Draft uploaded
Browse files- README.md +117 -3
- all_results.json +15 -0
- config.json +54 -0
- eval_results.json +10 -0
- merges.txt +0 -0
- pytorch_model.bin +3 -0
- runs/Aug03_12-27-18_DESKTOP-LE3I0SI/1659522461.8353674/events.out.tfevents.1659522461.DESKTOP-LE3I0SI.10376.1 +3 -0
- runs/Aug03_12-27-18_DESKTOP-LE3I0SI/events.out.tfevents.1659522461.DESKTOP-LE3I0SI.10376.0 +3 -0
- runs/Aug03_12-27-18_DESKTOP-LE3I0SI/events.out.tfevents.1659643194.DESKTOP-LE3I0SI.10376.2 +3 -0
- special_tokens_map.json +5 -0
- tokenizer.json +0 -0
- tokenizer_config.json +9 -0
- train_results.json +8 -0
- trainer_state.json +2809 -0
- training_args.bin +3 -0
- vocab.json +0 -0
README.md
CHANGED
|
@@ -1,3 +1,117 @@
|
|
| 1 |
-
---
|
| 2 |
-
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language: pl
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
- text-generation
|
| 6 |
+
widget:
|
| 7 |
+
- text: "Bolesław Leśmian - polski poeta"
|
| 8 |
+
datasets:
|
| 9 |
+
- wikipedia
|
| 10 |
+
metrics:
|
| 11 |
+
- accuracy
|
| 12 |
+
model-index:
|
| 13 |
+
- name: gpt_neo_pl_125M
|
| 14 |
+
results:
|
| 15 |
+
- task:
|
| 16 |
+
name: Causal Language Modeling
|
| 17 |
+
type: text-generation
|
| 18 |
+
dataset:
|
| 19 |
+
name: wikipedia 20220720.pl
|
| 20 |
+
type: wikipedia
|
| 21 |
+
args: 20220720.pl
|
| 22 |
+
metrics:
|
| 23 |
+
- name: Accuracy
|
| 24 |
+
type: accuracy
|
| 25 |
+
value: 0.4312838299951148
|
| 26 |
+
---
|
| 27 |
+
|
| 28 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 29 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 30 |
+
|
| 31 |
+
# gpt_neo_pl_125M_v2
|
| 32 |
+
|
| 33 |
+
This model was trained from scratch on the wikipedia 20220720.pl dataset.
|
| 34 |
+
It achieves the following results on the evaluation set:
|
| 35 |
+
- Loss: 3.3862
|
| 36 |
+
- Accuracy: 0.4313
|
| 37 |
+
|
| 38 |
+
## Model description
|
| 39 |
+
|
| 40 |
+
More information needed
|
| 41 |
+
|
| 42 |
+
## Intended uses & limitations
|
| 43 |
+
|
| 44 |
+
More information needed
|
| 45 |
+
|
| 46 |
+
## Training and evaluation data
|
| 47 |
+
|
| 48 |
+
More information needed
|
| 49 |
+
|
| 50 |
+
## Training procedure
|
| 51 |
+
|
| 52 |
+
### Training hyperparameters
|
| 53 |
+
|
| 54 |
+
The following hyperparameters were used during training:
|
| 55 |
+
- learning_rate: 0.0002
|
| 56 |
+
- train_batch_size: 1
|
| 57 |
+
- eval_batch_size: 2
|
| 58 |
+
- seed: 42
|
| 59 |
+
- gradient_accumulation_steps: 8
|
| 60 |
+
- total_train_batch_size: 8
|
| 61 |
+
- optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-08
|
| 62 |
+
- lr_scheduler_type: cosine
|
| 63 |
+
- lr_scheduler_warmup_steps: 1000
|
| 64 |
+
- num_epochs: 1.0
|
| 65 |
+
|
| 66 |
+
### Training results
|
| 67 |
+
|
| 68 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 69 |
+
|:-------------:|:-----:|:-----:|:---------------:|:--------:|
|
| 70 |
+
| 5.9469 | 0.02 | 1000 | 6.5843 | 0.1435 |
|
| 71 |
+
| 4.9953 | 0.05 | 2000 | 5.7709 | 0.1911 |
|
| 72 |
+
| 4.3754 | 0.07 | 3000 | 5.2624 | 0.2331 |
|
| 73 |
+
| 3.9795 | 0.1 | 4000 | 4.8752 | 0.2731 |
|
| 74 |
+
| 3.7099 | 0.12 | 5000 | 4.5927 | 0.3039 |
|
| 75 |
+
| 3.4747 | 0.15 | 6000 | 4.3942 | 0.3230 |
|
| 76 |
+
| 3.343 | 0.17 | 7000 | 4.2879 | 0.3349 |
|
| 77 |
+
| 3.2767 | 0.2 | 8000 | 4.1698 | 0.3459 |
|
| 78 |
+
| 3.1852 | 0.22 | 9000 | 4.0925 | 0.3534 |
|
| 79 |
+
| 3.0871 | 0.25 | 10000 | 4.0239 | 0.3608 |
|
| 80 |
+
| 3.0746 | 0.27 | 11000 | 3.9646 | 0.3664 |
|
| 81 |
+
| 2.9473 | 0.3 | 12000 | 3.9245 | 0.3706 |
|
| 82 |
+
| 2.9737 | 0.32 | 13000 | 3.8742 | 0.3754 |
|
| 83 |
+
| 2.9193 | 0.35 | 14000 | 3.8285 | 0.3796 |
|
| 84 |
+
| 2.8833 | 0.37 | 15000 | 3.7952 | 0.3837 |
|
| 85 |
+
| 2.8533 | 0.4 | 16000 | 3.7616 | 0.3873 |
|
| 86 |
+
| 2.8654 | 0.42 | 17000 | 3.7296 | 0.3907 |
|
| 87 |
+
| 2.8196 | 0.44 | 18000 | 3.7049 | 0.3936 |
|
| 88 |
+
| 2.7883 | 0.47 | 19000 | 3.6786 | 0.3966 |
|
| 89 |
+
| 2.747 | 0.49 | 20000 | 3.6488 | 0.3990 |
|
| 90 |
+
| 2.7355 | 0.52 | 21000 | 3.6243 | 0.4021 |
|
| 91 |
+
| 2.7355 | 0.54 | 22000 | 3.5982 | 0.4053 |
|
| 92 |
+
| 2.6999 | 0.57 | 23000 | 3.5765 | 0.4075 |
|
| 93 |
+
| 2.7243 | 0.59 | 24000 | 3.5558 | 0.4101 |
|
| 94 |
+
| 2.6526 | 0.62 | 25000 | 3.5371 | 0.4125 |
|
| 95 |
+
| 2.641 | 0.64 | 26000 | 3.5150 | 0.4146 |
|
| 96 |
+
| 2.6602 | 0.67 | 27000 | 3.4971 | 0.4168 |
|
| 97 |
+
| 2.644 | 0.69 | 28000 | 3.4812 | 0.4192 |
|
| 98 |
+
| 2.6558 | 0.72 | 29000 | 3.4622 | 0.4215 |
|
| 99 |
+
| 2.5664 | 0.74 | 30000 | 3.4504 | 0.4229 |
|
| 100 |
+
| 2.5669 | 0.77 | 31000 | 3.4376 | 0.4245 |
|
| 101 |
+
| 2.5498 | 0.79 | 32000 | 3.4263 | 0.4263 |
|
| 102 |
+
| 2.5874 | 0.82 | 33000 | 3.4169 | 0.4274 |
|
| 103 |
+
| 2.5555 | 0.84 | 34000 | 3.4067 | 0.4286 |
|
| 104 |
+
| 2.5502 | 0.86 | 35000 | 3.3997 | 0.4298 |
|
| 105 |
+
| 2.5232 | 0.89 | 36000 | 3.3946 | 0.4302 |
|
| 106 |
+
| 2.5369 | 0.91 | 37000 | 3.3898 | 0.4309 |
|
| 107 |
+
| 2.5335 | 0.94 | 38000 | 3.3869 | 0.4313 |
|
| 108 |
+
| 2.6032 | 0.96 | 39000 | 3.3853 | 0.4315 |
|
| 109 |
+
| 2.5244 | 0.99 | 40000 | 3.3850 | 0.4314 |
|
| 110 |
+
|
| 111 |
+
|
| 112 |
+
### Framework versions
|
| 113 |
+
|
| 114 |
+
- Transformers 4.22.0.dev0
|
| 115 |
+
- Pytorch 1.12.0
|
| 116 |
+
- Datasets 2.4.0
|
| 117 |
+
- Tokenizers 0.12.1
|
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_accuracy": 0.4312838299951148,
|
| 4 |
+
"eval_loss": 3.386244058609009,
|
| 5 |
+
"eval_runtime": 155.058,
|
| 6 |
+
"eval_samples": 1000,
|
| 7 |
+
"eval_samples_per_second": 6.449,
|
| 8 |
+
"eval_steps_per_second": 3.225,
|
| 9 |
+
"perplexity": 29.55473768081637,
|
| 10 |
+
"train_loss": 3.073313300175377,
|
| 11 |
+
"train_runtime": 120576.6781,
|
| 12 |
+
"train_samples": 323789,
|
| 13 |
+
"train_samples_per_second": 2.685,
|
| 14 |
+
"train_steps_per_second": 0.336
|
| 15 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,54 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": ".\\gpt_neo_pl_125M",
|
| 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 |
+
],
|
| 22 |
+
"attention_types": [
|
| 23 |
+
[
|
| 24 |
+
[
|
| 25 |
+
"global",
|
| 26 |
+
"local"
|
| 27 |
+
],
|
| 28 |
+
6
|
| 29 |
+
]
|
| 30 |
+
],
|
| 31 |
+
"bos_token_id": 0,
|
| 32 |
+
"embed_dropout": 0,
|
| 33 |
+
"eos_token_id": 0,
|
| 34 |
+
"gradient_checkpointing": false,
|
| 35 |
+
"hidden_size": 768,
|
| 36 |
+
"initializer_range": 0.02,
|
| 37 |
+
"intermediate_size": null,
|
| 38 |
+
"layer_norm_epsilon": 1e-05,
|
| 39 |
+
"max_position_embeddings": 2048,
|
| 40 |
+
"model_type": "gpt_neo",
|
| 41 |
+
"num_heads": 12,
|
| 42 |
+
"num_layers": 12,
|
| 43 |
+
"resid_dropout": 0,
|
| 44 |
+
"summary_activation": null,
|
| 45 |
+
"summary_first_dropout": 0.1,
|
| 46 |
+
"summary_proj_to_labels": true,
|
| 47 |
+
"summary_type": "cls_index",
|
| 48 |
+
"summary_use_proj": true,
|
| 49 |
+
"torch_dtype": "float32",
|
| 50 |
+
"transformers_version": "4.22.0.dev0",
|
| 51 |
+
"use_cache": true,
|
| 52 |
+
"vocab_size": 50265,
|
| 53 |
+
"window_size": 256
|
| 54 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_accuracy": 0.4312838299951148,
|
| 4 |
+
"eval_loss": 3.386244058609009,
|
| 5 |
+
"eval_runtime": 155.058,
|
| 6 |
+
"eval_samples": 1000,
|
| 7 |
+
"eval_samples_per_second": 6.449,
|
| 8 |
+
"eval_steps_per_second": 3.225,
|
| 9 |
+
"perplexity": 29.55473768081637
|
| 10 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cace4f7863e18c8f26e132b0668a2099885ef47a2dd36d3edb98c2c12edb7fed
|
| 3 |
+
size 551209681
|
runs/Aug03_12-27-18_DESKTOP-LE3I0SI/1659522461.8353674/events.out.tfevents.1659522461.DESKTOP-LE3I0SI.10376.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3ad44293aeba480bfe541bd31a7f5707db4e2fbee143248423177900f02bafc8
|
| 3 |
+
size 5449
|
runs/Aug03_12-27-18_DESKTOP-LE3I0SI/events.out.tfevents.1659522461.DESKTOP-LE3I0SI.10376.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7c929705e74a1243b51a245c638ddb5c1732c9c44d4cf6e9fcf0351287cab22f
|
| 3 |
+
size 81698
|
runs/Aug03_12-27-18_DESKTOP-LE3I0SI/events.out.tfevents.1659643194.DESKTOP-LE3I0SI.10376.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b75d4657da27465a378f8f1fb8e26eb9904e8354e4c85e2b5f9958a693035b6c
|
| 3 |
+
size 369
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<|endoftext|>",
|
| 3 |
+
"eos_token": "<|endoftext|>",
|
| 4 |
+
"unk_token": "<|endoftext|>"
|
| 5 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"bos_token": "<|endoftext|>",
|
| 4 |
+
"eos_token": "<|endoftext|>",
|
| 5 |
+
"name_or_path": ".\\gpt_neo_pl_125M",
|
| 6 |
+
"special_tokens_map_file": null,
|
| 7 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 8 |
+
"unk_token": "<|endoftext|>"
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 3.073313300175377,
|
| 4 |
+
"train_runtime": 120576.6781,
|
| 5 |
+
"train_samples": 323789,
|
| 6 |
+
"train_samples_per_second": 2.685,
|
| 7 |
+
"train_steps_per_second": 0.336
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2809 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9999845578447693,
|
| 5 |
+
"global_step": 40473,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.0,
|
| 12 |
+
"learning_rate": 2e-05,
|
| 13 |
+
"loss": 9.7353,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.0,
|
| 18 |
+
"learning_rate": 4e-05,
|
| 19 |
+
"loss": 8.2306,
|
| 20 |
+
"step": 200
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.01,
|
| 24 |
+
"learning_rate": 6e-05,
|
| 25 |
+
"loss": 7.5102,
|
| 26 |
+
"step": 300
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.01,
|
| 30 |
+
"learning_rate": 8e-05,
|
| 31 |
+
"loss": 7.2441,
|
| 32 |
+
"step": 400
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.01,
|
| 36 |
+
"learning_rate": 0.0001,
|
| 37 |
+
"loss": 6.9668,
|
| 38 |
+
"step": 500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.01,
|
| 42 |
+
"learning_rate": 0.00012,
|
| 43 |
+
"loss": 6.6706,
|
| 44 |
+
"step": 600
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.02,
|
| 48 |
+
"learning_rate": 0.00014,
|
| 49 |
+
"loss": 6.4207,
|
| 50 |
+
"step": 700
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.02,
|
| 54 |
+
"learning_rate": 0.00016,
|
| 55 |
+
"loss": 6.2927,
|
| 56 |
+
"step": 800
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.02,
|
| 60 |
+
"learning_rate": 0.00018,
|
| 61 |
+
"loss": 6.1555,
|
| 62 |
+
"step": 900
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.02,
|
| 66 |
+
"learning_rate": 0.0002,
|
| 67 |
+
"loss": 5.9469,
|
| 68 |
+
"step": 1000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.02,
|
| 72 |
+
"eval_accuracy": 0.14350219833903274,
|
| 73 |
+
"eval_loss": 6.5843048095703125,
|
| 74 |
+
"eval_runtime": 149.8081,
|
| 75 |
+
"eval_samples_per_second": 6.675,
|
| 76 |
+
"eval_steps_per_second": 3.338,
|
| 77 |
+
"step": 1000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.03,
|
| 81 |
+
"learning_rate": 0.00019999683286053426,
|
| 82 |
+
"loss": 5.8516,
|
| 83 |
+
"step": 1100
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.03,
|
| 87 |
+
"learning_rate": 0.0001999873316427525,
|
| 88 |
+
"loss": 5.6928,
|
| 89 |
+
"step": 1200
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.03,
|
| 93 |
+
"learning_rate": 0.00019997149694848834,
|
| 94 |
+
