Upload model finetuned on utg4java-220m using strategy src_fm_fc_ms_ff
Browse files- README.md +50 -3
- added_tokens.json +8 -0
- config.json +33 -0
- generation_config.json +8 -0
- merges.txt +0 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +59 -0
- tokenizer.json +0 -0
- tokenizer_config.json +917 -0
- trainer_state.json +1603 -0
- training_args.bin +3 -0
- vocab.json +0 -0
README.md
CHANGED
|
@@ -1,3 +1,50 @@
|
|
| 1 |
-
---
|
| 2 |
-
license:
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: bsd-3-clause
|
| 3 |
+
base_model: eljavatar/pretrain_utg4java_codet5p
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
model-index:
|
| 7 |
+
- name: finetuning_05_utg4java_src_fm_fc_ms_ff
|
| 8 |
+
results: []
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 12 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 13 |
+
|
| 14 |
+
# finetuning_05_utg4java_src_fm_fc_ms_ff
|
| 15 |
+
|
| 16 |
+
This model is a fine-tuned version of [eljavatar/pretrain_utg4java_codet5p](https://huggingface.co/eljavatar/pretrain_utg4java_codet5p) on an unknown dataset.
|
| 17 |
+
|
| 18 |
+
## Model description
|
| 19 |
+
|
| 20 |
+
More information needed
|
| 21 |
+
|
| 22 |
+
## Intended uses & limitations
|
| 23 |
+
|
| 24 |
+
More information needed
|
| 25 |
+
|
| 26 |
+
## Training and evaluation data
|
| 27 |
+
|
| 28 |
+
More information needed
|
| 29 |
+
|
| 30 |
+
## Training procedure
|
| 31 |
+
|
| 32 |
+
### Training hyperparameters
|
| 33 |
+
|
| 34 |
+
The following hyperparameters were used during training:
|
| 35 |
+
- learning_rate: 5e-05
|
| 36 |
+
- train_batch_size: 32
|
| 37 |
+
- eval_batch_size: 64
|
| 38 |
+
- seed: 42
|
| 39 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 40 |
+
- lr_scheduler_type: linear
|
| 41 |
+
- lr_scheduler_warmup_steps: 800
|
| 42 |
+
- num_epochs: 5
|
| 43 |
+
- mixed_precision_training: Native AMP
|
| 44 |
+
|
| 45 |
+
### Framework versions
|
| 46 |
+
|
| 47 |
+
- Transformers 4.38.2
|
| 48 |
+
- Pytorch 2.1.0
|
| 49 |
+
- Datasets 2.20.0
|
| 50 |
+
- Tokenizers 0.15.2
|
added_tokens.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</ECTX>": 32103,
|
| 3 |
+
"</FCTX>": 32101,
|
| 4 |
+
"</PRIVATE_FCTX>": 32105,
|
| 5 |
+
"<ECTX>": 32102,
|
| 6 |
+
"<FCTX>": 32100,
|
| 7 |
+
"<PRIVATE_FCTX>": 32104
|
| 8 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "eljavatar/pretrain_utg4java_codet5p",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"T5ForConditionalGeneration"
|
| 5 |
+
],
|
| 6 |
+
"bos_token_id": 1,
|
| 7 |
+
"classifier_dropout": 0.0,
|
| 8 |
+
"d_ff": 3072,
|
| 9 |
+
"d_kv": 64,
|
| 10 |
+
"d_model": 768,
|
| 11 |
+
"decoder_start_token_id": 0,
|
| 12 |
+
"dense_act_fn": "relu",
|
| 13 |
+
"dropout_rate": 0.1,
|
| 14 |
+
"eos_token_id": 2,
|
| 15 |
+
"feed_forward_proj": "relu",
|
| 16 |
+
"initializer_factor": 1.0,
|
| 17 |
+
"is_encoder_decoder": true,
|
| 18 |
+
"is_gated_act": false,
|
| 19 |
+
"layer_norm_epsilon": 1e-06,
|
| 20 |
+
"model_type": "t5",
|
| 21 |
+
"n_positions": 512,
|
| 22 |
+
"num_decoder_layers": 12,
|
| 23 |
+
"num_heads": 12,
|
| 24 |
+
"num_layers": 12,
|
| 25 |
+
"output_past": true,
|
| 26 |
+
"pad_token_id": 0,
|
| 27 |
+
"relative_attention_max_distance": 128,
|
| 28 |
+
"relative_attention_num_buckets": 32,
|
| 29 |
+
"torch_dtype": "float32",
|
| 30 |
+
"transformers_version": "4.38.2",
|
| 31 |
+
"use_cache": true,
|
| 32 |
+
"vocab_size": 32106
|
| 33 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"decoder_start_token_id": 0,
|
| 5 |
+
"eos_token_id": 2,
|
| 6 |
+
"pad_token_id": 0,
|
| 7 |
+
"transformers_version": "4.38.2"
|
| 8 |
+
}
|
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:e526c80833be7d4e59d6e49b68ca350831d4ba4bbde3fb8e571ddfb8d11a877f
|
| 3 |
+
size 891635790
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<FCTX>",
|
| 4 |
+
"</FCTX>",
|
| 5 |
+
"<ECTX>",
|
| 6 |
+
"</ECTX>",
|
| 7 |
+
"<PRIVATE_FCTX>",
|
| 8 |
+
"</PRIVATE_FCTX>"
|
| 9 |
+
],
|
| 10 |
+
"bos_token": {
|
| 11 |
+
"content": "<s>",
|
| 12 |
+
"lstrip": false,
|
| 13 |
+
"normalized": true,
|
| 14 |
+
"rstrip": false,
|
| 15 |
+
"single_word": false
|
| 16 |
+
},
|
| 17 |
+
"cls_token": {
|
| 18 |
+
"content": "<s>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": true,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
},
|
| 24 |
+
"eos_token": {
|
| 25 |
+
"content": "</s>",
|
| 26 |
+
"lstrip": false,
|
| 27 |
+
"normalized": true,
|
| 28 |
+
"rstrip": false,
|
| 29 |
+
"single_word": false
|
| 30 |
+
},
|
| 31 |
+
"mask_token": {
|
| 32 |
+
"content": "<mask>",
|
| 33 |
+
"lstrip": true,
|
| 34 |
+
"normalized": true,
|
| 35 |
+
"rstrip": false,
|
| 36 |
+
"single_word": false
|
| 37 |
+
},
|
| 38 |
+
"pad_token": {
|
| 39 |
+
"content": "<pad>",
|
| 40 |
+
"lstrip": false,
|
| 41 |
+
"normalized": true,
|
| 42 |
+
"rstrip": false,
|
| 43 |
+
"single_word": false
|
| 44 |
+
},
|
| 45 |
+
"sep_token": {
|
| 46 |
+
"content": "</s>",
|
| 47 |
+
"lstrip": false,
|
| 48 |
+
"normalized": true,
|
| 49 |
+
"rstrip": false,
|
| 50 |
+
"single_word": false
|
| 51 |
+
},
|
| 52 |
+
"unk_token": {
|
| 53 |
+
"content": "<unk>",
|
| 54 |
+
"lstrip": false,
|
| 55 |
+
"normalized": true,
|
| 56 |
+
"rstrip": false,
|
| 57 |
+
"single_word": false
|
| 58 |
+
}
|
| 59 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,917 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<pad>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<s>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"2": {
|
| 21 |
+
"content": "</s>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"3": {
|
| 29 |
+
"content": "<unk>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"4": {
|
| 37 |
+
"content": "<mask>",
|
| 38 |
+
"lstrip": true,
|
| 39 |
+
"normalized": true,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
},
|
| 44 |
+
"32000": {
|
| 45 |
+
"content": "<extra_id_99>",
|
| 46 |
+
"lstrip": true,
|
| 47 |
+
"normalized": true,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false,
|
| 50 |
+
"special": true
|
| 51 |
+
},
|
| 52 |
+
"32001": {
|
| 53 |
+
"content": "<extra_id_98>",
|
| 54 |
+
"lstrip": true,
|
| 55 |
+
"normalized": true,
|
| 56 |
+
"rstrip": false,
|
| 57 |
+
"single_word": false,
|
| 58 |
+
"special": true
|
| 59 |
+
},
|
| 60 |
+
"32002": {
|
| 61 |
+
"content": "<extra_id_97>",
|
| 62 |
+
"lstrip": true,
|
| 63 |
+
"normalized": true,
|
| 64 |
+
"rstrip": false,
|
| 65 |
+
"single_word": false,
|
| 66 |
+
"special": true
|
| 67 |
+
},
|
| 68 |
+
"32003": {
|
| 69 |
+
"content": "<extra_id_96>",
|
| 70 |
+
"lstrip": true,
|
| 71 |
+
"normalized": true,
|
| 72 |
+
"rstrip": false,
|
| 73 |
+
"single_word": false,
|
| 74 |
+
"special": true
|
| 75 |
+
},
|
| 76 |
+
"32004": {
|
| 77 |
+
"content": "<extra_id_95>",
|
| 78 |
+
"lstrip": true,
|
| 79 |
+
"normalized": true,
|
| 80 |
+
"rstrip": false,
|
| 81 |
+
"single_word": false,
|
| 82 |
+
"special": true
|
| 83 |
+
},
|
| 84 |
+
"32005": {
|
| 85 |
+
"content": "<extra_id_94>",
|
| 86 |
+
"lstrip": true,
|
| 87 |
+
"normalized": true,
|
| 88 |
+
"rstrip": false,
|
| 89 |
+
"single_word": false,
|
| 90 |
+
"special": true
|
| 91 |
+
},
|
| 92 |
+
"32006": {
|
| 93 |
+
"content": "<extra_id_93>",
|
| 94 |
+
"lstrip": true,
|
| 95 |
+
"normalized": true,
|
| 96 |
+
"rstrip": false,
|
| 97 |
+
"single_word": false,
|
| 98 |
+
"special": true
|
| 99 |
+
},
|
| 100 |
+
"32007": {
|
| 101 |
+
"content": "<extra_id_92>",
|
| 102 |
+
"lstrip": true,
|
| 103 |
+
"normalized": true,
|
| 104 |
+
"rstrip": false,
|
| 105 |
+
"single_word": false,
|
| 106 |
+
"special": true
|
| 107 |
+
},
|
| 108 |
+
"32008": {
|
| 109 |
+
"content": "<extra_id_91>",
|
| 110 |
+
"lstrip": true,
|
| 111 |
+
"normalized": true,
|
| 112 |
+
"rstrip": false,
|
| 113 |
+
"single_word": false,
|
| 114 |
+
"special": true
|
| 115 |
+
},
|
| 116 |
+
"32009": {
|
| 117 |
+
"content": "<extra_id_90>",
|
| 118 |
+
"lstrip": true,
|
| 119 |
+
"normalized": true,
|
| 120 |
+
"rstrip": false,
|
| 121 |
+
"single_word": false,
|
| 122 |
+
"special": true
|
| 123 |
+
},
|
| 124 |
+
"32010": {
|
| 125 |
+
"content": "<extra_id_89>",
|
| 126 |
+
"lstrip": true,
|
| 127 |
+
"normalized": true,
|
| 128 |
+
"rstrip": false,
|
| 129 |
+
"single_word": false,
|
| 130 |
+
"special": true
|
| 131 |
+
},
|
| 132 |
+
"32011": {
|
| 133 |
+
"content": "<extra_id_88>",
|
| 134 |
+
"lstrip": true,
|
| 135 |
+
"normalized": true,
|
| 136 |
+
"rstrip": false,
|
| 137 |
+
"single_word": false,
|
| 138 |
+
"special": true
|
| 139 |
+
},
|
| 140 |
+
"32012": {
|
| 141 |
+
"content": "<extra_id_87>",
|
| 142 |
+
"lstrip": true,
|
| 143 |
+
"normalized": true,
|
| 144 |
+
"rstrip": false,
|
| 145 |
+
"single_word": false,
|
| 146 |
+
"special": true
|
| 147 |
+
},
|
| 148 |
+
"32013": {
|
| 149 |
+
"content": "<extra_id_86>",
|
| 150 |
+
"lstrip": true,
|
| 151 |
+
"normalized": true,
|
| 152 |
+
"rstrip": false,
|
| 153 |
+
"single_word": false,
|
| 154 |
+
"special": true
|
| 155 |
+
},
|
| 156 |
+
"32014": {
|
| 157 |
+
"content": "<extra_id_85>",
|
| 158 |
+
"lstrip": true,
|
| 159 |
+
"normalized": true,
|
| 160 |
+
"rstrip": false,
|
| 161 |
+
"single_word": false,
|
| 162 |
+
"special": true
|
| 163 |
+
},
|
| 164 |
+
"32015": {
|
| 165 |
+
"content": "<extra_id_84>",
|
| 166 |
+
"lstrip": true,
|
| 167 |
+
"normalized": true,
|
| 168 |
+
"rstrip": false,
|
| 169 |
+
"single_word": false,
|
| 170 |
+
"special": true
|
| 171 |
+
},
|
| 172 |
+
"32016": {
|
| 173 |
+
"content": "<extra_id_83>",
|
| 174 |
+
"lstrip": true,
|
| 175 |
+
"normalized": true,
|
| 176 |
+
"rstrip": false,
|
| 177 |
+
"single_word": false,
|
| 178 |
+
"special": true
|
| 179 |
+
},
|
| 180 |
+
"32017": {
|
| 181 |
+
"content": "<extra_id_82>",
|
| 182 |
+
"lstrip": true,
|
| 183 |
+
"normalized": true,
|
| 184 |
+
"rstrip": false,
|
| 185 |
+
"single_word": false,
|
| 186 |
+
"special": true
|
| 187 |
+
},
|
| 188 |
+
"32018": {
|
| 189 |
+
"content": "<extra_id_81>",
|
| 190 |
