+
"loss": 0.2567,
|
| 2468 |
+
"step": 1740
|
| 2469 |
+
},
|
| 2470 |
+
{
|
| 2471 |
+
"epoch": 1.399498746867168,
|
| 2472 |
+
"grad_norm": 0.16949954628944397,
|
| 2473 |
+
"learning_rate": 0.00010290074331387617,
|
| 2474 |
+
"loss": 0.2462,
|
| 2475 |
+
"step": 1745
|
| 2476 |
+
},
|
| 2477 |
+
{
|
| 2478 |
+
"epoch": 1.4035087719298245,
|
| 2479 |
+
"grad_norm": 0.17096513509750366,
|
| 2480 |
+
"learning_rate": 0.00010162949448316089,
|
| 2481 |
+
"loss": 0.2818,
|
| 2482 |
+
"step": 1750
|
| 2483 |
+
},
|
| 2484 |
+
{
|
| 2485 |
+
"epoch": 1.4075187969924812,
|
| 2486 |
+
"grad_norm": 0.16146983206272125,
|
| 2487 |
+
"learning_rate": 0.00010036414073398479,
|
| 2488 |
+
"loss": 0.2691,
|
| 2489 |
+
"step": 1755
|
| 2490 |
+
},
|
| 2491 |
+
{
|
| 2492 |
+
"epoch": 1.4115288220551379,
|
| 2493 |
+
"grad_norm": 0.15868982672691345,
|
| 2494 |
+
"learning_rate": 9.910473234159286e-05,
|
| 2495 |
+
"loss": 0.2857,
|
| 2496 |
+
"step": 1760
|
| 2497 |
+
},
|
| 2498 |
+
{
|
| 2499 |
+
"epoch": 1.4155388471177945,
|
| 2500 |
+
"grad_norm": 0.1619461625814438,
|
| 2501 |
+
"learning_rate": 9.785131934500818e-05,
|
| 2502 |
+
"loss": 0.2807,
|
| 2503 |
+
"step": 1765
|
| 2504 |
+
},
|
| 2505 |
+
{
|
| 2506 |
+
"epoch": 1.4195488721804512,
|
| 2507 |
+
"grad_norm": 0.1557190865278244,
|
| 2508 |
+
"learning_rate": 9.660395154504401e-05,
|
| 2509 |
+
"loss": 0.2548,
|
| 2510 |
+
"step": 1770
|
| 2511 |
+
},
|
| 2512 |
+
{
|
| 2513 |
+
"epoch": 1.4235588972431077,
|
| 2514 |
+
"grad_norm": 0.17061957716941833,
|
| 2515 |
+
"learning_rate": 9.536267850232472e-05,
|
| 2516 |
+
"loss": 0.2801,
|
| 2517 |
+
"step": 1775
|
| 2518 |
+
},
|
| 2519 |
+
{
|
| 2520 |
+
"epoch": 1.4275689223057644,
|
| 2521 |
+
"grad_norm": 0.17641028761863708,
|
| 2522 |
+
"learning_rate": 9.412754953531663e-05,
|
| 2523 |
+
"loss": 0.2924,
|
| 2524 |
+
"step": 1780
|
| 2525 |
+
},
|
| 2526 |
+
{
|
| 2527 |
+
"epoch": 1.431578947368421,
|
| 2528 |
+
"grad_norm": 0.15920333564281464,
|
| 2529 |
+
"learning_rate": 9.289861371836885e-05,
|
| 2530 |
+
"loss": 0.2563,
|
| 2531 |
+
"step": 1785
|
| 2532 |
+
},
|
| 2533 |
+
{
|
| 2534 |
+
"epoch": 1.4355889724310777,
|
| 2535 |
+
"grad_norm": 0.15198518335819244,
|
| 2536 |
+
"learning_rate": 9.167591987976312e-05,
|
| 2537 |
+
"loss": 0.2357,
|
| 2538 |
+
"step": 1790
|
| 2539 |
+
},
|
| 2540 |
+
{
|
| 2541 |
+
"epoch": 1.4395989974937344,
|
| 2542 |
+
"grad_norm": 0.1758808046579361,
|
| 2543 |
+
"learning_rate": 9.045951659977397e-05,
|
| 2544 |
+
"loss": 0.2535,
|
| 2545 |
+
"step": 1795
|
| 2546 |
+
},
|
| 2547 |
+
{
|
| 2548 |
+
"epoch": 1.443609022556391,
|
| 2549 |
+
"grad_norm": 0.18127872049808502,
|
| 2550 |
+
"learning_rate": 8.924945220873823e-05,
|
| 2551 |
+
"loss": 0.2888,
|
| 2552 |
+
"step": 1800
|
| 2553 |
+
},
|
| 2554 |
+
{
|
| 2555 |
+
"epoch": 1.4476190476190476,
|
| 2556 |
+
"grad_norm": 0.1601109355688095,
|
| 2557 |
+
"learning_rate": 8.804577478513492e-05,
|
| 2558 |
+
"loss": 0.2528,
|
| 2559 |
+
"step": 1805
|
| 2560 |
+
},
|
| 2561 |
+
{
|
| 2562 |
+
"epoch": 1.4516290726817043,
|
| 2563 |
+
"grad_norm": 0.16406309604644775,
|
| 2564 |
+
"learning_rate": 8.684853215367522e-05,
|
| 2565 |
+
"loss": 0.2766,
|
| 2566 |
+
"step": 1810
|
| 2567 |
+
},
|
| 2568 |
+
{
|
| 2569 |
+
"epoch": 1.455639097744361,
|
| 2570 |
+
"grad_norm": 0.158805251121521,
|
| 2571 |
+
"learning_rate": 8.565777188340207e-05,
|
| 2572 |
+
"loss": 0.2747,
|
| 2573 |
+
"step": 1815
|
| 2574 |
+
},
|
| 2575 |
+
{
|
| 2576 |
+
"epoch": 1.4596491228070176,
|
| 2577 |
+
"grad_norm": 0.17480885982513428,
|
| 2578 |
+
"learning_rate": 8.44735412857999e-05,
|
| 2579 |
+
"loss": 0.3079,
|
| 2580 |
+
"step": 1820
|
| 2581 |
+
},
|
| 2582 |
+
{
|
| 2583 |
+
"epoch": 1.463659147869674,
|
| 2584 |
+
"grad_norm": 0.1617252379655838,
|
| 2585 |
+
"learning_rate": 8.329588741291535e-05,
|
| 2586 |
+
"loss": 0.2701,
|
| 2587 |
+
"step": 1825
|
| 2588 |
+
},
|
| 2589 |
+
{
|
| 2590 |
+
"epoch": 1.4676691729323308,
|
| 2591 |
+
"grad_norm": 0.1652277261018753,
|
| 2592 |
+
"learning_rate": 8.212485705548756e-05,
|
| 2593 |
+
"loss": 0.25,
|
| 2594 |
+
"step": 1830
|
| 2595 |
+
},
|
| 2596 |
+
{
|
| 2597 |
+
"epoch": 1.4716791979949875,
|
| 2598 |
+
"grad_norm": 0.16213206946849823,
|
| 2599 |
+
"learning_rate": 8.096049674108877e-05,
|
| 2600 |
+
"loss": 0.2683,
|
| 2601 |
+
"step": 1835
|
| 2602 |
+
},
|
| 2603 |
+
{
|
| 2604 |
+
"epoch": 1.4756892230576442,
|
| 2605 |
+
"grad_norm": 0.167341947555542,
|
| 2606 |
+
"learning_rate": 7.980285273227633e-05,