"loss": 5.5905,
|
| 95 |
+
"step": 1300
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.03,
|
| 99 |
+
"learning_rate": 0.00019994932978075548,
|
| 100 |
+
"loss": 5.4808,
|
| 101 |
+
"step": 1400
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.04,
|
| 105 |
+
"learning_rate": 0.00019992083154368414,
|
| 106 |
+
"loss": 5.3261,
|
| 107 |
+
"step": 1500
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.04,
|
| 111 |
+
"learning_rate": 0.00019988600404243222,
|
| 112 |
+
"loss": 5.2571,
|
| 113 |
+
"step": 1600
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.04,
|
| 117 |
+
"learning_rate": 0.0001998448494830707,
|
| 118 |
+
"loss": 5.1842,
|
| 119 |
+
"step": 1700
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.04,
|
| 123 |
+
"learning_rate": 0.00019979737047244424,
|
| 124 |
+
"loss": 5.1502,
|
| 125 |
+
"step": 1800
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.05,
|
| 129 |
+
"learning_rate": 0.00019974357001800575,
|
| 130 |
+
"loss": 5.0854,
|
| 131 |
+
"step": 1900
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.05,
|
| 135 |
+
"learning_rate": 0.00019968345152762614,
|
| 136 |
+
"loss": 4.9953,
|
| 137 |
+
"step": 2000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.05,
|
| 141 |
+
"eval_accuracy": 0.1911294577430386,
|
| 142 |
+
"eval_loss": 5.77092170715332,
|
| 143 |
+
"eval_runtime": 149.2322,
|
| 144 |
+
"eval_samples_per_second": 6.701,
|
| 145 |
+
"eval_steps_per_second": 3.35,
|
| 146 |
+
"step": 2000
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.05,
|
| 150 |
+
"learning_rate": 0.00019961701880937823,
|
| 151 |
+
"loss": 4.9123,
|
| 152 |
+
"step": 2100
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.05,
|
| 156 |
+
"learning_rate": 0.0001995442760712957,
|
| 157 |
+
"loss": 4.8373,
|
| 158 |
+
"step": 2200
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.06,
|
| 162 |
+
"learning_rate": 0.00019946522792110648,
|
| 163 |
+
"loss": 4.7841,
|
| 164 |
+
"step": 2300
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.06,
|
| 168 |
+
"learning_rate": 0.00019937987936594095,
|
| 169 |
+
"loss": 4.731,
|
| 170 |
+
"step": 2400
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.06,
|
| 174 |
+
"learning_rate": 0.00019928823581201455,
|
| 175 |
+
"loss": 4.6354,
|
| 176 |
+
"step": 2500
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.06,
|
| 180 |
+
"learning_rate": 0.0001991903030642857,
|
| 181 |
+
"loss": 4.6333,
|
| 182 |
+
"step": 2600
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.07,
|
| 186 |
+
"learning_rate": 0.0001990860873260878,
|
| 187 |
+
"loss": 4.5291,
|
| 188 |
+
"step": 2700
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.07,
|
| 192 |
+
"learning_rate": 0.00019897559519873635,
|
| 193 |
+
"loss": 4.4949,
|
| 194 |
+
"step": 2800
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.07,
|
| 198 |
+
"learning_rate": 0.00019885883368111093,
|
| 199 |
+
"loss": 4.4967,
|
| 200 |
+
"step": 2900
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.07,
|
| 204 |
+
"learning_rate": 0.00019873581016921173,
|
| 205 |
+
"loss": 4.3754,
|
| 206 |
+
"step": 3000
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.07,
|
| 210 |
+
"eval_accuracy": 0.23306155349291646,
|
| 211 |
+
"eval_loss": 5.262444019317627,
|
| 212 |
+
"eval_runtime": 151.6242,
|
| 213 |
+
"eval_samples_per_second": 6.595,
|
| 214 |
+
"eval_steps_per_second": 3.298,
|
| 215 |
+
"step": 3000
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 0.08,
|
| 219 |
+
"learning_rate": 0.0001986065324556912,
|
| 220 |
+
"loss": 4.3805,
|
| 221 |
+
"step": 3100
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.08,
|
| 225 |
+
"learning_rate": 0.00019847100872936023,
|
| 226 |
+
"loss": 4.2668,
|
| 227 |
+
"step": 3200
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.08,
|
| 231 |
+
"learning_rate": 0.00019832924757466974,
|
| 232 |
+
"loss": 4.289,
|
| 233 |
+
"step": 3300
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.08,
|
| 237 |
+
"learning_rate": 0.0001981812579711666,
|
| 238 |
+
"loss": 4.2334,
|
| 239 |
+
"step": 3400
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 0.09,
|
| 243 |
+
"learning_rate": 0.00019802704929292514,
|
| 244 |
+
"loss": 4.2227,
|
| 245 |
+
"step": 3500
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 0.09,
|
| 249 |
+
"learning_rate": 0.00019786663130795317,
|
| 250 |
+
"loss": 4.1494,
|
| 251 |
+
"step": 3600
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 0.09,
|
| 255 |
+
"learning_rate": 0.00019770001417757332,
|
| 256 |
+
"loss": 4.1242,
|
| 257 |
+
"step": 3700
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.09,
|
| 261 |
+
"learning_rate": 0.00019752720845577938,
|
| 262 |
+
"loss": 4.0128,
|
| 263 |
+
"step": 3800
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 0.1,
|
| 267 |
+
"learning_rate": 0.00019734822508856773,
|
| 268 |
+
"loss": 3.9907,
|
| 269 |
+
"step": 3900
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.1,
|
| 273 |
+
"learning_rate": 0.00019716307541324418,
|
| 274 |
+
"loss": 3.9795,
|
| 275 |
+
"step": 4000
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.1,
|
| 279 |
+
"eval_accuracy": 0.27312457254518807,
|
| 280 |
+
"eval_loss": 4.875171184539795,
|
| 281 |
+
"eval_runtime": 149.3435,
|
| 282 |
+
"eval_samples_per_second": 6.696,
|
| 283 |
+
"eval_steps_per_second": 3.348,
|
| 284 |
+
"step": 4000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.1,
|
| 288 |
+
"learning_rate": 0.00019697177115770548,
|
| 289 |
+
"loss": 3.9426,
|
| 290 |
+
"step": 4100
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.1,
|
| 294 |
+
"learning_rate": 0.0001967743244396969,
|
| 295 |
+
"loss": 3.8986,
|
| 296 |
+
"step": 4200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.11,
|
| 300 |
+
"learning_rate": 0.00019657074776604426,
|
| 301 |
+
"loss": 3.8835,
|
| 302 |
+
"step": 4300
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.11,
|
| 306 |
+
"learning_rate": 0.0001963610540318619,
|
| 307 |
+
"loss": 3.9364,
|
| 308 |
+
"step": 4400
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.11,
|
| 312 |
+
"learning_rate": 0.00019614525651973582,
|
| 313 |
+
"loss": 3.7685,
|
| 314 |
+
"step": 4500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.11,
|
| 318 |
+
"learning_rate": 0.00019592336889888244,
|
| 319 |
+
"loss": 3.7921,
|
| 320 |
+
"step": 4600
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.12,
|
| 324 |
+
"learning_rate": 0.0001956954052242825,
|
| 325 |
+
"loss": 3.7474,
|
| 326 |
+
"step": 4700
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.12,
|
| 330 |
+
"learning_rate": 0.00019546137993579107,
|
| 331 |
+
"loss": 3.7227,
|
| 332 |
+
"step": 4800
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.12,
|
| 336 |
+
"learning_rate": 0.00019522130785722266,
|
| 337 |
+
"loss": 3.7087,
|
| 338 |
+
"step": 4900
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.12,
|
| 342 |
+
"learning_rate": 0.00019497520419541235,
|
| 343 |
+
"loss": 3.7099,
|
| 344 |
+
"step": 5000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.12,
|
| 348 |
+
"eval_accuracy": 0.30389936492427944,
|
| 349 |
+
"eval_loss": 4.592658996582031,
|
| 350 |
+
"eval_runtime": 149.1172,
|
| 351 |
+
"eval_samples_per_second": 6.706,
|
| 352 |
+
"eval_steps_per_second": 3.353,
|
| 353 |
+
"step": 5000
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.13,
|
| 357 |
+
"learning_rate": 0.00019472308453925258,
|
| 358 |
+
"loss": 3.649,
|
| 359 |
+
"step": 5100
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.13,
|
| 363 |
+
"learning_rate": 0.0001944649648587056,
|
| 364 |
+
"loss": 3.6001,
|
| 365 |
+
"step": 5200
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 0.13,
|
| 369 |
+
"learning_rate": 0.00019420086150379192,
|
| 370 |
+
"loss": 3.6509,
|
| 371 |
+
"step": 5300
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 0.13,
|
| 375 |
+
"learning_rate": 0.00019393079120355475,
|
| 376 |
+
"loss": 3.598,
|
| 377 |
+
"step": 5400
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.14,
|
| 381 |
+
"learning_rate": 0.0001936547710650002,
|
| 382 |
+
"loss": 3.5339,
|
| 383 |
+
"step": 5500
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 0.14,
|
| 387 |
+
"learning_rate": 0.00019337281857201378,
|
| 388 |
+
"loss": 3.573,
|
| 389 |
+
"step": 5600
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.14,
|
| 393 |
+
"learning_rate": 0.0001930849515842528,
|
| 394 |
+
"loss": 3.5411,
|
| 395 |
+
"step": 5700
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.14,
|
| 399 |
+
"learning_rate": 0.00019279118833601523,
|
| 400 |
+
"loss": 3.5458,
|
| 401 |
+
"step": 5800
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.15,
|
| 405 |
+
"learning_rate": 0.00019249154743508463,
|
| 406 |
+
"loss": 3.5076,
|
| 407 |
+
"step": 5900
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 0.15,
|
| 411 |
+
"learning_rate": 0.00019218604786155148,
|
| 412 |
+
"loss": 3.4747,
|
| 413 |
+
"step": 6000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 0.15,
|
| 417 |
+
"eval_accuracy": 0.32299853444064486,
|
| 418 |
+
"eval_loss": 4.394214630126953,
|
| 419 |
+
"eval_runtime": 164.2578,
|
| 420 |
+
"eval_samples_per_second": 6.088,
|
| 421 |
+
"eval_steps_per_second": 3.044,
|
| 422 |
+
"step": 6000
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.15,
|
| 426 |
+
"learning_rate": 0.00019187470896661084,
|
| 427 |
+
"loss": 3.5194,
|
| 428 |
+
"step": 6100
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.15,
|
| 432 |
+
"learning_rate": 0.00019155755047133678,
|
| 433 |
+
"loss": 3.4677,
|
| 434 |
+
"step": 6200
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.16,
|
| 438 |
+
"learning_rate": 0.00019123459246543303,
|
| 439 |
+
"loss": 3.4157,
|
| 440 |
+
"step": 6300
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.16,
|
| 444 |
+
"learning_rate": 0.00019090585540596053,
|
| 445 |
+
"loss": 3.4342,
|
| 446 |
+
"step": 6400
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.16,
|
| 450 |
+
"learning_rate": 0.00019057136011604156,
|
| 451 |
+
"loss": 3.436,
|
| 452 |
+
"step": 6500
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.16,
|
| 456 |
+
"learning_rate": 0.00019023112778354082,
|
| 457 |
+
"loss": 3.394,
|
| 458 |
+
"step": 6600
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.17,
|
| 462 |
+
"learning_rate": 0.00018988517995972325,
|
| 463 |
+
"loss": 3.4081,
|
| 464 |
+
"step": 6700
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.17,
|
| 468 |
+
"learning_rate": 0.00018953353855788896,
|
| 469 |
+
"loss": 3.3736,
|
| 470 |
+
"step": 6800
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.17,
|
| 474 |
+
"learning_rate": 0.0001891762258519852,
|
| 475 |
+
"loss": 3.3793,
|
| 476 |
+
"step": 6900
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.17,
|
| 480 |
+
"learning_rate": 0.0001888132644751954,
|
| 481 |
+
"loss": 3.343,
|
| 482 |
+
"step": 7000
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.17,
|
| 486 |
+
"eval_accuracy": 0.334904738641915,
|
| 487 |
+
"eval_loss": 4.287891387939453,
|
| 488 |
+
"eval_runtime": 142.0586,
|
| 489 |
+
"eval_samples_per_second": 7.039,
|
| 490 |
+
"eval_steps_per_second": 3.52,
|
| 491 |