+
"lstrip": true,
|
| 191 |
+
"normalized": true,
|
| 192 |
+
"rstrip": false,
|
| 193 |
+
"single_word": false,
|
| 194 |
+
"special": true
|
| 195 |
+
},
|
| 196 |
+
"32019": {
|
| 197 |
+
"content": "<extra_id_80>",
|
| 198 |
+
"lstrip": true,
|
| 199 |
+
"normalized": true,
|
| 200 |
+
"rstrip": false,
|
| 201 |
+
"single_word": false,
|
| 202 |
+
"special": true
|
| 203 |
+
},
|
| 204 |
+
"32020": {
|
| 205 |
+
"content": "<extra_id_79>",
|
| 206 |
+
"lstrip": true,
|
| 207 |
+
"normalized": true,
|
| 208 |
+
"rstrip": false,
|
| 209 |
+
"single_word": false,
|
| 210 |
+
"special": true
|
| 211 |
+
},
|
| 212 |
+
"32021": {
|
| 213 |
+
"content": "<extra_id_78>",
|
| 214 |
+
"lstrip": true,
|
| 215 |
+
"normalized": true,
|
| 216 |
+
"rstrip": false,
|
| 217 |
+
"single_word": false,
|
| 218 |
+
"special": true
|
| 219 |
+
},
|
| 220 |
+
"32022": {
|
| 221 |
+
"content": "<extra_id_77>",
|
| 222 |
+
"lstrip": true,
|
| 223 |
+
"normalized": true,
|
| 224 |
+
"rstrip": false,
|
| 225 |
+
"single_word": false,
|
| 226 |
+
"special": true
|
| 227 |
+
},
|
| 228 |
+
"32023": {
|
| 229 |
+
"content": "<extra_id_76>",
|
| 230 |
+
"lstrip": true,
|
| 231 |
+
"normalized": true,
|
| 232 |
+
"rstrip": false,
|
| 233 |
+
"single_word": false,
|
| 234 |
+
"special": true
|
| 235 |
+
},
|
| 236 |
+
"32024": {
|
| 237 |
+
"content": "<extra_id_75>",
|
| 238 |
+
"lstrip": true,
|
| 239 |
+
"normalized": true,
|
| 240 |
+
"rstrip": false,
|
| 241 |
+
"single_word": false,
|
| 242 |
+
"special": true
|
| 243 |
+
},
|
| 244 |
+
"32025": {
|
| 245 |
+
"content": "<extra_id_74>",
|
| 246 |
+
"lstrip": true,
|
| 247 |
+
"normalized": true,
|
| 248 |
+
"rstrip": false,
|
| 249 |
+
"single_word": false,
|
| 250 |
+
"special": true
|
| 251 |
+
},
|
| 252 |
+
"32026": {
|
| 253 |
+
"content": "<extra_id_73>",
|
| 254 |
+
"lstrip": true,
|
| 255 |
+
"normalized": true,
|
| 256 |
+
"rstrip": false,
|
| 257 |
+
"single_word": false,
|
| 258 |
+
"special": true
|
| 259 |
+
},
|
| 260 |
+
"32027": {
|
| 261 |
+
"content": "<extra_id_72>",
|
| 262 |
+
"lstrip": true,
|
| 263 |
+
"normalized": true,
|
| 264 |
+
"rstrip": false,
|
| 265 |
+
"single_word": false,
|
| 266 |
+
"special": true
|
| 267 |
+
},
|
| 268 |
+
"32028": {
|
| 269 |
+
"content": "<extra_id_71>",
|
| 270 |
+
"lstrip": true,
|
| 271 |
+
"normalized": true,
|
| 272 |
+
"rstrip": false,
|
| 273 |
+
"single_word": false,
|
| 274 |
+
"special": true
|
| 275 |
+
},
|
| 276 |
+
"32029": {
|
| 277 |
+
"content": "<extra_id_70>",
|
| 278 |
+
"lstrip": true,
|
| 279 |
+
"normalized": true,
|
| 280 |
+
"rstrip": false,
|
| 281 |
+
"single_word": false,
|
| 282 |
+
"special": true
|
| 283 |
+
},
|
| 284 |
+
"32030": {
|
| 285 |
+
"content": "<extra_id_69>",
|
| 286 |
+
"lstrip": true,
|
| 287 |
+
"normalized": true,
|
| 288 |
+
"rstrip": false,
|
| 289 |
+
"single_word": false,
|
| 290 |
+
"special": true
|
| 291 |
+
},
|
| 292 |
+
"32031": {
|
| 293 |
+
"content": "<extra_id_68>",
|
| 294 |
+
"lstrip": true,
|
| 295 |
+
"normalized": true,
|
| 296 |
+
"rstrip": false,
|
| 297 |
+
"single_word": false,
|
| 298 |
+
"special": true
|
| 299 |
+
},
|
| 300 |
+
"32032": {
|
| 301 |
+
"content": "<extra_id_67>",
|
| 302 |
+
"lstrip": true,
|
| 303 |
+
"normalized": true,
|
| 304 |
+
"rstrip": false,
|
| 305 |
+
"single_word": false,
|
| 306 |
+
"special": true
|
| 307 |
+
},
|
| 308 |
+
"32033": {
|
| 309 |
+
"content": "<extra_id_66>",
|
| 310 |
+
"lstrip": true,
|
| 311 |
+
"normalized": true,
|
| 312 |
+
"rstrip": false,
|
| 313 |
+
"single_word": false,
|
| 314 |
+
"special": true
|
| 315 |
+
},
|
| 316 |
+
"32034": {
|
| 317 |
+
"content": "<extra_id_65>",
|
| 318 |
+
"lstrip": true,
|
| 319 |
+
"normalized": true,
|
| 320 |
+
"rstrip": false,
|
| 321 |
+
"single_word": false,
|
| 322 |
+
"special": true
|
| 323 |
+
},
|
| 324 |
+
"32035": {
|
| 325 |
+
"content": "<extra_id_64>",
|
| 326 |
+
"lstrip": true,
|
| 327 |
+
"normalized": true,
|
| 328 |
+
"rstrip": false,
|
| 329 |
+
"single_word": false,
|
| 330 |
+
"special": true
|
| 331 |
+
},
|
| 332 |
+
"32036": {
|
| 333 |
+
"content": "<extra_id_63>",
|
| 334 |
+
"lstrip": true,
|
| 335 |
+
"normalized": true,
|
| 336 |
+
"rstrip": false,
|
| 337 |
+
"single_word": false,
|
| 338 |
+
"special": true
|
| 339 |
+
},
|
| 340 |
+
"32037": {
|
| 341 |
+
"content": "<extra_id_62>",
|
| 342 |
+
"lstrip": true,
|
| 343 |
+
"normalized": true,
|
| 344 |
+
"rstrip": false,
|
| 345 |
+
"single_word": false,
|
| 346 |
+
"special": true
|
| 347 |
+
},
|
| 348 |
+
"32038": {
|
| 349 |
+
"content": "<extra_id_61>",
|
| 350 |
+
"lstrip": true,
|
| 351 |
+
"normalized": true,
|
| 352 |
+
"rstrip": false,
|
| 353 |
+
"single_word": false,
|
| 354 |
+
"special": true
|
| 355 |
+
},
|
| 356 |
+
"32039": {
|
| 357 |
+
"content": "<extra_id_60>",
|
| 358 |
+
"lstrip": true,
|
| 359 |
+
"normalized": true,
|
| 360 |
+
"rstrip": false,
|
| 361 |
+
"single_word": false,
|
| 362 |
+
"special": true
|
| 363 |
+
},
|
| 364 |
+
"32040": {
|
| 365 |
+
"content": "<extra_id_59>",
|
| 366 |
+
"lstrip": true,
|
| 367 |
+
"normalized": true,
|
| 368 |
+
"rstrip": false,
|
| 369 |
+
"single_word": false,
|
| 370 |
+
"special": true
|
| 371 |
+
},
|
| 372 |
+
"32041": {
|
| 373 |
+
"content": "<extra_id_58>",
|
| 374 |
+
"lstrip": true,
|
| 375 |
+
"normalized": true,
|
| 376 |
+
"rstrip": false,
|
| 377 |
+
"single_word": false,
|
| 378 |
+
"special": true
|
| 379 |
+
},
|
| 380 |
+
"32042": {
|
| 381 |
+
"content": "<extra_id_57>",
|
| 382 |
+
"lstrip": true,
|
| 383 |
+
"normalized": true,
|
| 384 |
+
"rstrip": false,
|
| 385 |
+
"single_word": false,
|
| 386 |
+
"special": true
|
| 387 |
+
},
|
| 388 |
+
"32043": {
|
| 389 |
+
"content": "<extra_id_56>",
|
| 390 |
+
"lstrip": true,
|
| 391 |
+
"normalized": true,
|
| 392 |
+
"rstrip": false,
|
| 393 |
+
"single_word": false,
|
| 394 |
+
"special": true
|
| 395 |
+
},
|
| 396 |
+
"32044": {
|
| 397 |
+
"content": "<extra_id_55>",
|
| 398 |
+
"lstrip": true,
|
| 399 |
+
"normalized": true,
|
| 400 |
+
"rstrip": false,
|
| 401 |
+
"single_word": false,
|
| 402 |
+
"special": true
|
| 403 |
+
},
|
| 404 |
+
"32045": {
|
| 405 |
+
"content": "<extra_id_54>",
|
| 406 |
+
"lstrip": true,
|
| 407 |
+
"normalized": true,
|
| 408 |
+
"rstrip": false,
|
| 409 |
+
"single_word": false,
|
| 410 |
+
"special": true
|
| 411 |
+
},
|
| 412 |
+
"32046": {
|
| 413 |
+
"content": "<extra_id_53>",
|
| 414 |
+
"lstrip": true,
|
| 415 |
+
"normalized": true,
|
| 416 |
+
"rstrip": false,
|
| 417 |
+
"single_word": false,
|
| 418 |
+
"special": true
|
| 419 |
+
},
|
| 420 |
+
"32047": {
|
| 421 |
+
"content": "<extra_id_52>",
|
| 422 |
+
"lstrip": true,
|
| 423 |
+
"normalized": true,
|
| 424 |
+
"rstrip": false,
|
| 425 |
+
"single_word": false,
|
| 426 |
+
"special": true
|
| 427 |
+
},
|
| 428 |
+
"32048": {
|
| 429 |
+
"content": "<extra_id_51>",
|
| 430 |
+
"lstrip": true,
|
| 431 |
+
"normalized": true,
|
| 432 |
+
"rstrip": false,
|
| 433 |
+
"single_word": false,
|
| 434 |
+
"special": true
|
| 435 |
+
},
|
| 436 |
+
"32049": {
|
| 437 |
+
"content": "<extra_id_50>",
|
| 438 |
+
"lstrip": true,
|
| 439 |
+
"normalized": true,
|
| 440 |
+
"rstrip": false,
|
| 441 |
+
"single_word": false,
|
| 442 |
+
"special": true
|
| 443 |
+
},
|
| 444 |
+
"32050": {
|
| 445 |
+
"content": "<extra_id_49>",
|
| 446 |
+
"lstrip": true,
|
| 447 |
+
"normalized": true,
|
| 448 |
+
"rstrip": false,
|
| 449 |
+
"single_word": false,
|
| 450 |
+
"special": true
|
| 451 |
+
},
|
| 452 |
+
"32051": {
|
| 453 |
+
"content": "<extra_id_48>",
|
| 454 |
+
"lstrip": true,
|
| 455 |
+
"normalized": true,
|
| 456 |
+
"rstrip": false,
|
| 457 |
+
"single_word": false,
|
| 458 |
+
"special": true
|
| 459 |
+
},
|
| 460 |
+
"32052": {
|
| 461 |
+
"content": "<extra_id_47>",
|
| 462 |
+
"lstrip": true,
|
| 463 |
+
"normalized": true,
|
| 464 |
+
"rstrip": false,
|
| 465 |
+
"single_word": false,
|
| 466 |
+
"special": true
|
| 467 |
+
},
|
| 468 |
+
"32053": {
|
| 469 |
+
"content": "<extra_id_46>",
|
| 470 |
+
"lstrip": true,
|
| 471 |
+
"normalized": true,
|
| 472 |
+
"rstrip": false,
|
| 473 |
+
"single_word": false,
|
| 474 |
+
"special": true
|
| 475 |
+
},
|
| 476 |
+
"32054": {
|
| 477 |
+
"content": "<extra_id_45>",
|
| 478 |
+
"lstrip": true,
|
| 479 |
+
"normalized": true,
|
| 480 |
+
"rstrip": false,
|
| 481 |
+
"single_word": false,
|
| 482 |
+
"special": true
|
| 483 |
+
},
|
| 484 |
+
"32055": {
|
| 485 |
+
"content": "<extra_id_44>",
|
| 486 |
+
"lstrip": true,
|
| 487 |
+
"normalized": true,
|
| 488 |
+
"rstrip": false,
|
| 489 |
+
"single_word": false,
|
| 490 |
+
"special": true
|
| 491 |
+
},
|
| 492 |
+
"32056": {
|
| 493 |
+
"content": "<extra_id_43>",
|
| 494 |
+
"lstrip": true,
|
| 495 |
+
"normalized": true,
|
| 496 |
+
"rstrip": false,
|
| 497 |
+
"single_word": false,
|
| 498 |
+
"special": true
|
| 499 |
+
},
|
| 500 |
+
"32057": {
|
| 501 |
+
"content": "<extra_id_42>",
|
| 502 |
+
"lstrip": true,
|
| 503 |
+
"normalized": true,
|
| 504 |
+
"rstrip": false,
|
| 505 |
+
"single_word": false,
|
| 506 |
+
"special": true
|
| 507 |
+
},
|
| 508 |
+
"32058": {
|
| 509 |
+
"content": "<extra_id_41>",
|
| 510 |
+
"lstrip": true,
|
| 511 |
+
"normalized": true,
|
| 512 |
+
"rstrip": false,
|
| 513 |
+
"single_word": false,
|
| 514 |
+
"special": true
|
| 515 |
+
},
|
| 516 |
+
"32059": {
|
| 517 |
+
"content": "<extra_id_40>",
|
| 518 |
+
"lstrip": true,
|
| 519 |
+
"normalized": true,
|
| 520 |
+
"rstrip": false,
|
| 521 |
+
"single_word": false,
|
| 522 |
+
"special": true
|
| 523 |
+
},
|
| 524 |
+
"32060": {
|
| 525 |
+
"content": "<extra_id_39>",
|
| 526 |
+
"lstrip": true,
|
| 527 |
+
"normalized": true,
|
| 528 |
+
"rstrip": false,
|
| 529 |
+
"single_word": false,
|
| 530 |
+
"special": true
|
| 531 |
+
},
|
| 532 |
+
"32061": {
|
| 533 |
+
"content": "<extra_id_38>",
|
| 534 |
+
"lstrip": true,
|
| 535 |
+
"normalized": true,
|
| 536 |
+
"rstrip": false,
|
| 537 |
+
"single_word": false,
