Upload utg4java model pretrained on codet5p-220m using batch_size 64
Browse files- README.md +50 -3
- all_results.json +11 -0
- codet5p_modelconfigs.txt +33 -0
- config.json +33 -0
- generation_config.json +8 -0
- merges.txt +0 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +753 -0
- test_results.json +6 -0
- tokenizer.json +0 -0
- tokenizer_config.json +959 -0
- train_results.json +7 -0
- trainer_state.json +1666 -0
- training_args.bin +3 -0
- utg4java_pretrain_bleu_metrics.txt +6 -0
- utg4java_pretrain_log_history.txt +230 -0
- utg4java_pretrain_results.txt +60 -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: Salesforce/codet5p-220m
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
model-index:
|
| 7 |
+
- name: pretrain_utg4java_02
|
| 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 |
+
# pretrain_utg4java_02
|
| 15 |
+
|
| 16 |
+
This model is a fine-tuned version of [Salesforce/codet5p-220m](https://huggingface.co/Salesforce/codet5p-220m) 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: 0.0002
|
| 36 |
+
- train_batch_size: 64
|
| 37 |
+
- eval_batch_size: 128
|
| 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: 1000
|
| 42 |
+
- num_epochs: 40
|
| 43 |
+
- mixed_precision_training: Native AMP
|
| 44 |
+
|
| 45 |
+
### Framework versions
|
| 46 |
+
|
| 47 |
+
- Transformers 4.38.2
|
| 48 |
+
- Pytorch 2.1.0
|
| 49 |
+
- Datasets 2.19.0
|
| 50 |
+
- Tokenizers 0.15.2
|
all_results.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"test_loss": 0.20962746441364288,
|
| 4 |
+
"test_runtime": 564.3642,
|
| 5 |
+
"test_samples_per_second": 33.338,
|
| 6 |
+
"test_steps_per_second": 0.26,
|
| 7 |
+
"train_loss": 0.2880888063569923,
|
| 8 |
+
"train_runtime": 54838.0441,
|
| 9 |
+
"train_samples_per_second": 109.795,
|
| 10 |
+
"train_steps_per_second": 1.715
|
| 11 |
+
}
|
codet5p_modelconfigs.txt
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
T5Config {
|
| 2 |
+
"_name_or_path": "Salesforce/codet5p-220m",
|
| 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": "float16",
|
| 30 |
+
"transformers_version": "4.38.2",
|
| 31 |
+
"use_cache": true,
|
| 32 |
+
"vocab_size": 32100
|
| 33 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "Salesforce/codet5p-220m",
|
| 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": 32100
|
| 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 sha256:8b08fd21737ea4cc2a217ba6f82e36b4588941558457fd45526d192185e3dc6a
|
| 3 |
+
size 891617358
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,753 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
{
|
| 4 |
+
"content": "<extra_id_99>",
|
| 5 |
+
"lstrip": true,
|
| 6 |
+
"normalized": true,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false
|
| 9 |
+
},
|
| 10 |
+
{
|
| 11 |
+
"content": "<extra_id_98>",
|
| 12 |
+
"lstrip": true,
|
| 13 |
+
"normalized": true,
|
| 14 |
+
"rstrip": false,
|
| 15 |
+
"single_word": false
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"content": "<extra_id_97>",
|
| 19 |
+
"lstrip": true,
|
| 20 |
+
"normalized": true,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"content": "<extra_id_96>",
|
| 26 |
+
"lstrip": true,
|
| 27 |
+
"normalized": true,
|
| 28 |
+
"rstrip": false,
|
| 29 |
+
"single_word": false
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"content": "<extra_id_95>",
|
| 33 |
+
"lstrip": true,
|
| 34 |
+
"normalized": true,
|
| 35 |
+
"rstrip": false,
|
| 36 |
+
"single_word": false
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"content": "<extra_id_94>",
|
| 40 |
+
"lstrip": true,
|
| 41 |
+
"normalized": true,
|
| 42 |
+
"rstrip": false,
|
| 43 |
+
"single_word": false
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"content": "<extra_id_93>",
|
| 47 |
+
"lstrip": true,
|
| 48 |
+
"normalized": true,
|
| 49 |
+
"rstrip": false,
|
| 50 |
+
"single_word": false
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"content": "<extra_id_92>",
|
| 54 |
+
"lstrip": true,
|
| 55 |
+
"normalized": true,
|
| 56 |
+
"rstrip": false,
|
| 57 |
+
"single_word": false
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"content": "<extra_id_91>",
|
| 61 |
+
"lstrip": true,
|
| 62 |
+
"normalized": true,
|
| 63 |
+
"rstrip": false,
|
| 64 |
+
"single_word": false
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"content": "<extra_id_90>",
|
| 68 |
+
"lstrip": true,
|
| 69 |
+
"normalized": true,
|
| 70 |
+
"rstrip": false,
|
| 71 |
+
"single_word": false
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"content": "<extra_id_89>",
|
| 75 |
+
"lstrip": true,
|
| 76 |
+
"normalized": true,
|
| 77 |
+
"rstrip": false,
|
| 78 |
+
"single_word": false
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"content": "<extra_id_88>",
|
| 82 |
+
"lstrip": true,
|
| 83 |
+
"normalized": true,
|
| 84 |
+
"rstrip": false,
|
| 85 |
+
"single_word": false
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"content": "<extra_id_87>",
|
| 89 |
+
"lstrip": true,
|
| 90 |
+
"normalized": true,
|
| 91 |
+
"rstrip": false,
|
| 92 |
+
"single_word": false
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"content": "<extra_id_86>",
|
| 96 |
+
"lstrip": true,
|
| 97 |
+
"normalized": true,
|
| 98 |
+
"rstrip": false,
|
| 99 |
+
"single_word": false