+
"step": 7000
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 0.18,
|
| 495 |
+
"learning_rate": 0.0001884446774185056,
|
| 496 |
+
"loss": 3.3829,
|
| 497 |
+
"step": 7100
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 0.18,
|
| 501 |
+
"learning_rate": 0.00018807048802924804,
|
| 502 |
+
"loss": 3.3683,
|
| 503 |
+
"step": 7200
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 0.18,
|
| 507 |
+
"learning_rate": 0.0001876907200096224,
|
| 508 |
+
"loss": 3.299,
|
| 509 |
+
"step": 7300
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 0.18,
|
| 513 |
+
"learning_rate": 0.00018730539741519436,
|
| 514 |
+
"loss": 3.3129,
|
| 515 |
+
"step": 7400
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 0.19,
|
| 519 |
+
"learning_rate": 0.0001869145446533718,
|
| 520 |
+
"loss": 3.3329,
|
| 521 |
+
"step": 7500
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.19,
|
| 525 |
+
"learning_rate": 0.00018651818648185886,
|
| 526 |
+
"loss": 3.3036,
|
| 527 |
+
"step": 7600
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.19,
|
| 531 |
+
"learning_rate": 0.00018611634800708767,
|
| 532 |
+
"loss": 3.2714,
|
| 533 |
+
"step": 7700
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 0.19,
|
| 537 |
+
"learning_rate": 0.00018570905468262816,
|
| 538 |
+
"loss": 3.2606,
|
| 539 |
+
"step": 7800
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 0.2,
|
| 543 |
+
"learning_rate": 0.00018529633230757547,
|
| 544 |
+
"loss": 3.2601,
|
| 545 |
+
"step": 7900
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 0.2,
|
| 549 |
+
"learning_rate": 0.00018487820702491615,
|
| 550 |
+
"loss": 3.2767,
|
| 551 |
+
"step": 8000
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 0.2,
|
| 555 |
+
"eval_accuracy": 0.3458881289692233,
|
| 556 |
+
"eval_loss": 4.169787883758545,
|
| 557 |
+
"eval_runtime": 147.5408,
|
| 558 |
+
"eval_samples_per_second": 6.778,
|
| 559 |
+
"eval_steps_per_second": 3.389,
|
| 560 |
+
"step": 8000
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.2,
|
| 564 |
+
"learning_rate": 0.00018445470531987178,
|
| 565 |
+
"loss": 3.2766,
|
| 566 |
+
"step": 8100
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 0.2,
|
| 570 |
+
"learning_rate": 0.00018402585401822172,
|
| 571 |
+
"loss": 3.2475,
|
| 572 |
+
"step": 8200
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 0.21,
|
| 576 |
+
"learning_rate": 0.0001835916802846036,
|
| 577 |
+
"loss": 3.2341,
|
| 578 |
+
"step": 8300
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.21,
|
| 582 |
+
"learning_rate": 0.0001831522116207927,
|
| 583 |
+
"loss": 3.2234,
|
| 584 |
+
"step": 8400
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.21,
|
| 588 |
+
"learning_rate": 0.00018270747586396008,
|
| 589 |
+
"loss": 3.2216,
|
| 590 |
+
"step": 8500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.21,
|
| 594 |
+
"learning_rate": 0.00018225750118490904,
|
| 595 |
+
"loss": 3.2029,
|
| 596 |
+
"step": 8600
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 0.21,
|
| 600 |
+
"learning_rate": 0.0001818023160862909,
|
| 601 |
+
"loss": 3.1778,
|
| 602 |
+
"step": 8700
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.22,
|
| 606 |
+
"learning_rate": 0.00018134194940079942,
|
| 607 |
+
"loss": 3.1699,
|
| 608 |
+
"step": 8800
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.22,
|
| 612 |
+
"learning_rate": 0.0001808764302893446,
|
| 613 |
+
"loss": 3.1801,
|
| 614 |
+
"step": 8900
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.22,
|
| 618 |
+
"learning_rate": 0.00018040578823920545,
|
| 619 |
+
"loss": 3.1852,
|
| 620 |
+
"step": 9000
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 0.22,
|
| 624 |
+
"eval_accuracy": 0.3533742061553493,
|
| 625 |
+
"eval_loss": 4.092532634735107,
|
| 626 |
+
"eval_runtime": 149.3684,
|
| 627 |
+
"eval_samples_per_second": 6.695,
|
| 628 |
+
"eval_steps_per_second": 3.347,
|
| 629 |
+
"step": 9000
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 0.22,
|
| 633 |
+
"learning_rate": 0.00017993005306216216,
|
| 634 |
+
"loss": 3.1856,
|
| 635 |
+
"step": 9100
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 0.23,
|
| 639 |
+
"learning_rate": 0.00017944925489260784,
|
| 640 |
+
"loss": 3.1604,
|
| 641 |
+
"step": 9200
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 0.23,
|
| 645 |
+
"learning_rate": 0.00017896342418563963,
|
| 646 |
+
"loss": 3.1725,
|
| 647 |
+
"step": 9300
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.23,
|
| 651 |
+
"learning_rate": 0.00017847259171512963,
|
| 652 |
+
"loss": 3.1451,
|
| 653 |
+
"step": 9400
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.23,
|
| 657 |
+
"learning_rate": 0.00017797678857177566,
|
| 658 |
+
"loss": 3.142,
|
| 659 |
+
"step": 9500
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 0.24,
|
| 663 |
+
"learning_rate": 0.00017747604616113175,
|
| 664 |
+
"loss": 3.1825,
|
| 665 |
+
"step": 9600
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 0.24,
|
| 669 |
+
"learning_rate": 0.00017697039620161894,
|
| 670 |
+
"loss": 3.1747,
|
| 671 |
+
"step": 9700
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 0.24,
|
| 675 |
+
"learning_rate": 0.000176459870722516,
|
| 676 |
+
"loss": 3.0989,
|
| 677 |
+
"step": 9800
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 0.24,
|
| 681 |
+
"learning_rate": 0.00017594450206193088,
|
| 682 |
+
"loss": 3.1139,
|
| 683 |
+
"step": 9900
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 0.25,
|
| 687 |
+
"learning_rate": 0.000175424322864752,
|
| 688 |
+
"loss": 3.0871,
|
| 689 |
+
"step": 10000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.25,
|
| 693 |
+
"eval_accuracy": 0.360838299951148,
|
| 694 |
+
"eval_loss": 4.023919105529785,
|
| 695 |
+
"eval_runtime": 149.3116,
|
| 696 |
+
"eval_samples_per_second": 6.697,
|
| 697 |
+
"eval_steps_per_second": 3.349,
|
| 698 |
+
"step": 10000
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 0.25,
|
| 702 |
+
"learning_rate": 0.00017489936608058073,
|
| 703 |
+
"loss": 3.1034,
|
| 704 |
+
"step": 10100
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.25,
|
| 708 |
+
"learning_rate": 0.00017436966496164397,
|
| 709 |
+
"loss": 3.0933,
|
| 710 |
+
"step": 10200
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.25,
|
| 714 |
+
"learning_rate": 0.00017383525306068816,
|
| 715 |
+
"loss": 3.1107,
|
| 716 |
+
"step": 10300
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.26,
|
| 720 |
+
"learning_rate": 0.00017329616422885372,
|
| 721 |
+
"loss": 3.0945,
|
| 722 |
+
"step": 10400
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 0.26,
|
| 726 |
+
"learning_rate": 0.00017275243261353094,
|
| 727 |
+
"loss": 3.0885,
|
| 728 |
+
"step": 10500
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 0.26,
|
| 732 |
+
"learning_rate": 0.00017220409265619697,
|
| 733 |
+
"loss": 3.0761,
|
| 734 |
+
"step": 10600
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 0.26,
|
| 738 |
+
"learning_rate": 0.00017165117909023425,
|
| 739 |
+
"loss": 3.0988,
|
| 740 |
+
"step": 10700
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 0.27,
|
| 744 |
+
"learning_rate": 0.00017109372693873022,
|
| 745 |
+
"loss": 3.0941,
|
| 746 |
+
"step": 10800
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.27,
|
| 750 |
+
"learning_rate": 0.00017053177151225913,
|
| 751 |
+
"loss": 3.1068,
|
| 752 |
+
"step": 10900
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.27,
|
| 756 |
+
"learning_rate": 0.00016996534840664514,
|
| 757 |
+
"loss": 3.0746,
|
| 758 |
+
"step": 11000
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.27,
|
| 762 |
+
"eval_accuracy": 0.36640889106008795,
|
| 763 |
+
"eval_loss": 3.9645512104034424,
|
| 764 |
+
"eval_runtime": 151.7078,
|
| 765 |
+
"eval_samples_per_second": 6.592,
|
| 766 |
+
"eval_steps_per_second": 3.296,
|
| 767 |
+
"step": 11000
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 0.27,
|
| 771 |
+
"learning_rate": 0.00016939449350070768,
|
| 772 |
+
"loss": 3.0761,
|
| 773 |
+
"step": 11100
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.28,
|
| 777 |
+
"learning_rate": 0.00016881924295398883,
|
| 778 |
+
"loss": 3.0362,
|
| 779 |
+
"step": 11200
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.28,
|
| 783 |
+
"learning_rate": 0.00016823963320446273,
|
| 784 |
+
"loss": 3.0012,
|
| 785 |
+
"step": 11300
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 0.28,
|
| 789 |
+
"learning_rate": 0.00016765570096622761,
|
| 790 |
+
"loss": 3.0244,
|
| 791 |
+
"step": 11400
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 0.28,
|
| 795 |
+
"learning_rate": 0.00016706748322718027,
|
| 796 |
+
"loss": 3.0108,
|
| 797 |
+
"step": 11500
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 0.29,
|
| 801 |
+
"learning_rate": 0.000166475017246673,
|
| 802 |
+
"loss": 3.0364,
|
| 803 |
+
"step": 11600
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 0.29,
|
| 807 |
+
"learning_rate": 0.00016587834055315358,
|
| 808 |
+
"loss": 3.0235,
|
| 809 |
+
"step": 11700
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 0.29,
|
| 813 |
+
"learning_rate": 0.00016527749094178806,
|
| 814 |
+
"loss": 3.0356,
|
| 815 |
+
"step": 11800
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.29,
|
| 819 |
+
"learning_rate": 0.00016467250647206685,
|
| 820 |
+
"loss": 3.0178,
|
| 821 |
+
"step": 11900
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.3,
|
| 825 |
+
"learning_rate": 0.0001640634254653937,
|
| 826 |
+
"loss": 2.9473,
|
| 827 |
+
"step": 12000
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 0.3,
|
| 831 |
+
"eval_accuracy": 0.37060478749389353,
|
| 832 |
+
"eval_loss": 3.9244790077209473,
|
| 833 |
+
"eval_runtime": 149.2451,
|
| 834 |
+
"eval_samples_per_second": 6.7,
|
| 835 |
+
"eval_steps_per_second": 3.35,
|
| 836 |
+
"step": 12000
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.3,
|
| 840 |
+
"learning_rate": 0.00016345028650265853,
|
| 841 |
+
"loss": 3.0204,
|
| 842 |
+
"step": 12100
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.3,
|
| 846 |
+
"learning_rate": 0.00016283312842179344,
|
| 847 |
+
"loss": 2.976,
|
| 848 |
+
"step": 12200
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 0.3,
|
| 852 |
+
"learning_rate": 0.00016221199031531277,
|
| 853 |
+
"loss": 3.0472,
|
| 854 |
+
"step": 12300
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 0.31,
|
| 858 |
+
"learning_rate": 0.00016158691152783666,
|
| 859 |
+
"loss": 2.9778,
|
| 860 |
+
"step": 12400
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 0.31,
|
| 864 |
+
"learning_rate": 0.0001609579316535991,
|
| 865 |
+
"loss": 2.9958,
|
| 866 |
+
"step": 12500
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 0.31,
|
| 870 |
+
"learning_rate": 0.00016032509053393978,
|
| 871 |
+
"loss": 2.9772,
|
| 872 |
+
"step": 12600
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 0.31,
|
| 876 |
+
"learning_rate": 0.00015968842825478033,
|
| 877 |
+
"loss": 3.0037,
|
| 878 |
+
"step": 12700
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.32,
|
| 882 |
+
"learning_rate": 0.00015904798514408545,
|
| 883 |
+
"loss": 2.9888,
|
| 884 |
+
"step": 12800
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.32,
