adding model files
Browse files- .gitignore +1 -0
- README.md +60 -0
- config.json +74 -0
- pytorch_model.bin +3 -0
- trainer_state.json +1931 -0
- training_args.bin +3 -0
.gitignore
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
checkpoint-*/
|
README.md
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
model-index:
|
| 6 |
+
- name: gpt-neo-1pt3B-emailgen_DS-multi-clean-100k_Ep-2_Bs-4
|
| 7 |
+
results: []
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 11 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 12 |
+
|
| 13 |
+
# gpt-neo-1pt3B-emailgen_DS-multi-clean-100k_Ep-2_Bs-4
|
| 14 |
+
|
| 15 |
+
This model is a fine-tuned version of [EleutherAI/gpt-neo-1.3B](https://huggingface.co/EleutherAI/gpt-neo-1.3B) on an unknown dataset.
|
| 16 |
+
It achieves the following results on the evaluation set:
|
| 17 |
+
- Loss: 1.6930
|
| 18 |
+
|
| 19 |
+
## Model description
|
| 20 |
+
|
| 21 |
+
More information needed
|
| 22 |
+
|
| 23 |
+
## Intended uses & limitations
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Training and evaluation data
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training procedure
|
| 32 |
+
|
| 33 |
+
### Training hyperparameters
|
| 34 |
+
|
| 35 |
+
The following hyperparameters were used during training:
|
| 36 |
+
- learning_rate: 0.0001
|
| 37 |
+
- train_batch_size: 4
|
| 38 |
+
- eval_batch_size: 4
|
| 39 |
+
- seed: 42
|
| 40 |
+
- distributed_type: multi-GPU
|
| 41 |
+
- gradient_accumulation_steps: 32
|
| 42 |
+
- total_train_batch_size: 128
|
| 43 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 44 |
+
- lr_scheduler_type: cosine
|
| 45 |
+
- lr_scheduler_warmup_ratio: 0.02
|
| 46 |
+
- num_epochs: 2
|
| 47 |
+
|
| 48 |
+
### Training results
|
| 49 |
+
|
| 50 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 51 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 52 |
+
| 1.8669 | 1.0 | 789 | 1.7866 |
|
| 53 |
+
| 1.4049 | 2.0 | 1578 | 1.6930 |
|
| 54 |
+
|
| 55 |
+
|
| 56 |
+
### Framework versions
|
| 57 |
+
|
| 58 |
+
- Transformers 4.22.2
|
| 59 |
+
- Pytorch 1.10.0+cu113
|
| 60 |
+
- Tokenizers 0.12.1
|
config.json
ADDED
|
@@ -0,0 +1,74 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "EleutherAI/gpt-neo-1.3B",
|
| 3 |
+
"activation_function": "gelu_new",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"GPTNeoForCausalLM"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout": 0,
|
| 8 |
+
"attention_layers": [
|
| 9 |
+
"global",
|
| 10 |
+
"local",
|
| 11 |
+
"global",
|
| 12 |
+
"local",
|
| 13 |
+
"global",
|
| 14 |
+
"local",
|
| 15 |
+
"global",
|
| 16 |
+
"local",
|
| 17 |
+
"global",
|
| 18 |
+
"local",
|
| 19 |
+
"global",
|
| 20 |
+
"local",
|
| 21 |
+
"global",
|
| 22 |
+
"local",
|
| 23 |
+
"global",
|
| 24 |
+
"local",
|
| 25 |
+
"global",
|
| 26 |
+
"local",
|
| 27 |
+
"global",
|
| 28 |
+
"local",
|
| 29 |
+
"global",
|
| 30 |
+
"local",
|
| 31 |
+
"global",
|
| 32 |
+
"local"
|
| 33 |
+
],
|
| 34 |
+
"attention_types": [
|
| 35 |
+
[
|
| 36 |
+
[
|
| 37 |
+
"global",
|
| 38 |
+
"local"
|
| 39 |
+
],
|
| 40 |
+
12
|
| 41 |
+
]
|
| 42 |
+
],
|
| 43 |
+
"bos_token_id": 50256,
|
| 44 |
+
"embed_dropout": 0,
|
| 45 |
+
"eos_token_id": 50256,
|
| 46 |
+
"gradient_checkpointing": false,
|
| 47 |
+
"hidden_size": 2048,
|
| 48 |
+
"initializer_range": 0.02,
|
| 49 |
+
"intermediate_size": null,
|
| 50 |
+
"layer_norm_epsilon": 1e-05,
|
| 51 |
+
"max_position_embeddings": 2048,
|
| 52 |
+
"model_type": "gpt_neo",
|
| 53 |
+
"num_heads": 16,
|
| 54 |
+
"num_layers": 24,
|
| 55 |
+
"resid_dropout": 0,
|
| 56 |
+
"summary_activation": null,
|
| 57 |
+
"summary_first_dropout": 0.1,
|
| 58 |
+
"summary_proj_to_labels": true,
|
| 59 |
+
"summary_type": "cls_index",
|
| 60 |
+
"summary_use_proj": true,
|
| 61 |
+
"task_specific_params": {
|
| 62 |
+
"text-generation": {
|
| 63 |
+
"do_sample": true,
|
| 64 |
+
"max_length": 50,
|
| 65 |
+
"temperature": 0.9
|
| 66 |
+
}
|
| 67 |
+
},
|
| 68 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 69 |
+
"torch_dtype": "float32",
|
| 70 |
+
"transformers_version": "4.22.2",
|
| 71 |
+
"use_cache": false,
|
| 72 |
+
"vocab_size": 50257,
|
| 73 |
+
"window_size": 256
|
| 74 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:08be47da43718499b006a4aff7f17a3e1174555f954555598a1e77fef80b34ad
|
| 3 |
+
size 5363085953
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1931 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.999920792079208,
|
| 5 |
+
"global_step": 1578,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.01,
|
| 12 |
+
"learning_rate": 1.5625e-05,
|
| 13 |
+
"loss": 4.3204,
|
| 14 |
+
"step": 5
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.01,
|
| 18 |
+
"learning_rate": 3.125e-05,
|
| 19 |
+
"loss": 3.777,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.02,
|
| 24 |
+
"learning_rate": 4.6875e-05,
|
| 25 |
+
"loss": 3.065,
|
| 26 |
+
"step": 15
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.03,
|
| 30 |
+
"learning_rate": 6.25e-05,
|
| 31 |
+
"loss": 2.8056,
|
| 32 |
+
"step": 20
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.03,
|
| 36 |
+
"learning_rate": 7.8125e-05,
|
| 37 |
+
"loss": 2.6582,
|
| 38 |
+
"step": 25
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.04,
|
| 42 |
+
"learning_rate": 9.375e-05,
|
| 43 |
+
"loss": 2.5482,
|
| 44 |
+
"step": 30
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.04,
|
| 48 |
+
"learning_rate": 9.999907090111407e-05,
|
| 49 |
+
"loss": 2.4959,
|
| 50 |
+
"step": 35
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.05,
|
| 54 |
+
"learning_rate": 9.999339319963168e-05,
|
| 55 |
+
"loss": 2.4399,
|
| 56 |
+
"step": 40
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.06,
|
| 60 |
+
"learning_rate": 9.998255454812832e-05,
|
| 61 |
+
"loss": 2.3978,
|
| 62 |
+
"step": 45
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.06,
|
| 66 |
+
"learning_rate": 9.996655606550656e-05,
|
| 67 |
+
"loss": 2.3887,
|
| 68 |
+
"step": 50
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.07,
|
| 72 |
+
"learning_rate": 9.994539940333206e-05,
|
| 73 |
+
"loss": 2.3651,
|
| 74 |
+
"step": 55
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.08,
|
| 78 |
+
"learning_rate": 9.99190867456629e-05,
|
| 79 |
+
"loss": 2.4034,
|
| 80 |
+
"step": 60
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.08,
|
| 84 |
+
"learning_rate": 9.988762080882424e-05,
|
| 85 |
+
"loss": 2.2935,
|
| 86 |
+
"step": 65
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.09,
|
| 90 |
+
"learning_rate": 9.985100484112785e-05,
|
| 91 |
+
"loss": 2.3129,
|
| 92 |
+
"step": 70
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.1,
|
| 96 |
+
"learning_rate": 9.980924262253681e-05,
|
| 97 |
+
"loss": 2.3247,
|
| 98 |
+
"step": 75
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.1,
|
| 102 |
+
"learning_rate": 9.97623384642752e-05,
|
| 103 |
+
"loss": 2.3264,
|
| 104 |
+
"step": 80
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.11,
|
| 108 |
+
"learning_rate": 9.971029720838318e-05,
|
| 109 |
+
"loss": 2.3193,
|
| 110 |
+
"step": 85
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.11,
|
| 114 |
+
"learning_rate": 9.965312422721704e-05,