|
| 2607 |
+
"loss": 0.2724,
|
| 2608 |
+
"step": 1840
|
| 2609 |
+
},
|
| 2610 |
+
{
|
| 2611 |
+
"epoch": 1.4796992481203008,
|
| 2612 |
+
"grad_norm": 0.1590610146522522,
|
| 2613 |
+
"learning_rate": 7.865197102475388e-05,
|
| 2614 |
+
"loss": 0.2518,
|
| 2615 |
+
"step": 1845
|
| 2616 |
+
},
|
| 2617 |
+
{
|
| 2618 |
+
"epoch": 1.4837092731829573,
|
| 2619 |
+
"grad_norm": 0.16797657310962677,
|
| 2620 |
+
"learning_rate": 7.75078973455445e-05,
|
| 2621 |
+
"loss": 0.2651,
|
| 2622 |
+
"step": 1850
|
| 2623 |
+
},
|
| 2624 |
+
{
|
| 2625 |
+
"epoch": 1.487719298245614,
|
| 2626 |
+
"grad_norm": 0.17504721879959106,
|
| 2627 |
+
"learning_rate": 7.637067715117327e-05,
|
| 2628 |
+
"loss": 0.3017,
|
| 2629 |
+
"step": 1855
|
| 2630 |
+
},
|
| 2631 |
+
{
|
| 2632 |
+
"epoch": 1.4917293233082707,
|
| 2633 |
+
"grad_norm": 0.16186365485191345,
|
| 2634 |
+
"learning_rate": 7.52403556258617e-05,
|
| 2635 |
+
"loss": 0.2344,
|
| 2636 |
+
"step": 1860
|
| 2637 |
+
},
|
| 2638 |
+
{
|
| 2639 |
+
"epoch": 1.4957393483709274,
|
| 2640 |
+
"grad_norm": 0.17528776824474335,
|
| 2641 |
+
"learning_rate": 7.41169776797322e-05,
|
| 2642 |
+
"loss": 0.264,
|
| 2643 |
+
"step": 1865
|
| 2644 |
+
},
|
| 2645 |
+
{
|
| 2646 |
+
"epoch": 1.4989473684210526,
|
| 2647 |
+
"eval_loss": 0.2417936623096466,
|
| 2648 |
+
"eval_runtime": 8.3922,
|
| 2649 |
+
"eval_samples_per_second": 6.554,
|
| 2650 |
+
"eval_steps_per_second": 0.834,
|
| 2651 |
+
"step": 1869
|
| 2652 |
+
},
|
| 2653 |
+
{
|
| 2654 |
+
"epoch": 1.499749373433584,
|
| 2655 |
+
"grad_norm": 0.1840500384569168,
|
| 2656 |
+
"learning_rate": 7.300058794702352e-05,
|
| 2657 |
+
"loss": 0.27,
|
| 2658 |
+
"step": 1870
|
| 2659 |
+
},
|
| 2660 |
+
{
|
| 2661 |
+
"epoch": 1.5037593984962405,
|
| 2662 |
+
"grad_norm": 0.1910168081521988,
|
| 2663 |
+
"learning_rate": 7.189123078431784e-05,
|
| 2664 |
+
"loss": 0.2823,
|
| 2665 |
+
"step": 1875
|
| 2666 |
+
},
|
| 2667 |
+
{
|
| 2668 |
+
"epoch": 1.5077694235588972,
|
| 2669 |
+
"grad_norm": 0.16170190274715424,
|
| 2670 |
+
"learning_rate": 7.078895026877804e-05,
|
| 2671 |
+
"loss": 0.253,
|
| 2672 |
+
"step": 1880
|
| 2673 |
+
},
|
| 2674 |
+
{
|
| 2675 |
+
"epoch": 1.511779448621554,
|
| 2676 |
+
"grad_norm": 0.17806147038936615,
|
| 2677 |
+
"learning_rate": 6.969379019639635e-05,
|
| 2678 |
+
"loss": 0.2836,
|
| 2679 |
+
"step": 1885
|
| 2680 |
+
},
|
| 2681 |
+
{
|
| 2682 |
+
"epoch": 1.5157894736842106,
|
| 2683 |
+
"grad_norm": 0.15123490989208221,
|
| 2684 |
+
"learning_rate": 6.860579408025436e-05,
|
| 2685 |
+
"loss": 0.2495,
|
| 2686 |
+
"step": 1890
|
| 2687 |
+
},
|
| 2688 |
+
{
|
| 2689 |
+
"epoch": 1.5197994987468673,
|
| 2690 |
+
"grad_norm": 0.16695557534694672,
|
| 2691 |
+
"learning_rate": 6.752500514879437e-05,
|
| 2692 |
+
"loss": 0.2773,
|
| 2693 |
+
"step": 1895
|
| 2694 |
+
},
|
| 2695 |
+
{
|
| 2696 |
+
"epoch": 1.5238095238095237,
|
| 2697 |
+
"grad_norm": 0.16708271205425262,
|
| 2698 |
+
"learning_rate": 6.645146634410151e-05,
|
| 2699 |
+
"loss": 0.252,
|
| 2700 |
+
"step": 1900
|
| 2701 |
+
},
|
| 2702 |
+
{
|
| 2703 |
+
"epoch": 1.5278195488721804,
|
| 2704 |
+
"grad_norm": 0.1510598361492157,
|
| 2705 |
+
"learning_rate": 6.538522032019759e-05,
|
| 2706 |
+
"loss": 0.2373,
|
| 2707 |
+
"step": 1905
|
| 2708 |
+
},
|
| 2709 |
+
{
|
| 2710 |
+
"epoch": 1.531829573934837,
|
| 2711 |
+
"grad_norm": 0.1686723679304123,
|
| 2712 |
+
"learning_rate": 6.432630944134654e-05,
|
| 2713 |
+
"loss": 0.2743,
|
| 2714 |
+
"step": 1910
|
| 2715 |
+
},
|
| 2716 |
+
{
|
| 2717 |
+
"epoch": 1.5358395989974938,
|
| 2718 |
+
"grad_norm": 0.16884025931358337,
|
| 2719 |
+
"learning_rate": 6.327477578037106e-05,
|
| 2720 |
+
"loss": 0.2673,
|
| 2721 |
+
"step": 1915
|
| 2722 |
+
},
|
| 2723 |
+
{
|
| 2724 |
+
"epoch": 1.5398496240601505,
|
| 2725 |
+
"grad_norm": 0.16477200388908386,
|
| 2726 |
+
"learning_rate": 6.223066111698111e-05,
|
| 2727 |
+
"loss": 0.2713,
|
| 2728 |
+
"step": 1920
|
| 2729 |
+
},
|
| 2730 |
+
{
|
| 2731 |
+
"epoch": 1.543859649122807,
|
| 2732 |
+
"grad_norm": 0.18342123925685883,
|
| 2733 |
+
"learning_rate": 6.119400693611357e-05,
|
| 2734 |
+
"loss": 0.2791,
|
| 2735 |
+
"step": 1925
|
| 2736 |
+
},
|
| 2737 |
+
{
|
| 2738 |
+
"epoch": 1.5478696741854636,
|
| 2739 |
+
"grad_norm": 0.16829492151737213,
|
| 2740 |
+
"learning_rate": 6.0164854426284447e-05,
|
| 2741 |
+
"loss": 0.2702,
|
| 2742 |
+
"step": 1930
|
| 2743 |
+
},
|
| 2744 |
+
{
|
| 2745 |
+
"epoch": 1.5518796992481203,
|
| 2746 |
+