|
| 888 |
+
"learning_rate": 0.00015840380176930808,
|
| 889 |
+
"loss": 2.9614,
|
| 890 |
+
"step": 12900
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 0.32,
|
| 894 |
+
"learning_rate": 0.0001577559189348201,
|
| 895 |
+
"loss": 2.9737,
|
| 896 |
+
"step": 13000
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 0.32,
|
| 900 |
+
"eval_accuracy": 0.375352711284807,
|
| 901 |
+
"eval_loss": 3.8742177486419678,
|
| 902 |
+
"eval_runtime": 149.4881,
|
| 903 |
+
"eval_samples_per_second": 6.689,
|
| 904 |
+
"eval_steps_per_second": 3.345,
|
| 905 |
+
"step": 13000
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.32,
|
| 909 |
+
"learning_rate": 0.0001571043776793273,
|
| 910 |
+
"loss": 2.9724,
|
| 911 |
+
"step": 13100
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 0.33,
|
| 915 |
+
"learning_rate": 0.00015644921927327024,
|
| 916 |
+
"loss": 2.9695,
|
| 917 |
+
"step": 13200
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 0.33,
|
| 921 |
+
"learning_rate": 0.00015579048521620977,
|
| 922 |
+
"loss": 3.0015,
|
| 923 |
+
"step": 13300
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 0.33,
|
| 927 |
+
"learning_rate": 0.0001551282172341985,
|
| 928 |
+
"loss": 2.9804,
|
| 929 |
+
"step": 13400
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 0.33,
|
| 933 |
+
"learning_rate": 0.00015446245727713758,
|
| 934 |
+
"loss": 2.9711,
|
| 935 |
+
"step": 13500
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 0.34,
|
| 939 |
+
"learning_rate": 0.00015379324751611985,
|
| 940 |
+
"loss": 2.9746,
|
| 941 |
+
"step": 13600
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.34,
|
| 945 |
+
"learning_rate": 0.00015312063034075816,
|
| 946 |
+
"loss": 2.9414,
|
| 947 |
+
"step": 13700
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.34,
|
| 951 |
+
"learning_rate": 0.00015244464835650052,
|
| 952 |
+
"loss": 2.9365,
|
| 953 |
+
"step": 13800
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 0.34,
|
| 957 |
+
"learning_rate": 0.0001517653443819313,
|
| 958 |
+
"loss": 2.9674,
|
| 959 |
+
"step": 13900
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 0.35,
|
| 963 |
+
"learning_rate": 0.00015108276144605912,
|
| 964 |
+
"loss": 2.9193,
|
| 965 |
+
"step": 14000
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 0.35,
|
| 969 |
+
"eval_accuracy": 0.37964826575476307,
|
| 970 |
+
"eval_loss": 3.8284761905670166,
|
| 971 |
+
"eval_runtime": 149.3853,
|
| 972 |
+
"eval_samples_per_second": 6.694,
|
| 973 |
+
"eval_steps_per_second": 3.347,
|
| 974 |
+
"step": 14000
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 0.35,
|
| 978 |
+
"learning_rate": 0.0001503969427855911,
|
| 979 |
+
"loss": 2.9236,
|
| 980 |
+
"step": 14100
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 0.35,
|
| 984 |
+
"learning_rate": 0.00014970793184219402,
|
| 985 |
+
"loss": 2.8793,
|
| 986 |
+
"step": 14200
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 0.35,
|
| 990 |
+
"learning_rate": 0.000149015772259743,
|
| 991 |
+
"loss": 2.9347,
|
| 992 |
+
"step": 14300
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 0.36,
|
| 996 |
+
"learning_rate": 0.00014832050788155668,
|
| 997 |
+
"loss": 2.908,
|
| 998 |
+
"step": 14400
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 0.36,
|
| 1002 |
+
"learning_rate": 0.00014762218274762,
|
| 1003 |
+
"loss": 2.9523,
|
| 1004 |
+
"step": 14500
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.36,
|
| 1008 |
+
"learning_rate": 0.00014692084109179483,
|
| 1009 |
+
"loss": 2.9633,
|
| 1010 |
+
"step": 14600
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.36,
|
| 1014 |
+
"learning_rate": 0.0001462165273390179,
|
| 1015 |
+
"loss": 2.8945,
|
| 1016 |
+
"step": 14700
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 0.37,
|
| 1020 |
+
"learning_rate": 0.00014550928610248688,
|
| 1021 |
+
"loss": 2.9083,
|
| 1022 |
+
"step": 14800
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 0.37,
|
| 1026 |
+
"learning_rate": 0.0001447991621808344,
|
| 1027 |
+
"loss": 2.881,
|
| 1028 |
+
"step": 14900
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 0.37,
|
| 1032 |
+
"learning_rate": 0.00014408620055529045,
|
| 1033 |
+
"loss": 2.8833,
|
| 1034 |
+
"step": 15000
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 0.37,
|
| 1038 |
+
"eval_accuracy": 0.3837151929653151,
|
| 1039 |
+
"eval_loss": 3.7952051162719727,
|
| 1040 |
+
"eval_runtime": 142.0749,
|
| 1041 |
+
"eval_samples_per_second": 7.039,
|
| 1042 |
+
"eval_steps_per_second": 3.519,
|
| 1043 |
+
"step": 15000
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 0.37,
|
| 1047 |
+
"learning_rate": 0.00014337044638683297,
|
| 1048 |
+
"loss": 2.8999,
|
| 1049 |
+
"step": 15100
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 0.38,
|
| 1053 |
+
"learning_rate": 0.00014265194501332756,
|
| 1054 |
+
"loss": 2.8795,
|
| 1055 |
+
"step": 15200
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 0.38,
|
| 1059 |
+
"learning_rate": 0.0001419307419466553,
|
| 1060 |
+
"loss": 2.8995,
|
| 1061 |
+
"step": 15300
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 0.38,
|
| 1065 |
+
"learning_rate": 0.00014120688286983012,
|
| 1066 |
+
"loss": 2.9001,
|
| 1067 |
+
"step": 15400
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.38,
|
| 1071 |
+
"learning_rate": 0.00014048041363410498,
|
| 1072 |
+
"loss": 2.9089,
|
| 1073 |
+
"step": 15500
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.39,
|
| 1077 |
+
"learning_rate": 0.00013975138025606765,
|
| 1078 |
+
"loss": 2.9049,
|
| 1079 |
+
"step": 15600
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 0.39,
|
| 1083 |
+
"learning_rate": 0.0001390198289147258,
|
| 1084 |
+
"loss": 2.8915,
|
| 1085 |
+
"step": 15700
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 0.39,
|
| 1089 |
+
"learning_rate": 0.0001382858059485819,
|
| 1090 |
+
"loss": 2.8382,
|
| 1091 |
+
"step": 15800
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 0.39,
|
| 1095 |
+
"learning_rate": 0.00013754935785269802,
|
| 1096 |
+
"loss": 2.8535,
|
| 1097 |
+
"step": 15900
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 0.4,
|
| 1101 |
+
"learning_rate": 0.00013681053127575078,
|
| 1102 |
+
"loss": 2.8533,
|
| 1103 |
+
"step": 16000
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 0.4,
|
| 1107 |
+
"eval_accuracy": 0.3872970200293112,
|
| 1108 |
+
"eval_loss": 3.761565923690796,
|
| 1109 |
+
"eval_runtime": 141.7939,
|
| 1110 |
+
"eval_samples_per_second": 7.052,
|
| 1111 |
+
"eval_steps_per_second": 3.526,
|
| 1112 |
+
"step": 16000
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 0.4,
|
| 1116 |
+
"learning_rate": 0.0001360693730170764,
|
| 1117 |
+
"loss": 2.8792,
|
| 1118 |
+
"step": 16100
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 0.4,
|
| 1122 |
+
"learning_rate": 0.0001353259300237063,
|
| 1123 |
+
"loss": 2.8601,
|
| 1124 |
+
"step": 16200
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 0.4,
|
| 1128 |
+
"learning_rate": 0.00013458024938739328,
|
| 1129 |
+
"loss": 2.8794,
|
| 1130 |
+
"step": 16300
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.41,
|
| 1134 |
+
"learning_rate": 0.00013383237834162893,
|
| 1135 |
+
"loss": 2.8738,
|
| 1136 |
+
"step": 16400
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.41,
|
| 1140 |
+
"learning_rate": 0.00013308236425865123,
|
| 1141 |
+
"loss": 2.8475,
|
| 1142 |
+
"step": 16500
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 0.41,
|
| 1146 |
+
"learning_rate": 0.00013233025464644426,
|
| 1147 |
+
"loss": 2.8692,
|
| 1148 |
+
"step": 16600
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 0.41,
|
| 1152 |
+
"learning_rate": 0.00013157609714572877,
|
| 1153 |
+
"loss": 2.8282,
|
| 1154 |
+
"step": 16700
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 0.42,
|
| 1158 |
+
"learning_rate": 0.00013081993952694434,
|
| 1159 |
+
"loss": 2.8956,
|
| 1160 |
+
"step": 16800
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 0.42,
|
| 1164 |
+
"learning_rate": 0.00013006182968722382,
|
| 1165 |
+
"loss": 2.8338,
|
| 1166 |
+
"step": 16900
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 0.42,
|
| 1170 |
+
"learning_rate": 0.00012930181564735897,
|
| 1171 |
+
"loss": 2.8654,
|
| 1172 |
+
"step": 17000
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.42,
|
| 1176 |
+
"eval_accuracy": 0.39067855398143625,
|
| 1177 |
+
"eval_loss": 3.72955322265625,
|
| 1178 |
+
"eval_runtime": 141.9976,
|
| 1179 |
+
"eval_samples_per_second": 7.042,
|
| 1180 |
+
"eval_steps_per_second": 3.521,
|
| 1181 |
+
"step": 17000
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"epoch": 0.42,
|
| 1185 |
+
"learning_rate": 0.00012853994554875902,
|
| 1186 |
+
"loss": 2.8337,
|
| 1187 |
+
"step": 17100
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 0.42,
|
| 1191 |
+
"learning_rate": 0.00012777626765040116,
|
| 1192 |
+
"loss": 2.811,
|
| 1193 |
+
"step": 17200
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.43,
|
| 1197 |
+
"learning_rate": 0.00012701083032577354,
|
| 1198 |
+
"loss": 2.827,
|
| 1199 |
+
"step": 17300
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.43,
|
| 1203 |
+
"learning_rate": 0.00012624368205981136,
|
| 1204 |
+
"loss": 2.8085,
|
| 1205 |
+
"step": 17400
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 0.43,
|
| 1209 |
+
"learning_rate": 0.00012547487144582564,
|
| 1210 |
+
"loss": 2.8328,
|
| 1211 |
+
"step": 17500
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 0.43,
|
| 1215 |
+
"learning_rate": 0.0001247044471824251,
|
| 1216 |
+
"loss": 2.8269,
|
| 1217 |
+
"step": 17600
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 0.44,
|
| 1221 |
+
"learning_rate": 0.00012393245807043154,
|
| 1222 |
+
"loss": 2.8141,
|
| 1223 |
+
"step": 17700
|
| 1224 |
+
},
|
| 1225 |
+
{
|
| 1226 |
+
"epoch": 0.44,
|
| 1227 |
+
"learning_rate": 0.00012315895300978863,
|
| 1228 |
+
"loss": 2.8054,
|
| 1229 |
+
"step": 17800
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 0.44,
|
| 1233 |
+
"learning_rate": 0.0001223839809964645,
|
| 1234 |
+
"loss": 2.8176,
|
| 1235 |
+
"step": 17900
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.44,
|
| 1239 |
+
"learning_rate": 0.0001216075911193481,
|
| 1240 |
+
"loss": 2.8196,
|
| 1241 |
+
"step": 18000
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.44,
|
| 1245 |
+
"eval_accuracy": 0.3935666829506595,
|
| 1246 |
+
"eval_loss": 3.7048707008361816,
|
| 1247 |
+
"eval_runtime": 141.7514,
|
| 1248 |
+
"eval_samples_per_second": 7.055,
|
| 1249 |
+
"eval_steps_per_second": 3.527,
|
| 1250 |
+
"step": 18000
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 0.45,
|
| 1254 |
+
"learning_rate": 0.0001208298325571398,
|
| 1255 |
+
"loss": 2.8031,
|
| 1256 |
+
"step": 18100
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.45,
|
| 1260 |
+
"learning_rate": 0.00012005075457523635,
|
| 1261 |
+
"loss": 2.8213,
|
| 1262 |
+
"step": 18200
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.45,
|
| 1266 |
+
"learning_rate": 0.00011927040652261028,
|
| 1267 |
+
"loss": 2.823,
|
| 1268 |
+
"step": 18300
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 0.45,
|
| 1272 |
+
"learning_rate": 0.00011848883782868381,
|