|
| 538 |
+
"special": true
|
| 539 |
+
},
|
| 540 |
+
"32062": {
|
| 541 |
+
"content": "<extra_id_37>",
|
| 542 |
+
"lstrip": true,
|
| 543 |
+
"normalized": true,
|
| 544 |
+
"rstrip": false,
|
| 545 |
+
"single_word": false,
|
| 546 |
+
"special": true
|
| 547 |
+
},
|
| 548 |
+
"32063": {
|
| 549 |
+
"content": "<extra_id_36>",
|
| 550 |
+
"lstrip": true,
|
| 551 |
+
"normalized": true,
|
| 552 |
+
"rstrip": false,
|
| 553 |
+
"single_word": false,
|
| 554 |
+
"special": true
|
| 555 |
+
},
|
| 556 |
+
"32064": {
|
| 557 |
+
"content": "<extra_id_35>",
|
| 558 |
+
"lstrip": true,
|
| 559 |
+
"normalized": true,
|
| 560 |
+
"rstrip": false,
|
| 561 |
+
"single_word": false,
|
| 562 |
+
"special": true
|
| 563 |
+
},
|
| 564 |
+
"32065": {
|
| 565 |
+
"content": "<extra_id_34>",
|
| 566 |
+
"lstrip": true,
|
| 567 |
+
"normalized": true,
|
| 568 |
+
"rstrip": false,
|
| 569 |
+
"single_word": false,
|
| 570 |
+
"special": true
|
| 571 |
+
},
|
| 572 |
+
"32066": {
|
| 573 |
+
"content": "<extra_id_33>",
|
| 574 |
+
"lstrip": true,
|
| 575 |
+
"normalized": true,
|
| 576 |
+
"rstrip": false,
|
| 577 |
+
"single_word": false,
|
| 578 |
+
"special": true
|
| 579 |
+
},
|
| 580 |
+
"32067": {
|
| 581 |
+
"content": "<extra_id_32>",
|
| 582 |
+
"lstrip": true,
|
| 583 |
+
"normalized": true,
|
| 584 |
+
"rstrip": false,
|
| 585 |
+
"single_word": false,
|
| 586 |
+
"special": true
|
| 587 |
+
},
|
| 588 |
+
"32068": {
|
| 589 |
+
"content": "<extra_id_31>",
|
| 590 |
+
"lstrip": true,
|
| 591 |
+
"normalized": true,
|
| 592 |
+
"rstrip": false,
|
| 593 |
+
"single_word": false,
|
| 594 |
+
"special": true
|
| 595 |
+
},
|
| 596 |
+
"32069": {
|
| 597 |
+
"content": "<extra_id_30>",
|
| 598 |
+
"lstrip": true,
|
| 599 |
+
"normalized": true,
|
| 600 |
+
"rstrip": false,
|
| 601 |
+
"single_word": false,
|
| 602 |
+
"special": true
|
| 603 |
+
},
|
| 604 |
+
"32070": {
|
| 605 |
+
"content": "<extra_id_29>",
|
| 606 |
+
"lstrip": true,
|
| 607 |
+
"normalized": true,
|
| 608 |
+
"rstrip": false,
|
| 609 |
+
"single_word": false,
|
| 610 |
+
"special": true
|
| 611 |
+
},
|
| 612 |
+
"32071": {
|
| 613 |
+
"content": "<extra_id_28>",
|
| 614 |
+
"lstrip": true,
|
| 615 |
+
"normalized": true,
|
| 616 |
+
"rstrip": false,
|
| 617 |
+
"single_word": false,
|
| 618 |
+
"special": true
|
| 619 |
+
},
|
| 620 |
+
"32072": {
|
| 621 |
+
"content": "<extra_id_27>",
|
| 622 |
+
"lstrip": true,
|
| 623 |
+
"normalized": true,
|
| 624 |
+
"rstrip": false,
|
| 625 |
+
"single_word": false,
|
| 626 |
+
"special": true
|
| 627 |
+
},
|
| 628 |
+
"32073": {
|
| 629 |
+
"content": "<extra_id_26>",
|
| 630 |
+
"lstrip": true,
|
| 631 |
+
"normalized": true,
|
| 632 |
+
"rstrip": false,
|
| 633 |
+
"single_word": false,
|
| 634 |
+
"special": true
|
| 635 |
+
},
|
| 636 |
+
"32074": {
|
| 637 |
+
"content": "<extra_id_25>",
|
| 638 |
+
"lstrip": true,
|
| 639 |
+
"normalized": true,
|
| 640 |
+
"rstrip": false,
|
| 641 |
+
"single_word": false,
|
| 642 |
+
"special": true
|
| 643 |
+
},
|
| 644 |
+
"32075": {
|
| 645 |
+
"content": "<extra_id_24>",
|
| 646 |
+
"lstrip": true,
|
| 647 |
+
"normalized": true,
|
| 648 |
+
"rstrip": false,
|
| 649 |
+
"single_word": false,
|
| 650 |
+
"special": true
|
| 651 |
+
},
|
| 652 |
+
"32076": {
|
| 653 |
+
"content": "<extra_id_23>",
|
| 654 |
+
"lstrip": true,
|
| 655 |
+
"normalized": true,
|
| 656 |
+
"rstrip": false,
|
| 657 |
+
"single_word": false,
|
| 658 |
+
"special": true
|
| 659 |
+
},
|
| 660 |
+
"32077": {
|
| 661 |
+
"content": "<extra_id_22>",
|
| 662 |
+
"lstrip": true,
|
| 663 |
+
"normalized": true,
|
| 664 |
+
"rstrip": false,
|
| 665 |
+
"single_word": false,
|
| 666 |
+
"special": true
|
| 667 |
+
},
|
| 668 |
+
"32078": {
|
| 669 |
+
"content": "<extra_id_21>",
|
| 670 |
+
"lstrip": true,
|
| 671 |
+
"normalized": true,
|
| 672 |
+
"rstrip": false,
|
| 673 |
+
"single_word": false,
|
| 674 |
+
"special": true
|
| 675 |
+
},
|
| 676 |
+
"32079": {
|
| 677 |
+
"content": "<extra_id_20>",
|
| 678 |
+
"lstrip": true,
|
| 679 |
+
"normalized": true,
|
| 680 |
+
"rstrip": false,
|
| 681 |
+
"single_word": false,
|
| 682 |
+
"special": true
|
| 683 |
+
},
|
| 684 |
+
"32080": {
|
| 685 |
+
"content": "<extra_id_19>",
|
| 686 |
+
"lstrip": true,
|
| 687 |
+
"normalized": true,
|
| 688 |
+
"rstrip": false,
|
| 689 |
+
"single_word": false,
|
| 690 |
+
"special": true
|
| 691 |
+
},
|
| 692 |
+
"32081": {
|
| 693 |
+
"content": "<extra_id_18>",
|
| 694 |
+
"lstrip": true,
|
| 695 |
+
"normalized": true,
|
| 696 |
+
"rstrip": false,
|
| 697 |
+
"single_word": false,
|
| 698 |
+
"special": true
|
| 699 |
+
},
|
| 700 |
+
"32082": {
|
| 701 |
+
"content": "<extra_id_17>",
|
| 702 |
+
"lstrip": true,
|
| 703 |
+
"normalized": true,
|
| 704 |
+
"rstrip": false,
|
| 705 |
+
"single_word": false,
|
| 706 |
+
"special": true
|
| 707 |
+
},
|
| 708 |
+
"32083": {
|
| 709 |
+
"content": "<extra_id_16>",
|
| 710 |
+
"lstrip": true,
|
| 711 |
+
"normalized": true,
|
| 712 |
+
"rstrip": false,
|
| 713 |
+
"single_word": false,
|
| 714 |
+
"special": true
|
| 715 |
+
},
|
| 716 |
+
"32084": {
|
| 717 |
+
"content": "<extra_id_15>",
|
| 718 |
+
"lstrip": true,
|
| 719 |
+
"normalized": true,
|
| 720 |
+
"rstrip": false,
|
| 721 |
+
"single_word": false,
|
| 722 |
+
"special": true
|
| 723 |
+
},
|
| 724 |
+
"32085": {
|
| 725 |
+
"content": "<extra_id_14>",
|
| 726 |
+
"lstrip": true,
|
| 727 |
+
"normalized": true,
|
| 728 |
+
"rstrip": false,
|
| 729 |
+
"single_word": false,
|
| 730 |
+
"special": true
|
| 731 |
+
},
|
| 732 |
+
"32086": {
|
| 733 |
+
"content": "<extra_id_13>",
|
| 734 |
+
"lstrip": true,
|
| 735 |
+
"normalized": true,
|
| 736 |
+
"rstrip": false,
|
| 737 |
+
"single_word": false,
|
| 738 |
+
"special": true
|
| 739 |
+
},
|
| 740 |
+
"32087": {
|
| 741 |
+
"content": "<extra_id_12>",
|
| 742 |
+
"lstrip": true,
|
| 743 |
+
"normalized": true,
|
| 744 |
+
"rstrip": false,
|
| 745 |
+
"single_word": false,
|
| 746 |
+
"special": true
|
| 747 |
+
},
|
| 748 |
+
"32088": {
|
| 749 |
+
"content": "<extra_id_11>",
|
| 750 |
+
"lstrip": true,
|
| 751 |
+
"normalized": true,
|
| 752 |
+
"rstrip": false,
|
| 753 |
+
"single_word": false,
|
| 754 |
+
"special": true
|
| 755 |
+
},
|
| 756 |
+
"32089": {
|
| 757 |
+
"content": "<extra_id_10>",
|
| 758 |
+
"lstrip": true,
|
| 759 |
+
"normalized": true,
|
| 760 |
+
"rstrip": false,
|
| 761 |
+
"single_word": false,
|
| 762 |
+
"special": true
|
| 763 |
+
},
|
| 764 |
+
"32090": {
|
| 765 |
+
"content": "<extra_id_9>",
|
| 766 |
+
"lstrip": true,
|
| 767 |
+
"normalized": true,
|
| 768 |
+
"rstrip": false,
|
| 769 |
+
"single_word": false,
|
| 770 |
+
"special": true
|
| 771 |
+
},
|
| 772 |
+
"32091": {
|
| 773 |
+
"content": "<extra_id_8>",
|
| 774 |
+
"lstrip": true,
|
| 775 |
+
"normalized": true,
|
| 776 |
+
"rstrip": false,
|
| 777 |
+
"single_word": false,
|
| 778 |
+
"special": true
|
| 779 |
+
},
|
| 780 |
+
"32092": {
|
| 781 |
+
"content": "<extra_id_7>",
|
| 782 |
+
"lstrip": true,
|
| 783 |
+
"normalized": true,
|
| 784 |
+
"rstrip": false,
|
| 785 |
+
"single_word": false,
|
| 786 |
+
"special": true
|
| 787 |
+
},
|
| 788 |
+
"32093": {
|
| 789 |
+
"content": "<extra_id_6>",
|
| 790 |
+
"lstrip": true,
|
| 791 |
+
"normalized": true,
|
| 792 |
+
"rstrip": false,
|
| 793 |
+
"single_word": false,
|
| 794 |
+
"special": true
|
| 795 |
+
},
|
| 796 |
+
"32094": {
|
| 797 |
+
"content": "<extra_id_5>",
|
| 798 |
+
"lstrip": true,
|
| 799 |
+
"normalized": true,
|
| 800 |
+
"rstrip": false,
|
| 801 |
+
"single_word": false,
|
| 802 |
+
"special": true
|
| 803 |
+
},
|
| 804 |
+
"32095": {
|
| 805 |
+
"content": "<extra_id_4>",
|
| 806 |
+
"lstrip": true,
|
| 807 |
+
"normalized": true,
|
| 808 |
+
"rstrip": false,
|
| 809 |
+
"single_word": false,
|
| 810 |
+
"special": true
|
| 811 |
+
},
|
| 812 |
+
"32096": {
|
| 813 |
+
"content": "<extra_id_3>",
|
| 814 |
+
"lstrip": true,
|
| 815 |
+
"normalized": true,
|
| 816 |
+
"rstrip": false,
|
| 817 |
+
"single_word": false,
|
| 818 |
+
"special": true
|
| 819 |
+
},
|
| 820 |
+
"32097": {
|
| 821 |
+
"content": "<extra_id_2>",
|
| 822 |
+
"lstrip": true,
|
| 823 |
+
"normalized": true,
|
| 824 |
+
"rstrip": false,
|
| 825 |
+
"single_word": false,
|
| 826 |
+
"special": true
|
| 827 |
+
},
|
| 828 |
+
"32098": {
|
| 829 |
+
"content": "<extra_id_1>",
|
| 830 |
+
"lstrip": true,
|
| 831 |
+
"normalized": true,
|
| 832 |
+
"rstrip": false,
|
| 833 |
+
"single_word": false,
|
| 834 |
+
"special": true
|
| 835 |
+
},
|
| 836 |
+
"32099": {
|
| 837 |
+
"content": "<extra_id_0>",
|
| 838 |
+
"lstrip": true,
|
| 839 |
+
"normalized": true,
|
| 840 |
+
"rstrip": false,
|
| 841 |
+
"single_word": false,
|
| 842 |
+
"special": true
|
| 843 |
+
},
|
| 844 |
+
"32100": {
|
| 845 |
+
"content": "<FCTX>",
|
| 846 |
+
"lstrip": false,
|
| 847 |
+
"normalized": false,
|
| 848 |
+
"rstrip": false,
|
| 849 |
+
"single_word": false,
|
| 850 |
+
"special": true
|
| 851 |
+
},
|
| 852 |
+
"32101": {
|
| 853 |
+
"content": "</FCTX>",
|
| 854 |
+
"lstrip": false,
|
| 855 |
+
"normalized": false,
|
| 856 |
+
"rstrip": false,
|
| 857 |
+
"single_word": false,
|
| 858 |
+
"special": true
|
| 859 |
+
},
|
| 860 |
+
"32102": {
|
| 861 |
+
"content": "<ECTX>",
|
| 862 |
+
"lstrip": false,
|
| 863 |
+
"normalized": false,
|
| 864 |
+
"rstrip": false,
|
| 865 |
+
"single_word": false,
|
| 866 |
+
"special": true
|
| 867 |
+
},
|
| 868 |
+
"32103": {
|
| 869 |
+
"content": "</ECTX>",
|
| 870 |
+
"lstrip": false,
|
| 871 |
+
"normalized": false,
|
| 872 |
+
"rstrip": false,
|
| 873 |
+
"single_word": false,
|
| 874 |
+
"special": true
|
| 875 |
+
},
|
| 876 |
+
"32104": {
|
| 877 |
+
"content": "<PRIVATE_FCTX>",
|
| 878 |
+
"lstrip": false,
|
| 879 |
+
"normalized": false,
|
| 880 |
+
"rstrip": false,
|
| 881 |
+
"single_word": false,
|
| 882 |
+
"special": true
|
| 883 |
+
},
|
| 884 |
+
"32105": {
|
| 885 |
+
"content": "</PRIVATE_FCTX>",
|
| 886 |
+
"lstrip": false,