"grad_norm": 0.16608929634094238,
|
| 2747 |
+
"learning_rate": 5.914324447795186e-05,
|
| 2748 |
+
"loss": 0.2741,
|
| 2749 |
+
"step": 1935
|
| 2750 |
+
},
|
| 2751 |
+
{
|
| 2752 |
+
"epoch": 1.555889724310777,
|
| 2753 |
+
"grad_norm": 0.16408754885196686,
|
| 2754 |
+
"learning_rate": 5.8129217681891886e-05,
|
| 2755 |
+
"loss": 0.2661,
|
| 2756 |
+
"step": 1940
|
| 2757 |
+
},
|
| 2758 |
+
{
|
| 2759 |
+
"epoch": 1.5598997493734337,
|
| 2760 |
+
"grad_norm": 0.1709819734096527,
|
| 2761 |
+
"learning_rate": 5.7122814327585316e-05,
|
| 2762 |
+
"loss": 0.2557,
|
| 2763 |
+
"step": 1945
|
| 2764 |
+
},
|
| 2765 |
+
{
|
| 2766 |
+
"epoch": 1.5639097744360901,
|
| 2767 |
+
"grad_norm": 0.187583327293396,
|
| 2768 |
+
"learning_rate": 5.612407440161721e-05,
|
| 2769 |
+
"loss": 0.2623,
|
| 2770 |
+
"step": 1950
|
| 2771 |
+
},
|
| 2772 |
+
{
|
| 2773 |
+
"epoch": 1.5679197994987468,
|
| 2774 |
+
"grad_norm": 0.16979658603668213,
|
| 2775 |
+
"learning_rate": 5.513303758608804e-05,
|
| 2776 |
+
"loss": 0.2615,
|
| 2777 |
+
"step": 1955
|
| 2778 |
+
},
|
| 2779 |
+
{
|
| 2780 |
+
"epoch": 1.5719298245614035,
|
| 2781 |
+
"grad_norm": 0.13367925584316254,
|
| 2782 |
+
"learning_rate": 5.414974325703686e-05,
|
| 2783 |
+
"loss": 0.2276,
|
| 2784 |
+
"step": 1960
|
| 2785 |
+
},
|
| 2786 |
+
{
|
| 2787 |
+
"epoch": 1.5759398496240602,
|
| 2788 |
+
"grad_norm": 0.17686474323272705,
|
| 2789 |
+
"learning_rate": 5.317423048287717e-05,
|
| 2790 |
+
"loss": 0.2444,
|
| 2791 |
+
"step": 1965
|
| 2792 |
+
},
|
| 2793 |
+
{
|
| 2794 |
+
"epoch": 1.5799498746867169,
|
| 2795 |
+
"grad_norm": 0.17005467414855957,
|
| 2796 |
+
"learning_rate": 5.220653802284439e-05,
|
| 2797 |
+
"loss": 0.2581,
|
| 2798 |
+
"step": 1970
|
| 2799 |
+
},
|
| 2800 |
+
{
|
| 2801 |
+
"epoch": 1.5839598997493733,
|
| 2802 |
+
"grad_norm": 0.1664769947528839,
|
| 2803 |
+
"learning_rate": 5.124670432545578e-05,
|
| 2804 |
+
"loss": 0.2628,
|
| 2805 |
+
"step": 1975
|
| 2806 |
+
},
|
| 2807 |
+
{
|
| 2808 |
+
"epoch": 1.58796992481203,
|
| 2809 |
+
"grad_norm": 0.17945070564746857,
|
| 2810 |
+
"learning_rate": 5.0294767526983006e-05,
|
| 2811 |
+
"loss": 0.276,
|
| 2812 |
+
"step": 1980
|
| 2813 |
+
},
|
| 2814 |
+
{
|
| 2815 |
+
"epoch": 1.5919799498746867,
|
| 2816 |
+
"grad_norm": 0.1758703887462616,
|
| 2817 |
+
"learning_rate": 4.9350765449936915e-05,
|
| 2818 |
+
"loss": 0.2744,
|
| 2819 |
+
"step": 1985
|
| 2820 |
+
},
|
| 2821 |
+
{
|
| 2822 |
+
"epoch": 1.5959899749373434,
|
| 2823 |
+
"grad_norm": 0.1651277393102646,
|
| 2824 |
+
"learning_rate": 4.84147356015647e-05,
|
| 2825 |
+
"loss": 0.2872,
|
| 2826 |
+
"step": 1990
|
| 2827 |
+
},
|
| 2828 |
+
{
|
| 2829 |
+
"epoch": 1.6,
|
| 2830 |
+
"grad_norm": 0.16701501607894897,
|
| 2831 |
+
"learning_rate": 4.748671517235948e-05,
|
| 2832 |
+
"loss": 0.2676,
|
| 2833 |
+
"step": 1995
|
| 2834 |
+
},
|
| 2835 |
+
{
|
| 2836 |
+
"epoch": 1.6040100250626566,
|
| 2837 |
+
"grad_norm": 0.16929472982883453,
|
| 2838 |
+
"learning_rate": 4.656674103458291e-05,
|
| 2839 |
+
"loss": 0.2498,
|
| 2840 |
+
"step": 2000
|
| 2841 |
+
},
|
| 2842 |
+
{
|
| 2843 |
+
"epoch": 1.6080200501253132,
|
| 2844 |
+
"grad_norm": 0.15207117795944214,
|
| 2845 |
+
"learning_rate": 4.5654849740800145e-05,
|
| 2846 |
+
"loss": 0.26,
|
| 2847 |
+
"step": 2005
|
| 2848 |
+
},
|
| 2849 |
+
{
|
| 2850 |
+
"epoch": 1.61203007518797,
|
| 2851 |
+
"grad_norm": 0.15770703554153442,
|
| 2852 |
+
"learning_rate": 4.47510775224273e-05,
|
| 2853 |
+
"loss": 0.2644,
|
| 2854 |
+
"step": 2010
|
| 2855 |
+
},
|
| 2856 |
+
{
|
| 2857 |
+
"epoch": 1.6160401002506266,
|
| 2858 |
+
"grad_norm": 0.16916294395923615,
|
| 2859 |
+
"learning_rate": 4.38554602882921e-05,
|
| 2860 |
+
"loss": 0.3026,
|
| 2861 |
+
"step": 2015
|
| 2862 |
+
},
|
| 2863 |
+
{
|
| 2864 |
+
"epoch": 1.6200501253132833,
|
| 2865 |
+
"grad_norm": 0.16087712347507477,
|
| 2866 |
+
"learning_rate": 4.2968033623206974e-05,
|
| 2867 |
+
"loss": 0.2632,
|
| 2868 |
+
"step": 2020
|
| 2869 |
+
},
|
| 2870 |
+
{
|
| 2871 |
+
"epoch": 1.6240601503759398,
|
| 2872 |
+
"grad_norm": 0.18064051866531372,
|
| 2873 |
+
"learning_rate": 4.2088832786555456e-05,
|
| 2874 |
+
"loss": 0.2654,
|
| 2875 |
+
"step": 2025
|
| 2876 |
+
},
|
| 2877 |
+
{
|
| 2878 |
+
"epoch": 1.6280701754385964,
|
| 2879 |
+
"grad_norm": 0.17459240555763245,
|
| 2880 |
+
"learning_rate": 4.121789271089113e-05,
|
| 2881 |
+
"loss": 0.276,
|
| 2882 |
+
"step": 2030
|
| 2883 |
+
},
|
| 2884 |
+
{
|
| 2885 |
+
"epoch": 1.6320802005012531,
|
| 2886 |
+