|
| 115 |
+
"loss": 2.2133,
|
| 116 |
+
"step": 90
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.12,
|
| 120 |
+
"learning_rate": 9.959082542289465e-05,
|
| 121 |
+
"loss": 2.1766,
|
| 122 |
+
"step": 95
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.13,
|
| 126 |
+
"learning_rate": 9.952340722668609e-05,
|
| 127 |
+
"loss": 2.2439,
|
| 128 |
+
"step": 100
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.13,
|
| 132 |
+
"learning_rate": 9.945087659834985e-05,
|
| 133 |
+
"loss": 2.1283,
|
| 134 |
+
"step": 105
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.14,
|
| 138 |
+
"learning_rate": 9.937324102541423e-05,
|
| 139 |
+
"loss": 2.1173,
|
| 140 |
+
"step": 110
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.15,
|
| 144 |
+
"learning_rate": 9.929050852240449e-05,
|
| 145 |
+
"loss": 2.1494,
|
| 146 |
+
"step": 115
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.15,
|
| 150 |
+
"learning_rate": 9.920268763001542e-05,
|
| 151 |
+
"loss": 2.2087,
|
| 152 |
+
"step": 120
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.16,
|
| 156 |
+
"learning_rate": 9.910978741422969e-05,
|
| 157 |
+
"loss": 2.2189,
|
| 158 |
+
"step": 125
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.16,
|
| 162 |
+
"learning_rate": 9.901181746538196e-05,
|
| 163 |
+
"loss": 2.12,
|
| 164 |
+
"step": 130
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.17,
|
| 168 |
+
"learning_rate": 9.890878789716877e-05,
|
| 169 |
+
"loss": 2.0899,
|
| 170 |
+
"step": 135
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.18,
|
| 174 |
+
"learning_rate": 9.880070934560458e-05,
|
| 175 |
+
"loss": 2.1151,
|
| 176 |
+
"step": 140
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.18,
|
| 180 |
+
"learning_rate": 9.868759296792373e-05,
|
| 181 |
+
"loss": 2.1848,
|
| 182 |
+
"step": 145
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.19,
|
| 186 |
+
"learning_rate": 9.856945044142865e-05,
|
| 187 |
+
"loss": 2.0883,
|
| 188 |
+
"step": 150
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.2,
|
| 192 |
+
"learning_rate": 9.844629396228438e-05,
|
| 193 |
+
"loss": 2.1461,
|
| 194 |
+
"step": 155
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.2,
|
| 198 |
+
"learning_rate": 9.831813624425952e-05,
|
| 199 |
+
"loss": 2.0882,
|
| 200 |
+
"step": 160
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.21,
|
| 204 |
+
"learning_rate": 9.818499051741381e-05,
|
| 205 |
+
"loss": 2.1026,
|
| 206 |
+
"step": 165
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.22,
|
| 210 |
+
"learning_rate": 9.804687052673229e-05,
|
| 211 |
+
"loss": 2.0476,
|
| 212 |
+
"step": 170
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.22,
|
| 216 |
+
"learning_rate": 9.790379053070638e-05,
|
| 217 |
+
"loss": 2.118,
|
| 218 |
+
"step": 175
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.23,
|
| 222 |
+
"learning_rate": 9.775576529986199e-05,
|
| 223 |
+
"loss": 2.1244,
|
| 224 |
+
"step": 180
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.23,
|
| 228 |
+
"learning_rate": 9.760281011523465e-05,
|
| 229 |
+
"loss": 2.1312,
|
| 230 |
+
"step": 185
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.24,
|
| 234 |
+
"learning_rate": 9.744494076679205e-05,
|
| 235 |
+
"loss": 2.032,
|
| 236 |
+
"step": 190
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.25,
|
| 240 |
+
"learning_rate": 9.7282173551804e-05,
|
| 241 |
+
"loss": 2.0547,
|
| 242 |
+
"step": 195
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.25,
|
| 246 |
+
"learning_rate": 9.711452527315998e-05,
|
| 247 |
+
"loss": 2.087,
|
| 248 |
+
"step": 200
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.26,
|
| 252 |
+
"learning_rate": 9.694201323763459e-05,
|
| 253 |
+
"loss": 1.9386,
|
| 254 |
+
"step": 205
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.27,
|
| 258 |
+
"learning_rate": 9.676465525410088e-05,
|
| 259 |
+
"loss": 2.0093,
|
| 260 |
+
"step": 210
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.27,
|
| 264 |
+
"learning_rate": 9.658246963169195e-05,
|
| 265 |
+
"loss": 2.0762,
|
| 266 |
+
"step": 215
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.28,
|
| 270 |
+
"learning_rate": 9.639547517791076e-05,
|
| 271 |
+
"loss": 2.0568,
|
| 272 |
+
"step": 220
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.29,
|
| 276 |
+
"learning_rate": 9.620369119668866e-05,
|
| 277 |
+
"loss": 2.0782,
|
| 278 |
+
"step": 225
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.29,
|
| 282 |
+
"learning_rate": 9.600713748639258e-05,
|
| 283 |
+
"loss": 2.0198,
|
| 284 |
+
"step": 230
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.3,
|
| 288 |
+
"learning_rate": 9.580583433778116e-05,
|
| 289 |
+
"loss": 2.042,
|
| 290 |
+
"step": 235
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.3,
|
| 294 |
+
"learning_rate": 9.55998025319101e-05,
|
| 295 |
+
"loss": 2.0037,
|
| 296 |
+
"step": 240
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.31,
|
| 300 |
+
"learning_rate": 9.538906333798689e-05,
|
| 301 |
+
"loss": 1.953,
|
| 302 |
+
"step": 245
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.32,
|
| 306 |
+
"learning_rate": 9.517363851117512e-05,
|
| 307 |
+
"loss": 2.0362,
|
| 308 |
+
"step": 250
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.32,
|
| 312 |
+
"learning_rate": 9.49535502903486e-05,
|
| 313 |
+
"loss": 2.0131,
|
| 314 |
+
"step": 255
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.33,
|
| 318 |
+
"learning_rate": 9.472882139579572e-05,
|
| 319 |
+
"loss": 2.0257,
|
| 320 |
+
"step": 260
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.34,
|
| 324 |
+
"learning_rate": 9.449947502687376e-05,
|
| 325 |
+
"loss": 2.0066,
|
| 326 |
+
"step": 265
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.34,
|
| 330 |
+
"learning_rate": 9.426553485961415e-05,
|
| 331 |
+
"loss": 2.0057,
|
| 332 |
+
"step": 270
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.35,
|
| 336 |
+
"learning_rate": 9.402702504427823e-05,
|
| 337 |
+
"loss": 1.9991,
|
| 338 |
+
"step": 275
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.35,
|
| 342 |
+
"learning_rate": 9.378397020286417e-05,
|
| 343 |
+
"loss": 2.021,
|
| 344 |
+
"step": 280
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.36,
|
| 348 |
+
"learning_rate": 9.353639542656521e-05,
|
| 349 |
+
"loss": 1.9371,
|
| 350 |
+
"step": 285
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.37,
|
| 354 |
+
"learning_rate": 9.328432627317938e-05,
|
| 355 |
+
"loss": 1.9793,
|
| 356 |
+
"step": 290
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.37,
|
| 360 |
+
"learning_rate": 9.302778876447115e-05,
|
| 361 |
+
"loss": 1.9013,
|
| 362 |
+
"step": 295
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.38,
|
| 366 |
+
"learning_rate": 9.276680938348512e-05,
|
| 367 |
+
"loss": 1.9733,
|
| 368 |
+
"step": 300
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.39,
|
| 372 |
+
"learning_rate": 9.250141507181206e-05,
|
| 373 |
+
"loss": 1.9828,
|
| 374 |
+
"step": 305
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.39,
|
| 378 |
+
"learning_rate": 9.223163322680772e-05,
|
| 379 |
+
"loss": 1.9855,
|
| 380 |
+
"step": 310