|
| 887 |
+
"normalized": false,
|
| 888 |
+
"rstrip": false,
|
| 889 |
+
"single_word": false,
|
| 890 |
+
"special": true
|
| 891 |
+
}
|
| 892 |
+
},
|
| 893 |
+
"additional_special_tokens": [
|
| 894 |
+
"<FCTX>",
|
| 895 |
+
"</FCTX>",
|
| 896 |
+
"<ECTX>",
|
| 897 |
+
"</ECTX>",
|
| 898 |
+
"<PRIVATE_FCTX>",
|
| 899 |
+
"</PRIVATE_FCTX>"
|
| 900 |
+
],
|
| 901 |
+
"bos_token": "<s>",
|
| 902 |
+
"clean_up_tokenization_spaces": true,
|
| 903 |
+
"cls_token": "<s>",
|
| 904 |
+
"eos_token": "</s>",
|
| 905 |
+
"errors": "replace",
|
| 906 |
+
"mask_token": "<mask>",
|
| 907 |
+
"max_length": 512,
|
| 908 |
+
"model_max_length": 512,
|
| 909 |
+
"pad_token": "<pad>",
|
| 910 |
+
"sep_token": "</s>",
|
| 911 |
+
"stride": 0,
|
| 912 |
+
"tokenizer_class": "RobertaTokenizer",
|
| 913 |
+
"trim_offsets": true,
|
| 914 |
+
"truncation_side": "right",
|
| 915 |
+
"truncation_strategy": "longest_first",
|
| 916 |
+
"unk_token": "<unk>"
|
| 917 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1603 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.177372694015503,
|
| 3 |
+
"best_model_checkpoint": "/root/finetuning_executions/finetuning_05_utg4java_src_fm_fc_ms_ff/checkpoint-52644",
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 87740,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02,
|
| 13 |
+
"grad_norm": 2.229098081588745,
|
| 14 |
+
"learning_rate": 2.4750000000000002e-05,
|
| 15 |
+
"loss": 3.3229,
|
| 16 |
+
"step": 400
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.05,
|
| 20 |
+
"grad_norm": 1.5058149099349976,
|
| 21 |
+
"learning_rate": 4.975e-05,
|
| 22 |
+
"loss": 1.748,
|
| 23 |
+
"step": 800
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.07,
|
| 27 |
+
"grad_norm": 1.120527982711792,
|
| 28 |
+
"learning_rate": 4.977225672877847e-05,
|
| 29 |
+
"loss": 1.5462,
|
| 30 |
+
"step": 1200
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.09,
|
| 34 |
+
"grad_norm": 1.5305182933807373,
|
| 35 |
+
"learning_rate": 4.9542213020473895e-05,
|
| 36 |
+
"loss": 1.4719,
|
| 37 |
+
"step": 1600
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.11,
|
| 41 |
+
"grad_norm": 1.0366812944412231,
|
| 42 |
+
"learning_rate": 4.931216931216932e-05,
|
| 43 |
+
"loss": 1.4154,
|
| 44 |
+
"step": 2000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.14,
|
| 48 |
+
"grad_norm": 1.1470097303390503,
|
| 49 |
+
"learning_rate": 4.9082125603864734e-05,
|
| 50 |
+
"loss": 1.3773,
|
| 51 |
+
"step": 2400
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.16,
|
| 55 |
+
"grad_norm": 1.1128191947937012,
|
| 56 |
+
"learning_rate": 4.885208189556016e-05,
|
| 57 |
+
"loss": 1.3428,
|
| 58 |
+
"step": 2800
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.18,
|
| 62 |
+
"grad_norm": 1.253928303718567,
|
| 63 |
+
"learning_rate": 4.862203818725558e-05,
|
| 64 |
+
"loss": 1.3226,
|
| 65 |
+
"step": 3200
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.21,
|
| 69 |
+
"grad_norm": 1.0254652500152588,
|
| 70 |
+
"learning_rate": 4.8391994478951e-05,
|
| 71 |
+
"loss": 1.291,
|
| 72 |
+
"step": 3600
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.23,
|
| 76 |
+
"grad_norm": 1.1027791500091553,
|
| 77 |
+
"learning_rate": 4.8161950770646426e-05,
|
| 78 |
+
"loss": 1.2715,
|
| 79 |
+
"step": 4000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.25,
|
| 83 |
+
"grad_norm": 1.1553912162780762,
|
| 84 |
+
"learning_rate": 4.793190706234185e-05,
|
| 85 |
+
"loss": 1.2662,
|
| 86 |
+
"step": 4400
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.27,
|
| 90 |
+
"grad_norm": 1.144874930381775,
|
| 91 |
+
"learning_rate": 4.770186335403727e-05,
|
| 92 |
+
"loss": 1.242,
|
| 93 |
+
"step": 4800
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.3,
|
| 97 |
+
"grad_norm": 1.1357582807540894,
|
| 98 |
+
"learning_rate": 4.747181964573269e-05,
|
| 99 |
+
"loss": 1.2392,
|
| 100 |
+
"step": 5200
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.32,
|
| 104 |
+
"grad_norm": 0.8985121250152588,
|
| 105 |
+
"learning_rate": 4.724177593742811e-05,
|
| 106 |
+
"loss": 1.2093,
|
| 107 |
+
"step": 5600
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.34,
|
| 111 |
+
"grad_norm": 0.9756821990013123,
|
| 112 |
+
"learning_rate": 4.7011732229123534e-05,
|
| 113 |
+
"loss": 1.1905,
|
| 114 |
+
"step": 6000
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.36,
|
| 118 |
+
"grad_norm": 1.1240946054458618,
|
| 119 |
+
"learning_rate": 4.678168852081896e-05,
|
| 120 |
+
"loss": 1.1769,
|
| 121 |
+
"step": 6400
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.39,
|
| 125 |
+
"grad_norm": 0.9794073700904846,
|
| 126 |
+
"learning_rate": 4.655164481251438e-05,
|
| 127 |
+
"loss": 1.1658,
|
| 128 |
+
"step": 6800
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.41,
|
| 132 |
+
"grad_norm": 0.977505624294281,
|
| 133 |
+
"learning_rate": 4.63216011042098e-05,
|
| 134 |
+
"loss": 1.1749,
|
| 135 |
+
"step": 7200
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.43,
|
| 139 |
+
"grad_norm": 1.1371684074401855,
|
| 140 |
+
"learning_rate": 4.609155739590522e-05,
|
| 141 |
+
"loss": 1.1424,
|
| 142 |
+
"step": 7600
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.46,
|
| 146 |
+
"grad_norm": 1.0095725059509277,
|
| 147 |
+
"learning_rate": 4.586151368760065e-05,
|
| 148 |
+
"loss": 1.1417,
|
| 149 |
+
"step": 8000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.48,
|
| 153 |
+
"grad_norm": 1.0289288759231567,
|
| 154 |
+
"learning_rate": 4.563146997929607e-05,
|
| 155 |
+
"loss": 1.1408,
|
| 156 |
+
"step": 8400
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.5,
|
| 160 |
+
"grad_norm": 1.1291728019714355,
|
| 161 |
+
"learning_rate": 4.5401426270991495e-05,
|
| 162 |
+
"loss": 1.1243,
|
| 163 |
+
"step": 8800
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.52,
|
| 167 |
+
"grad_norm": 1.1059273481369019,
|
| 168 |
+
"learning_rate": 4.517138256268692e-05,
|
| 169 |
+
"loss": 1.1143,
|
| 170 |
+
"step": 9200
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.55,
|
| 174 |
+
"grad_norm": 0.8934476971626282,
|
| 175 |
+
"learning_rate": 4.4941338854382334e-05,
|
| 176 |
+
"loss": 1.1019,
|
| 177 |
+
"step": 9600
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.57,
|
| 181 |
+
"grad_norm": 1.0957863330841064,
|
| 182 |
+
"learning_rate": 4.471129514607776e-05,
|
| 183 |
+
"loss": 1.094,
|
| 184 |
+
"step": 10000
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.59,
|
| 188 |
+
"grad_norm": 1.2800277471542358,
|
| 189 |
+
"learning_rate": 4.448125143777318e-05,
|
| 190 |
+
"loss": 1.0773,
|
| 191 |
+
"step": 10400
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.62,
|
| 195 |
+
"grad_norm": 0.9538272619247437,
|
| 196 |
+
"learning_rate": 4.42512077294686e-05,
|
| 197 |
+
"loss": 1.0952,
|
| 198 |
+
"step": 10800
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.64,
|
| 202 |
+
"grad_norm": 0.9216693639755249,
|
| 203 |
+
"learning_rate": 4.4021164021164026e-05,
|
| 204 |
+
"loss": 1.0725,
|
| 205 |
+
"step": 11200
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.66,
|
| 209 |
+
"grad_norm": 0.9263585805892944,
|
| 210 |
+
"learning_rate": 4.379112031285945e-05,
|
| 211 |
+
"loss": 1.0693,
|
| 212 |
+
"step": 11600
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.68,
|
| 216 |
+
"grad_norm": 0.9969348311424255,
|
| 217 |
+
"learning_rate": 4.3561076604554865e-05,
|
| 218 |
+
"loss": 1.0582,
|
| 219 |
+
"step": 12000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.71,
|
| 223 |
+
"grad_norm": 0.9046760201454163,
|
| 224 |
+
"learning_rate": 4.333103289625029e-05,
|
| 225 |
+
"loss": 1.062,
|
| 226 |
+
"step": 12400
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.73,
|
| 230 |
+
"grad_norm": 1.0008882284164429,
|
| 231 |
+
"learning_rate": 4.310098918794571e-05,
|
| 232 |
+
"loss": 1.0676,
|
| 233 |
+
"step": 12800
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.75,
|
| 237 |
+
"grad_norm": 1.0041269063949585,
|
| 238 |
+
"learning_rate": 4.2870945479641134e-05,
|
| 239 |
+
"loss": 1.0428,
|
| 240 |
+
"step": 13200
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.78,
|
| 244 |
+
"grad_norm": 1.0424519777297974,
|
| 245 |
+
"learning_rate": 4.264090177133656e-05,
|
| 246 |
+
"loss": 1.0403,
|
| 247 |
+
"step": 13600
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.8,
|
| 251 |
+
"grad_norm": 0.9091747999191284,
|
| 252 |
+
"learning_rate": 4.241143317230274e-05,
|
| 253 |
+
"loss": 1.04,
|
| 254 |
+
"step": 14000
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.82,
|
| 258 |
+
"grad_norm": 1.0176922082901,
|
| 259 |
+
"learning_rate": 4.218138946399816e-05,
|
| 260 |
+
"loss": 1.0271,
|
| 261 |
+
"step": 14400
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.84,
|
| 265 |
+
"grad_norm": 1.1477534770965576,
|
| 266 |
+
"learning_rate": 4.1951345755693586e-05,
|
| 267 |
+
"loss": 1.0325,
|
| 268 |
+
"step": 14800
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.87,
|
| 272 |
+
"grad_norm": 0.9284051060676575,
|
| 273 |
+
"learning_rate": 4.1721302047389e-05,
|
| 274 |
+
"loss": 1.0331,
|
| 275 |
+
"step": 15200
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.89,
|
| 279 |
+
"grad_norm": 0.9155886173248291,
|
| 280 |
+
"learning_rate": 4.1491258339084425e-05,
|
| 281 |
+
"loss": 1.0081,
|
| 282 |
+
"step": 15600
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.91,
|
| 286 |
+
"grad_norm": 0.9425845146179199,
|
| 287 |
+
"learning_rate": 4.126121463077985e-05,
|
| 288 |
+
"loss": 1.0014,
|
| 289 |
+
"step": 16000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.93,
|
| 293 |
+
"grad_norm": 0.9099463224411011,
|
| 294 |
+
"learning_rate": 4.103117092247527e-05,
|
| 295 |
+
"loss": 1.0117,
|
| 296 |
+
"step": 16400
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.96,
|
| 300 |
+
"grad_norm": 0.941250205039978,
|
| 301 |
+
"learning_rate": 4.0801127214170694e-05,
|
| 302 |
+
"loss": 1.0021,
|
| 303 |
+
"step": 16800