"grad_norm": 0.15657363831996918,
|
| 2887 |
+
"learning_rate": 4.0355248000549434e-05,
|
| 2888 |
+
"loss": 0.241,
|
| 2889 |
+
"step": 2035
|
| 2890 |
+
},
|
| 2891 |
+
{
|
| 2892 |
+
"epoch": 1.6360902255639098,
|
| 2893 |
+
"grad_norm": 0.1620418131351471,
|
| 2894 |
+
"learning_rate": 3.950093293027318e-05,
|
| 2895 |
+
"loss": 0.2821,
|
| 2896 |
+
"step": 2040
|
| 2897 |
+
},
|
| 2898 |
+
{
|
| 2899 |
+
"epoch": 1.6401002506265665,
|
| 2900 |
+
"grad_norm": 0.18404340744018555,
|
| 2901 |
+
"learning_rate": 3.865498144385049e-05,
|
| 2902 |
+
"loss": 0.2935,
|
| 2903 |
+
"step": 2045
|
| 2904 |
+
},
|
| 2905 |
+
{
|
| 2906 |
+
"epoch": 1.644110275689223,
|
| 2907 |
+
"grad_norm": 0.16540312767028809,
|
| 2908 |
+
"learning_rate": 3.7817427152766094e-05,
|
| 2909 |
+
"loss": 0.2583,
|
| 2910 |
+
"step": 2050
|
| 2911 |
+
},
|
| 2912 |
+
{
|
| 2913 |
+
"epoch": 1.6481203007518797,
|
| 2914 |
+
"grad_norm": 0.16649490594863892,
|
| 2915 |
+
"learning_rate": 3.6988303334866054e-05,
|
| 2916 |
+
"loss": 0.2774,
|
| 2917 |
+
"step": 2055
|
| 2918 |
+
},
|
| 2919 |
+
{
|
| 2920 |
+
"epoch": 1.6521303258145363,
|
| 2921 |
+
"grad_norm": 0.1626511514186859,
|
| 2922 |
+
"learning_rate": 3.6167642933035486e-05,
|
| 2923 |
+
"loss": 0.2467,
|
| 2924 |
+
"step": 2060
|
| 2925 |
+
},
|
| 2926 |
+
{
|
| 2927 |
+
"epoch": 1.656140350877193,
|
| 2928 |
+
"grad_norm": 0.16354456543922424,
|
| 2929 |
+
"learning_rate": 3.535547855388963e-05,
|
| 2930 |
+
"loss": 0.2544,
|
| 2931 |
+
"step": 2065
|
| 2932 |
+
},
|
| 2933 |
+
{
|
| 2934 |
+
"epoch": 1.6601503759398497,
|
| 2935 |
+
"grad_norm": 0.16961799561977386,
|
| 2936 |
+
"learning_rate": 3.455184246647822e-05,
|
| 2937 |
+
"loss": 0.2389,
|
| 2938 |
+
"step": 2070
|
| 2939 |
+
},
|
| 2940 |
+
{
|
| 2941 |
+
"epoch": 1.6641604010025062,
|
| 2942 |
+
"grad_norm": 0.17093317210674286,
|
| 2943 |
+
"learning_rate": 3.375676660100366e-05,
|
| 2944 |
+
"loss": 0.2767,
|
| 2945 |
+
"step": 2075
|
| 2946 |
+
},
|
| 2947 |
+
{
|
| 2948 |
+
"epoch": 1.6681704260651629,
|
| 2949 |
+
"grad_norm": 0.1904112696647644,
|
| 2950 |
+
"learning_rate": 3.297028254755221e-05,
|
| 2951 |
+
"loss": 0.29,
|
| 2952 |
+
"step": 2080
|
| 2953 |
+
},
|
| 2954 |
+
{
|
| 2955 |
+
"epoch": 1.6721804511278195,
|
| 2956 |
+
"grad_norm": 0.18005013465881348,
|
| 2957 |
+
"learning_rate": 3.219242155483868e-05,
|
| 2958 |
+
"loss": 0.249,
|
| 2959 |
+
"step": 2085
|
| 2960 |
+
},
|
| 2961 |
+
{
|
| 2962 |
+
"epoch": 1.6761904761904762,
|
| 2963 |
+
"grad_norm": 0.1755589246749878,
|
| 2964 |
+
"learning_rate": 3.142321452896504e-05,
|
| 2965 |
+
"loss": 0.2956,
|
| 2966 |
+
"step": 2090
|
| 2967 |
+
},
|
| 2968 |
+
{
|
| 2969 |
+
"epoch": 1.680200501253133,
|
| 2970 |
+
"grad_norm": 0.15832673013210297,
|
| 2971 |
+
"learning_rate": 3.0662692032192516e-05,
|
| 2972 |
+
"loss": 0.2561,
|
| 2973 |
+
"step": 2095
|
| 2974 |
+
},
|
| 2975 |
+
{
|
| 2976 |
+
"epoch": 1.6842105263157894,
|
| 2977 |
+
"grad_norm": 0.15485481917858124,
|
| 2978 |
+
"learning_rate": 2.9910884281727225e-05,
|
| 2979 |
+
"loss": 0.2502,
|
| 2980 |
+
"step": 2100
|
| 2981 |
+
},
|
| 2982 |
+
{
|
| 2983 |
+
"epoch": 1.688220551378446,
|
| 2984 |
+
"grad_norm": 0.16627748310565948,
|
| 2985 |
+
"learning_rate": 2.916782114851918e-05,
|
| 2986 |
+
"loss": 0.2939,
|
| 2987 |
+
"step": 2105
|
| 2988 |
+
},
|
| 2989 |
+
{
|
| 2990 |
+
"epoch": 1.6922305764411028,
|
| 2991 |
+
"grad_norm": 0.17444783449172974,
|
| 2992 |
+
"learning_rate": 2.843353215607619e-05,
|
| 2993 |
+
"loss": 0.2731,
|
| 2994 |
+
"step": 2110
|
| 2995 |
+
},
|
| 2996 |
+
{
|
| 2997 |
+
"epoch": 1.6962406015037594,
|
| 2998 |
+
"grad_norm": 0.17445623874664307,
|
| 2999 |
+
"learning_rate": 2.7708046479290316e-05,
|
| 3000 |
+
"loss": 0.2752,
|
| 3001 |
+
"step": 2115
|
| 3002 |
+
},
|
| 3003 |
+
{
|
| 3004 |
+
"epoch": 1.7002506265664161,
|
| 3005 |
+
"grad_norm": 0.17704740166664124,
|
| 3006 |
+
"learning_rate": 2.6991392943278874e-05,
|
| 3007 |
+
"loss": 0.2423,
|
| 3008 |
+
"step": 2120
|
| 3009 |
+
},
|
| 3010 |
+
{
|
| 3011 |
+
"epoch": 1.7042606516290726,
|
| 3012 |
+
"grad_norm": 0.16555817425251007,
|
| 3013 |
+
"learning_rate": 2.6283600022238925e-05,
|
| 3014 |
+
"loss": 0.2811,
|
| 3015 |
+
"step": 2125
|
| 3016 |
+
},
|
| 3017 |
+
{
|
| 3018 |
+
"epoch": 1.7082706766917293,
|
| 3019 |
+
"grad_norm": 0.1633770912885666,
|
| 3020 |
+
"learning_rate": 2.5584695838316323e-05,
|
| 3021 |
+
"loss": 0.2463,
|
| 3022 |
+
"step": 2130
|
| 3023 |
+
},
|
| 3024 |
+
{
|
| 3025 |
+