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.4,
|
| 384 |
+
"learning_rate": 9.195749169876448e-05,
|
| 385 |
+
"loss": 1.9448,
|
| 386 |
+
"step": 315
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.41,
|
| 390 |
+
"learning_rate": 9.167901878803638e-05,
|
| 391 |
+
"loss": 1.9595,
|
| 392 |
+
"step": 320
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.41,
|
| 396 |
+
"learning_rate": 9.139624324211744e-05,
|
| 397 |
+
"loss": 1.9461,
|
| 398 |
+
"step": 325
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.42,
|
| 402 |
+
"learning_rate": 9.110919425267415e-05,
|
| 403 |
+
"loss": 2.0007,
|
| 404 |
+
"step": 330
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 0.42,
|
| 408 |
+
"learning_rate": 9.081790145253182e-05,
|
| 409 |
+
"loss": 1.9819,
|
| 410 |
+
"step": 335
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.43,
|
| 414 |
+
"learning_rate": 9.052239491261559e-05,
|
| 415 |
+
"loss": 1.8421,
|
| 416 |
+
"step": 340
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.44,
|
| 420 |
+
"learning_rate": 9.022270513884601e-05,
|
| 421 |
+
"loss": 1.8801,
|
| 422 |
+
"step": 345
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.44,
|
| 426 |
+
"learning_rate": 8.991886306899002e-05,
|
| 427 |
+
"loss": 1.9804,
|
| 428 |
+
"step": 350
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.45,
|
| 432 |
+
"learning_rate": 8.961090006946693e-05,
|
| 433 |
+
"loss": 1.9277,
|
| 434 |
+
"step": 355
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.46,
|
| 438 |
+
"learning_rate": 8.92988479321106e-05,
|
| 439 |
+
"loss": 1.9278,
|
| 440 |
+
"step": 360
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.46,
|
| 444 |
+
"learning_rate": 8.898273887088735e-05,
|
| 445 |
+
"loss": 1.9603,
|
| 446 |
+
"step": 365
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.47,
|
| 450 |
+
"learning_rate": 8.866260551857045e-05,
|
| 451 |
+
"loss": 1.8603,
|
| 452 |
+
"step": 370
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.48,
|
| 456 |
+
"learning_rate": 8.83384809233714e-05,
|
| 457 |
+
"loss": 1.9064,
|
| 458 |
+
"step": 375
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.48,
|
| 462 |
+
"learning_rate": 8.801039854552821e-05,
|
| 463 |
+
"loss": 2.0242,
|
| 464 |
+
"step": 380
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.49,
|
| 468 |
+
"learning_rate": 8.76783922538513e-05,
|
| 469 |
+
"loss": 1.9399,
|
| 470 |
+
"step": 385
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.49,
|
| 474 |
+
"learning_rate": 8.734249632222702e-05,
|
| 475 |
+
"loss": 1.9208,
|
| 476 |
+
"step": 390
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.5,
|
| 480 |
+
"learning_rate": 8.700274542607955e-05,
|
| 481 |
+
"loss": 1.9567,
|
| 482 |
+
"step": 395
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.51,
|
| 486 |
+
"learning_rate": 8.665917463879125e-05,
|
| 487 |
+
"loss": 1.7881,
|
| 488 |
+
"step": 400
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 0.51,
|
| 492 |
+
"learning_rate": 8.631181942808191e-05,
|
| 493 |
+
"loss": 1.8781,
|
| 494 |
+
"step": 405
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.52,
|
| 498 |
+
"learning_rate": 8.596071565234733e-05,
|
| 499 |
+
"loss": 1.8789,
|
| 500 |
+
"step": 410
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.53,
|
| 504 |
+
"learning_rate": 8.560589955695768e-05,
|
| 505 |
+
"loss": 1.8068,
|
| 506 |
+
"step": 415
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.53,
|
| 510 |
+
"learning_rate": 8.524740777051555e-05,
|
| 511 |
+
"loss": 1.9045,
|
| 512 |
+
"step": 420
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.54,
|
| 516 |
+
"learning_rate": 8.48852773010749e-05,
|
| 517 |
+
"loss": 1.8835,
|
| 518 |
+
"step": 425
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 0.54,
|
| 522 |
+
"learning_rate": 8.451954553232055e-05,
|
| 523 |
+
"loss": 1.9104,
|
| 524 |
+
"step": 430
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 0.55,
|
| 528 |
+
"learning_rate": 8.415025021970897e-05,
|
| 529 |
+
"loss": 1.8687,
|
| 530 |
+
"step": 435
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 0.56,
|
| 534 |
+
"learning_rate": 8.37774294865706e-05,
|
| 535 |
+
"loss": 1.8392,
|
| 536 |
+
"step": 440
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.56,
|
| 540 |
+
"learning_rate": 8.340112182017445e-05,
|
| 541 |
+
"loss": 1.952,
|
| 542 |
+
"step": 445
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.57,
|
| 546 |
+
"learning_rate": 8.30213660677548e-05,
|
| 547 |
+
"loss": 1.9331,
|
| 548 |
+
"step": 450
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.58,
|
| 552 |
+
"learning_rate": 8.263820143250096e-05,
|
| 553 |
+
"loss": 1.8135,
|
| 554 |
+
"step": 455
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.58,
|
| 558 |
+
"learning_rate": 8.225166746951023e-05,
|
| 559 |
+
"loss": 1.9092,
|
| 560 |
+
"step": 460
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.59,
|
| 564 |
+
"learning_rate": 8.186180408170453e-05,
|
| 565 |
+
"loss": 1.8667,
|
| 566 |
+
"step": 465
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 0.6,
|
| 570 |
+
"learning_rate": 8.146865151571108e-05,
|
| 571 |
+
"loss": 1.7837,
|
| 572 |
+
"step": 470
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 0.6,
|
| 576 |
+
"learning_rate": 8.107225035770763e-05,
|
| 577 |
+
"loss": 1.8695,
|
| 578 |
+
"step": 475
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.61,
|
| 582 |
+
"learning_rate": 8.067264152923268e-05,
|
| 583 |
+
"loss": 1.8875,
|
| 584 |
+
"step": 480
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.61,
|
| 588 |
+
"learning_rate": 8.026986628296104e-05,
|
| 589 |
+
"loss": 1.8252,
|
| 590 |
+
"step": 485
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.62,
|
| 594 |
+
"learning_rate": 7.986396619844519e-05,
|
| 595 |
+
"loss": 1.8115,
|
| 596 |
+
"step": 490
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 0.63,
|
| 600 |
+
"learning_rate": 7.945498317782289e-05,
|
| 601 |
+
"loss": 1.8459,
|
| 602 |
+
"step": 495
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.63,
|
| 606 |
+
"learning_rate": 7.904295944149157e-05,
|
| 607 |
+
"loss": 1.7909,
|
| 608 |
+
"step": 500
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.64,
|
| 612 |
+
"learning_rate": 7.86279375237497e-05,
|
| 613 |
+
"loss": 1.8135,
|
| 614 |
+
"step": 505
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.65,
|
| 618 |
+
"learning_rate": 7.820996026840607e-05,
|
| 619 |
+
"loss": 1.8054,
|
| 620 |
+
"step": 510
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 0.65,
|
| 624 |
+
"learning_rate": 7.778907082435664e-05,
|
| 625 |
+
"loss": 1.7808,
|
| 626 |
+
"step": 515
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.66,
|
| 630 |
+
"learning_rate": 7.736531264113041e-05,
|
| 631 |
+
"loss": 1.812,
|
| 632 |
+
"step": 520
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.67,
|
| 636 |
+
"learning_rate": 7.693872946440381e-05,
|
| 637 |
+
"loss": 1.8145,
|
| 638 |
+
"step": 525
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 0.67,
|
| 642 |
+
"learning_rate": 7.650936533148485e-05,
|
| 643 |
+
"loss": 1.8522,
|
| 644 |
+
"step": 530
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 0.68,
|
| 648 |