| 1273 |
+
"loss": 2.8006,
|
| 1274 |
+
"step": 18400
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 0.46,
|
| 1278 |
+
"learning_rate": 0.0001177060980001981,
|
| 1279 |
+
"loss": 2.7874,
|
| 1280 |
+
"step": 18500
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 0.46,
|
| 1284 |
+
"learning_rate": 0.00011692223661807716,
|
| 1285 |
+
"loss": 2.7917,
|
| 1286 |
+
"step": 18600
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 0.46,
|
| 1290 |
+
"learning_rate": 0.00011613730333428738,
|
| 1291 |
+
"loss": 2.7983,
|
| 1292 |
+
"step": 18700
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 0.46,
|
| 1296 |
+
"learning_rate": 0.00011535134786869242,
|
| 1297 |
+
"loss": 2.8175,
|
| 1298 |
+
"step": 18800
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.47,
|
| 1302 |
+
"learning_rate": 0.00011456442000590371,
|
| 1303 |
+
"loss": 2.7732,
|
| 1304 |
+
"step": 18900
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.47,
|
| 1308 |
+
"learning_rate": 0.00011377656959212703,
|
| 1309 |
+
"loss": 2.7883,
|
| 1310 |
+
"step": 19000
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 0.47,
|
| 1314 |
+
"eval_accuracy": 0.3966370297997069,
|
| 1315 |
+
"eval_loss": 3.678572177886963,
|
| 1316 |
+
"eval_runtime": 141.8348,
|
| 1317 |
+
"eval_samples_per_second": 7.05,
|
| 1318 |
+
"eval_steps_per_second": 3.525,
|
| 1319 |
+
"step": 19000
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.47,
|
| 1323 |
+
"learning_rate": 0.00011298784653200525,
|
| 1324 |
+
"loss": 2.779,
|
| 1325 |
+
"step": 19100
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.47,
|
| 1329 |
+
"learning_rate": 0.0001121983007854569,
|
| 1330 |
+
"loss": 2.7928,
|
| 1331 |
+
"step": 19200
|
| 1332 |
+
},
|
| 1333 |
+
{
|
| 1334 |
+
"epoch": 0.48,
|
| 1335 |
+
"learning_rate": 0.0001114079823645119,
|
| 1336 |
+
"loss": 2.7889,
|
| 1337 |
+
"step": 19300
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 0.48,
|
| 1341 |
+
"learning_rate": 0.00011061694133014351,
|
| 1342 |
+
"loss": 2.7992,
|
| 1343 |
+
"step": 19400
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 0.48,
|
| 1347 |
+
"learning_rate": 0.00010982522778909725,
|
| 1348 |
+
"loss": 2.8229,
|
| 1349 |
+
"step": 19500
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 0.48,
|
| 1353 |
+
"learning_rate": 0.00010903289189071721,
|
| 1354 |
+
"loss": 2.7589,
|
| 1355 |
+
"step": 19600
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 0.49,
|
| 1359 |
+
"learning_rate": 0.00010823998382376919,
|
| 1360 |
+
"loss": 2.7828,
|
| 1361 |
+
"step": 19700
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.49,
|
| 1365 |
+
"learning_rate": 0.00010744655381326188,
|
| 1366 |
+
"loss": 2.7468,
|
| 1367 |
+
"step": 19800
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.49,
|
| 1371 |
+
"learning_rate": 0.00010665265211726528,
|
| 1372 |
+
"loss": 2.7686,
|
| 1373 |
+
"step": 19900
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 0.49,
|
| 1377 |
+
"learning_rate": 0.00010585832902372721,
|
| 1378 |
+
"loss": 2.747,
|
| 1379 |
+
"step": 20000
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 0.49,
|
| 1383 |
+
"eval_accuracy": 0.3989570102589155,
|
| 1384 |
+
"eval_loss": 3.6488404273986816,
|
| 1385 |
+
"eval_runtime": 141.8989,
|
| 1386 |
+
"eval_samples_per_second": 7.047,
|
| 1387 |
+
"eval_steps_per_second": 3.524,
|
| 1388 |
+
"step": 20000
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.5,
|
| 1392 |
+
"learning_rate": 0.00010506363484728803,
|
| 1393 |
+
"loss": 2.7899,
|
| 1394 |
+
"step": 20100
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 0.5,
|
| 1398 |
+
"learning_rate": 0.00010426861992609356,
|
| 1399 |
+
"loss": 2.7468,
|
| 1400 |
+
"step": 20200
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 0.5,
|
| 1404 |
+
"learning_rate": 0.00010347333461860647,
|
| 1405 |
+
"loss": 2.7607,
|
| 1406 |
+
"step": 20300
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 0.5,
|
| 1410 |
+
"learning_rate": 0.0001026778293004164,
|
| 1411 |
+
"loss": 2.7659,
|
| 1412 |
+
"step": 20400
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 0.51,
|
| 1416 |
+
"learning_rate": 0.00010188215436104913,
|
| 1417 |
+
"loss": 2.7515,
|
| 1418 |
+
"step": 20500
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 0.51,
|
| 1422 |
+
"learning_rate": 0.00010108636020077471,
|
| 1423 |
+
"loss": 2.786,
|
| 1424 |
+
"step": 20600
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.51,
|
| 1428 |
+
"learning_rate": 0.000100290497227415,
|
| 1429 |
+
"loss": 2.7153,
|
| 1430 |
+
"step": 20700
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.51,
|
| 1434 |
+
"learning_rate": 9.94946158531506e-05,
|
| 1435 |
+
"loss": 2.7299,
|
| 1436 |
+
"step": 20800
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 0.52,
|
| 1440 |
+
"learning_rate": 9.869876649132773e-05,
|
| 1441 |
+
"loss": 2.7673,
|
| 1442 |
+
"step": 20900
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 0.52,
|
| 1446 |
+
"learning_rate": 9.790299955326484e-05,
|
| 1447 |
+
"loss": 2.7355,
|
| 1448 |
+
"step": 21000
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 0.52,
|
| 1452 |
+
"eval_accuracy": 0.4021494870542257,
|
| 1453 |
+
"eval_loss": 3.6243462562561035,
|
| 1454 |
+
"eval_runtime": 142.1627,
|
| 1455 |
+
"eval_samples_per_second": 7.034,
|
| 1456 |
+
"eval_steps_per_second": 3.517,
|
| 1457 |
+
"step": 21000
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 0.52,
|
| 1461 |
+
"learning_rate": 9.710736544505947e-05,
|
| 1462 |
+
"loss": 2.7641,
|
| 1463 |
+
"step": 21100
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 0.52,
|
| 1467 |
+
"learning_rate": 9.631191456439525e-05,
|
| 1468 |
+
"loss": 2.7809,
|
| 1469 |
+
"step": 21200
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 0.53,
|
| 1473 |
+
"learning_rate": 9.551669729734978e-05,
|
| 1474 |
+
"loss": 2.7434,
|
| 1475 |
+
"step": 21300
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 0.53,
|
| 1479 |
+
"learning_rate": 9.472176401520279e-05,
|
| 1480 |
+
"loss": 2.7287,
|
| 1481 |
+
"step": 21400
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 0.53,
|
| 1485 |
+
"learning_rate": 9.392716507124576e-05,
|
| 1486 |
+
"loss": 2.7536,
|
| 1487 |
+
"step": 21500
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.53,
|
| 1491 |
+
"learning_rate": 9.313295079759219e-05,
|
| 1492 |
+
"loss": 2.7539,
|
| 1493 |
+
"step": 21600
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.54,
|
| 1497 |
+
"learning_rate": 9.233917150198943e-05,
|
| 1498 |
+
"loss": 2.7753,
|
| 1499 |
+
"step": 21700
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 0.54,
|
| 1503 |
+
"learning_rate": 9.154587746463225e-05,
|
| 1504 |
+
"loss": 2.7551,
|
| 1505 |
+
"step": 21800
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 0.54,
|
| 1509 |
+
"learning_rate": 9.07531189349776e-05,
|
| 1510 |
+
"loss": 2.706,
|
| 1511 |
+
"step": 21900
|
| 1512 |
+
},
|
| 1513 |
+
{
|
| 1514 |
+
"epoch": 0.54,
|
| 1515 |
+
"learning_rate": 8.996094612856212e-05,
|
| 1516 |
+
"loss": 2.7355,
|
| 1517 |
+
"step": 22000
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 0.54,
|
| 1521 |
+
"eval_accuracy": 0.40534440644846115,
|
| 1522 |
+
"eval_loss": 3.5981943607330322,
|
| 1523 |
+
"eval_runtime": 141.9963,
|
| 1524 |
+
"eval_samples_per_second": 7.042,
|
| 1525 |
+
"eval_steps_per_second": 3.521,
|
| 1526 |
+
"step": 22000
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 0.55,
|
| 1530 |
+
"learning_rate": 8.916940922382091e-05,
|
| 1531 |
+
"loss": 2.7117,
|
| 1532 |
+
"step": 22100
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 0.55,
|
| 1536 |
+
"learning_rate": 8.837855835890943e-05,
|
| 1537 |
+
"loss": 2.717,
|
| 1538 |
+
"step": 22200
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 0.55,
|
| 1542 |
+
"learning_rate": 8.758844362852732e-05,
|
| 1543 |
+
"loss": 2.733,
|
| 1544 |
+
"step": 22300
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 0.55,
|
| 1548 |
+
"learning_rate": 8.679911508074552e-05,
|
| 1549 |
+
"loss": 2.7257,
|
| 1550 |
+
"step": 22400
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.56,
|
| 1554 |
+
"learning_rate": 8.601062271383592e-05,
|
| 1555 |
+
"loss": 2.6842,
|
| 1556 |
+
"step": 22500
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.56,
|
| 1560 |
+
"learning_rate": 8.522301647310445e-05,
|
| 1561 |
+
"loss": 2.7144,
|
| 1562 |
+
"step": 22600
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 0.56,
|
| 1566 |
+
"learning_rate": 8.443634624772719e-05,
|
| 1567 |
+
"loss": 2.7521,
|
| 1568 |
+
"step": 22700
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 0.56,
|
| 1572 |
+
"learning_rate": 8.365066186759055e-05,
|
| 1573 |
+
"loss": 2.7484,
|
| 1574 |
+
"step": 22800
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 0.57,
|
| 1578 |
+
"learning_rate": 8.286601310013463e-05,
|
| 1579 |
+
"loss": 2.7177,
|
| 1580 |
+
"step": 22900
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 0.57,
|
| 1584 |
+
"learning_rate": 8.208244964720108e-05,
|
| 1585 |
+
"loss": 2.6999,
|
| 1586 |
+
"step": 23000
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 0.57,
|
| 1590 |
+
"eval_accuracy": 0.4074880312652662,
|
| 1591 |
+
"eval_loss": 3.5765340328216553,
|
| 1592 |
+
"eval_runtime": 142.0816,
|
| 1593 |
+
"eval_samples_per_second": 7.038,
|
| 1594 |
+
"eval_steps_per_second": 3.519,
|
| 1595 |
+
"step": 23000
|
| 1596 |
+
},
|
| 1597 |
+
{
|
| 1598 |
+
"epoch": 0.57,
|
| 1599 |
+
"learning_rate": 8.130002114188453e-05,
|
| 1600 |
+
"loss": 2.6884,
|
| 1601 |
+
"step": 23100
|
| 1602 |
+
},
|
| 1603 |
+
{
|
| 1604 |
+
"epoch": 0.57,
|
| 1605 |
+
"learning_rate": 8.0518777145389e-05,
|
| 1606 |
+
"loss": 2.6783,
|
| 1607 |
+
"step": 23200
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 0.58,
|
| 1611 |
+
"learning_rate": 7.97387671438883e-05,
|
| 1612 |
+
"loss": 2.6857,
|
| 1613 |
+
"step": 23300
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.58,
|
| 1617 |
+
"learning_rate": 7.896004054539173e-05,
|
| 1618 |
+
"loss": 2.6957,
|
| 1619 |
+
"step": 23400
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.58,
|
| 1623 |
+
"learning_rate": 7.818264667661405e-05,
|
| 1624 |
+
"loss": 2.7092,
|
| 1625 |
+
"step": 23500
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 0.58,
|
| 1629 |
+
"learning_rate": 7.740663477985133e-05,
|
| 1630 |
+
"loss": 2.6602,
|
| 1631 |
+
"step": 23600
|
| 1632 |
+
},
|
| 1633 |
+
{
|
| 1634 |
+
"epoch": 0.59,
|
| 1635 |
+
"learning_rate": 7.663205400986167e-05,
|
| 1636 |
+
"loss": 2.7544,
|
| 1637 |
+
"step": 23700
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"epoch": 0.59,
|
| 1641 |
+
"learning_rate": 7.585895343075162e-05,
|
| 1642 |
+
"loss": 2.6894,
|
| 1643 |
+
"step": 23800
|
| 1644 |
+
},
|
| 1645 |
+
{
|
| 1646 |
+
"epoch": 0.59,
|
| 1647 |
+
"learning_rate": 7.508738201286819e-05,
|
| 1648 |
+
"loss": 2.6764,
|
| 1649 |
+
"step": 23900
|
| 1650 |
+
},
|
| 1651 |
+
{
|
| 1652 |
+
"epoch": 0.59,
|
| 1653 |
+
"learning_rate": 7.431738862969718e-05,
|
| 1654 |
+
"loss": 2.7243,
|
| 1655 |
+
"step": 24000
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.59,
|
| 1659 |
+
"eval_accuracy": 0.41014020517830974,
|
| 1660 |
+