"epoch": 1.712280701754386,
|
| 3026 |
+
"grad_norm": 0.17787782847881317,
|
| 3027 |
+
"learning_rate": 2.4894708160488057e-05,
|
| 3028 |
+
"loss": 0.266,
|
| 3029 |
+
"step": 2135
|
| 3030 |
+
},
|
| 3031 |
+
{
|
| 3032 |
+
"epoch": 1.7162907268170426,
|
| 3033 |
+
"grad_norm": 0.15796466171741486,
|
| 3034 |
+
"learning_rate": 2.4213664403458903e-05,
|
| 3035 |
+
"loss": 0.2574,
|
| 3036 |
+
"step": 2140
|
| 3037 |
+
},
|
| 3038 |
+
{
|
| 3039 |
+
"epoch": 1.7203007518796993,
|
| 3040 |
+
"grad_norm": 0.1675783395767212,
|
| 3041 |
+
"learning_rate": 2.3541591626572518e-05,
|
| 3042 |
+
"loss": 0.251,
|
| 3043 |
+
"step": 2145
|
| 3044 |
+
},
|
| 3045 |
+
{
|
| 3046 |
+
"epoch": 1.7243107769423558,
|
| 3047 |
+
"grad_norm": 0.17418669164180756,
|
| 3048 |
+
"learning_rate": 2.287851653273587e-05,
|
| 3049 |
+
"loss": 0.2552,
|
| 3050 |
+
"step": 2150
|
| 3051 |
+
},
|
| 3052 |
+
{
|
| 3053 |
+
"epoch": 1.7283208020050125,
|
| 3054 |
+
"grad_norm": 0.17587542533874512,
|
| 3055 |
+
"learning_rate": 2.222446546735868e-05,
|
| 3056 |
+
"loss": 0.272,
|
| 3057 |
+
"step": 2155
|
| 3058 |
+
},
|
| 3059 |
+
{
|
| 3060 |
+
"epoch": 1.7323308270676692,
|
| 3061 |
+
"grad_norm": 0.17269235849380493,
|
| 3062 |
+
"learning_rate": 2.1579464417306266e-05,
|
| 3063 |
+
"loss": 0.2586,
|
| 3064 |
+
"step": 2160
|
| 3065 |
+
},
|
| 3066 |
+
{
|
| 3067 |
+
"epoch": 1.7363408521303259,
|
| 3068 |
+
"grad_norm": 0.1612161546945572,
|
| 3069 |
+
"learning_rate": 2.0943539009867423e-05,
|
| 3070 |
+
"loss": 0.2783,
|
| 3071 |
+
"step": 2165
|
| 3072 |
+
},
|
| 3073 |
+
{
|
| 3074 |
+
"epoch": 1.7403508771929825,
|
| 3075 |
+
"grad_norm": 0.15821614861488342,
|
| 3076 |
+
"learning_rate": 2.0316714511736002e-05,
|
| 3077 |
+
"loss": 0.2634,
|
| 3078 |
+
"step": 2170
|
| 3079 |
+
},
|
| 3080 |
+
{
|
| 3081 |
+
"epoch": 1.744360902255639,
|
| 3082 |
+
"grad_norm": 0.17940925061702728,
|
| 3083 |
+
"learning_rate": 1.9699015828006788e-05,
|
| 3084 |
+
"loss": 0.2632,
|
| 3085 |
+
"step": 2175
|
| 3086 |
+
},
|
| 3087 |
+
{
|
| 3088 |
+
"epoch": 1.7483709273182957,
|
| 3089 |
+
"grad_norm": 0.1658666729927063,
|
| 3090 |
+
"learning_rate": 1.909046750118648e-05,
|
| 3091 |
+
"loss": 0.2666,
|
| 3092 |
+
"step": 2180
|
| 3093 |
+
},
|
| 3094 |
+
{
|
| 3095 |
+
"epoch": 1.7523809523809524,
|
| 3096 |
+
"grad_norm": 0.16111963987350464,
|
| 3097 |
+
"learning_rate": 1.849109371021815e-05,
|
| 3098 |
+
"loss": 0.2561,
|
| 3099 |
+
"step": 2185
|
| 3100 |
+
},
|
| 3101 |
+
{
|
| 3102 |
+
"epoch": 1.756390977443609,
|
| 3103 |
+
"grad_norm": 0.17400680482387543,
|
| 3104 |
+
"learning_rate": 1.7900918269520672e-05,
|
| 3105 |
+
"loss": 0.2499,
|
| 3106 |
+
"step": 2190
|
| 3107 |
+
},
|
| 3108 |
+
{
|
| 3109 |
+
"epoch": 1.7604010025062657,
|
| 3110 |
+
"grad_norm": 0.16988424956798553,
|
| 3111 |
+
"learning_rate": 1.73199646280425e-05,
|
| 3112 |
+
"loss": 0.2719,
|
| 3113 |
+
"step": 2195
|
| 3114 |
+
},
|
| 3115 |
+
{
|
| 3116 |
+
"epoch": 1.7644110275689222,
|
| 3117 |
+
"grad_norm": 0.16524524986743927,
|
| 3118 |
+
"learning_rate": 1.6748255868330148e-05,
|
| 3119 |
+
"loss": 0.2296,
|
| 3120 |
+
"step": 2200
|
| 3121 |
+
},
|
| 3122 |
+
{
|
| 3123 |
+
"epoch": 1.768421052631579,
|
| 3124 |
+
"grad_norm": 0.18039652705192566,
|
| 3125 |
+
"learning_rate": 1.6185814705610925e-05,
|
| 3126 |
+
"loss": 0.2677,
|
| 3127 |
+
"step": 2205
|
| 3128 |
+
},
|
| 3129 |
+
{
|
| 3130 |
+
"epoch": 1.7724310776942356,
|
| 3131 |
+
"grad_norm": 0.1674271821975708,
|
| 3132 |
+
"learning_rate": 1.5632663486890407e-05,
|
| 3133 |
+
"loss": 0.2644,
|
| 3134 |
+
"step": 2210
|
| 3135 |
+
},
|
| 3136 |
+
{
|
| 3137 |
+
"epoch": 1.7764411027568923,
|
| 3138 |
+
"grad_norm": 0.1731635481119156,
|
| 3139 |
+
"learning_rate": 1.5088824190064521e-05,
|
| 3140 |
+
"loss": 0.2676,
|
| 3141 |
+
"step": 2215
|
| 3142 |
+
},
|
| 3143 |
+
{
|
| 3144 |
+
"epoch": 1.780451127819549,
|
| 3145 |
+
"grad_norm": 0.1482301503419876,
|
| 3146 |
+
"learning_rate": 1.455431842304647e-05,
|
| 3147 |
+
"loss": 0.2497,
|
| 3148 |
+
"step": 2220
|
| 3149 |
+
},
|
| 3150 |
+
{
|
| 3151 |
+
"epoch": 1.7844611528822054,
|
| 3152 |
+
"grad_norm": 0.16372230648994446,
|
| 3153 |
+
"learning_rate": 1.4029167422908107e-05,
|
| 3154 |
+
"loss": 0.266,
|
| 3155 |
+
"step": 2225
|
| 3156 |
+
},
|
| 3157 |
+
{
|
| 3158 |
+
"epoch": 1.788471177944862,
|
| 3159 |
+
"grad_norm": 0.17107130587100983,
|
| 3160 |
+
"learning_rate": 1.3513392055036072e-05,
|
| 3161 |
+
"loss": 0.279,
|
| 3162 |
+
"step": 2230
|
| 3163 |
+
},
|