+
"learning_rate": 7.607726456676701e-05,
|
| 649 |
+
"loss": 1.8313,
|
| 650 |
+
"step": 535
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 0.68,
|
| 654 |
+
"learning_rate": 7.56424717771535e-05,
|
| 655 |
+
"loss": 1.8377,
|
| 656 |
+
"step": 540
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 0.69,
|
| 660 |
+
"learning_rate": 7.520503184745231e-05,
|
| 661 |
+
"loss": 1.8008,
|
| 662 |
+
"step": 545
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 0.7,
|
| 666 |
+
"learning_rate": 7.476498993574277e-05,
|
| 667 |
+
"loss": 1.7855,
|
| 668 |
+
"step": 550
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.7,
|
| 672 |
+
"learning_rate": 7.432239146871364e-05,
|
| 673 |
+
"loss": 1.8009,
|
| 674 |
+
"step": 555
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.71,
|
| 678 |
+
"learning_rate": 7.387728213697365e-05,
|
| 679 |
+
"loss": 1.9104,
|
| 680 |
+
"step": 560
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.72,
|
| 684 |
+
"learning_rate": 7.342970789033473e-05,
|
| 685 |
+
"loss": 1.8005,
|
| 686 |
+
"step": 565
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 0.72,
|
| 690 |
+
"learning_rate": 7.297971493306848e-05,
|
| 691 |
+
"loss": 1.8387,
|
| 692 |
+
"step": 570
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 0.73,
|
| 696 |
+
"learning_rate": 7.252734971913636e-05,
|
| 697 |
+
"loss": 1.8135,
|
| 698 |
+
"step": 575
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 0.74,
|
| 702 |
+
"learning_rate": 7.20726589473942e-05,
|
| 703 |
+
"loss": 1.7277,
|
| 704 |
+
"step": 580
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.74,
|
| 708 |
+
"learning_rate": 7.161568955677129e-05,
|
| 709 |
+
"loss": 1.8434,
|
| 710 |
+
"step": 585
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.75,
|
| 714 |
+
"learning_rate": 7.115648872142475e-05,
|
| 715 |
+
"loss": 1.8187,
|
| 716 |
+
"step": 590
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.75,
|
| 720 |
+
"learning_rate": 7.069510384586962e-05,
|
| 721 |
+
"loss": 1.7758,
|
| 722 |
+
"step": 595
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 0.76,
|
| 726 |
+
"learning_rate": 7.023158256008521e-05,
|
| 727 |
+
"loss": 1.6581,
|
| 728 |
+
"step": 600
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 0.77,
|
| 732 |
+
"learning_rate": 6.976597271459811e-05,
|
| 733 |
+
"loss": 1.793,
|
| 734 |
+
"step": 605
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 0.77,
|
| 738 |
+
"learning_rate": 6.929832237554241e-05,
|
| 739 |
+
"loss": 1.7647,
|
| 740 |
+
"step": 610
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 0.78,
|
| 744 |
+
"learning_rate": 6.882867981969777e-05,
|
| 745 |
+
"loss": 1.7386,
|
| 746 |
+
"step": 615
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.79,
|
| 750 |
+
"learning_rate": 6.835709352950557e-05,
|
| 751 |
+
"loss": 1.7605,
|
| 752 |
+
"step": 620
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.79,
|
| 756 |
+
"learning_rate": 6.788361218806416e-05,
|
| 757 |
+
"loss": 1.782,
|
| 758 |
+
"step": 625
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.8,
|
| 762 |
+
"learning_rate": 6.740828467410294e-05,
|
| 763 |
+
"loss": 1.7424,
|
| 764 |
+
"step": 630
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 0.8,
|
| 768 |
+
"learning_rate": 6.693116005693659e-05,
|
| 769 |
+
"loss": 1.6963,
|
| 770 |
+
"step": 635
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 0.81,
|
| 774 |
+
"learning_rate": 6.64522875913995e-05,
|
| 775 |
+
"loss": 1.7481,
|
| 776 |
+
"step": 640
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 0.82,
|
| 780 |
+
"learning_rate": 6.597171671276109e-05,
|
| 781 |
+
"loss": 1.7889,
|
| 782 |
+
"step": 645
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 0.82,
|
| 786 |
+
"learning_rate": 6.54894970316224e-05,
|
| 787 |
+
"loss": 1.7209,
|
| 788 |
+
"step": 650
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 0.83,
|
| 792 |
+
"learning_rate": 6.500567832879478e-05,
|
| 793 |
+
"loss": 1.7533,
|
| 794 |
+
"step": 655
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.84,
|
| 798 |
+
"learning_rate": 6.452031055016073e-05,
|
| 799 |
+
"loss": 1.8314,
|
| 800 |
+
"step": 660
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.84,
|
| 804 |
+
"learning_rate": 6.41309346585278e-05,
|
| 805 |
+
"loss": 1.7855,
|
| 806 |
+
"step": 665
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 0.85,
|
| 810 |
+
"learning_rate": 6.364290491421375e-05,
|
| 811 |
+
"loss": 1.8475,
|
| 812 |
+
"step": 670
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 0.86,
|
| 816 |
+
"learning_rate": 6.315346677679218e-05,
|
| 817 |
+
"loss": 1.7069,
|
| 818 |
+
"step": 675
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 0.86,
|
| 822 |
+
"learning_rate": 6.266267077225471e-05,
|
| 823 |
+
"loss": 1.7635,
|
| 824 |
+
"step": 680
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 0.87,
|
| 828 |
+
"learning_rate": 6.217056756676917e-05,
|
| 829 |
+
"loss": 1.7859,
|
| 830 |
+
"step": 685
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 0.87,
|
| 834 |
+
"learning_rate": 6.16772079614492e-05,
|
| 835 |
+
"loss": 1.7227,
|
| 836 |
+
"step": 690
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.88,
|
| 840 |
+
"learning_rate": 6.118264288710988e-05,
|
| 841 |
+
"loss": 1.796,
|
| 842 |
+
"step": 695
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.89,
|
| 846 |
+
"learning_rate": 6.068692339901006e-05,
|
| 847 |
+
"loss": 1.7148,
|
| 848 |
+
"step": 700
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 0.89,
|
| 852 |
+
"learning_rate": 6.019010067158181e-05,
|
| 853 |
+
"loss": 1.7648,
|
| 854 |
+
"step": 705
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 0.9,
|
| 858 |
+
"learning_rate": 5.9692225993147444e-05,
|
| 859 |
+
"loss": 1.7023,
|
| 860 |
+
"step": 710
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 0.91,
|
| 864 |
+
"learning_rate": 5.9193350760625014e-05,
|
| 865 |
+
"loss": 1.7477,
|
| 866 |
+
"step": 715
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 0.91,
|
| 870 |
+
"learning_rate": 5.869352647422235e-05,
|
| 871 |
+
"loss": 1.7712,
|
| 872 |
+
"step": 720
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 0.92,
|
| 876 |
+
"learning_rate": 5.81928047321207e-05,
|
| 877 |
+
"loss": 1.7387,
|
| 878 |
+
"step": 725
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.93,
|
| 882 |
+
"learning_rate": 5.769123722514795e-05,
|
| 883 |
+
"loss": 1.7172,
|
| 884 |
+
"step": 730
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.93,
|
| 888 |
+
"learning_rate": 5.7188875731442605e-05,
|
| 889 |
+
"loss": 1.6791,
|
| 890 |
+
"step": 735
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 0.94,
|
| 894 |
+
"learning_rate": 5.668577211110848e-05,
|
| 895 |
+
"loss": 1.8073,
|
| 896 |
+
"step": 740
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 0.94,
|
| 900 |
+
"learning_rate": 5.6181978300861046e-05,
|
| 901 |
+
"loss": 1.7626,
|
| 902 |
+
"step": 745
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 0.95,
|
| 906 |
+
"learning_rate": 5.5677546308665985e-05,
|
| 907 |
+
"loss": 1.8242,
|
| 908 |
+
"step": 750
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 0.96,
|
| 912 |
+