"eval_loss": 3.5558369159698486,
|
| 1661 |
+
"eval_runtime": 141.9008,
|
| 1662 |
+
"eval_samples_per_second": 7.047,
|
| 1663 |
+
"eval_steps_per_second": 3.524,
|
| 1664 |
+
"step": 24000
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 0.6,
|
| 1668 |
+
"learning_rate": 7.354902205476732e-05,
|
| 1669 |
+
"loss": 2.7029,
|
| 1670 |
+
"step": 24100
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 0.6,
|
| 1674 |
+
"learning_rate": 7.278233095856057e-05,
|
| 1675 |
+
"loss": 2.6547,
|
| 1676 |
+
"step": 24200
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.6,
|
| 1680 |
+
"learning_rate": 7.201736390542958e-05,
|
| 1681 |
+
"loss": 2.6755,
|
| 1682 |
+
"step": 24300
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.6,
|
| 1686 |
+
"learning_rate": 7.125416935052119e-05,
|
| 1687 |
+
"loss": 2.6666,
|
| 1688 |
+
"step": 24400
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 0.61,
|
| 1692 |
+
"learning_rate": 7.049279563670732e-05,
|
| 1693 |
+
"loss": 2.6753,
|
| 1694 |
+
"step": 24500
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 0.61,
|
| 1698 |
+
"learning_rate": 6.973329099152268e-05,
|
| 1699 |
+
"loss": 2.6411,
|
| 1700 |
+
"step": 24600
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 0.61,
|
| 1704 |
+
"learning_rate": 6.897570352411003e-05,
|
| 1705 |
+
"loss": 2.6519,
|
| 1706 |
+
"step": 24700
|
| 1707 |
+
},
|
| 1708 |
+
{
|
| 1709 |
+
"epoch": 0.61,
|
| 1710 |
+
"learning_rate": 6.822008122217267e-05,
|
| 1711 |
+
"loss": 2.6922,
|
| 1712 |
+
"step": 24800
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 0.62,
|
| 1716 |
+
"learning_rate": 6.746647194893494e-05,
|
| 1717 |
+
"loss": 2.6432,
|
| 1718 |
+
"step": 24900
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.62,
|
| 1722 |
+
"learning_rate": 6.671492344011017e-05,
|
| 1723 |
+
"loss": 2.6526,
|
| 1724 |
+
"step": 25000
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.62,
|
| 1728 |
+
"eval_accuracy": 0.41251294577430386,
|
| 1729 |
+
"eval_loss": 3.537142753601074,
|
| 1730 |
+
"eval_runtime": 149.0732,
|
| 1731 |
+
"eval_samples_per_second": 6.708,
|
| 1732 |
+
"eval_steps_per_second": 3.354,
|
| 1733 |
+
"step": 25000
|
| 1734 |
+
},
|
| 1735 |
+
{
|
| 1736 |
+
"epoch": 0.62,
|
| 1737 |
+
"learning_rate": 6.596548330087729e-05,
|
| 1738 |
+
"loss": 2.6583,
|
| 1739 |
+
"step": 25100
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.62,
|
| 1743 |
+
"learning_rate": 6.521819900286509e-05,
|
| 1744 |
+
"loss": 2.7001,
|
| 1745 |
+
"step": 25200
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.63,
|
| 1749 |
+
"learning_rate": 6.447311788114547e-05,
|
| 1750 |
+
"loss": 2.6591,
|
| 1751 |
+
"step": 25300
|
| 1752 |
+
},
|
| 1753 |
+
{
|
| 1754 |
+
"epoch": 0.63,
|
| 1755 |
+
"learning_rate": 6.373028713123489e-05,
|
| 1756 |
+
"loss": 2.6622,
|
| 1757 |
+
"step": 25400
|
| 1758 |
+
},
|
| 1759 |
+
{
|
| 1760 |
+
"epoch": 0.63,
|
| 1761 |
+
"learning_rate": 6.298975380610508e-05,
|
| 1762 |
+
"loss": 2.7062,
|
| 1763 |
+
"step": 25500
|
| 1764 |
+
},
|
| 1765 |
+
{
|
| 1766 |
+
"epoch": 0.63,
|
| 1767 |
+
"learning_rate": 6.22515648132024e-05,
|
| 1768 |
+
"loss": 2.679,
|
| 1769 |
+
"step": 25600
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 0.63,
|
| 1773 |
+
"learning_rate": 6.151576691147676e-05,
|
| 1774 |
+
"loss": 2.6539,
|
| 1775 |
+
"step": 25700
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 0.64,
|
| 1779 |
+
"learning_rate": 6.0782406708419535e-05,
|
| 1780 |
+
"loss": 2.6741,
|
| 1781 |
+
"step": 25800
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.64,
|
| 1785 |
+
"learning_rate": 6.005153065711164e-05,
|
| 1786 |
+
"loss": 2.6285,
|
| 1787 |
+
"step": 25900
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.64,
|
| 1791 |
+
"learning_rate": 5.932318505328076e-05,
|
| 1792 |
+
"loss": 2.641,
|
| 1793 |
+
"step": 26000
|
| 1794 |
+
},
|
| 1795 |
+
{
|
| 1796 |
+
"epoch": 0.64,
|
| 1797 |
+
"eval_accuracy": 0.4146458231558378,
|
| 1798 |
+
"eval_loss": 3.5149853229522705,
|
| 1799 |
+
"eval_runtime": 181.5716,
|
| 1800 |
+
"eval_samples_per_second": 5.507,
|
| 1801 |
+
"eval_steps_per_second": 2.754,
|
| 1802 |
+
"step": 26000
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.64,
|
| 1806 |
+
"learning_rate": 5.859741603236909e-05,
|
| 1807 |
+
"loss": 2.6834,
|
| 1808 |
+
"step": 26100
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.65,
|
| 1812 |
+
"learning_rate": 5.787426956661071e-05,
|
| 1813 |
+
"loss": 2.6712,
|
| 1814 |
+
"step": 26200
|
| 1815 |
+
},
|
| 1816 |
+
{
|
| 1817 |
+
"epoch": 0.65,
|
| 1818 |
+
"learning_rate": 5.715379146211998e-05,
|
| 1819 |
+
"loss": 2.6772,
|
| 1820 |
+
"step": 26300
|
| 1821 |
+
},
|
| 1822 |
+
{
|
| 1823 |
+
"epoch": 0.65,
|
| 1824 |
+
"learning_rate": 5.6436027355989517e-05,
|
| 1825 |
+
"loss": 2.6672,
|
| 1826 |
+
"step": 26400
|
| 1827 |
+
},
|
| 1828 |
+
{
|
| 1829 |
+
"epoch": 0.65,
|
| 1830 |
+
"learning_rate": 5.5721022713399965e-05,
|
| 1831 |
+
"loss": 2.6001,
|
| 1832 |
+
"step": 26500
|
| 1833 |
+
},
|
| 1834 |
+
{
|
| 1835 |
+
"epoch": 0.66,
|
| 1836 |
+
"learning_rate": 5.500882282473983e-05,
|
| 1837 |
+
"loss": 2.6681,
|
| 1838 |
+
"step": 26600
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 0.66,
|
| 1842 |
+
"learning_rate": 5.429947280273646e-05,
|
| 1843 |
+
"loss": 2.6229,
|
| 1844 |
+
"step": 26700
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 0.66,
|
| 1848 |
+
"learning_rate": 5.359301757959896e-05,
|
| 1849 |
+
"loss": 2.6637,
|
| 1850 |
+
"step": 26800
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.66,
|
| 1854 |
+
"learning_rate": 5.2889501904171633e-05,
|
| 1855 |
+
"loss": 2.617,
|
| 1856 |
+
"step": 26900
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 0.67,
|
| 1860 |
+
"learning_rate": 5.2188970339099716e-05,
|
| 1861 |
+
"loss": 2.6602,
|
| 1862 |
+
"step": 27000
|
| 1863 |
+
},
|
| 1864 |
+
{
|
| 1865 |
+
"epoch": 0.67,
|
| 1866 |
+
"eval_accuracy": 0.4168324377137274,
|
| 1867 |
+
"eval_loss": 3.497141122817993,
|
| 1868 |
+
"eval_runtime": 143.8542,
|
| 1869 |
+
"eval_samples_per_second": 6.951,
|
| 1870 |
+
"eval_steps_per_second": 3.476,
|
| 1871 |
+
"step": 27000
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.67,
|
| 1875 |
+
"learning_rate": 5.149146725800653e-05,
|
| 1876 |
+
"loss": 2.6309,
|
| 1877 |
+
"step": 27100
|
| 1878 |
+
},
|
| 1879 |
+
{
|
| 1880 |
+
"epoch": 0.67,
|
| 1881 |
+
"learning_rate": 5.0797036842682796e-05,
|
| 1882 |
+
"loss": 2.6206,
|
| 1883 |
+
"step": 27200
|
| 1884 |
+
},
|
| 1885 |
+
{
|
| 1886 |
+
"epoch": 0.67,
|
| 1887 |
+
"learning_rate": 5.0105723080287956e-05,
|
| 1888 |
+
"loss": 2.6232,
|
| 1889 |
+
"step": 27300
|
| 1890 |
+
},
|
| 1891 |
+
{
|
| 1892 |
+
"epoch": 0.68,
|
| 1893 |
+
"learning_rate": 4.941756976056414e-05,
|
| 1894 |
+
"loss": 2.6094,
|
| 1895 |
+
"step": 27400
|
| 1896 |
+
},
|
| 1897 |
+
{
|
| 1898 |
+
"epoch": 0.68,
|
| 1899 |
+
"learning_rate": 4.873262047306193e-05,
|
| 1900 |
+
"loss": 2.6533,
|
| 1901 |
+
"step": 27500
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 0.68,
|
| 1905 |
+
"learning_rate": 4.805091860437985e-05,
|
| 1906 |
+
"loss": 2.611,
|
| 1907 |
+
"step": 27600
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 0.68,
|
| 1911 |
+
"learning_rate": 4.7372507335415716e-05,
|
| 1912 |
+
"loss": 2.6003,
|
| 1913 |
+
"step": 27700
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.69,
|
| 1917 |
+
"learning_rate": 4.6697429638631594e-05,
|
| 1918 |
+
"loss": 2.6118,
|
| 1919 |
+
"step": 27800
|
| 1920 |
+
},
|
| 1921 |
+
{
|
| 1922 |
+
"epoch": 0.69,
|
| 1923 |
+
"learning_rate": 4.6025728275331805e-05,
|
| 1924 |
+
"loss": 2.5899,
|
| 1925 |
+
"step": 27900
|
| 1926 |
+
},
|
| 1927 |
+
{
|
| 1928 |
+
"epoch": 0.69,
|
| 1929 |
+
"learning_rate": 4.53574457929543e-05,
|
| 1930 |
+
"loss": 2.644,
|
| 1931 |
+
"step": 28000
|
| 1932 |
+
},
|
| 1933 |
+
{
|
| 1934 |
+
"epoch": 0.69,
|
| 1935 |
+
"eval_accuracy": 0.4192027357107963,
|
| 1936 |
+
"eval_loss": 3.4811973571777344,
|
| 1937 |
+
"eval_runtime": 151.4293,
|
| 1938 |
+
"eval_samples_per_second": 6.604,
|
| 1939 |
+
"eval_steps_per_second": 3.302,
|
| 1940 |
+
"step": 28000
|
| 1941 |
+
},
|
| 1942 |
+
{
|
| 1943 |
+
"epoch": 0.69,
|
| 1944 |
+
"learning_rate": 4.469262452237552e-05,
|
| 1945 |
+
"loss": 2.6313,
|
| 1946 |
+
"step": 28100
|
| 1947 |
+
},
|
| 1948 |
+
{
|
| 1949 |
+
"epoch": 0.7,
|
| 1950 |
+
"learning_rate": 4.403130657522924e-05,
|
| 1951 |
+
"loss": 2.5829,
|
| 1952 |
+
"step": 28200
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"epoch": 0.7,
|
| 1956 |
+
"learning_rate": 4.337353384123871e-05,
|
| 1957 |
+
"loss": 2.6009,
|
| 1958 |
+
"step": 28300
|
| 1959 |
+
},
|
| 1960 |
+
{
|
| 1961 |
+
"epoch": 0.7,
|
| 1962 |
+
"learning_rate": 4.2719347985563754e-05,
|
| 1963 |
+
"loss": 2.6282,
|
| 1964 |
+
"step": 28400
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 0.7,
|
| 1968 |
+
"learning_rate": 4.2068790446161166e-05,
|
| 1969 |
+
"loss": 2.6724,
|
| 1970 |
+
"step": 28500
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 0.71,
|
| 1974 |
+
"learning_rate": 4.1421902431160095e-05,
|
| 1975 |
+
"loss": 2.6432,
|
| 1976 |
+
"step": 28600
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.71,
|
| 1980 |
+
"learning_rate": 4.07787249162518e-05,
|
| 1981 |
+
"loss": 2.6422,
|
| 1982 |
+
"step": 28700
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"epoch": 0.71,
|
| 1986 |
+
"learning_rate": 4.013929864209407e-05,
|
| 1987 |
+
"loss": 2.6249,
|
| 1988 |
+
"step": 28800
|
| 1989 |
+
},
|
| 1990 |
+
{
|
| 1991 |
+
"epoch": 0.71,
|
| 1992 |
+
"learning_rate": 3.950366411173067e-05,
|
| 1993 |
+
"loss": 2.6187,
|
| 1994 |
+
"step": 28900
|
| 1995 |
+
},
|
| 1996 |
+
{
|
| 1997 |
+
"epoch": 0.72,
|
| 1998 |
+
"learning_rate": 3.887186158802581e-05,
|
| 1999 |
+
"loss": 2.6558,
|
| 2000 |
+
"step": 29000
|
| 2001 |
+
},
|
| 2002 |
+
{
|
| 2003 |
+
"epoch": 0.72,
|
| 2004 |
+
"eval_accuracy": 0.42146458231558376,
|
| 2005 |
+
"eval_loss": 3.462165594100952,
|
| 2006 |
+
"eval_runtime": 159.3822,
|
| 2007 |
+
"eval_samples_per_second": 6.274,
|
| 2008 |
+
"eval_steps_per_second": 3.137,
|
| 2009 |
+
"step": 29000
|
| 2010 |
+
},
|
| 2011 |
+
{
|
| 2012 |
+
"epoch": 0.72,
|
| 2013 |
+
"learning_rate": 3.8243931091113525e-05,
|
| 2014 |
+
"loss": 2.6114,
|
| 2015 |
+
"step": 29100
|
| 2016 |
+
},
|
| 2017 |
+
{
|
| 2018 |
+
"epoch": 0.72,
|
| 2019 |
+
"learning_rate": 3.7619912395863045e-05,
|
| 2020 |
+
"loss": 2.6093,
|
| 2021 |
+
"step": 29200
|
| 2022 |
+
},
|
| 2023 |
+
{
|
| 2024 |
+
"epoch": 0.72,
|
| 2025 |
+
"learning_rate": 3.699984502935918e-05,
|
| 2026 |
+
"loss": 2.641,
|
| 2027 |
+
"step": 29300
|
| 2028 |
+
},
|
| 2029 |
+
{
|
| 2030 |
+
"epoch": 0.73,
|
| 2031 |
+
"learning_rate": 3.6383768268398356e-05,
|
| 2032 |
+
"loss": 2.618,
|
| 2033 |
+
"step": 29400
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 0.73,
|
| 2037 |
+
"learning_rate": 3.5771721137001134e-05,
|
| 2038 |
+
"loss": 2.5961,
|
| 2039 |
+
"step": 29500
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.73,
|
| 2043 |
+
"learning_rate": 3.516374240393999e-05,
|
| 2044 |
+
"loss": 2.5759,
|
| 2045 |
+
"step": 29600