| 3164 |
+
{
|
| 3165 |
+
"epoch": 1.7924812030075188,
|
| 3166 |
+
"grad_norm": 0.15426160395145416,
|
| 3167 |
+
"learning_rate": 1.3007012812302737e-05,
|
| 3168 |
+
"loss": 0.2512,
|
| 3169 |
+
"step": 2235
|
| 3170 |
+
},
|
| 3171 |
+
{
|
| 3172 |
+
"epoch": 1.7964912280701755,
|
| 3173 |
+
"grad_norm": 0.16106005012989044,
|
| 3174 |
+
"learning_rate": 1.25100498142523e-05,
|
| 3175 |
+
"loss": 0.2639,
|
| 3176 |
+
"step": 2240
|
| 3177 |
+
},
|
| 3178 |
+
{
|
| 3179 |
+
"epoch": 1.8005012531328322,
|
| 3180 |
+
"grad_norm": 0.16801631450653076,
|
| 3181 |
+
"learning_rate": 1.2022522806301005e-05,
|
| 3182 |
+
"loss": 0.2504,
|
| 3183 |
+
"step": 2245
|
| 3184 |
+
},
|
| 3185 |
+
{
|
| 3186 |
+
"epoch": 1.8045112781954886,
|
| 3187 |
+
"grad_norm": 0.15076345205307007,
|
| 3188 |
+
"learning_rate": 1.1544451158952807e-05,
|
| 3189 |
+
"loss": 0.2516,
|
| 3190 |
+
"step": 2250
|
| 3191 |
+
},
|
| 3192 |
+
{
|
| 3193 |
+
"epoch": 1.8085213032581455,
|
| 3194 |
+
"grad_norm": 0.16535955667495728,
|
| 3195 |
+
"learning_rate": 1.1075853867029694e-05,
|
| 3196 |
+
"loss": 0.2541,
|
| 3197 |
+
"step": 2255
|
| 3198 |
+
},
|
| 3199 |
+
{
|
| 3200 |
+
"epoch": 1.812531328320802,
|
| 3201 |
+
"grad_norm": 0.14483940601348877,
|
| 3202 |
+
"learning_rate": 1.0616749548917059e-05,
|
| 3203 |
+
"loss": 0.2392,
|
| 3204 |
+
"step": 2260
|
| 3205 |
+
},
|
| 3206 |
+
{
|
| 3207 |
+
"epoch": 1.8165413533834587,
|
| 3208 |
+
"grad_norm": 0.1692323386669159,
|
| 3209 |
+
"learning_rate": 1.0167156445823777e-05,
|
| 3210 |
+
"loss": 0.2535,
|
| 3211 |
+
"step": 2265
|
| 3212 |
+
},
|
| 3213 |
+
{
|
| 3214 |
+
"epoch": 1.8205513784461154,
|
| 3215 |
+
"grad_norm": 0.15716001391410828,
|
| 3216 |
+
"learning_rate": 9.727092421057627e-06,
|
| 3217 |
+
"loss": 0.2369,
|
| 3218 |
+
"step": 2270
|
| 3219 |
+
},
|
| 3220 |
+
{
|
| 3221 |
+
"epoch": 1.8245614035087718,
|
| 3222 |
+
"grad_norm": 0.14877931773662567,
|
| 3223 |
+
"learning_rate": 9.296574959315463e-06,
|
| 3224 |
+
"loss": 0.2565,
|
| 3225 |
+
"step": 2275
|
| 3226 |
+
},
|
| 3227 |
+
{
|
| 3228 |
+
"epoch": 1.8285714285714287,
|
| 3229 |
+
"grad_norm": 0.1587458699941635,
|
| 3230 |
+
"learning_rate": 8.875621165988474e-06,
|
| 3231 |
+
"loss": 0.269,
|
| 3232 |
+
"step": 2280
|
| 3233 |
+
},
|
| 3234 |
+
{
|
| 3235 |
+
"epoch": 1.8325814536340852,
|
| 3236 |
+
"grad_norm": 0.17435844242572784,
|
| 3237 |
+
"learning_rate": 8.46424776648258e-06,
|
| 3238 |
+
"loss": 0.2456,
|
| 3239 |
+
"step": 2285
|
| 3240 |
+
},
|
| 3241 |
+
{
|
| 3242 |
+
"epoch": 1.8365914786967419,
|
| 3243 |
+
"grad_norm": 0.16262808442115784,
|
| 3244 |
+
"learning_rate": 8.06247110555397e-06,
|
| 3245 |
+
"loss": 0.2582,
|
| 3246 |
+
"step": 2290
|
| 3247 |
+
},
|
| 3248 |
+
{
|
| 3249 |
+
"epoch": 1.8406015037593986,
|
| 3250 |
+
"grad_norm": 0.15774986147880554,
|
| 3251 |
+
"learning_rate": 7.670307146659588e-06,
|
| 3252 |
+
"loss": 0.2713,
|
| 3253 |
+
"step": 2295
|
| 3254 |
+
},
|
| 3255 |
+
{
|
| 3256 |
+
"epoch": 1.844611528822055,
|
| 3257 |
+
"grad_norm": 0.16883234679698944,
|
| 3258 |
+
"learning_rate": 7.287771471322951e-06,
|
| 3259 |
+
"loss": 0.2732,
|
| 3260 |
+
"step": 2300
|
| 3261 |
+
},
|
| 3262 |
+
{
|
| 3263 |
+
"epoch": 1.848621553884712,
|
| 3264 |
+
"grad_norm": 0.16868174076080322,
|
| 3265 |
+
"learning_rate": 6.91487927851489e-06,
|
| 3266 |
+
"loss": 0.2495,
|
| 3267 |
+
"step": 2305
|
| 3268 |
+
},
|
| 3269 |
+
{
|
| 3270 |
+
"epoch": 1.8526315789473684,
|
| 3271 |
+
"grad_norm": 0.16309338808059692,
|
| 3272 |
+
"learning_rate": 6.551645384049898e-06,
|
| 3273 |
+
"loss": 0.2684,
|
| 3274 |
+
"step": 2310
|
| 3275 |
+
},
|
| 3276 |
+
{
|
| 3277 |
+
"epoch": 1.856641604010025,
|
| 3278 |
+
"grad_norm": 0.1636061668395996,
|
| 3279 |
+
"learning_rate": 6.198084219997374e-06,
|
| 3280 |
+
"loss": 0.2668,
|
| 3281 |
+
"step": 2315
|
| 3282 |
+
},
|
| 3283 |
+
{
|
| 3284 |
+
"epoch": 1.8606516290726818,
|
| 3285 |
+
"grad_norm": 0.17537201941013336,
|
| 3286 |
+
"learning_rate": 5.854209834108087e-06,
|
| 3287 |
+
"loss": 0.2743,
|
| 3288 |
+
"step": 2320
|
| 3289 |
+
},
|
| 3290 |
+
{
|
| 3291 |
+
"epoch": 1.8646616541353382,
|
| 3292 |
+
"grad_norm": 0.18073910474777222,
|
| 3293 |
+
"learning_rate": 5.5200358892561755e-06,
|
| 3294 |
+
"loss": 0.2738,
|
| 3295 |
+
"step": 2325
|
| 3296 |
+
},
|
| 3297 |
+
{
|
| 3298 |
+
"epoch": 1.8686716791979952,
|
| 3299 |
+
"grad_norm": 0.16718195378780365,
|
| 3300 |
+
"learning_rate": 5.195575662896301e-06,
|
| 3301 |
+
"loss": 0.2668,
|
| 3302 |
+