"learning_rate": 5.517252820837011e-05,
|
| 913 |
+
"loss": 1.6738,
|
| 914 |
+
"step": 755
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 0.96,
|
| 918 |
+
"learning_rate": 5.4666976134325776e-05,
|
| 919 |
+
"loss": 1.7522,
|
| 920 |
+
"step": 760
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.97,
|
| 924 |
+
"learning_rate": 5.416094227600881e-05,
|
| 925 |
+
"loss": 1.72,
|
| 926 |
+
"step": 765
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.98,
|
| 930 |
+
"learning_rate": 5.365447887263093e-05,
|
| 931 |
+
"loss": 1.6826,
|
| 932 |
+
"step": 770
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 0.98,
|
| 936 |
+
"learning_rate": 5.314763820774695e-05,
|
| 937 |
+
"loss": 1.7828,
|
| 938 |
+
"step": 775
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 0.99,
|
| 942 |
+
"learning_rate": 5.264047260385734e-05,
|
| 943 |
+
"loss": 1.7497,
|
| 944 |
+
"step": 780
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 0.99,
|
| 948 |
+
"learning_rate": 5.213303441700691e-05,
|
| 949 |
+
"loss": 1.8669,
|
| 950 |
+
"step": 785
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 1.0,
|
| 954 |
+
"eval_loss": 1.786563515663147,
|
| 955 |
+
"eval_runtime": 649.3288,
|
| 956 |
+
"eval_samples_per_second": 19.723,
|
| 957 |
+
"eval_steps_per_second": 4.931,
|
| 958 |
+
"step": 789
|
| 959 |
+
},
|
| 960 |
+
{
|
| 961 |
+
"epoch": 1.0,
|
| 962 |
+
"learning_rate": 5.162537603137989e-05,
|
| 963 |
+
"loss": 1.7179,
|
| 964 |
+
"step": 790
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 1.01,
|
| 968 |
+
"learning_rate": 5.1117549853892254e-05,
|
| 969 |
+
"loss": 1.5756,
|
| 970 |
+
"step": 795
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 1.01,
|
| 974 |
+
"learning_rate": 5.060960830878158e-05,
|
| 975 |
+
"loss": 1.5276,
|
| 976 |
+
"step": 800
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.02,
|
| 980 |
+
"learning_rate": 5.01016038321951e-05,
|
| 981 |
+
"loss": 1.4801,
|
| 982 |
+
"step": 805
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 1.03,
|
| 986 |
+
"learning_rate": 4.959358886677667e-05,
|
| 987 |
+
"loss": 1.5357,
|
| 988 |
+
"step": 810
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 1.03,
|
| 992 |
+
"learning_rate": 4.908561585625293e-05,
|
| 993 |
+
"loss": 1.508,
|
| 994 |
+
"step": 815
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 1.04,
|
| 998 |
+
"learning_rate": 4.857773724001939e-05,
|
| 999 |
+
"loss": 1.4764,
|
| 1000 |
+
"step": 820
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 1.05,
|
| 1004 |
+
"learning_rate": 4.807000544772698e-05,
|
| 1005 |
+
"loss": 1.509,
|
| 1006 |
+
"step": 825
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 1.05,
|
| 1010 |
+
"learning_rate": 4.756247289386967e-05,
|
| 1011 |
+
"loss": 1.4372,
|
| 1012 |
+
"step": 830
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 1.06,
|
| 1016 |
+
"learning_rate": 4.705519197237344e-05,
|
| 1017 |
+
"loss": 1.475,
|
| 1018 |
+
"step": 835
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.06,
|
| 1022 |
+
"learning_rate": 4.664958360274928e-05,
|
| 1023 |
+
"loss": 1.4519,
|
| 1024 |
+
"step": 840
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 1.07,
|
| 1028 |
+
"learning_rate": 4.6142887565827685e-05,
|
| 1029 |
+
"loss": 1.5711,
|
| 1030 |
+
"step": 845
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 1.08,
|
| 1034 |
+
"learning_rate": 4.563658970880996e-05,
|
| 1035 |
+
"loss": 1.597,
|
| 1036 |
+
"step": 850
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 1.08,
|
| 1040 |
+
"learning_rate": 4.513074229816111e-05,
|
| 1041 |
+
"loss": 1.5804,
|
| 1042 |
+
"step": 855
|
| 1043 |
+
},
|
| 1044 |
+
{
|
| 1045 |
+
"epoch": 1.09,
|
| 1046 |
+
"learning_rate": 4.462539755384534e-05,
|
| 1047 |
+
"loss": 1.5397,
|
| 1048 |
+
"step": 860
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 1.1,
|
| 1052 |
+
"learning_rate": 4.4120607643935336e-05,
|
| 1053 |
+
"loss": 1.4828,
|
| 1054 |
+
"step": 865
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 1.1,
|
| 1058 |
+
"learning_rate": 4.3616424679226705e-05,
|
| 1059 |
+
"loss": 1.4731,
|
| 1060 |
+
"step": 870
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.11,
|
| 1064 |
+
"learning_rate": 4.3112900707858534e-05,
|
| 1065 |
+
"loss": 1.4711,
|
| 1066 |
+
"step": 875
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 1.12,
|
| 1070 |
+
"learning_rate": 4.26100877099403e-05,
|
| 1071 |
+
"loss": 1.4487,
|
| 1072 |
+
"step": 880
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 1.12,
|
| 1076 |
+
"learning_rate": 4.210803759218575e-05,
|
| 1077 |
+
"loss": 1.4584,
|
| 1078 |
+
"step": 885
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 1.13,
|
| 1082 |
+
"learning_rate": 4.160680218255458e-05,
|
| 1083 |
+
"loss": 1.468,
|
| 1084 |
+
"step": 890
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 1.13,
|
| 1088 |
+
"learning_rate": 4.1106433224901966e-05,
|
| 1089 |
+
"loss": 1.4801,
|
| 1090 |
+
"step": 895
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 1.14,
|
| 1094 |
+
"learning_rate": 4.060698237363695e-05,
|
| 1095 |
+
"loss": 1.5046,
|
| 1096 |
+
"step": 900
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 1.15,
|
| 1100 |
+
"learning_rate": 4.0108501188390055e-05,
|
| 1101 |
+
"loss": 1.474,
|
| 1102 |
+
"step": 905
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.15,
|
| 1106 |
+
"learning_rate": 3.9611041128690586e-05,
|
| 1107 |
+
"loss": 1.4256,
|
| 1108 |
+
"step": 910
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 1.16,
|
| 1112 |
+
"learning_rate": 3.911465354865441e-05,
|
| 1113 |
+
"loss": 1.4885,
|
| 1114 |
+
"step": 915
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 1.17,
|
| 1118 |
+
"learning_rate": 3.861938969168245e-05,
|
| 1119 |
+
"loss": 1.5432,
|
| 1120 |
+
"step": 920
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 1.17,
|
| 1124 |
+
"learning_rate": 3.812530068517073e-05,
|
| 1125 |
+
"loss": 1.4848,
|
| 1126 |
+
"step": 925
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"epoch": 1.18,
|
| 1130 |
+
"learning_rate": 3.763243753523237e-05,
|
| 1131 |
+
"loss": 1.5015,
|
| 1132 |
+
"step": 930
|
| 1133 |
+
},
|
| 1134 |
+
{
|
| 1135 |
+
"epoch": 1.19,
|
| 1136 |
+
"learning_rate": 3.7140851121432054e-05,
|
| 1137 |
+
"loss": 1.4893,
|
| 1138 |
+
"step": 935
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 1.19,
|
| 1142 |
+
"learning_rate": 3.665059219153366e-05,
|
| 1143 |
+
"loss": 1.4881,
|
| 1144 |
+
"step": 940
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.2,
|
| 1148 |
+
"learning_rate": 3.616171135626139e-05,
|
| 1149 |
+
"loss": 1.4386,
|
| 1150 |
+
"step": 945
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 1.2,
|
| 1154 |
+
"learning_rate": 3.567425908407507e-05,
|
| 1155 |
+
"loss": 1.4236,
|
| 1156 |
+
"step": 950
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 1.21,
|
| 1160 |
+
"learning_rate": 3.5188285695960185e-05,
|
| 1161 |
+
"loss": 1.4149,
|
| 1162 |
+
"step": 955
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"epoch": 1.22,
|
| 1166 |
+
"learning_rate": 3.470384136023316e-05,
|
| 1167 |
+
"loss": 1.5044,
|
| 1168 |
+
"step": 960
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 1.22,