|
| 2046 |
+
},
|
| 2047 |
+
{
|
| 2048 |
+
"epoch": 0.73,
|
| 2049 |
+
"learning_rate": 3.455987058028372e-05,
|
| 2050 |
+
"loss": 2.5699,
|
| 2051 |
+
"step": 29700
|
| 2052 |
+
},
|
| 2053 |
+
{
|
| 2054 |
+
"epoch": 0.74,
|
| 2055 |
+
"learning_rate": 3.3960143916957995e-05,
|
| 2056 |
+
"loss": 2.6057,
|
| 2057 |
+
"step": 29800
|
| 2058 |
+
},
|
| 2059 |
+
{
|
| 2060 |
+
"epoch": 0.74,
|
| 2061 |
+
"learning_rate": 3.336460040232252e-05,
|
| 2062 |
+
"loss": 2.6013,
|
| 2063 |
+
"step": 29900
|
| 2064 |
+
},
|
| 2065 |
+
{
|
| 2066 |
+
"epoch": 0.74,
|
| 2067 |
+
"learning_rate": 3.2773277759764644e-05,
|
| 2068 |
+
"loss": 2.5664,
|
| 2069 |
+
"step": 30000
|
| 2070 |
+
},
|
| 2071 |
+
{
|
| 2072 |
+
"epoch": 0.74,
|
| 2073 |
+
"eval_accuracy": 0.42288812896922323,
|
| 2074 |
+
"eval_loss": 3.4503753185272217,
|
| 2075 |
+
"eval_runtime": 152.9493,
|
| 2076 |
+
"eval_samples_per_second": 6.538,
|
| 2077 |
+
"eval_steps_per_second": 3.269,
|
| 2078 |
+
"step": 30000
|
| 2079 |
+
},
|
| 2080 |
+
{
|
| 2081 |
+
"epoch": 0.74,
|
| 2082 |
+
"learning_rate": 3.218621344531002e-05,
|
| 2083 |
+
"loss": 2.601,
|
| 2084 |
+
"step": 30100
|
| 2085 |
+
},
|
| 2086 |
+
{
|
| 2087 |
+
"epoch": 0.75,
|
| 2088 |
+
"learning_rate": 3.16034446452497e-05,
|
| 2089 |
+
"loss": 2.5845,
|
| 2090 |
+
"step": 30200
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"epoch": 0.75,
|
| 2094 |
+
"learning_rate": 3.102500827378511e-05,
|
| 2095 |
+
"loss": 2.5578,
|
| 2096 |
+
"step": 30300
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 0.75,
|
| 2100 |
+
"learning_rate": 3.0450940970689413e-05,
|
| 2101 |
+
"loss": 2.6059,
|
| 2102 |
+
"step": 30400
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.75,
|
| 2106 |
+
"learning_rate": 2.988127909898686e-05,
|
| 2107 |
+
"loss": 2.5967,
|
| 2108 |
+
"step": 30500
|
| 2109 |
+
},
|
| 2110 |
+
{
|
| 2111 |
+
"epoch": 0.76,
|
| 2112 |
+
"learning_rate": 2.9316058742649354e-05,
|
| 2113 |
+
"loss": 2.5889,
|
| 2114 |
+
"step": 30600
|
| 2115 |
+
},
|
| 2116 |
+
{
|
| 2117 |
+
"epoch": 0.76,
|
| 2118 |
+
"learning_rate": 2.875531570431086e-05,
|
| 2119 |
+
"loss": 2.5626,
|
| 2120 |
+
"step": 30700
|
| 2121 |
+
},
|
| 2122 |
+
{
|
| 2123 |
+
"epoch": 0.76,
|
| 2124 |
+
"learning_rate": 2.819908550299947e-05,
|
| 2125 |
+
"loss": 2.6232,
|
| 2126 |
+
"step": 30800
|
| 2127 |
+
},
|
| 2128 |
+
{
|
| 2129 |
+
"epoch": 0.76,
|
| 2130 |
+
"learning_rate": 2.7647403371887736e-05,
|
| 2131 |
+
"loss": 2.5861,
|
| 2132 |
+
"step": 30900
|
| 2133 |
+
},
|
| 2134 |
+
{
|
| 2135 |
+
"epoch": 0.77,
|
| 2136 |
+
"learning_rate": 2.7100304256060526e-05,
|
| 2137 |
+
"loss": 2.5669,
|
| 2138 |
+
"step": 31000
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 0.77,
|
| 2142 |
+
"eval_accuracy": 0.424466536394724,
|
| 2143 |
+
"eval_loss": 3.4375574588775635,
|
| 2144 |
+
"eval_runtime": 144.0616,
|
| 2145 |
+
"eval_samples_per_second": 6.941,
|
| 2146 |
+
"eval_steps_per_second": 3.471,
|
| 2147 |
+
"step": 31000
|
| 2148 |
+
},
|
| 2149 |
+
{
|
| 2150 |
+
"epoch": 0.77,
|
| 2151 |
+
"learning_rate": 2.655782281030198e-05,
|
| 2152 |
+
"loss": 2.5697,
|
| 2153 |
+
"step": 31100
|
| 2154 |
+
},
|
| 2155 |
+
{
|
| 2156 |
+
"epoch": 0.77,
|
| 2157 |
+
"learning_rate": 2.6019993396899977e-05,
|
| 2158 |
+
"loss": 2.613,
|
| 2159 |
+
"step": 31200
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 0.77,
|
| 2163 |
+
"learning_rate": 2.548685008346974e-05,
|
| 2164 |
+
"loss": 2.5858,
|
| 2165 |
+
"step": 31300
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.78,
|
| 2169 |
+
"learning_rate": 2.495842664079584e-05,
|
| 2170 |
+
"loss": 2.534,
|
| 2171 |
+
"step": 31400
|
| 2172 |
+
},
|
| 2173 |
+
{
|
| 2174 |
+
"epoch": 0.78,
|
| 2175 |
+
"learning_rate": 2.4434756540693082e-05,
|
| 2176 |
+
"loss": 2.5843,
|
| 2177 |
+
"step": 31500
|
| 2178 |
+
},
|
| 2179 |
+
{
|
| 2180 |
+
"epoch": 0.78,
|
| 2181 |
+
"learning_rate": 2.3915872953886265e-05,
|
| 2182 |
+
"loss": 2.5988,
|
| 2183 |
+
"step": 31600
|
| 2184 |
+
},
|
| 2185 |
+
{
|
| 2186 |
+
"epoch": 0.78,
|
| 2187 |
+
"learning_rate": 2.340180874790916e-05,
|
| 2188 |
+
"loss": 2.5684,
|
| 2189 |
+
"step": 31700
|
| 2190 |
+
},
|
| 2191 |
+
{
|
| 2192 |
+
"epoch": 0.79,
|
| 2193 |
+
"learning_rate": 2.2892596485022423e-05,
|
| 2194 |
+
"loss": 2.5921,
|
| 2195 |
+
"step": 31800
|
| 2196 |
+
},
|
| 2197 |
+
{
|
| 2198 |
+
"epoch": 0.79,
|
| 2199 |
+
"learning_rate": 2.2388268420151127e-05,
|
| 2200 |
+
"loss": 2.5761,
|
| 2201 |
+
"step": 31900
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 0.79,
|
| 2205 |
+
"learning_rate": 2.1888856498841704e-05,
|
| 2206 |
+
"loss": 2.5498,
|
| 2207 |
+
"step": 32000
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.79,
|
| 2211 |
+
"eval_accuracy": 0.42633659013190034,
|
| 2212 |
+
"eval_loss": 3.4263486862182617,
|
| 2213 |
+
"eval_runtime": 144.2437,
|
| 2214 |
+
"eval_samples_per_second": 6.933,
|
| 2215 |
+
"eval_steps_per_second": 3.466,
|
| 2216 |
+
"step": 32000
|
| 2217 |
+
},
|
| 2218 |
+
{
|
| 2219 |
+
"epoch": 0.79,
|
| 2220 |
+
"learning_rate": 2.1394392355238145e-05,
|
| 2221 |
+
"loss": 2.5563,
|
| 2222 |
+
"step": 32100
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 0.8,
|
| 2226 |
+
"learning_rate": 2.0904907310078626e-05,
|
| 2227 |
+
"loss": 2.574,
|
| 2228 |
+
"step": 32200
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.8,
|
| 2232 |
+
"learning_rate": 2.0420432368711195e-05,
|
| 2233 |
+
"loss": 2.5599,
|
| 2234 |
+
"step": 32300
|
| 2235 |
+
},
|
| 2236 |
+
{
|
| 2237 |
+
"epoch": 0.8,
|
| 2238 |
+
"learning_rate": 1.9940998219130004e-05,
|
| 2239 |
+
"loss": 2.5991,
|
| 2240 |
+
"step": 32400
|
| 2241 |
+
},
|
| 2242 |
+
{
|
| 2243 |
+
"epoch": 0.8,
|
| 2244 |
+
"learning_rate": 1.946663523003136e-05,
|
| 2245 |
+
"loss": 2.5622,
|
| 2246 |
+
"step": 32500
|
| 2247 |
+
},
|
| 2248 |
+
{
|
| 2249 |
+
"epoch": 0.81,
|
| 2250 |
+
"learning_rate": 1.8997373448890154e-05,
|
| 2251 |
+
"loss": 2.5672,
|
| 2252 |
+
"step": 32600
|
| 2253 |
+
},
|
| 2254 |
+
{
|
| 2255 |
+
"epoch": 0.81,
|
| 2256 |
+
"learning_rate": 1.853324260005649e-05,
|
| 2257 |
+
"loss": 2.5733,
|
| 2258 |
+
"step": 32700
|
| 2259 |
+
},
|
| 2260 |
+
{
|
| 2261 |
+
"epoch": 0.81,
|
| 2262 |
+
"learning_rate": 1.8074272082873023e-05,
|
| 2263 |
+
"loss": 2.5602,
|
| 2264 |
+
"step": 32800
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 0.81,
|
| 2268 |
+
"learning_rate": 1.762049096981243e-05,
|
| 2269 |
+
"loss": 2.6033,
|
| 2270 |
+
"step": 32900
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.82,
|
| 2274 |
+
"learning_rate": 1.7171928004636205e-05,
|
| 2275 |
+
"loss": 2.5874,
|
| 2276 |
+
"step": 33000
|
| 2277 |
+
},
|
| 2278 |
+
{
|
| 2279 |
+
"epoch": 0.82,
|
| 2280 |
+
"eval_accuracy": 0.42740302882266734,
|
| 2281 |
+
"eval_loss": 3.416874885559082,
|
| 2282 |
+
"eval_runtime": 166.1443,
|
| 2283 |
+
"eval_samples_per_second": 6.019,
|
| 2284 |
+
"eval_steps_per_second": 3.009,
|
| 2285 |
+
"step": 33000
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 0.82,
|
| 2289 |
+
"learning_rate": 1.672861160057373e-05,
|
| 2290 |
+
"loss": 2.5585,
|
| 2291 |
+
"step": 33100
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.82,
|
| 2295 |
+
"learning_rate": 1.6290569838522605e-05,
|
| 2296 |
+
"loss": 2.6195,
|
| 2297 |
+
"step": 33200
|
| 2298 |
+
},
|
| 2299 |
+
{
|
| 2300 |
+
"epoch": 0.82,
|
| 2301 |
+
"learning_rate": 1.585783046526984e-05,
|
| 2302 |
+
"loss": 2.5723,
|
| 2303 |
+
"step": 33300
|
| 2304 |
+
},
|
| 2305 |
+
{
|
| 2306 |
+
"epoch": 0.83,
|
| 2307 |
+
"learning_rate": 1.5430420891734422e-05,
|
| 2308 |
+
"loss": 2.5354,
|
| 2309 |
+
"step": 33400
|
| 2310 |
+
},
|
| 2311 |
+
{
|
| 2312 |
+
"epoch": 0.83,
|
| 2313 |
+
"learning_rate": 1.500836819123086e-05,
|
| 2314 |
+
"loss": 2.5736,
|
| 2315 |
+
"step": 33500
|
| 2316 |
+
},
|
| 2317 |
+
{
|
| 2318 |
+
"epoch": 0.83,
|
| 2319 |
+
"learning_rate": 1.4591699097754541e-05,
|
| 2320 |
+
"loss": 2.5509,
|
| 2321 |
+
"step": 33600
|
| 2322 |
+
},
|
| 2323 |
+
{
|
| 2324 |
+
"epoch": 0.83,
|
| 2325 |
+
"learning_rate": 1.4180440004287954e-05,
|
| 2326 |
+
"loss": 2.5864,
|
| 2327 |
+
"step": 33700
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 0.84,
|
| 2331 |
+
"learning_rate": 1.3774616961129284e-05,
|
| 2332 |
+
"loss": 2.5337,
|
| 2333 |
+
"step": 33800
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.84,
|
| 2337 |
+
"learning_rate": 1.3374255674242031e-05,
|
| 2338 |
+
"loss": 2.5268,
|
| 2339 |
+
"step": 33900
|
| 2340 |
+
},
|
| 2341 |
+
{
|
| 2342 |
+
"epoch": 0.84,
|
| 2343 |
+
"learning_rate": 1.2979381503626842e-05,
|
| 2344 |
+
"loss": 2.5555,
|
| 2345 |
+
"step": 34000
|
| 2346 |
+
},
|
| 2347 |
+
{
|
| 2348 |
+
"epoch": 0.84,
|
| 2349 |
+
"eval_accuracy": 0.42858915486077187,
|
| 2350 |
+
"eval_loss": 3.4067070484161377,
|
| 2351 |
+
"eval_runtime": 152.5339,
|
| 2352 |
+
"eval_samples_per_second": 6.556,
|
| 2353 |
+
"eval_steps_per_second": 3.278,
|
| 2354 |
+
"step": 34000
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.84,
|
| 2358 |
+
"learning_rate": 1.2590019461715118e-05,
|
| 2359 |
+
"loss": 2.5291,
|
| 2360 |
+
"step": 34100
|
| 2361 |
+
},
|
| 2362 |
+
{
|
| 2363 |
+
"epoch": 0.84,
|
| 2364 |
+
"learning_rate": 1.2206194211784615e-05,
|
| 2365 |
+
"loss": 2.5592,
|
| 2366 |
+
"step": 34200
|
| 2367 |
+
},
|
| 2368 |
+
{
|
| 2369 |
+
"epoch": 0.85,
|
| 2370 |
+
"learning_rate": 1.182793006639732e-05,
|
| 2371 |
+
"loss": 2.535,
|
| 2372 |
+
"step": 34300
|
| 2373 |
+
},
|
| 2374 |
+
{
|
| 2375 |
+
"epoch": 0.85,
|
| 2376 |
+
"learning_rate": 1.1455250985859277e-05,
|
| 2377 |
+
"loss": 2.5476,
|
| 2378 |
+
"step": 34400
|
| 2379 |
+
},
|
| 2380 |
+
{
|
| 2381 |
+
"epoch": 0.85,
|
| 2382 |
+
"learning_rate": 1.1088180576702967e-05,
|
| 2383 |
+
"loss": 2.5339,
|
| 2384 |
+
"step": 34500
|
| 2385 |
+
},
|
| 2386 |
+
{
|
| 2387 |
+
"epoch": 0.85,
|
| 2388 |
+
"learning_rate": 1.0726742090191965e-05,
|
| 2389 |
+
"loss": 2.5451,
|
| 2390 |
+
"step": 34600
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 0.86,
|
| 2394 |
+
"learning_rate": 1.037095842084822e-05,
|
| 2395 |
+
"loss": 2.5518,
|
| 2396 |
+
"step": 34700
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.86,
|
| 2400 |
+
"learning_rate": 1.0020852105001677e-05,
|
| 2401 |
+
"loss": 2.5064,
|
| 2402 |
+
"step": 34800
|
| 2403 |
+
},
|
| 2404 |
+
{
|
| 2405 |
+
"epoch": 0.86,
|
| 2406 |
+
"learning_rate": 9.676445319362993e-06,
|
| 2407 |
+
"loss": 2.5714,
|
| 2408 |
+
"step": 34900
|
| 2409 |
+
},
|
| 2410 |
+
{
|
| 2411 |
+
"epoch": 0.86,
|
| 2412 |
+
"learning_rate": 9.337759879618601e-06,
|
| 2413 |
+
"loss": 2.5502,
|
| 2414 |
+
"step": 35000
|
| 2415 |
+
},
|
| 2416 |
+
{
|
| 2417 |
+
"epoch": 0.86,
|
| 2418 |
+
"eval_accuracy": 0.42984025403028825,
|
| 2419 |
+
"eval_loss": 3.3996939659118652,
|
| 2420 |
+
"eval_runtime": 152.8547,
|
| 2421 |
+
"eval_samples_per_second": 6.542,
|
| 2422 |
+
"eval_steps_per_second": 3.271,
|
| 2423 |
+
"step": 35000
|
| 2424 |
+
},
|
| 2425 |
+
{
|
| 2426 |
+
"epoch": 0.87,
|
| 2427 |
+
"learning_rate": 9.004817239048968e-06,
|
| 2428 |
+
"loss": 2.5244,
|
| 2429 |
+
"step": 35100
|
| 2430 |
+
},
|
| 2431 |
+