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.98,
|
| 307 |
+
"grad_norm": 0.8845394849777222,
|
| 308 |
+
"learning_rate": 4.0571083505866117e-05,
|
| 309 |
+
"loss": 0.9984,
|
| 310 |
+
"step": 17200
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 1.0,
|
| 314 |
+
"eval_loss": 1.2029205560684204,
|
| 315 |
+
"eval_runtime": 240.0634,
|
| 316 |
+
"eval_samples_per_second": 251.08,
|
| 317 |
+
"eval_steps_per_second": 3.924,
|
| 318 |
+
"step": 17548
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.0,
|
| 322 |
+
"grad_norm": 0.8803815841674805,
|
| 323 |
+
"learning_rate": 4.034103979756153e-05,
|
| 324 |
+
"loss": 0.9937,
|
| 325 |
+
"step": 17600
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.03,
|
| 329 |
+
"grad_norm": 0.9823230504989624,
|
| 330 |
+
"learning_rate": 4.0110996089256956e-05,
|
| 331 |
+
"loss": 0.9468,
|
| 332 |
+
"step": 18000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.05,
|
| 336 |
+
"grad_norm": 0.9398742914199829,
|
| 337 |
+
"learning_rate": 3.9880952380952386e-05,
|
| 338 |
+
"loss": 0.9499,
|
| 339 |
+
"step": 18400
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.07,
|
| 343 |
+
"grad_norm": 1.0421141386032104,
|
| 344 |
+
"learning_rate": 3.965090867264781e-05,
|
| 345 |
+
"loss": 0.9471,
|
| 346 |
+
"step": 18800
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.09,
|
| 350 |
+
"grad_norm": 1.0032306909561157,
|
| 351 |
+
"learning_rate": 3.9422015182884744e-05,
|
| 352 |
+
"loss": 0.9375,
|
| 353 |
+
"step": 19200
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.12,
|
| 357 |
+
"grad_norm": 1.4174487590789795,
|
| 358 |
+
"learning_rate": 3.9191971474580174e-05,
|
| 359 |
+
"loss": 0.9417,
|
| 360 |
+
"step": 19600
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.14,
|
| 364 |
+
"grad_norm": 1.037479281425476,
|
| 365 |
+
"learning_rate": 3.89619277662756e-05,
|
| 366 |
+
"loss": 0.925,
|
| 367 |
+
"step": 20000
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.16,
|
| 371 |
+
"grad_norm": 1.0530787706375122,
|
| 372 |
+
"learning_rate": 3.873245916724178e-05,
|
| 373 |
+
"loss": 0.9378,
|
| 374 |
+
"step": 20400
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.19,
|
| 378 |
+
"grad_norm": 1.1429826021194458,
|
| 379 |
+
"learning_rate": 3.85024154589372e-05,
|
| 380 |
+
"loss": 0.9386,
|
| 381 |
+
"step": 20800
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.21,
|
| 385 |
+
"grad_norm": 0.8666102886199951,
|
| 386 |
+
"learning_rate": 3.8272371750632626e-05,
|
| 387 |
+
"loss": 0.9169,
|
| 388 |
+
"step": 21200
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.23,
|
| 392 |
+
"grad_norm": 1.0205742120742798,
|
| 393 |
+
"learning_rate": 3.804232804232805e-05,
|
| 394 |
+
"loss": 0.9247,
|
| 395 |
+
"step": 21600
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.25,
|
| 399 |
+
"grad_norm": 1.046938180923462,
|
| 400 |
+
"learning_rate": 3.7812284334023465e-05,
|
| 401 |
+
"loss": 0.9168,
|
| 402 |
+
"step": 22000
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.28,
|
| 406 |
+
"grad_norm": 1.0534473657608032,
|
| 407 |
+
"learning_rate": 3.758224062571889e-05,
|
| 408 |
+
"loss": 0.9179,
|
| 409 |
+
"step": 22400
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 1.3,
|
| 413 |
+
"grad_norm": 0.9643887877464294,
|
| 414 |
+
"learning_rate": 3.735219691741431e-05,
|
| 415 |
+
"loss": 0.9147,
|
| 416 |
+
"step": 22800
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.32,
|
| 420 |
+
"grad_norm": 1.162414312362671,
|
| 421 |
+
"learning_rate": 3.7122153209109734e-05,
|
| 422 |
+
"loss": 0.9252,
|
| 423 |
+
"step": 23200
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.34,
|
| 427 |
+
"grad_norm": 1.0543121099472046,
|
| 428 |
+
"learning_rate": 3.689210950080516e-05,
|
| 429 |
+
"loss": 0.9002,
|
| 430 |
+
"step": 23600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 1.37,
|
| 434 |
+
"grad_norm": 1.095170021057129,
|
| 435 |
+
"learning_rate": 3.666206579250058e-05,
|
| 436 |
+
"loss": 0.9078,
|
| 437 |
+
"step": 24000
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 1.39,
|
| 441 |
+
"grad_norm": 0.938887357711792,
|
| 442 |
+
"learning_rate": 3.6432022084195996e-05,
|
| 443 |
+
"loss": 0.9074,
|
| 444 |
+
"step": 24400
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.41,
|
| 448 |
+
"grad_norm": 1.1190577745437622,
|
| 449 |
+
"learning_rate": 3.620197837589142e-05,
|
| 450 |
+
"loss": 0.9087,
|
| 451 |
+
"step": 24800
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 1.44,
|
| 455 |
+
"grad_norm": 0.9494956731796265,
|
| 456 |
+
"learning_rate": 3.597193466758684e-05,
|
| 457 |
+
"loss": 0.8903,
|
| 458 |
+
"step": 25200
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 1.46,
|
| 462 |
+
"grad_norm": 1.1163825988769531,
|
| 463 |
+
"learning_rate": 3.5741890959282265e-05,
|
| 464 |
+
"loss": 0.8934,
|
| 465 |
+
"step": 25600
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 1.48,
|
| 469 |
+
"grad_norm": 0.9037898182868958,
|
| 470 |
+
"learning_rate": 3.551184725097769e-05,
|
| 471 |
+
"loss": 0.9022,
|
| 472 |
+
"step": 26000
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 1.5,
|
| 476 |
+
"grad_norm": 1.0662837028503418,
|
| 477 |
+
"learning_rate": 3.528237865194387e-05,
|
| 478 |
+
"loss": 0.8859,
|
| 479 |
+
"step": 26400
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 1.53,
|
| 483 |
+
"grad_norm": 0.8364838361740112,
|
| 484 |
+
"learning_rate": 3.5052334943639294e-05,
|
| 485 |
+
"loss": 0.8829,
|
| 486 |
+
"step": 26800
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 1.55,
|
| 490 |
+
"grad_norm": 1.109578013420105,
|
| 491 |
+
"learning_rate": 3.482229123533472e-05,
|
| 492 |
+
"loss": 0.8784,
|
| 493 |
+
"step": 27200
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 1.57,
|
| 497 |
+
"grad_norm": 0.8675488829612732,
|
| 498 |
+
"learning_rate": 3.459224752703013e-05,
|
| 499 |
+
"loss": 0.8822,
|
| 500 |
+
"step": 27600
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 1.6,
|
| 504 |
+
"grad_norm": 1.089260458946228,
|
| 505 |
+
"learning_rate": 3.4362203818725556e-05,
|
| 506 |
+
"loss": 0.8864,
|
| 507 |
+
"step": 28000
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 1.62,
|
| 511 |
+
"grad_norm": 1.2143120765686035,
|
| 512 |
+
"learning_rate": 3.413216011042098e-05,
|
| 513 |
+
"loss": 0.8872,
|
| 514 |
+
"step": 28400
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 1.64,
|
| 518 |
+
"grad_norm": 0.969947099685669,
|
| 519 |
+
"learning_rate": 3.39021164021164e-05,
|
| 520 |
+
"loss": 0.8888,
|
| 521 |
+
"step": 28800
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 1.66,
|
| 525 |
+
"grad_norm": 0.9949952960014343,
|
| 526 |
+
"learning_rate": 3.3672647803082585e-05,
|
| 527 |
+
"loss": 0.8837,
|
| 528 |
+
"step": 29200
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 1.69,
|
| 532 |
+
"grad_norm": 1.0020053386688232,
|
| 533 |
+
"learning_rate": 3.344260409477801e-05,
|
| 534 |
+
"loss": 0.8832,
|
| 535 |
+
"step": 29600
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 1.71,
|
| 539 |
+
"grad_norm": 0.9824060797691345,
|
| 540 |
+
"learning_rate": 3.321256038647343e-05,
|
| 541 |
+
"loss": 0.8772,
|
| 542 |
+
"step": 30000
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.73,
|
| 546 |
+
"grad_norm": 0.9114558100700378,
|
| 547 |
+
"learning_rate": 3.2982516678168854e-05,
|
| 548 |
+
"loss": 0.8618,
|
| 549 |
+
"step": 30400
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.76,
|
| 553 |
+
"grad_norm": 1.1614044904708862,
|
| 554 |
+
"learning_rate": 3.275247296986428e-05,
|
| 555 |
+
"loss": 0.8735,
|
| 556 |
+
"step": 30800
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 1.78,
|
| 560 |
+
"grad_norm": 1.030380129814148,
|
| 561 |
+
"learning_rate": 3.25224292615597e-05,
|
| 562 |
+
"loss": 0.8657,
|
| 563 |
+
"step": 31200
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 1.8,
|
| 567 |
+
"grad_norm": 0.9505811929702759,
|
| 568 |
+
"learning_rate": 3.229238555325512e-05,
|
| 569 |
+
"loss": 0.8783,
|
| 570 |
+
"step": 31600
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 1.82,
|
| 574 |
+
"grad_norm": 1.2939780950546265,
|
| 575 |
+
"learning_rate": 3.2062916954221306e-05,
|
| 576 |
+
"loss": 0.8635,
|
| 577 |
+
"step": 32000
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 1.85,
|
| 581 |
+
"grad_norm": 1.1592987775802612,
|
| 582 |
+
"learning_rate": 3.183287324591673e-05,
|
| 583 |
+
"loss": 0.8616,
|
| 584 |
+
"step": 32400
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 1.87,
|
| 588 |
+
"grad_norm": 1.104283332824707,
|
| 589 |
+
"learning_rate": 3.160340464688291e-05,
|
| 590 |
+
"loss": 0.8613,
|
| 591 |
+
"step": 32800
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 1.89,
|
| 595 |
+
"grad_norm": 0.9907343983650208,
|
| 596 |
+
"learning_rate": 3.1373360938578334e-05,
|
| 597 |
+
"loss": 0.8623,
|
| 598 |
+
"step": 33200
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 1.91,
|
| 602 |
+
"grad_norm": 0.96369868516922,
|
| 603 |
+
"learning_rate": 3.114331723027376e-05,
|
| 604 |
+
"loss": 0.8467,
|
| 605 |
+
"step": 33600
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 1.94,
|
| 609 |
+
"grad_norm": 1.026637315750122,
|
| 610 |
+
"learning_rate": 3.091327352196918e-05,
|
| 611 |
+
"loss": 0.8563,
|
| 612 |
+
"step": 34000
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 1.96,
|
| 616 |
+
"grad_norm": 1.00987708568573,
|
| 617 |
+
"learning_rate": 3.0683229813664596e-05,
|
| 618 |
+
"loss": 0.8489,
|
| 619 |
+
"step": 34400
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 1.98,
|
| 623 |
+
"grad_norm": 1.0772453546524048,
|
| 624 |
+
"learning_rate": 3.0453761214630783e-05,
|
| 625 |
+
"loss": 0.8597,
|
| 626 |
+
"step": 34800
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 2.0,
|
| 630 |
+
"eval_loss": 1.1818490028381348,
|
| 631 |
+
"eval_runtime": 240.2583,
|
| 632 |
+
"eval_samples_per_second": 250.876,
|
| 633 |
+
"eval_steps_per_second": 3.921,
|
| 634 |
+
"step": 35096