|
| 1172 |
+
"learning_rate": 3.422097608736226e-05,
|
| 1173 |
+
"loss": 1.4973,
|
| 1174 |
+
"step": 965
|
| 1175 |
+
},
|
| 1176 |
+
{
|
| 1177 |
+
"epoch": 1.23,
|
| 1178 |
+
"learning_rate": 3.373973972480493e-05,
|
| 1179 |
+
"loss": 1.514,
|
| 1180 |
+
"step": 970
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 1.24,
|
| 1184 |
+
"learning_rate": 3.32601819518619e-05,
|
| 1185 |
+
"loss": 1.5015,
|
| 1186 |
+
"step": 975
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.24,
|
| 1190 |
+
"learning_rate": 3.278235227454865e-05,
|
| 1191 |
+
"loss": 1.471,
|
| 1192 |
+
"step": 980
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 1.25,
|
| 1196 |
+
"learning_rate": 3.230630002048485e-05,
|
| 1197 |
+
"loss": 1.4508,
|
| 1198 |
+
"step": 985
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 1.25,
|
| 1202 |
+
"learning_rate": 3.1832074333801954e-05,
|
| 1203 |
+
"loss": 1.5247,
|
| 1204 |
+
"step": 990
|
| 1205 |
+
},
|
| 1206 |
+
{
|
| 1207 |
+
"epoch": 1.26,
|
| 1208 |
+
"learning_rate": 3.135972417007016e-05,
|
| 1209 |
+
"loss": 1.49,
|
| 1210 |
+
"step": 995
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 1.27,
|
| 1214 |
+
"learning_rate": 3.08892982912444e-05,
|
| 1215 |
+
"loss": 1.4816,
|
| 1216 |
+
"step": 1000
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 1.27,
|
| 1220 |
+
"learning_rate": 3.0420845260630616e-05,
|
| 1221 |
+
"loss": 1.3529,
|
| 1222 |
+
"step": 1005
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 1.28,
|
| 1226 |
+
"learning_rate": 2.9954413437872464e-05,
|
| 1227 |
+
"loss": 1.3932,
|
| 1228 |
+
"step": 1010
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.29,
|
| 1232 |
+
"learning_rate": 2.949005097395887e-05,
|
| 1233 |
+
"loss": 1.3895,
|
| 1234 |
+
"step": 1015
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 1.29,
|
| 1238 |
+
"learning_rate": 2.9027805806253505e-05,
|
| 1239 |
+
"loss": 1.4641,
|
| 1240 |
+
"step": 1020
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 1.3,
|
| 1244 |
+
"learning_rate": 2.856772565354583e-05,
|
| 1245 |
+
"loss": 1.4891,
|
| 1246 |
+
"step": 1025
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 1.31,
|
| 1250 |
+
"learning_rate": 2.810985801112521e-05,
|
| 1251 |
+
"loss": 1.3609,
|
| 1252 |
+
"step": 1030
|
| 1253 |
+
},
|
| 1254 |
+
{
|
| 1255 |
+
"epoch": 1.31,
|
| 1256 |
+
"learning_rate": 2.765425014587769e-05,
|
| 1257 |
+
"loss": 1.4134,
|
| 1258 |
+
"step": 1035
|
| 1259 |
+
},
|
| 1260 |
+
{
|
| 1261 |
+
"epoch": 1.32,
|
| 1262 |
+
"learning_rate": 2.7200949091406496e-05,
|
| 1263 |
+
"loss": 1.4344,
|
| 1264 |
+
"step": 1040
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 1.32,
|
| 1268 |
+
"learning_rate": 2.675000164317672e-05,
|
| 1269 |
+
"loss": 1.4872,
|
| 1270 |
+
"step": 1045
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.33,
|
| 1274 |
+
"learning_rate": 2.630145435368453e-05,
|
| 1275 |
+
"loss": 1.5234,
|
| 1276 |
+
"step": 1050
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 1.34,
|
| 1280 |
+
"learning_rate": 2.585535352765125e-05,
|
| 1281 |
+
"loss": 1.463,
|
| 1282 |
+
"step": 1055
|
| 1283 |
+
},
|
| 1284 |
+
{
|
| 1285 |
+
"epoch": 1.34,
|
| 1286 |
+
"learning_rate": 2.541174521724341e-05,
|
| 1287 |
+
"loss": 1.3836,
|
| 1288 |
+
"step": 1060
|
| 1289 |
+
},
|
| 1290 |
+
{
|
| 1291 |
+
"epoch": 1.35,
|
| 1292 |
+
"learning_rate": 2.4970675217318516e-05,
|
| 1293 |
+
"loss": 1.415,
|
| 1294 |
+
"step": 1065
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 1.36,
|
| 1298 |
+
"learning_rate": 2.4532189060697624e-05,
|
| 1299 |
+
"loss": 1.4798,
|
| 1300 |
+
"step": 1070
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 1.36,
|
| 1304 |
+
"learning_rate": 2.4096332013464784e-05,
|
| 1305 |
+
"loss": 1.4155,
|
| 1306 |
+
"step": 1075
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 1.37,
|
| 1310 |
+
"learning_rate": 2.366314907029416e-05,
|
| 1311 |
+
"loss": 1.3719,
|
| 1312 |
+
"step": 1080
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.38,
|
| 1316 |
+
"learning_rate": 2.32326849498051e-05,
|
| 1317 |
+
"loss": 1.4487,
|
| 1318 |
+
"step": 1085
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 1.38,
|
| 1322 |
+
"learning_rate": 2.280498408994565e-05,
|
| 1323 |
+
"loss": 1.4238,
|
| 1324 |
+
"step": 1090
|
| 1325 |
+
},
|
| 1326 |
+
{
|
| 1327 |
+
"epoch": 1.39,
|
| 1328 |
+
"learning_rate": 2.2380090643405216e-05,
|
| 1329 |
+
"loss": 1.4892,
|
| 1330 |
+
"step": 1095
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 1.39,
|
| 1334 |
+
"learning_rate": 2.195804847305654e-05,
|
| 1335 |
+
"loss": 1.4752,
|
| 1336 |
+
"step": 1100
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 1.4,
|
| 1340 |
+
"learning_rate": 2.1538901147427508e-05,
|
| 1341 |
+
"loss": 1.36,
|
| 1342 |
+
"step": 1105
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 1.41,
|
| 1346 |
+
"learning_rate": 2.1122691936203598e-05,
|
| 1347 |
+
"loss": 1.3771,
|
| 1348 |
+
"step": 1110
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"epoch": 1.41,
|
| 1352 |
+
"learning_rate": 2.0709463805760966e-05,
|
| 1353 |
+
"loss": 1.4437,
|
| 1354 |
+
"step": 1115
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 1.42,
|
| 1358 |
+
"learning_rate": 2.0299259414730914e-05,
|
| 1359 |
+
"loss": 1.4338,
|
| 1360 |
+
"step": 1120
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 1.43,
|
| 1364 |
+
"learning_rate": 1.9892121109596173e-05,
|
| 1365 |
+
"loss": 1.5161,
|
| 1366 |
+
"step": 1125
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 1.43,
|
| 1370 |
+
"learning_rate": 1.9488090920319247e-05,
|
| 1371 |
+
"loss": 1.5169,
|
| 1372 |
+
"step": 1130
|
| 1373 |
+
},
|
| 1374 |
+
{
|
| 1375 |
+
"epoch": 1.44,
|
| 1376 |
+
"learning_rate": 1.9087210556003743e-05,
|
| 1377 |
+
"loss": 1.3437,
|
| 1378 |
+
"step": 1135
|
| 1379 |
+
},
|
| 1380 |
+
{
|
| 1381 |
+
"epoch": 1.44,
|
| 1382 |
+
"learning_rate": 1.8689521400588463e-05,
|
| 1383 |
+
"loss": 1.5007,
|
| 1384 |
+
"step": 1140
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"epoch": 1.45,
|
| 1388 |
+
"learning_rate": 1.8295064508575223e-05,
|
| 1389 |
+
"loss": 1.394,
|
| 1390 |
+
"step": 1145
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 1.46,
|
| 1394 |
+
"learning_rate": 1.790388060079089e-05,
|
| 1395 |
+
"loss": 1.3743,
|
| 1396 |
+
"step": 1150
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.46,
|
| 1400 |
+
"learning_rate": 1.7516010060183396e-05,
|
| 1401 |
+
"loss": 1.4011,
|
| 1402 |
+
"step": 1155
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 1.47,
|
| 1406 |
+
"learning_rate": 1.7131492927653098e-05,
|
| 1407 |
+
"loss": 1.3722,
|
| 1408 |
+
"step": 1160
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 1.48,
|
| 1412 |
+
"learning_rate": 1.6750368897919157e-05,
|
| 1413 |
+
"loss": 1.4329,
|
| 1414 |
+
"step": 1165
|
| 1415 |
+
},
|
| 1416 |
+
{
|
| 1417 |
+
"epoch": 1.48,
|
| 1418 |
+
"learning_rate": 1.6372677315421768e-05,
|
| 1419 |
+
"loss": 1.3472,
|
| 1420 |
+
"step": 1170
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 1.49,
|
| 1424 |
+
"learning_rate": 1.5998457170260577e-05,
|
| 1425 |
+
"loss": 1.3845,
|
| 1426 |
+
"step": 1175
|