{
|
| 2432 |
+
"epoch": 0.87,
|
| 2433 |
+
"learning_rate": 8.677638487169614e-06,
|
| 2434 |
+
"loss": 2.5521,
|
| 2435 |
+
"step": 35200
|
| 2436 |
+
},
|
| 2437 |
+
{
|
| 2438 |
+
"epoch": 0.87,
|
| 2439 |
+
"learning_rate": 8.356244348395304e-06,
|
| 2440 |
+
"loss": 2.5726,
|
| 2441 |
+
"step": 35300
|
| 2442 |
+
},
|
| 2443 |
+
{
|
| 2444 |
+
"epoch": 0.87,
|
| 2445 |
+
"learning_rate": 8.04065518072723e-06,
|
| 2446 |
+
"loss": 2.538,
|
| 2447 |
+
"step": 35400
|
| 2448 |
+
},
|
| 2449 |
+
{
|
| 2450 |
+
"epoch": 0.88,
|
| 2451 |
+
"learning_rate": 7.730890974463601e-06,
|
| 2452 |
+
"loss": 2.5026,
|
| 2453 |
+
"step": 35500
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 0.88,
|
| 2457 |
+
"learning_rate": 7.42697135093321e-06,
|
| 2458 |
+
"loss": 2.5445,
|
| 2459 |
+
"step": 35600
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.88,
|
| 2463 |
+
"learning_rate": 7.1289155612527805e-06,
|
| 2464 |
+
"loss": 2.5331,
|
| 2465 |
+
"step": 35700
|
| 2466 |
+
},
|
| 2467 |
+
{
|
| 2468 |
+
"epoch": 0.88,
|
| 2469 |
+
"learning_rate": 6.836742485107383e-06,
|
| 2470 |
+
"loss": 2.5293,
|
| 2471 |
+
"step": 35800
|
| 2472 |
+
},
|
| 2473 |
+
{
|
| 2474 |
+
"epoch": 0.89,
|
| 2475 |
+
"learning_rate": 6.55047062955465e-06,
|
| 2476 |
+
"loss": 2.4755,
|
| 2477 |
+
"step": 35900
|
| 2478 |
+
},
|
| 2479 |
+
{
|
| 2480 |
+
"epoch": 0.89,
|
| 2481 |
+
"learning_rate": 6.270118127852376e-06,
|
| 2482 |
+
"loss": 2.5232,
|
| 2483 |
+
"step": 36000
|
| 2484 |
+
},
|
| 2485 |
+
{
|
| 2486 |
+
"epoch": 0.89,
|
| 2487 |
+
"eval_accuracy": 0.4302408402540303,
|
| 2488 |
+
"eval_loss": 3.3945846557617188,
|
| 2489 |
+
"eval_runtime": 153.7116,
|
| 2490 |
+
"eval_samples_per_second": 6.506,
|
| 2491 |
+
"eval_steps_per_second": 3.253,
|
| 2492 |
+
"step": 36000
|
| 2493 |
+
},
|
| 2494 |
+
{
|
| 2495 |
+
"epoch": 0.89,
|
| 2496 |
+
"learning_rate": 5.995702738310038e-06,
|
| 2497 |
+
"loss": 2.5161,
|
| 2498 |
+
"step": 36100
|
| 2499 |
+
},
|
| 2500 |
+
{
|
| 2501 |
+
"epoch": 0.89,
|
| 2502 |
+
"learning_rate": 5.727241843163822e-06,
|
| 2503 |
+
"loss": 2.5739,
|
| 2504 |
+
"step": 36200
|
| 2505 |
+
},
|
| 2506 |
+
{
|
| 2507 |
+
"epoch": 0.9,
|
| 2508 |
+
"learning_rate": 5.464752447475685e-06,
|
| 2509 |
+
"loss": 2.5553,
|
| 2510 |
+
"step": 36300
|
| 2511 |
+
},
|
| 2512 |
+
{
|
| 2513 |
+
"epoch": 0.9,
|
| 2514 |
+
"learning_rate": 5.208251178056056e-06,
|
| 2515 |
+
"loss": 2.5469,
|
| 2516 |
+
"step": 36400
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 0.9,
|
| 2520 |
+
"learning_rate": 4.9577542824108535e-06,
|
| 2521 |
+
"loss": 2.5504,
|
| 2522 |
+
"step": 36500
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.9,
|
| 2526 |
+
"learning_rate": 4.713277627712143e-06,
|
| 2527 |
+
"loss": 2.57,
|
| 2528 |
+
"step": 36600
|
| 2529 |
+
},
|
| 2530 |
+
{
|
| 2531 |
+
"epoch": 0.91,
|
| 2532 |
+
"learning_rate": 4.474836699793139e-06,
|
| 2533 |
+
"loss": 2.5595,
|
| 2534 |
+
"step": 36700
|
| 2535 |
+
},
|
| 2536 |
+
{
|
| 2537 |
+
"epoch": 0.91,
|
| 2538 |
+
"learning_rate": 4.242446602167361e-06,
|
| 2539 |
+
"loss": 2.5876,
|
| 2540 |
+
"step": 36800
|
| 2541 |
+
},
|
| 2542 |
+
{
|
| 2543 |
+
"epoch": 0.91,
|
| 2544 |
+
"learning_rate": 4.016122055071725e-06,
|
| 2545 |
+
"loss": 2.5513,
|
| 2546 |
+
"step": 36900
|
| 2547 |
+
},
|
| 2548 |
+
{
|
| 2549 |
+
"epoch": 0.91,
|
| 2550 |
+
"learning_rate": 3.795877394534386e-06,
|
| 2551 |
+
"loss": 2.5369,
|
| 2552 |
+
"step": 37000
|
| 2553 |
+
},
|
| 2554 |
+
{
|
| 2555 |
+
"epoch": 0.91,
|
| 2556 |
+
"eval_accuracy": 0.4308510014655594,
|
| 2557 |
+
"eval_loss": 3.3898189067840576,
|
| 2558 |
+
"eval_runtime": 153.5134,
|
| 2559 |
+
"eval_samples_per_second": 6.514,
|
| 2560 |
+
"eval_steps_per_second": 3.257,
|
| 2561 |
+
"step": 37000
|
| 2562 |
+
},
|
| 2563 |
+
{
|
| 2564 |
+
"epoch": 0.92,
|
| 2565 |
+
"learning_rate": 3.5817265714664592e-06,
|
| 2566 |
+
"loss": 2.5822,
|
| 2567 |
+
"step": 37100
|
| 2568 |
+
},
|
| 2569 |
+
{
|
| 2570 |
+
"epoch": 0.92,
|
| 2571 |
+
"learning_rate": 3.373683150778406e-06,
|
| 2572 |
+
"loss": 2.5611,
|
| 2573 |
+
"step": 37200
|
| 2574 |
+
},
|
| 2575 |
+
{
|
| 2576 |
+
"epoch": 0.92,
|
| 2577 |
+
"learning_rate": 3.171760310520788e-06,
|
| 2578 |
+
"loss": 2.5626,
|
| 2579 |
+
"step": 37300
|
| 2580 |
+
},
|
| 2581 |
+
{
|
| 2582 |
+
"epoch": 0.92,
|
| 2583 |
+
"learning_rate": 2.975970841049569e-06,
|
| 2584 |
+
"loss": 2.5263,
|
| 2585 |
+
"step": 37400
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.93,
|
| 2589 |
+
"learning_rate": 2.786327144215817e-06,
|
| 2590 |
+
"loss": 2.5412,
|
| 2591 |
+
"step": 37500
|
| 2592 |
+
},
|
| 2593 |
+
{
|
| 2594 |
+
"epoch": 0.93,
|
| 2595 |
+
"learning_rate": 2.6028412325802887e-06,
|
| 2596 |
+
"loss": 2.5721,
|
| 2597 |
+
"step": 37600
|
| 2598 |
+
},
|
| 2599 |
+
{
|
| 2600 |
+
"epoch": 0.93,
|
| 2601 |
+
"learning_rate": 2.4255247286524286e-06,
|
| 2602 |
+
"loss": 2.5137,
|
| 2603 |
+
"step": 37700
|
| 2604 |
+
},
|
| 2605 |
+
{
|
| 2606 |
+
"epoch": 0.93,
|
| 2607 |
+
"learning_rate": 2.254388864154189e-06,
|
| 2608 |
+
"loss": 2.5477,
|
| 2609 |
+
"step": 37800
|
| 2610 |
+
},
|
| 2611 |
+
{
|
| 2612 |
+
"epoch": 0.94,
|
| 2613 |
+
"learning_rate": 2.0894444793085664e-06,
|
| 2614 |
+
"loss": 2.5345,
|
| 2615 |
+
"step": 37900
|
| 2616 |
+
},
|
| 2617 |
+
{
|
| 2618 |
+
"epoch": 0.94,
|
| 2619 |
+
"learning_rate": 1.930702022152975e-06,
|
| 2620 |
+
"loss": 2.5335,
|
| 2621 |
+
"step": 38000
|
| 2622 |
+
},
|
| 2623 |
+
{
|
| 2624 |
+
"epoch": 0.94,
|
| 2625 |
+
"eval_accuracy": 0.43129213483146067,
|
| 2626 |
+
"eval_loss": 3.386915445327759,
|
| 2627 |
+
"eval_runtime": 152.6319,
|
| 2628 |
+
"eval_samples_per_second": 6.552,
|
| 2629 |
+
"eval_steps_per_second": 3.276,
|
| 2630 |
+
"step": 38000
|
| 2631 |
+
},
|
| 2632 |
+
{
|
| 2633 |
+
"epoch": 0.94,
|
| 2634 |
+
"learning_rate": 1.7781715478774497e-06,
|
| 2635 |
+
"loss": 2.5494,
|
| 2636 |
+
"step": 38100
|
| 2637 |
+
},
|
| 2638 |
+
{
|
| 2639 |
+
"epoch": 0.94,
|
| 2640 |
+
"learning_rate": 1.6318627181876933e-06,
|
| 2641 |
+
"loss": 2.5837,
|
| 2642 |
+
"step": 38200
|
| 2643 |
+
},
|
| 2644 |
+
{
|
| 2645 |
+
"epoch": 0.95,
|
| 2646 |
+
"learning_rate": 1.4917848006930413e-06,
|
| 2647 |
+
"loss": 2.5224,
|
| 2648 |
+
"step": 38300
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 0.95,
|
| 2652 |
+
"learning_rate": 1.357946668319554e-06,
|
| 2653 |
+
"loss": 2.5507,
|
| 2654 |
+
"step": 38400
|
| 2655 |
+
},
|
| 2656 |
+
{
|
| 2657 |
+
"epoch": 0.95,
|
| 2658 |
+
"learning_rate": 1.2303567987478336e-06,
|
| 2659 |
+
"loss": 2.5112,
|
| 2660 |
+
"step": 38500
|
| 2661 |
+
},
|
| 2662 |
+
{
|
| 2663 |
+
"epoch": 0.95,
|
| 2664 |
+
"learning_rate": 1.1090232738760975e-06,
|
| 2665 |
+
"loss": 2.5139,
|
| 2666 |
+
"step": 38600
|
| 2667 |
+
},
|
| 2668 |
+
{
|
| 2669 |
+
"epoch": 0.96,
|
| 2670 |
+
"learning_rate": 9.939537793082654e-07,
|
| 2671 |
+
"loss": 2.5593,
|
| 2672 |
+
"step": 38700
|
| 2673 |
+
},
|
| 2674 |
+
{
|
| 2675 |
+
"epoch": 0.96,
|
| 2676 |
+
"learning_rate": 8.851556038670828e-07,
|
| 2677 |
+
"loss": 2.5327,
|
| 2678 |
+
"step": 38800
|
| 2679 |
+
},
|
| 2680 |
+
{
|
| 2681 |
+
"epoch": 0.96,
|
| 2682 |
+
"learning_rate": 7.826356391324341e-07,
|
| 2683 |
+
"loss": 2.5504,
|
| 2684 |
+
"step": 38900
|
| 2685 |
+
},
|
| 2686 |
+
{
|
| 2687 |
+
"epoch": 0.96,
|
| 2688 |
+
"learning_rate": 6.864003790048923e-07,
|
| 2689 |
+
"loss": 2.6032,
|
| 2690 |
+
"step": 39000
|
| 2691 |
+
},
|
| 2692 |
+
{
|
| 2693 |
+
"epoch": 0.96,
|
| 2694 |
+
"eval_accuracy": 0.431466536394724,
|
| 2695 |
+
"eval_loss": 3.38529109954834,
|
| 2696 |
+
"eval_runtime": 155.9485,
|
| 2697 |
+
"eval_samples_per_second": 6.412,
|
| 2698 |
+
"eval_steps_per_second": 3.206,
|
| 2699 |
+
"step": 39000
|
| 2700 |
+
},
|
| 2701 |
+
{
|
| 2702 |
+
"epoch": 0.97,
|
| 2703 |
+
"learning_rate": 5.964559192942142e-07,
|
| 2704 |
+
"loss": 2.5262,
|
| 2705 |
+
"step": 39100
|
| 2706 |
+
},
|
| 2707 |
+
{
|
| 2708 |
+
"epoch": 0.97,
|
| 2709 |
+
"learning_rate": 5.128079573333943e-07,
|
| 2710 |
+
"loss": 2.5585,
|
| 2711 |
+
"step": 39200
|
| 2712 |
+
},
|
| 2713 |
+
{
|
| 2714 |
+
"epoch": 0.97,
|
| 2715 |
+
"learning_rate": 4.3546179161765287e-07,
|
| 2716 |
+
"loss": 2.4912,
|
| 2717 |
+
"step": 39300
|
| 2718 |
+
},
|
| 2719 |
+
{
|
| 2720 |
+
"epoch": 0.97,
|
| 2721 |
+
"learning_rate": 3.6442232146886024e-07,
|
| 2722 |
+
"loss": 2.5106,
|
| 2723 |
+
"step": 39400
|
| 2724 |
+
},
|
| 2725 |
+
{
|
| 2726 |
+
"epoch": 0.98,
|
| 2727 |
+
"learning_rate": 2.996940467252185e-07,
|
| 2728 |
+
"loss": 2.5446,
|
| 2729 |
+
"step": 39500
|
| 2730 |
+
},
|
| 2731 |
+
{
|
| 2732 |
+
"epoch": 0.98,
|
| 2733 |
+
"learning_rate": 2.412810674562005e-07,
|
| 2734 |
+
"loss": 2.5244,
|
| 2735 |
+
"step": 39600
|
| 2736 |
+
},
|
| 2737 |
+
{
|
| 2738 |
+
"epoch": 0.98,
|
| 2739 |
+
"learning_rate": 1.8918708370283532e-07,
|
| 2740 |
+
"loss": 2.5743,
|
| 2741 |
+
"step": 39700
|
| 2742 |
+
},
|
| 2743 |
+
{
|
| 2744 |
+
"epoch": 0.98,
|
| 2745 |
+
"learning_rate": 1.4341539524337366e-07,
|
| 2746 |
+
"loss": 2.5332,
|
| 2747 |
+
"step": 39800
|
| 2748 |
+
},
|
| 2749 |
+
{
|
| 2750 |
+
"epoch": 0.99,
|
| 2751 |
+
"learning_rate": 1.0396890138422155e-07,
|
| 2752 |
+
"loss": 2.5329,
|
| 2753 |
+
"step": 39900
|
| 2754 |
+
},
|
| 2755 |
+
{
|
| 2756 |
+
"epoch": 0.99,
|
| 2757 |
+
"learning_rate": 7.08501007763207e-08,
|
| 2758 |
+
"loss": 2.5244,
|
| 2759 |
+
"step": 40000
|
| 2760 |
+
},
|
| 2761 |
+
{
|
| 2762 |
+
"epoch": 0.99,
|
| 2763 |
+
"eval_accuracy": 0.43141182217879825,
|
| 2764 |
+
"eval_loss": 3.3849878311157227,
|
| 2765 |
+
"eval_runtime": 153.9769,
|
| 2766 |
+
"eval_samples_per_second": 6.494,
|
| 2767 |
+
"eval_steps_per_second": 3.247,
|
| 2768 |
+
"step": 40000
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 0.99,
|
| 2772 |
+
"learning_rate": 4.406109125690838e-08,
|
| 2773 |
+
"loss": 2.5568,
|
| 2774 |
+
"step": 40100
|
| 2775 |
+
},
|
| 2776 |
+
{
|
| 2777 |
+
"epoch": 0.99,
|
| 2778 |
+
"learning_rate": 2.3603569716545982e-08,
|
| 2779 |
+
"loss": 2.5507,
|
| 2780 |
+
"step": 40200
|
| 2781 |
+
},
|
| 2782 |
+
{
|
| 2783 |
+
"epoch": 1.0,
|
| 2784 |
+
"learning_rate": 9.478831991716064e-09,
|
| 2785 |
+
"loss": 2.5186,
|
| 2786 |
+
"step": 40300
|
| 2787 |
+
},
|
| 2788 |
+
{
|
| 2789 |
+
"epoch": 1.0,
|
| 2790 |
+
"learning_rate": 1.687772782721364e-09,
|
| 2791 |
+
"loss": 2.5194,
|
| 2792 |
+
"step": 40400
|
| 2793 |
+
},
|
| 2794 |
+
{
|
| 2795 |
+
"epoch": 1.0,
|
| 2796 |
+
"step": 40473,
|
| 2797 |
+
"total_flos": 3.382987152257188e+17,
|
| 2798 |
+
"train_loss": 3.073313300175377,
|
| 2799 |
+
"train_runtime": 120576.6781,
|
| 2800 |
+
"train_samples_per_second": 2.685,
|
| 2801 |
+
"train_steps_per_second": 0.336
|
| 2802 |
+
}
|
| 2803 |
+
],
|
| 2804 |
+
"max_steps": 40473,
|
| 2805 |
+
"num_train_epochs": 1,
|
| 2806 |
+
"total_flos": 3.382987152257188e+17,
|
| 2807 |
+
"trial_name": null,
|
| 2808 |
+
"trial_params": null
|
| 2809 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:334c2dbae5702869853de9c53c7e112befb726d3fb0d836f99de2e59a78d541a
|
| 3 |
+
size 3375
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|