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 2.01,
|
| 638 |
+
"grad_norm": 1.0454893112182617,
|
| 639 |
+
"learning_rate": 3.0223717506326206e-05,
|
| 640 |
+
"loss": 0.8354,
|
| 641 |
+
"step": 35200
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 2.03,
|
| 645 |
+
"grad_norm": 1.0722404718399048,
|
| 646 |
+
"learning_rate": 2.9993673798021625e-05,
|
| 647 |
+
"loss": 0.8027,
|
| 648 |
+
"step": 35600
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 2.05,
|
| 652 |
+
"grad_norm": 1.144631266593933,
|
| 653 |
+
"learning_rate": 2.9763630089717048e-05,
|
| 654 |
+
"loss": 0.7981,
|
| 655 |
+
"step": 36000
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 2.07,
|
| 659 |
+
"grad_norm": 1.0820245742797852,
|
| 660 |
+
"learning_rate": 2.953358638141247e-05,
|
| 661 |
+
"loss": 0.7989,
|
| 662 |
+
"step": 36400
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 2.1,
|
| 666 |
+
"grad_norm": 1.1228435039520264,
|
| 667 |
+
"learning_rate": 2.930354267310789e-05,
|
| 668 |
+
"loss": 0.7908,
|
| 669 |
+
"step": 36800
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 2.12,
|
| 673 |
+
"grad_norm": 1.0229965448379517,
|
| 674 |
+
"learning_rate": 2.9073498964803314e-05,
|
| 675 |
+
"loss": 0.792,
|
| 676 |
+
"step": 37200
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 2.14,
|
| 680 |
+
"grad_norm": 1.0550298690795898,
|
| 681 |
+
"learning_rate": 2.8843455256498737e-05,
|
| 682 |
+
"loss": 0.8046,
|
| 683 |
+
"step": 37600
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 2.17,
|
| 687 |
+
"grad_norm": 1.1629900932312012,
|
| 688 |
+
"learning_rate": 2.8613411548194156e-05,
|
| 689 |
+
"loss": 0.8019,
|
| 690 |
+
"step": 38000
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 2.19,
|
| 694 |
+
"grad_norm": 1.0293209552764893,
|
| 695 |
+
"learning_rate": 2.838336783988958e-05,
|
| 696 |
+
"loss": 0.7967,
|
| 697 |
+
"step": 38400
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 2.21,
|
| 701 |
+
"grad_norm": 1.0259321928024292,
|
| 702 |
+
"learning_rate": 2.8153324131585002e-05,
|
| 703 |
+
"loss": 0.8035,
|
| 704 |
+
"step": 38800
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 2.23,
|
| 708 |
+
"grad_norm": 1.2149560451507568,
|
| 709 |
+
"learning_rate": 2.7923280423280422e-05,
|
| 710 |
+
"loss": 0.7979,
|
| 711 |
+
"step": 39200
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 2.26,
|
| 715 |
+
"grad_norm": 1.082778811454773,
|
| 716 |
+
"learning_rate": 2.7693236714975845e-05,
|
| 717 |
+
"loss": 0.7913,
|
| 718 |
+
"step": 39600
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 2.28,
|
| 722 |
+
"grad_norm": 0.9237461686134338,
|
| 723 |
+
"learning_rate": 2.7463768115942028e-05,
|
| 724 |
+
"loss": 0.7868,
|
| 725 |
+
"step": 40000
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 2.3,
|
| 729 |
+
"grad_norm": 0.9772054553031921,
|
| 730 |
+
"learning_rate": 2.723372440763745e-05,
|
| 731 |
+
"loss": 0.8001,
|
| 732 |
+
"step": 40400
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 2.33,
|
| 736 |
+
"grad_norm": 0.9553162455558777,
|
| 737 |
+
"learning_rate": 2.7003680699332874e-05,
|
| 738 |
+
"loss": 0.7898,
|
| 739 |
+
"step": 40800
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 2.35,
|
| 743 |
+
"grad_norm": 0.9960311055183411,
|
| 744 |
+
"learning_rate": 2.6773636991028293e-05,
|
| 745 |
+
"loss": 0.7935,
|
| 746 |
+
"step": 41200
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 2.37,
|
| 750 |
+
"grad_norm": 1.439426302909851,
|
| 751 |
+
"learning_rate": 2.654474350126524e-05,
|
| 752 |
+
"loss": 0.7809,
|
| 753 |
+
"step": 41600
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 2.39,
|
| 757 |
+
"grad_norm": 0.9864240288734436,
|
| 758 |
+
"learning_rate": 2.6315850011502185e-05,
|
| 759 |
+
"loss": 0.7857,
|
| 760 |
+
"step": 42000
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 2.42,
|
| 764 |
+
"grad_norm": 1.095920205116272,
|
| 765 |
+
"learning_rate": 2.608580630319761e-05,
|
| 766 |
+
"loss": 0.791,
|
| 767 |
+
"step": 42400
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 2.44,
|
| 771 |
+
"grad_norm": 0.9908497929573059,
|
| 772 |
+
"learning_rate": 2.5855762594893028e-05,
|
| 773 |
+
"loss": 0.7902,
|
| 774 |
+
"step": 42800
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 2.46,
|
| 778 |
+
"grad_norm": 1.1290860176086426,
|
| 779 |
+
"learning_rate": 2.562571888658845e-05,
|
| 780 |
+
"loss": 0.779,
|
| 781 |
+
"step": 43200
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 2.48,
|
| 785 |
+
"grad_norm": 1.1015818119049072,
|
| 786 |
+
"learning_rate": 2.5395675178283874e-05,
|
| 787 |
+
"loss": 0.794,
|
| 788 |
+
"step": 43600
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 2.51,
|
| 792 |
+
"grad_norm": 0.9187479615211487,
|
| 793 |
+
"learning_rate": 2.5165631469979294e-05,
|
| 794 |
+
"loss": 0.791,
|
| 795 |
+
"step": 44000
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 2.53,
|
| 799 |
+
"grad_norm": 1.1698708534240723,
|
| 800 |
+
"learning_rate": 2.493558776167472e-05,
|
| 801 |
+
"loss": 0.788,
|
| 802 |
+
"step": 44400
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 2.55,
|
| 806 |
+
"grad_norm": 0.9840272068977356,
|
| 807 |
+
"learning_rate": 2.4705544053370143e-05,
|
| 808 |
+
"loss": 0.7855,
|
| 809 |
+
"step": 44800
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 2.58,
|
| 813 |
+
"grad_norm": 1.0132951736450195,
|
| 814 |
+
"learning_rate": 2.4475500345065562e-05,
|
| 815 |
+
"loss": 0.7701,
|
| 816 |
+
"step": 45200
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 2.6,
|
| 820 |
+
"grad_norm": 1.0666146278381348,
|
| 821 |
+
"learning_rate": 2.4245456636760985e-05,
|
| 822 |
+
"loss": 0.7799,
|
| 823 |
+
"step": 45600
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 2.62,
|
| 827 |
+
"grad_norm": 1.20628023147583,
|
| 828 |
+
"learning_rate": 2.401541292845641e-05,
|
| 829 |
+
"loss": 0.7764,
|
| 830 |
+
"step": 46000
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 2.64,
|
| 834 |
+
"grad_norm": 0.8813405632972717,
|
| 835 |
+
"learning_rate": 2.3785369220151828e-05,
|
| 836 |
+
"loss": 0.7746,
|
| 837 |
+
"step": 46400
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 2.67,
|
| 841 |
+
"grad_norm": 1.2259961366653442,
|
| 842 |
+
"learning_rate": 2.355532551184725e-05,
|
| 843 |
+
"loss": 0.7734,
|
| 844 |
+
"step": 46800
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 2.69,
|
| 848 |
+
"grad_norm": 0.9694178700447083,
|
| 849 |
+
"learning_rate": 2.3325281803542674e-05,
|
| 850 |
+
"loss": 0.7784,
|
| 851 |
+
"step": 47200
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 2.71,
|
| 855 |
+
"grad_norm": 1.1529206037521362,
|
| 856 |
+
"learning_rate": 2.3095238095238097e-05,
|
| 857 |
+
"loss": 0.7711,
|
| 858 |
+
"step": 47600
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 2.74,
|
| 862 |
+
"grad_norm": 0.9313886165618896,
|
| 863 |
+
"learning_rate": 2.286519438693352e-05,
|
| 864 |
+
"loss": 0.7797,
|
| 865 |
+
"step": 48000
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 2.76,
|
| 869 |
+
"grad_norm": 1.1616884469985962,
|
| 870 |
+
"learning_rate": 2.2635150678628943e-05,
|
| 871 |
+
"loss": 0.7694,
|
| 872 |
+
"step": 48400
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 2.78,
|
| 876 |
+
"grad_norm": 1.096562147140503,
|
| 877 |
+
"learning_rate": 2.2405106970324362e-05,
|
| 878 |
+
"loss": 0.7627,
|
| 879 |
+
"step": 48800
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 2.8,
|
| 883 |
+
"grad_norm": 0.990018904209137,
|
| 884 |
+
"learning_rate": 2.2175063262019785e-05,
|
| 885 |
+
"loss": 0.7674,
|
| 886 |
+
"step": 49200
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 2.83,
|
| 890 |
+
"grad_norm": 1.0565036535263062,
|
| 891 |
+
"learning_rate": 2.194501955371521e-05,
|
| 892 |
+
"loss": 0.7606,
|
| 893 |
+
"step": 49600
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 2.85,
|
| 897 |
+
"grad_norm": 1.1336785554885864,
|
| 898 |
+
"learning_rate": 2.1714975845410628e-05,
|
| 899 |
+
"loss": 0.759,
|
| 900 |
+
"step": 50000
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 2.87,
|
| 904 |
+
"grad_norm": 1.0509406328201294,
|
| 905 |
+
"learning_rate": 2.148493213710605e-05,
|
| 906 |
+
"loss": 0.777,
|
| 907 |
+
"step": 50400
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 2.89,
|
| 911 |
+
"grad_norm": 1.0363755226135254,
|
| 912 |
+
"learning_rate": 2.1254888428801474e-05,
|
| 913 |
+
"loss": 0.7692,
|
| 914 |
+
"step": 50800
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 2.92,
|
| 918 |
+
"grad_norm": 1.204566240310669,
|
| 919 |
+
"learning_rate": 2.1024844720496894e-05,
|
| 920 |
+
"loss": 0.7695,
|
| 921 |
+
"step": 51200
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 2.94,
|
| 925 |
+
"grad_norm": 1.0907776355743408,
|
| 926 |
+
"learning_rate": 2.0794801012192317e-05,
|
| 927 |
+
"loss": 0.7718,
|
| 928 |
+
"step": 51600
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 2.96,
|
| 932 |
+
"grad_norm": 1.0475547313690186,
|
| 933 |
+
"learning_rate": 2.05653324131585e-05,
|
| 934 |
+
"loss": 0.7688,
|
| 935 |
+
"step": 52000
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 2.99,
|
| 939 |
+
"grad_norm": 1.2301433086395264,
|
| 940 |
+
"learning_rate": 2.0335288704853922e-05,
|
| 941 |
+
"loss": 0.7614,
|
| 942 |
+
"step": 52400
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 3.0,
|
| 946 |
+
"eval_loss": 1.177372694015503,
|
| 947 |
+
"eval_runtime": 239.9376,
|
| 948 |
+
"eval_samples_per_second": 251.211,
|
| 949 |
+
"eval_steps_per_second": 3.926,
|
| 950 |
+
"step": 52644
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 3.01,
|
| 954 |
+
"grad_norm": 1.330518364906311,
|
| 955 |
+
"learning_rate": 2.0105820105820105e-05,
|
| 956 |
+
"loss": 0.7462,
|
| 957 |
+
"step": 52800
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 3.03,
|
| 961 |
+
"grad_norm": 1.0969927310943604,
|
| 962 |
+
"learning_rate": 1.9875776397515528e-05,
|
| 963 |
+
"loss": 0.7272,
|
| 964 |
+
"step": 53200