| 1427 |
+
},
|
| 1428 |
+
{
|
| 1429 |
+
"epoch": 1.5,
|
| 1430 |
+
"learning_rate": 1.5627747094169482e-05,
|
| 1431 |
+
"loss": 1.3696,
|
| 1432 |
+
"step": 1180
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 1.5,
|
| 1436 |
+
"learning_rate": 1.52605853565288e-05,
|
| 1437 |
+
"loss": 1.4439,
|
| 1438 |
+
"step": 1185
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 1.51,
|
| 1442 |
+
"learning_rate": 1.4897009860414441e-05,
|
| 1443 |
+
"loss": 1.4946,
|
| 1444 |
+
"step": 1190
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 1.51,
|
| 1448 |
+
"learning_rate": 1.4537058138685106e-05,
|
| 1449 |
+
"loss": 1.382,
|
| 1450 |
+
"step": 1195
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"epoch": 1.52,
|
| 1454 |
+
"learning_rate": 1.4180767350107733e-05,
|
| 1455 |
+
"loss": 1.4348,
|
| 1456 |
+
"step": 1200
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 1.53,
|
| 1460 |
+
"learning_rate": 1.3828174275521455e-05,
|
| 1461 |
+
"loss": 1.3328,
|
| 1462 |
+
"step": 1205
|
| 1463 |
+
},
|
| 1464 |
+
{
|
| 1465 |
+
"epoch": 1.53,
|
| 1466 |
+
"learning_rate": 1.3479315314040619e-05,
|
| 1467 |
+
"loss": 1.3845,
|
| 1468 |
+
"step": 1210
|
| 1469 |
+
},
|
| 1470 |
+
{
|
| 1471 |
+
"epoch": 1.54,
|
| 1472 |
+
"learning_rate": 1.3134226479297213e-05,
|
| 1473 |
+
"loss": 1.4184,
|
| 1474 |
+
"step": 1215
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 1.55,
|
| 1478 |
+
"learning_rate": 1.2792943395723083e-05,
|
| 1479 |
+
"loss": 1.4065,
|
| 1480 |
+
"step": 1220
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.55,
|
| 1484 |
+
"learning_rate": 1.2455501294872341e-05,
|
| 1485 |
+
"loss": 1.4034,
|
| 1486 |
+
"step": 1225
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 1.56,
|
| 1490 |
+
"learning_rate": 1.2121935011784286e-05,
|
| 1491 |
+
"loss": 1.4754,
|
| 1492 |
+
"step": 1230
|
| 1493 |
+
},
|
| 1494 |
+
{
|
| 1495 |
+
"epoch": 1.57,
|
| 1496 |
+
"learning_rate": 1.1792278981387338e-05,
|
| 1497 |
+
"loss": 1.4132,
|
| 1498 |
+
"step": 1235
|
| 1499 |
+
},
|
| 1500 |
+
{
|
| 1501 |
+
"epoch": 1.57,
|
| 1502 |
+
"learning_rate": 1.1466567234944203e-05,
|
| 1503 |
+
"loss": 1.4578,
|
| 1504 |
+
"step": 1240
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 1.58,
|
| 1508 |
+
"learning_rate": 1.1144833396538695e-05,
|
| 1509 |
+
"loss": 1.3845,
|
| 1510 |
+
"step": 1245
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"epoch": 1.58,
|
| 1514 |
+
"learning_rate": 1.0827110679604712e-05,
|
| 1515 |
+
"loss": 1.3815,
|
| 1516 |
+
"step": 1250
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 1.59,
|
| 1520 |
+
"learning_rate": 1.0513431883497516e-05,
|
| 1521 |
+
"loss": 1.458,
|
| 1522 |
+
"step": 1255
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 1.6,
|
| 1526 |
+
"learning_rate": 1.0203829390107673e-05,
|
| 1527 |
+
"loss": 1.3987,
|
| 1528 |
+
"step": 1260
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 1.6,
|
| 1532 |
+
"learning_rate": 9.898335160518295e-06,
|
| 1533 |
+
"loss": 1.476,
|
| 1534 |
+
"step": 1265
|
| 1535 |
+
},
|
| 1536 |
+
{
|
| 1537 |
+
"epoch": 1.61,
|
| 1538 |
+
"learning_rate": 9.596980731705574e-06,
|
| 1539 |
+
"loss": 1.4043,
|
| 1540 |
+
"step": 1270
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 1.62,
|
| 1544 |
+
"learning_rate": 9.299797213283118e-06,
|
| 1545 |
+
"loss": 1.3216,
|
| 1546 |
+
"step": 1275
|
| 1547 |
+
},
|
| 1548 |
+
{
|
| 1549 |
+
"epoch": 1.62,
|
| 1550 |
+
"learning_rate": 9.006815284290443e-06,
|
| 1551 |
+
"loss": 1.4191,
|
| 1552 |
+
"step": 1280
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 1.63,
|
| 1556 |
+
"learning_rate": 8.718065190025899e-06,
|
| 1557 |
+
"loss": 1.3306,
|
| 1558 |
+
"step": 1285
|
| 1559 |
+
},
|
| 1560 |
+
{
|
| 1561 |
+
"epoch": 1.63,
|
| 1562 |
+
"learning_rate": 8.433576738924353e-06,
|
| 1563 |
+
"loss": 1.3633,
|
| 1564 |
+
"step": 1290
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 1.64,
|
| 1568 |
+
"learning_rate": 8.153379299480013e-06,
|
| 1569 |
+
"loss": 1.4831,
|
| 1570 |
+
"step": 1295
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 1.65,
|
| 1574 |
+
"learning_rate": 7.877501797214553e-06,
|
| 1575 |
+
"loss": 1.2842,
|
| 1576 |
+
"step": 1300
|
| 1577 |
+
},
|
| 1578 |
+
{
|
| 1579 |
+
"epoch": 1.65,
|
| 1580 |
+
"learning_rate": 7.605972711691228e-06,
|
| 1581 |
+
"loss": 1.4293,
|
| 1582 |
+
"step": 1305
|
| 1583 |
+
},
|
| 1584 |
+
{
|
| 1585 |
+
"epoch": 1.66,
|
| 1586 |
+
"learning_rate": 7.3388200735746486e-06,
|
| 1587 |
+
"loss": 1.3349,
|
| 1588 |
+
"step": 1310
|
| 1589 |
+
},
|
| 1590 |
+
{
|
| 1591 |
+
"epoch": 1.67,
|
| 1592 |
+
"learning_rate": 7.0760714617372605e-06,
|
| 1593 |
+
"loss": 1.4032,
|
| 1594 |
+
"step": 1315
|
| 1595 |
+
},
|
| 1596 |
+
{
|
| 1597 |
+
"epoch": 1.67,
|
| 1598 |
+
"learning_rate": 6.817754000412219e-06,
|
| 1599 |
+
"loss": 1.5321,
|
| 1600 |
+
"step": 1320
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 1.68,
|
| 1604 |
+
"learning_rate": 6.56389435639333e-06,
|
| 1605 |
+
"loss": 1.3822,
|
| 1606 |
+
"step": 1325
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 1.69,
|
| 1610 |
+
"learning_rate": 6.314518736282132e-06,
|
| 1611 |
+
"loss": 1.4665,
|
| 1612 |
+
"step": 1330
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 1.69,
|
| 1616 |
+
"learning_rate": 6.069652883782501e-06,
|
| 1617 |
+
"loss": 1.4638,
|
| 1618 |
+
"step": 1335
|
| 1619 |
+
},
|
| 1620 |
+
{
|
| 1621 |
+
"epoch": 1.7,
|
| 1622 |
+
"learning_rate": 5.829322077043159e-06,
|
| 1623 |
+
"loss": 1.417,
|
| 1624 |
+
"step": 1340
|
| 1625 |
+
},
|
| 1626 |
+
{
|
| 1627 |
+
"epoch": 1.7,
|
| 1628 |
+
"learning_rate": 5.593551126048041e-06,
|
| 1629 |
+
"loss": 1.4486,
|
| 1630 |
+
"step": 1345
|
| 1631 |
+
},
|
| 1632 |
+
{
|
| 1633 |
+
"epoch": 1.71,
|
| 1634 |
+
"learning_rate": 5.36236437005509e-06,
|
| 1635 |
+
"loss": 1.4988,
|
| 1636 |
+
"step": 1350
|
| 1637 |
+
},
|
| 1638 |
+
{
|
| 1639 |
+
"epoch": 1.72,
|
| 1640 |
+
"learning_rate": 5.1357856750837765e-06,
|
| 1641 |
+
"loss": 1.4268,
|
| 1642 |
+
"step": 1355
|
| 1643 |
+
},
|
| 1644 |
+
{
|
| 1645 |
+
"epoch": 1.72,
|
| 1646 |
+
"learning_rate": 4.913838431451184e-06,
|
| 1647 |
+
"loss": 1.392,
|
| 1648 |
+
"step": 1360
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 1.73,
|
| 1652 |
+
"learning_rate": 4.696545551357484e-06,
|
| 1653 |
+
"loss": 1.3894,
|
| 1654 |
+
"step": 1365
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 1.74,
|
| 1658 |
+
"learning_rate": 4.483929466520592e-06,
|
| 1659 |
+
"loss": 1.4446,
|
| 1660 |
+
"step": 1370
|
| 1661 |
+
},
|
| 1662 |
+
{
|
| 1663 |
+
"epoch": 1.74,
|
| 1664 |
+
"learning_rate": 4.276012125860507e-06,
|
| 1665 |
+
"loss": 1.3722,
|
| 1666 |
+
"step": 1375
|
| 1667 |
+
},
|
| 1668 |
+
{
|
| 1669 |
+
"epoch": 1.75,
|
| 1670 |
+
"learning_rate": 4.072814993233442e-06,
|
| 1671 |
+
"loss": 1.4785,
|
| 1672 |
+
"step": 1380
|
| 1673 |
+
},
|
| 1674 |
+
{
|
| 1675 |
+
"epoch": 1.76,
|
| 1676 |
+
"learning_rate": 3.874359045216086e-06,
|
| 1677 |
+
"loss": 1.4526,
|
| 1678 |
+
"step": 1385
|
| 1679 |
+
},
|
| 1680 |
+
{
|
| 1681 |
+
"epoch": 1.76,
|
| 1682 |
+
"learning_rate": 3.680664768940101e-06,
|
| 1683 |
+
"loss": 1.3871,