"step": 2330
|
| 3303 |
+
},
|
| 3304 |
+
{
|
| 3305 |
+
"epoch": 1.8726817042606516,
|
| 3306 |
+
"grad_norm": 0.15906399488449097,
|
| 3307 |
+
"learning_rate": 4.880842046536021e-06,
|
| 3308 |
+
"loss": 0.2522,
|
| 3309 |
+
"step": 2335
|
| 3310 |
+
},
|
| 3311 |
+
{
|
| 3312 |
+
"epoch": 1.8766917293233083,
|
| 3313 |
+
"grad_norm": 0.1543634831905365,
|
| 3314 |
+
"learning_rate": 4.575847545223499e-06,
|
| 3315 |
+
"loss": 0.2325,
|
| 3316 |
+
"step": 2340
|
| 3317 |
+
},
|
| 3318 |
+
{
|
| 3319 |
+
"epoch": 1.880701754385965,
|
| 3320 |
+
"grad_norm": 0.15531331300735474,
|
| 3321 |
+
"learning_rate": 4.2806042770509316e-06,
|
| 3322 |
+
"loss": 0.2503,
|
| 3323 |
+
"step": 2345
|
| 3324 |
+
},
|
| 3325 |
+
{
|
| 3326 |
+
"epoch": 1.8847117794486214,
|
| 3327 |
+
"grad_norm": 0.17484238743782043,
|
| 3328 |
+
"learning_rate": 3.9951239726728485e-06,
|
| 3329 |
+
"loss": 0.2491,
|
| 3330 |
+
"step": 2350
|
| 3331 |
+
},
|
| 3332 |
+
{
|
| 3333 |
+
"epoch": 1.8887218045112784,
|
| 3334 |
+
"grad_norm": 0.15897777676582336,
|
| 3335 |
+
"learning_rate": 3.719417974839989e-06,
|
| 3336 |
+
"loss": 0.2708,
|
| 3337 |
+
"step": 2355
|
| 3338 |
+
},
|
| 3339 |
+
{
|
| 3340 |
+
"epoch": 1.8927318295739348,
|
| 3341 |
+
"grad_norm": 0.16040709614753723,
|
| 3342 |
+
"learning_rate": 3.453497237948855e-06,
|
| 3343 |
+
"loss": 0.2655,
|
| 3344 |
+
"step": 2360
|
| 3345 |
+
},
|
| 3346 |
+
{
|
| 3347 |
+
"epoch": 1.8967418546365915,
|
| 3348 |
+
"grad_norm": 0.14991337060928345,
|
| 3349 |
+
"learning_rate": 3.1973723276062517e-06,
|
| 3350 |
+
"loss": 0.2275,
|
| 3351 |
+
"step": 2365
|
| 3352 |
+
},
|
| 3353 |
+
{
|
| 3354 |
+
"epoch": 1.9007518796992482,
|
| 3355 |
+
"grad_norm": 0.1545797735452652,
|
| 3356 |
+
"learning_rate": 2.9510534202096263e-06,
|
| 3357 |
+
"loss": 0.2576,
|
| 3358 |
+
"step": 2370
|
| 3359 |
+
},
|
| 3360 |
+
{
|
| 3361 |
+
"epoch": 1.9047619047619047,
|
| 3362 |
+
"grad_norm": 0.16641472280025482,
|
| 3363 |
+
"learning_rate": 2.71455030254264e-06,
|
| 3364 |
+
"loss": 0.2607,
|
| 3365 |
+
"step": 2375
|
| 3366 |
+
},
|
| 3367 |
+
{
|
| 3368 |
+
"epoch": 1.9087719298245616,
|
| 3369 |
+
"grad_norm": 0.1749178171157837,
|
| 3370 |
+
"learning_rate": 2.487872371386424e-06,
|
| 3371 |
+
"loss": 0.2679,
|
| 3372 |
+
"step": 2380
|
| 3373 |
+
},
|
| 3374 |
+
{
|
| 3375 |
+
"epoch": 1.912781954887218,
|
| 3376 |
+
"grad_norm": 0.1886933445930481,
|
| 3377 |
+
"learning_rate": 2.271028633146127e-06,
|
| 3378 |
+
"loss": 0.2631,
|
| 3379 |
+
"step": 2385
|
| 3380 |
+
},
|
| 3381 |
+
{
|
| 3382 |
+
"epoch": 1.9167919799498747,
|
| 3383 |
+
"grad_norm": 0.15673983097076416,
|
| 3384 |
+
"learning_rate": 2.064027703493149e-06,
|
| 3385 |
+
"loss": 0.2502,
|
| 3386 |
+
"step": 2390
|
| 3387 |
+
},
|
| 3388 |
+
{
|
| 3389 |
+
"epoch": 1.9208020050125314,
|
| 3390 |
+
"grad_norm": 0.16062475740909576,
|
| 3391 |
+
"learning_rate": 1.866877807022771e-06,
|
| 3392 |
+
"loss": 0.2295,
|
| 3393 |
+
"step": 2395
|
| 3394 |
+
},
|
| 3395 |
+
{
|
| 3396 |
+
"epoch": 1.9248120300751879,
|
| 3397 |
+
"grad_norm": 0.16452881693840027,
|
| 3398 |
+
"learning_rate": 1.6795867769273665e-06,
|
| 3399 |
+
"loss": 0.2585,
|
| 3400 |
+
"step": 2400
|
| 3401 |
+
},
|
| 3402 |
+
{
|
| 3403 |
+
"epoch": 1.9288220551378448,
|
| 3404 |
+
"grad_norm": 0.17597217857837677,
|
| 3405 |
+
"learning_rate": 1.50216205468523e-06,
|
| 3406 |
+
"loss": 0.2586,
|
| 3407 |
+
"step": 2405
|
| 3408 |
+
},
|
| 3409 |
+
{
|
| 3410 |
+
"epoch": 1.9328320802005012,
|
| 3411 |
+
"grad_norm": 0.1579117476940155,
|
| 3412 |
+
"learning_rate": 1.3346106897648424e-06,
|
| 3413 |
+
"loss": 0.2442,
|
| 3414 |
+
"step": 2410
|
| 3415 |
+
},
|
| 3416 |
+
{
|
| 3417 |
+
"epoch": 1.936842105263158,
|
| 3418 |
+
"grad_norm": 0.16751398146152496,
|
| 3419 |
+
"learning_rate": 1.1769393393448457e-06,
|
| 3420 |
+
"loss": 0.2649,
|
| 3421 |
+
"step": 2415
|
| 3422 |
+
},
|
| 3423 |
+
{
|
| 3424 |
+
"epoch": 1.9408521303258146,
|
| 3425 |
+
"grad_norm": 0.17506803572177887,
|
| 3426 |
+
"learning_rate": 1.0291542680494758e-06,
|
| 3427 |
+
"loss": 0.2709,
|
| 3428 |
+
"step": 2420
|
| 3429 |
+
},
|
| 3430 |
+
{
|
| 3431 |
+
"epoch": 1.944862155388471,
|
| 3432 |
+
"grad_norm": 0.17089371383190155,
|
| 3433 |
+
"learning_rate": 8.912613476997067e-07,
|
| 3434 |
+
"loss": 0.2739,
|
| 3435 |
+
"step": 2425
|
| 3436 |
+
},
|
| 3437 |
+
{
|
| 3438 |
+
"epoch": 1.948872180451128,
|
| 3439 |
+
"grad_norm": 0.16731411218643188,
|
| 3440 |
+
"learning_rate": 7.632660570799366e-07,
|
| 3441 |
+