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 3.05,
|
| 968 |
+
"grad_norm": 1.0054128170013428,
|
| 969 |
+
"learning_rate": 1.964573268921095e-05,
|
| 970 |
+
"loss": 0.7164,
|
| 971 |
+
"step": 53600
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 3.08,
|
| 975 |
+
"grad_norm": 0.9748343229293823,
|
| 976 |
+
"learning_rate": 1.9416264090177137e-05,
|
| 977 |
+
"loss": 0.7222,
|
| 978 |
+
"step": 54000
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 3.1,
|
| 982 |
+
"grad_norm": 0.9346728920936584,
|
| 983 |
+
"learning_rate": 1.9186220381872557e-05,
|
| 984 |
+
"loss": 0.7262,
|
| 985 |
+
"step": 54400
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 3.12,
|
| 989 |
+
"grad_norm": 1.0906745195388794,
|
| 990 |
+
"learning_rate": 1.895617667356798e-05,
|
| 991 |
+
"loss": 0.7224,
|
| 992 |
+
"step": 54800
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 3.15,
|
| 996 |
+
"grad_norm": 1.0387004613876343,
|
| 997 |
+
"learning_rate": 1.8726132965263403e-05,
|
| 998 |
+
"loss": 0.7319,
|
| 999 |
+
"step": 55200
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 3.17,
|
| 1003 |
+
"grad_norm": 1.0948784351348877,
|
| 1004 |
+
"learning_rate": 1.8496089256958822e-05,
|
| 1005 |
+
"loss": 0.7118,
|
| 1006 |
+
"step": 55600
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 3.19,
|
| 1010 |
+
"grad_norm": 0.9571365118026733,
|
| 1011 |
+
"learning_rate": 1.8266045548654245e-05,
|
| 1012 |
+
"loss": 0.7127,
|
| 1013 |
+
"step": 56000
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 3.21,
|
| 1017 |
+
"grad_norm": 1.0155813694000244,
|
| 1018 |
+
"learning_rate": 1.803600184034967e-05,
|
| 1019 |
+
"loss": 0.7185,
|
| 1020 |
+
"step": 56400
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 3.24,
|
| 1024 |
+
"grad_norm": 1.0468432903289795,
|
| 1025 |
+
"learning_rate": 1.780653324131585e-05,
|
| 1026 |
+
"loss": 0.7223,
|
| 1027 |
+
"step": 56800
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 3.26,
|
| 1031 |
+
"grad_norm": 1.0302894115447998,
|
| 1032 |
+
"learning_rate": 1.7576489533011274e-05,
|
| 1033 |
+
"loss": 0.7179,
|
| 1034 |
+
"step": 57200
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 3.28,
|
| 1038 |
+
"grad_norm": 1.0335286855697632,
|
| 1039 |
+
"learning_rate": 1.7346445824706694e-05,
|
| 1040 |
+
"loss": 0.7144,
|
| 1041 |
+
"step": 57600
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 3.31,
|
| 1045 |
+
"grad_norm": 1.2124297618865967,
|
| 1046 |
+
"learning_rate": 1.7116402116402117e-05,
|
| 1047 |
+
"loss": 0.7152,
|
| 1048 |
+
"step": 58000
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 3.33,
|
| 1052 |
+
"grad_norm": 1.2023513317108154,
|
| 1053 |
+
"learning_rate": 1.688635840809754e-05,
|
| 1054 |
+
"loss": 0.7184,
|
| 1055 |
+
"step": 58400
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 3.35,
|
| 1059 |
+
"grad_norm": 1.0327694416046143,
|
| 1060 |
+
"learning_rate": 1.665631469979296e-05,
|
| 1061 |
+
"loss": 0.7128,
|
| 1062 |
+
"step": 58800
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 3.37,
|
| 1066 |
+
"grad_norm": 0.9827908873558044,
|
| 1067 |
+
"learning_rate": 1.6426270991488382e-05,
|
| 1068 |
+
"loss": 0.7162,
|
| 1069 |
+
"step": 59200
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"epoch": 3.4,
|
| 1073 |
+
"grad_norm": 1.189833641052246,
|
| 1074 |
+
"learning_rate": 1.6196227283183805e-05,
|
| 1075 |
+
"loss": 0.7235,
|
| 1076 |
+
"step": 59600
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 3.42,
|
| 1080 |
+
"grad_norm": 1.0521001815795898,
|
| 1081 |
+
"learning_rate": 1.5966183574879228e-05,
|
| 1082 |
+
"loss": 0.7045,
|
| 1083 |
+
"step": 60000
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 3.44,
|
| 1087 |
+
"grad_norm": 1.0198911428451538,
|
| 1088 |
+
"learning_rate": 1.573613986657465e-05,
|
| 1089 |
+
"loss": 0.711,
|
| 1090 |
+
"step": 60400
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 3.46,
|
| 1094 |
+
"grad_norm": 1.1130218505859375,
|
| 1095 |
+
"learning_rate": 1.5506096158270074e-05,
|
| 1096 |
+
"loss": 0.7158,
|
| 1097 |
+
"step": 60800
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 3.49,
|
| 1101 |
+
"grad_norm": 0.8894535303115845,
|
| 1102 |
+
"learning_rate": 1.5276052449965494e-05,
|
| 1103 |
+
"loss": 0.7202,
|
| 1104 |
+
"step": 61200
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 3.51,
|
| 1108 |
+
"grad_norm": 1.1167349815368652,
|
| 1109 |
+
"learning_rate": 1.5046008741660917e-05,
|
| 1110 |
+
"loss": 0.7139,
|
| 1111 |
+
"step": 61600
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 3.53,
|
| 1115 |
+
"grad_norm": 1.2239285707473755,
|
| 1116 |
+
"learning_rate": 1.4815965033356338e-05,
|
| 1117 |
+
"loss": 0.7163,
|
| 1118 |
+
"step": 62000
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 3.56,
|
| 1122 |
+
"grad_norm": 1.1781020164489746,
|
| 1123 |
+
"learning_rate": 1.458592132505176e-05,
|
| 1124 |
+
"loss": 0.7193,
|
| 1125 |
+
"step": 62400
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 3.58,
|
| 1129 |
+
"grad_norm": 1.0834379196166992,
|
| 1130 |
+
"learning_rate": 1.4356452726017944e-05,
|
| 1131 |
+
"loss": 0.7147,
|
| 1132 |
+
"step": 62800
|
| 1133 |
+
},
|
| 1134 |
+
{
|
| 1135 |
+
"epoch": 3.6,
|
| 1136 |
+
"grad_norm": 0.991364061832428,
|
| 1137 |
+
"learning_rate": 1.4126409017713365e-05,
|
| 1138 |
+
"loss": 0.7022,
|
| 1139 |
+
"step": 63200
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 3.62,
|
| 1143 |
+
"grad_norm": 1.1527454853057861,
|
| 1144 |
+
"learning_rate": 1.3896365309408788e-05,
|
| 1145 |
+
"loss": 0.7171,
|
| 1146 |
+
"step": 63600
|
| 1147 |
+
},
|
| 1148 |
+
{
|
| 1149 |
+
"epoch": 3.65,
|
| 1150 |
+
"grad_norm": 0.9275304675102234,
|
| 1151 |
+
"learning_rate": 1.366632160110421e-05,
|
| 1152 |
+
"loss": 0.7083,
|
| 1153 |
+
"step": 64000
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 3.67,
|
| 1157 |
+
"grad_norm": 1.2628445625305176,
|
| 1158 |
+
"learning_rate": 1.3436853002070392e-05,
|
| 1159 |
+
"loss": 0.7069,
|
| 1160 |
+
"step": 64400
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 3.69,
|
| 1164 |
+
"grad_norm": 1.100540280342102,
|
| 1165 |
+
"learning_rate": 1.3206809293765815e-05,
|
| 1166 |
+
"loss": 0.7062,
|
| 1167 |
+
"step": 64800
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 3.72,
|
| 1171 |
+
"grad_norm": 1.1310228109359741,
|
| 1172 |
+
"learning_rate": 1.2976765585461237e-05,
|
| 1173 |
+
"loss": 0.6933,
|
| 1174 |
+
"step": 65200
|
| 1175 |
+
},
|
| 1176 |
+
{
|
| 1177 |
+
"epoch": 3.74,
|
| 1178 |
+
"grad_norm": 1.2022795677185059,
|
| 1179 |
+
"learning_rate": 1.2746721877156661e-05,
|
| 1180 |
+
"loss": 0.7096,
|
| 1181 |
+
"step": 65600
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"epoch": 3.76,
|
| 1185 |
+
"grad_norm": 2.5451197624206543,
|
| 1186 |
+
"learning_rate": 1.2516678168852084e-05,
|
| 1187 |
+
"loss": 0.7087,
|
| 1188 |
+
"step": 66000
|
| 1189 |
+
},
|
| 1190 |
+
{
|
| 1191 |
+
"epoch": 3.78,
|
| 1192 |
+
"grad_norm": 1.125082015991211,
|
| 1193 |
+
"learning_rate": 1.2286634460547504e-05,
|
| 1194 |
+
"loss": 0.6946,
|
| 1195 |
+
"step": 66400
|
| 1196 |
+
},
|
| 1197 |
+
{
|
| 1198 |
+
"epoch": 3.81,
|
| 1199 |
+
"grad_norm": 1.3816386461257935,
|
| 1200 |
+
"learning_rate": 1.2056590752242927e-05,
|
| 1201 |
+
"loss": 0.7091,
|
| 1202 |
+
"step": 66800
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 3.83,
|
| 1206 |
+
"grad_norm": 1.197379231452942,
|
| 1207 |
+
"learning_rate": 1.182654704393835e-05,
|
| 1208 |
+
"loss": 0.7195,
|
| 1209 |
+
"step": 67200
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 3.85,
|
| 1213 |
+
"grad_norm": 1.1002612113952637,
|
| 1214 |
+
"learning_rate": 1.1597078444904533e-05,
|
| 1215 |
+
"loss": 0.7021,
|
| 1216 |
+
"step": 67600
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 3.88,
|
| 1220 |
+
"grad_norm": 1.1073203086853027,
|
| 1221 |
+
"learning_rate": 1.1367034736599954e-05,
|
| 1222 |
+
"loss": 0.7033,
|
| 1223 |
+
"step": 68000
|
| 1224 |
+
},
|
| 1225 |
+
{
|
| 1226 |
+
"epoch": 3.9,
|
| 1227 |
+
"grad_norm": 1.056327223777771,
|
| 1228 |
+
"learning_rate": 1.1136991028295377e-05,
|
| 1229 |
+
"loss": 0.7127,
|
| 1230 |
+
"step": 68400
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"epoch": 3.92,
|
| 1234 |
+
"grad_norm": 0.9232028722763062,
|
| 1235 |
+
"learning_rate": 1.090752242926156e-05,
|
| 1236 |
+
"loss": 0.7055,
|
| 1237 |
+
"step": 68800
|
| 1238 |
+
},
|
| 1239 |
+
{
|
| 1240 |
+
"epoch": 3.94,
|
| 1241 |
+
"grad_norm": 1.2083985805511475,
|
| 1242 |
+
"learning_rate": 1.0678053830227744e-05,
|
| 1243 |
+
"loss": 0.702,
|
| 1244 |
+
"step": 69200
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"epoch": 3.97,
|
| 1248 |
+
"grad_norm": 1.1936637163162231,
|
| 1249 |
+
"learning_rate": 1.0448010121923165e-05,
|
| 1250 |
+
"loss": 0.6961,
|
| 1251 |
+
"step": 69600
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"epoch": 3.99,
|
| 1255 |
+
"grad_norm": 1.1006180047988892,
|
| 1256 |
+
"learning_rate": 1.0217966413618588e-05,
|
| 1257 |
+
"loss": 0.6965,
|
| 1258 |
+
"step": 70000
|
| 1259 |
+
},
|
| 1260 |
+
{
|
| 1261 |
+
"epoch": 4.0,
|
| 1262 |
+
"eval_loss": 1.182397723197937,
|
| 1263 |
+
"eval_runtime": 239.7583,
|
| 1264 |
+
"eval_samples_per_second": 251.399,
|
| 1265 |
+
"eval_steps_per_second": 3.929,
|
| 1266 |
+
"step": 70192
|
| 1267 |
+
},
|
| 1268 |
+
{
|
| 1269 |
+
"epoch": 4.01,
|
| 1270 |
+
"grad_norm": 0.9501661658287048,
|
| 1271 |
+
"learning_rate": 9.988497814584771e-06,
|
| 1272 |
+
"loss": 0.681,
|
| 1273 |
+
"step": 70400
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 4.03,
|
| 1277 |
+
"grad_norm": 1.1048214435577393,
|
| 1278 |
+
"learning_rate": 9.759029215550954e-06,
|
| 1279 |
+
"loss": 0.6718,
|
| 1280 |
+
"step": 70800
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 4.06,
|
| 1284 |
+
"grad_norm": 1.1402852535247803,
|
| 1285 |
+
"learning_rate": 9.528985507246377e-06,
|
| 1286 |
+
"loss": 0.6646,
|
| 1287 |
+