|
| 1684 |
+
"step": 1390
|
| 1685 |
+
},
|
| 1686 |
+
{
|
| 1687 |
+
"epoch": 1.77,
|
| 1688 |
+
"learning_rate": 3.491752159977213e-06,
|
| 1689 |
+
"loss": 1.4023,
|
| 1690 |
+
"step": 1395
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 1.77,
|
| 1694 |
+
"learning_rate": 3.3076407202749903e-06,
|
| 1695 |
+
"loss": 1.3858,
|
| 1696 |
+
"step": 1400
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 1.78,
|
| 1700 |
+
"learning_rate": 3.128349456143609e-06,
|
| 1701 |
+
"loss": 1.3945,
|
| 1702 |
+
"step": 1405
|
| 1703 |
+
},
|
| 1704 |
+
{
|
| 1705 |
+
"epoch": 1.79,
|
| 1706 |
+
"learning_rate": 2.9538968762938377e-06,
|
| 1707 |
+
"loss": 1.4255,
|
| 1708 |
+
"step": 1410
|
| 1709 |
+
},
|
| 1710 |
+
{
|
| 1711 |
+
"epoch": 1.79,
|
| 1712 |
+
"learning_rate": 2.7843009899262253e-06,
|
| 1713 |
+
"loss": 1.34,
|
| 1714 |
+
"step": 1415
|
| 1715 |
+
},
|
| 1716 |
+
{
|
| 1717 |
+
"epoch": 1.8,
|
| 1718 |
+
"learning_rate": 2.619579304872061e-06,
|
| 1719 |
+
"loss": 1.3443,
|
| 1720 |
+
"step": 1420
|
| 1721 |
+
},
|
| 1722 |
+
{
|
| 1723 |
+
"epoch": 1.81,
|
| 1724 |
+
"learning_rate": 2.4597488257859457e-06,
|
| 1725 |
+
"loss": 1.2774,
|
| 1726 |
+
"step": 1425
|
| 1727 |
+
},
|
| 1728 |
+
{
|
| 1729 |
+
"epoch": 1.81,
|
| 1730 |
+
"learning_rate": 2.304826052390385e-06,
|
| 1731 |
+
"loss": 1.4238,
|
| 1732 |
+
"step": 1430
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 1.82,
|
| 1736 |
+
"learning_rate": 2.1548269777724505e-06,
|
| 1737 |
+
"loss": 1.383,
|
| 1738 |
+
"step": 1435
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 1.83,
|
| 1742 |
+
"learning_rate": 2.0097670867327835e-06,
|
| 1743 |
+
"loss": 1.4196,
|
| 1744 |
+
"step": 1440
|
| 1745 |
+
},
|
| 1746 |
+
{
|
| 1747 |
+
"epoch": 1.83,
|
| 1748 |
+
"learning_rate": 1.8696613541870745e-06,
|
| 1749 |
+
"loss": 1.4295,
|
| 1750 |
+
"step": 1445
|
| 1751 |
+
},
|
| 1752 |
+
{
|
| 1753 |
+
"epoch": 1.84,
|
| 1754 |
+
"learning_rate": 1.734524243620117e-06,
|
| 1755 |
+
"loss": 1.3821,
|
| 1756 |
+
"step": 1450
|
| 1757 |
+
},
|
| 1758 |
+
{
|
| 1759 |
+
"epoch": 1.84,
|
| 1760 |
+
"learning_rate": 1.6043697055927676e-06,
|
| 1761 |
+
"loss": 1.4504,
|
| 1762 |
+
"step": 1455
|
| 1763 |
+
},
|
| 1764 |
+
{
|
| 1765 |
+
"epoch": 1.85,
|
| 1766 |
+
"learning_rate": 1.4792111763017536e-06,
|
| 1767 |
+
"loss": 1.4768,
|
| 1768 |
+
"step": 1460
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"epoch": 1.86,
|
| 1772 |
+
"learning_rate": 1.3590615761926051e-06,
|
| 1773 |
+
"loss": 1.3775,
|
| 1774 |
+
"step": 1465
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 1.86,
|
| 1778 |
+
"learning_rate": 1.243933308625883e-06,
|
| 1779 |
+
"loss": 1.4304,
|
| 1780 |
+
"step": 1470
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 1.87,
|
| 1784 |
+
"learning_rate": 1.1338382585967255e-06,
|
| 1785 |
+
"loss": 1.4992,
|
| 1786 |
+
"step": 1475
|
| 1787 |
+
},
|
| 1788 |
+
{
|
| 1789 |
+
"epoch": 1.88,
|
| 1790 |
+
"learning_rate": 1.028787791507918e-06,
|
| 1791 |
+
"loss": 1.4249,
|
| 1792 |
+
"step": 1480
|
| 1793 |
+
},
|
| 1794 |
+
{
|
| 1795 |
+
"epoch": 1.88,
|
| 1796 |
+
"learning_rate": 9.287927519966378e-07,
|
| 1797 |
+
"loss": 1.3822,
|
| 1798 |
+
"step": 1485
|
| 1799 |
+
},
|
| 1800 |
+
{
|
| 1801 |
+
"epoch": 1.89,
|
| 1802 |
+
"learning_rate": 8.338634628149211e-07,
|
| 1803 |
+
"loss": 1.3807,
|
| 1804 |
+
"step": 1490
|
| 1805 |
+
},
|
| 1806 |
+
{
|
| 1807 |
+
"epoch": 1.89,
|
| 1808 |
+
"learning_rate": 7.440097237640109e-07,
|
| 1809 |
+
"loss": 1.5082,
|
| 1810 |
+
"step": 1495
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 1.9,
|
| 1814 |
+
"learning_rate": 6.592408106827152e-07,
|
| 1815 |
+
"loss": 1.3872,
|
| 1816 |
+
"step": 1500
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 1.91,
|
| 1820 |
+
"learning_rate": 5.795654744898016e-07,
|
| 1821 |
+
"loss": 1.3692,
|
| 1822 |
+
"step": 1505
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 1.91,
|
| 1826 |
+
"learning_rate": 5.049919402806802e-07,
|
| 1827 |
+
"loss": 1.3778,
|
| 1828 |
+
"step": 1510
|
| 1829 |
+
},
|
| 1830 |
+
{
|
| 1831 |
+
"epoch": 1.92,
|
| 1832 |
+
"learning_rate": 4.355279064782447e-07,
|
| 1833 |
+
"loss": 1.4629,
|
| 1834 |
+
"step": 1515
|
| 1835 |
+
},
|
| 1836 |
+
{
|
| 1837 |
+
"epoch": 1.93,
|
| 1838 |
+
"learning_rate": 3.7118054403816326e-07,
|
| 1839 |
+
"loss": 1.3796,
|
| 1840 |
+
"step": 1520
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 1.93,
|
| 1844 |
+
"learning_rate": 3.119564957086207e-07,
|
| 1845 |
+
"loss": 1.462,
|
| 1846 |
+
"step": 1525
|
| 1847 |
+
},
|
| 1848 |
+
{
|
| 1849 |
+
"epoch": 1.94,
|
| 1850 |
+
"learning_rate": 2.578618753445394e-07,
|
| 1851 |
+
"loss": 1.3725,
|
| 1852 |
+
"step": 1530
|
| 1853 |
+
},
|
| 1854 |
+
{
|
| 1855 |
+
"epoch": 1.95,
|
| 1856 |
+
"learning_rate": 2.0890226727646733e-07,
|
| 1857 |
+
"loss": 1.3738,
|
| 1858 |
+
"step": 1535
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 1.95,
|
| 1862 |
+
"learning_rate": 1.6508272573403373e-07,
|
| 1863 |
+
"loss": 1.3906,
|
| 1864 |
+
"step": 1540
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 1.96,
|
| 1868 |
+
"learning_rate": 1.2640777432426642e-07,
|
| 1869 |
+
"loss": 1.4195,
|
| 1870 |
+
"step": 1545
|
| 1871 |
+
},
|
| 1872 |
+
{
|
| 1873 |
+
"epoch": 1.96,
|
| 1874 |
+
"learning_rate": 9.288140556453773e-08,
|
| 1875 |
+
"loss": 1.436,
|
| 1876 |
+
"step": 1550
|
| 1877 |
+
},
|
| 1878 |
+
{
|
| 1879 |
+
"epoch": 1.97,
|
| 1880 |
+
"learning_rate": 6.450708047044396e-08,
|
| 1881 |
+
"loss": 1.4758,
|
| 1882 |
+
"step": 1555
|
| 1883 |
+
},
|
| 1884 |
+
{
|
| 1885 |
+
"epoch": 1.98,
|
| 1886 |
+
"learning_rate": 4.128772819850801e-08,
|
| 1887 |
+
"loss": 1.4119,
|
| 1888 |
+
"step": 1560
|
| 1889 |
+
},
|
| 1890 |
+
{
|
| 1891 |
+
"epoch": 1.98,
|
| 1892 |
+
"learning_rate": 2.3225745743787887e-08,
|
| 1893 |
+
"loss": 1.4828,
|
| 1894 |
+
"step": 1565
|
| 1895 |
+
},
|
| 1896 |
+
{
|
| 1897 |
+
"epoch": 1.99,
|
| 1898 |
+
"learning_rate": 1.0322997692441272e-08,
|
| 1899 |
+
"loss": 1.4087,
|
| 1900 |
+
"step": 1570
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 2.0,
|
| 1904 |
+
"learning_rate": 2.580816029223998e-09,
|
| 1905 |
+
"loss": 1.4049,
|
| 1906 |
+
"step": 1575
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 2.0,
|
| 1910 |
+
"eval_loss": 1.6930410861968994,
|
| 1911 |
+
"eval_runtime": 651.437,
|
| 1912 |
+
"eval_samples_per_second": 19.66,
|
| 1913 |
+
"eval_steps_per_second": 4.915,
|
| 1914 |
+
"step": 1578
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 2.0,
|
| 1918 |
+
"step": 1578,
|
| 1919 |
+
"total_flos": 1.4997323849651978e+18,
|
| 1920 |
+
"train_loss": 1.7094378060412196,
|
| 1921 |
+
"train_runtime": 39550.5514,
|
| 1922 |
+
"train_samples_per_second": 5.107,
|
| 1923 |
+
"train_steps_per_second": 0.04
|
| 1924 |
+
}
|
| 1925 |
+
],
|
| 1926 |
+
"max_steps": 1578,
|
| 1927 |
+
"num_train_epochs": 2,
|
| 1928 |
+
"total_flos": 1.4997323849651978e+18,
|
| 1929 |
+
"trial_name": null,
|
| 1930 |
+
"trial_params": null
|
| 1931 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3ce776571759962d4b12d03006933a5edd8b22f7993a95076b55f3224c66ca30
|
| 3 |
+
size 3567
|