"loss": 0.2617,
|
| 3442 |
+
"step": 2430
|
| 3443 |
+
},
|
| 3444 |
+
{
|
| 3445 |
+
"epoch": 1.9528822055137844,
|
| 3446 |
+
"grad_norm": 0.15917718410491943,
|
| 3447 |
+
"learning_rate": 6.451734817202736e-07,
|
| 3448 |
+
"loss": 0.2783,
|
| 3449 |
+
"step": 2435
|
| 3450 |
+
},
|
| 3451 |
+
{
|
| 3452 |
+
"epoch": 1.9568922305764411,
|
| 3453 |
+
"grad_norm": 0.17604759335517883,
|
| 3454 |
+
"learning_rate": 5.369883136945309e-07,
|
| 3455 |
+
"loss": 0.2538,
|
| 3456 |
+
"step": 2440
|
| 3457 |
+
},
|
| 3458 |
+
{
|
| 3459 |
+
"epoch": 1.9609022556390978,
|
| 3460 |
+
"grad_norm": 0.1530042290687561,
|
| 3461 |
+
"learning_rate": 4.387148514337358e-07,
|
| 3462 |
+
"loss": 0.2652,
|
| 3463 |
+
"step": 2445
|
| 3464 |
+
},
|
| 3465 |
+
{
|
| 3466 |
+
"epoch": 1.9649122807017543,
|
| 3467 |
+
"grad_norm": 0.15754002332687378,
|
| 3468 |
+
"learning_rate": 3.503569995554068e-07,
|
| 3469 |
+
"loss": 0.2615,
|
| 3470 |
+
"step": 2450
|
| 3471 |
+
},
|
| 3472 |
+
{
|
| 3473 |
+
"epoch": 1.9689223057644112,
|
| 3474 |
+
"grad_norm": 0.1827680766582489,
|
| 3475 |
+
"learning_rate": 2.7191826870834325e-07,
|
| 3476 |
+
"loss": 0.2742,
|
| 3477 |
+
"step": 2455
|
| 3478 |
+
},
|
| 3479 |
+
{
|
| 3480 |
+
"epoch": 1.9729323308270676,
|
| 3481 |
+
"grad_norm": 0.15967601537704468,
|
| 3482 |
+
"learning_rate": 2.034017754332651e-07,
|
| 3483 |
+
"loss": 0.2466,
|
| 3484 |
+
"step": 2460
|
| 3485 |
+
},
|
| 3486 |
+
{
|
| 3487 |
+
"epoch": 1.9769423558897243,
|
| 3488 |
+
"grad_norm": 0.17333175241947174,
|
| 3489 |
+
"learning_rate": 1.4481024203877313e-07,
|
| 3490 |
+
"loss": 0.2764,
|
| 3491 |
+
"step": 2465
|
| 3492 |
+
},
|
| 3493 |
+
{
|
| 3494 |
+
"epoch": 1.980952380952381,
|
| 3495 |
+
"grad_norm": 0.16165874898433685,
|
| 3496 |
+
"learning_rate": 9.6145996493463e-08,
|
| 3497 |
+
"loss": 0.2568,
|
| 3498 |
+
"step": 2470
|
| 3499 |
+
},
|
| 3500 |
+
{
|
| 3501 |
+
"epoch": 1.9849624060150375,
|
| 3502 |
+
"grad_norm": 0.16772541403770447,
|
| 3503 |
+
"learning_rate": 5.7410972333193924e-08,
|
| 3504 |
+
"loss": 0.2407,
|
| 3505 |
+
"step": 2475
|
| 3506 |
+
},
|
| 3507 |
+
{
|
| 3508 |
+
"epoch": 1.9889724310776944,
|
| 3509 |
+
"grad_norm": 0.14310236275196075,
|
| 3510 |
+
"learning_rate": 2.8606708584344445e-08,
|
| 3511 |
+
"loss": 0.2468,
|
| 3512 |
+
"step": 2480
|
| 3513 |
+
},
|
| 3514 |
+
{
|
| 3515 |
+
"epoch": 1.9929824561403509,
|
| 3516 |
+
"grad_norm": 0.1455429494380951,
|
| 3517 |
+
"learning_rate": 9.734349702722467e-09,
|
| 3518 |
+
"loss": 0.2635,
|
| 3519 |
+
"step": 2485
|
| 3520 |
+
},
|
| 3521 |
+
{
|
| 3522 |
+
"epoch": 1.9969924812030075,
|
| 3523 |
+
"grad_norm": 0.16766951978206635,
|
| 3524 |
+
"learning_rate": 7.94645528018334e-10,
|
| 3525 |
+
"loss": 0.2535,
|
| 3526 |
+
"step": 2490
|
| 3527 |
+
},
|
| 3528 |
+
{
|
| 3529 |
+
"epoch": 1.9985964912280703,
|
| 3530 |
+
"eval_loss": 0.22910664975643158,
|
| 3531 |
+
"eval_runtime": 8.4374,
|
| 3532 |
+
"eval_samples_per_second": 6.519,
|
| 3533 |
+
"eval_steps_per_second": 0.83,
|
| 3534 |
+
"step": 2492
|
| 3535 |
+
},
|
| 3536 |
+
{
|
| 3537 |
+
"epoch": 1.9985964912280703,
|
| 3538 |
+
"step": 2492,
|
| 3539 |
+
"total_flos": 8.025299762167153e+17,
|
| 3540 |
+
"train_loss": 0.37913749282088366,
|
| 3541 |
+
"train_runtime": 28381.1385,
|
| 3542 |
+
"train_samples_per_second": 2.812,
|
| 3543 |
+
"train_steps_per_second": 0.088
|
| 3544 |
+
},
|
| 3545 |
+
{
|
| 3546 |
+
"epoch": 1.9985964912280703,
|
| 3547 |
+
"eval_loss": 0.22910664975643158,
|
| 3548 |
+
"eval_runtime": 8.378,
|
| 3549 |
+
"eval_samples_per_second": 6.565,
|
| 3550 |
+
"eval_steps_per_second": 0.836,
|
| 3551 |
+
"step": 2492
|
| 3552 |
+
}
|
| 3553 |
+
],
|
| 3554 |
+
"logging_steps": 5,
|
| 3555 |
+
"max_steps": 2492,
|
| 3556 |
+
"num_input_tokens_seen": 0,
|
| 3557 |
+
"num_train_epochs": 2,
|
| 3558 |
+
"save_steps": 500,
|
| 3559 |
+
"stateful_callbacks": {
|
| 3560 |
+
"TrainerControl": {
|
| 3561 |
+
"args": {
|
| 3562 |
+
"should_epoch_stop": false,
|
| 3563 |
+
"should_evaluate": false,
|
| 3564 |
+
"should_log": false,
|
| 3565 |
+
"should_save": false,
|
| 3566 |
+
"should_training_stop": false
|
| 3567 |
+
},
|
| 3568 |
+
"attributes": {}
|
| 3569 |
+
}
|
| 3570 |
+
},
|
| 3571 |
+
"total_flos": 8.025299762167153e+17,
|
| 3572 |
+
"train_batch_size": 2,
|
| 3573 |
+
"trial_name": null,
|
| 3574 |
+
"trial_params": null
|
| 3575 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a817357f263a1bc6bd2f6b0a6a3b250ba364a4585866c89a1431fb091d8ee37e
|
| 3 |
+
size 5624
|