"step": 71200
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 4.08,
|
| 1291 |
+
"grad_norm": 1.1181689500808716,
|
| 1292 |
+
"learning_rate": 9.298941798941798e-06,
|
| 1293 |
+
"loss": 0.6803,
|
| 1294 |
+
"step": 71600
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 4.1,
|
| 1298 |
+
"grad_norm": 1.0445771217346191,
|
| 1299 |
+
"learning_rate": 9.068898090637221e-06,
|
| 1300 |
+
"loss": 0.6681,
|
| 1301 |
+
"step": 72000
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 4.13,
|
| 1305 |
+
"grad_norm": 1.2578911781311035,
|
| 1306 |
+
"learning_rate": 8.838854382332644e-06,
|
| 1307 |
+
"loss": 0.6743,
|
| 1308 |
+
"step": 72400
|
| 1309 |
+
},
|
| 1310 |
+
{
|
| 1311 |
+
"epoch": 4.15,
|
| 1312 |
+
"grad_norm": 1.0545594692230225,
|
| 1313 |
+
"learning_rate": 8.608810674028066e-06,
|
| 1314 |
+
"loss": 0.6721,
|
| 1315 |
+
"step": 72800
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 4.17,
|
| 1319 |
+
"grad_norm": 1.0575259923934937,
|
| 1320 |
+
"learning_rate": 8.378766965723487e-06,
|
| 1321 |
+
"loss": 0.6692,
|
| 1322 |
+
"step": 73200
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 4.19,
|
| 1326 |
+
"grad_norm": 1.0617632865905762,
|
| 1327 |
+
"learning_rate": 8.14872325741891e-06,
|
| 1328 |
+
"loss": 0.6634,
|
| 1329 |
+
"step": 73600
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 4.22,
|
| 1333 |
+
"grad_norm": 1.1915861368179321,
|
| 1334 |
+
"learning_rate": 7.918679549114331e-06,
|
| 1335 |
+
"loss": 0.6712,
|
| 1336 |
+
"step": 74000
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 4.24,
|
| 1340 |
+
"grad_norm": 1.16647207736969,
|
| 1341 |
+
"learning_rate": 7.688635840809754e-06,
|
| 1342 |
+
"loss": 0.6663,
|
| 1343 |
+
"step": 74400
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 4.26,
|
| 1347 |
+
"grad_norm": 1.0123538970947266,
|
| 1348 |
+
"learning_rate": 7.458592132505176e-06,
|
| 1349 |
+
"loss": 0.6785,
|
| 1350 |
+
"step": 74800
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 4.29,
|
| 1354 |
+
"grad_norm": 1.0426361560821533,
|
| 1355 |
+
"learning_rate": 7.228548424200598e-06,
|
| 1356 |
+
"loss": 0.6825,
|
| 1357 |
+
"step": 75200
|
| 1358 |
+
},
|
| 1359 |
+
{
|
| 1360 |
+
"epoch": 4.31,
|
| 1361 |
+
"grad_norm": 1.0516417026519775,
|
| 1362 |
+
"learning_rate": 6.9985047158960205e-06,
|
| 1363 |
+
"loss": 0.6662,
|
| 1364 |
+
"step": 75600
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"epoch": 4.33,
|
| 1368 |
+
"grad_norm": 1.0297150611877441,
|
| 1369 |
+
"learning_rate": 6.768461007591443e-06,
|
| 1370 |
+
"loss": 0.667,
|
| 1371 |
+
"step": 76000
|
| 1372 |
+
},
|
| 1373 |
+
{
|
| 1374 |
+
"epoch": 4.35,
|
| 1375 |
+
"grad_norm": 1.1151374578475952,
|
| 1376 |
+
"learning_rate": 6.538417299286864e-06,
|
| 1377 |
+
"loss": 0.6612,
|
| 1378 |
+
"step": 76400
|
| 1379 |
+
},
|
| 1380 |
+
{
|
| 1381 |
+
"epoch": 4.38,
|
| 1382 |
+
"grad_norm": 1.100305438041687,
|
| 1383 |
+
"learning_rate": 6.308373590982288e-06,
|
| 1384 |
+
"loss": 0.6731,
|
| 1385 |
+
"step": 76800
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 4.4,
|
| 1389 |
+
"grad_norm": 1.0381314754486084,
|
| 1390 |
+
"learning_rate": 6.078904991948471e-06,
|
| 1391 |
+
"loss": 0.6816,
|
| 1392 |
+
"step": 77200
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 4.42,
|
| 1396 |
+
"grad_norm": 1.166762351989746,
|
| 1397 |
+
"learning_rate": 5.848861283643893e-06,
|
| 1398 |
+
"loss": 0.6674,
|
| 1399 |
+
"step": 77600
|
| 1400 |
+
},
|
| 1401 |
+
{
|
| 1402 |
+
"epoch": 4.44,
|
| 1403 |
+
"grad_norm": 1.1503565311431885,
|
| 1404 |
+
"learning_rate": 5.619392684610076e-06,
|
| 1405 |
+
"loss": 0.6711,
|
| 1406 |
+
"step": 78000
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 4.47,
|
| 1410 |
+
"grad_norm": 1.2317191362380981,
|
| 1411 |
+
"learning_rate": 5.3893489763054985e-06,
|
| 1412 |
+
"loss": 0.6754,
|
| 1413 |
+
"step": 78400
|
| 1414 |
+
},
|
| 1415 |
+
{
|
| 1416 |
+
"epoch": 4.49,
|
| 1417 |
+
"grad_norm": 1.1069433689117432,
|
| 1418 |
+
"learning_rate": 5.159305268000921e-06,
|
| 1419 |
+
"loss": 0.6657,
|
| 1420 |
+
"step": 78800
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 4.51,
|
| 1424 |
+
"grad_norm": 1.081761121749878,
|
| 1425 |
+
"learning_rate": 4.929261559696342e-06,
|
| 1426 |
+
"loss": 0.6632,
|
| 1427 |
+
"step": 79200
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 4.54,
|
| 1431 |
+
"grad_norm": 1.0425435304641724,
|
| 1432 |
+
"learning_rate": 4.699217851391765e-06,
|
| 1433 |
+
"loss": 0.6607,
|
| 1434 |
+
"step": 79600
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 4.56,
|
| 1438 |
+
"grad_norm": 1.1054120063781738,
|
| 1439 |
+
"learning_rate": 4.469174143087187e-06,
|
| 1440 |
+
"loss": 0.6747,
|
| 1441 |
+
"step": 80000
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 4.58,
|
| 1445 |
+
"grad_norm": 1.1128557920455933,
|
| 1446 |
+
"learning_rate": 4.239130434782608e-06,
|
| 1447 |
+
"loss": 0.6686,
|
| 1448 |
+
"step": 80400
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 4.6,
|
| 1452 |
+
"grad_norm": 1.2730941772460938,
|
| 1453 |
+
"learning_rate": 4.009086726478031e-06,
|
| 1454 |
+
"loss": 0.6733,
|
| 1455 |
+
"step": 80800
|
| 1456 |
+
},
|
| 1457 |
+
{
|
| 1458 |
+
"epoch": 4.63,
|
| 1459 |
+
"grad_norm": 1.291892647743225,
|
| 1460 |
+
"learning_rate": 3.779043018173453e-06,
|
| 1461 |
+
"loss": 0.6631,
|
| 1462 |
+
"step": 81200
|
| 1463 |
+
},
|
| 1464 |
+
{
|
| 1465 |
+
"epoch": 4.65,
|
| 1466 |
+
"grad_norm": 0.9876201152801514,
|
| 1467 |
+
"learning_rate": 3.548999309868875e-06,
|
| 1468 |
+
"loss": 0.6691,
|
| 1469 |
+
"step": 81600
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 4.67,
|
| 1473 |
+
"grad_norm": 1.2430497407913208,
|
| 1474 |
+
"learning_rate": 3.3189556015642977e-06,
|
| 1475 |
+
"loss": 0.6579,
|
| 1476 |
+
"step": 82000
|
| 1477 |
+
},
|
| 1478 |
+
{
|
| 1479 |
+
"epoch": 4.7,
|
| 1480 |
+
"grad_norm": 1.012856364250183,
|
| 1481 |
+
"learning_rate": 3.0889118932597194e-06,
|
| 1482 |
+
"loss": 0.6637,
|
| 1483 |
+
"step": 82400
|
| 1484 |
+
},
|
| 1485 |
+
{
|
| 1486 |
+
"epoch": 4.72,
|
| 1487 |
+
"grad_norm": 1.1547410488128662,
|
| 1488 |
+
"learning_rate": 2.8588681849551415e-06,
|
| 1489 |
+
"loss": 0.6801,
|
| 1490 |
+
"step": 82800
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 4.74,
|
| 1494 |
+
"grad_norm": 1.1335712671279907,
|
| 1495 |
+
"learning_rate": 2.6288244766505636e-06,
|
| 1496 |
+
"loss": 0.6717,
|
| 1497 |
+
"step": 83200
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 4.76,
|
| 1501 |
+
"grad_norm": 1.0243676900863647,
|
| 1502 |
+
"learning_rate": 2.3987807683459858e-06,
|
| 1503 |
+
"loss": 0.6639,
|
| 1504 |
+
"step": 83600
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 4.79,
|
| 1508 |
+
"grad_norm": 1.0435352325439453,
|
| 1509 |
+
"learning_rate": 2.1693121693121695e-06,
|
| 1510 |
+
"loss": 0.677,
|
| 1511 |
+
"step": 84000
|
| 1512 |
+
},
|
| 1513 |
+
{
|
| 1514 |
+
"epoch": 4.81,
|
| 1515 |
+
"grad_norm": 1.1295571327209473,
|
| 1516 |
+
"learning_rate": 1.9398435702783527e-06,
|
| 1517 |
+
"loss": 0.6614,
|
| 1518 |
+
"step": 84400
|
| 1519 |
+
},
|
| 1520 |
+
{
|
| 1521 |
+
"epoch": 4.83,
|
| 1522 |
+
"grad_norm": 1.0412089824676514,
|
| 1523 |
+
"learning_rate": 1.7103749712445366e-06,
|
| 1524 |
+
"loss": 0.6613,
|
| 1525 |
+
"step": 84800
|
| 1526 |
+
},
|
| 1527 |
+
{
|
| 1528 |
+
"epoch": 4.86,
|
| 1529 |
+
"grad_norm": 1.0020956993103027,
|
| 1530 |
+
"learning_rate": 1.4803312629399585e-06,
|
| 1531 |
+
"loss": 0.6729,
|
| 1532 |
+
"step": 85200
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 4.88,
|
| 1536 |
+
"grad_norm": 1.2422608137130737,
|
| 1537 |
+
"learning_rate": 1.2502875546353809e-06,
|
| 1538 |
+
"loss": 0.6599,
|
| 1539 |
+
"step": 85600
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 4.9,
|
| 1543 |
+
"grad_norm": 1.1637760400772095,
|
| 1544 |
+
"learning_rate": 1.0202438463308028e-06,
|
| 1545 |
+
"loss": 0.664,
|
| 1546 |
+
"step": 86000
|
| 1547 |
+
},
|
| 1548 |
+
{
|
| 1549 |
+
"epoch": 4.92,
|
| 1550 |
+
"grad_norm": 1.0108635425567627,
|
| 1551 |
+
"learning_rate": 7.902001380262249e-07,
|
| 1552 |
+
"loss": 0.6627,
|
| 1553 |
+
"step": 86400
|
| 1554 |
+
},
|
| 1555 |
+
{
|
| 1556 |
+
"epoch": 4.95,
|
| 1557 |
+
"grad_norm": 1.2056249380111694,
|
| 1558 |
+
"learning_rate": 5.601564297216472e-07,
|
| 1559 |
+
"loss": 0.67,
|
| 1560 |
+
"step": 86800
|
| 1561 |
+
},
|
| 1562 |
+
{
|
| 1563 |
+
"epoch": 4.97,
|
| 1564 |
+
"grad_norm": 3.1152091026306152,
|
| 1565 |
+
"learning_rate": 3.301127214170693e-07,
|
| 1566 |
+
"loss": 0.6619,
|
| 1567 |
+
"step": 87200
|
| 1568 |
+
},
|
| 1569 |
+
{
|
| 1570 |
+
"epoch": 4.99,
|
| 1571 |
+
"grad_norm": 1.2400481700897217,
|
| 1572 |
+
"learning_rate": 1.0006901311249138e-07,
|
| 1573 |
+
"loss": 0.6636,
|
| 1574 |
+
"step": 87600
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 5.0,
|
| 1578 |
+
"eval_loss": 1.1956850290298462,
|
| 1579 |
+
"eval_runtime": 239.8036,
|
| 1580 |
+
"eval_samples_per_second": 251.351,
|
| 1581 |
+
"eval_steps_per_second": 3.928,
|
| 1582 |
+
"step": 87740
|
| 1583 |
+
},
|
| 1584 |
+
{
|
| 1585 |
+
"epoch": 5.0,
|
| 1586 |
+
"step": 87740,
|
| 1587 |
+
"total_flos": 1.7097588901675008e+18,
|
| 1588 |
+
"train_loss": 0.8546056313098415,
|
| 1589 |
+
"train_runtime": 31454.7003,
|
| 1590 |
+
"train_samples_per_second": 89.265,
|
| 1591 |
+
"train_steps_per_second": 2.789
|
| 1592 |
+
}
|
| 1593 |
+
],
|
| 1594 |
+
"logging_steps": 400,
|
| 1595 |
+
"max_steps": 87740,
|
| 1596 |
+
"num_input_tokens_seen": 0,
|
| 1597 |
+
"num_train_epochs": 5,
|
| 1598 |
+
"save_steps": 500,
|
| 1599 |
+
"total_flos": 1.7097588901675008e+18,
|
| 1600 |
+
"train_batch_size": 32,
|
| 1601 |
+
"trial_name": null,
|
| 1602 |
+
"trial_params": null
|
| 1603 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:863f5047ffb666088f1d950674e23c896cce9636765fe3e2c40b67ff0d9a2b52
|
| 3 |
+
size 5176
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|