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"content": "<extra_id_85>",
|
| 103 |
+
"lstrip": true,
|
| 104 |
+
"normalized": true,
|
| 105 |
+
"rstrip": false,
|
| 106 |
+
"single_word": false
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"content": "<extra_id_84>",
|
| 110 |
+
"lstrip": true,
|
| 111 |
+
"normalized": true,
|
| 112 |
+
"rstrip": false,
|
| 113 |
+
"single_word": false
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"content": "<extra_id_83>",
|
| 117 |
+
"lstrip": true,
|
| 118 |
+
"normalized": true,
|
| 119 |
+
"rstrip": false,
|
| 120 |
+
"single_word": false
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"content": "<extra_id_82>",
|
| 124 |
+
"lstrip": true,
|
| 125 |
+
"normalized": true,
|
| 126 |
+
"rstrip": false,
|
| 127 |
+
"single_word": false
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"content": "<extra_id_81>",
|
| 131 |
+
"lstrip": true,
|
| 132 |
+
"normalized": true,
|
| 133 |
+
"rstrip": false,
|
| 134 |
+
"single_word": false
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"content": "<extra_id_80>",
|
| 138 |
+
"lstrip": true,
|
| 139 |
+
"normalized": true,
|
| 140 |
+
"rstrip": false,
|
| 141 |
+
"single_word": false
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"content": "<extra_id_79>",
|
| 145 |
+
"lstrip": true,
|
| 146 |
+
"normalized": true,
|
| 147 |
+
"rstrip": false,
|
| 148 |
+
"single_word": false
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"content": "<extra_id_78>",
|
| 152 |
+
"lstrip": true,
|
| 153 |
+
"normalized": true,
|
| 154 |
+
"rstrip": false,
|
| 155 |
+
"single_word": false
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"content": "<extra_id_77>",
|
| 159 |
+
"lstrip": true,
|
| 160 |
+
"normalized": true,
|
| 161 |
+
"rstrip": false,
|
| 162 |
+
"single_word": false
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"content": "<extra_id_76>",
|
| 166 |
+
"lstrip": true,
|
| 167 |
+
"normalized": true,
|
| 168 |
+
"rstrip": false,
|
| 169 |
+
"single_word": false
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"content": "<extra_id_75>",
|
| 173 |
+
"lstrip": true,
|
| 174 |
+
"normalized": true,
|
| 175 |
+
"rstrip": false,
|
| 176 |
+
"single_word": false
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"content": "<extra_id_74>",
|
| 180 |
+
"lstrip": true,
|
| 181 |
+
"normalized": true,
|
| 182 |
+
"rstrip": false,
|
| 183 |
+
"single_word": false
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"content": "<extra_id_73>",
|
| 187 |
+
"lstrip": true,
|
| 188 |
+
"normalized": true,
|
| 189 |
+
"rstrip": false,
|
| 190 |
+
"single_word": false
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"content": "<extra_id_72>",
|
| 194 |
+
"lstrip": true,
|
| 195 |
+
"normalized": true,
|
| 196 |
+
"rstrip": false,
|
| 197 |
+
"single_word": false
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"content": "<extra_id_71>",
|
| 201 |
+
"lstrip": true,
|
| 202 |
+
"normalized": true,
|
| 203 |
+
"rstrip": false,
|
| 204 |
+
"single_word": false
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"content": "<extra_id_70>",
|
| 208 |
+
"lstrip": true,
|
| 209 |
+
"normalized": true,
|
| 210 |
+
"rstrip": false,
|
| 211 |
+
"single_word": false
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"content": "<extra_id_69>",
|
| 215 |
+
"lstrip": true,
|
| 216 |
+
"normalized": true,
|
| 217 |
+
"rstrip": false,
|
| 218 |
+
"single_word": false
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"content": "<extra_id_68>",
|
| 222 |
+
"lstrip": true,
|
| 223 |
+
"normalized": true,
|
| 224 |
+
"rstrip": false,
|
| 225 |
+
"single_word": false
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"content": "<extra_id_67>",
|
| 229 |
+
"lstrip": true,
|
| 230 |
+
"normalized": true,
|
| 231 |
+
"rstrip": false,
|
| 232 |
+
"single_word": false
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"content": "<extra_id_66>",
|
| 236 |
+
"lstrip": true,
|
| 237 |
+
"normalized": true,
|
| 238 |
+
"rstrip": false,
|
| 239 |
+
"single_word": false
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"content": "<extra_id_65>",
|
| 243 |
+
"lstrip": true,
|
| 244 |
+
"normalized": true,
|
| 245 |
+
"rstrip": false,
|
| 246 |
+
"single_word": false
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"content": "<extra_id_64>",
|
| 250 |
+
"lstrip": true,
|
| 251 |
+
"normalized": true,
|
| 252 |
+
"rstrip": false,
|
| 253 |
+
"single_word": false
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"content": "<extra_id_63>",
|
| 257 |
+
"lstrip": true,
|
| 258 |
+
"normalized": true,
|
| 259 |
+
"rstrip": false,
|
| 260 |
+
"single_word": false
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"content": "<extra_id_62>",
|
| 264 |
+
"lstrip": true,
|
| 265 |
+
"normalized": true,
|
| 266 |
+
"rstrip": false,
|
| 267 |
+
"single_word": false
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"content": "<extra_id_61>",
|
| 271 |
+
"lstrip": true,
|
| 272 |
+
"normalized": true,
|
| 273 |
+
"rstrip": false,
|
| 274 |
+
"single_word": false
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"content": "<extra_id_60>",
|
| 278 |
+
"lstrip": true,
|
| 279 |
+
"normalized": true,
|
| 280 |
+
"rstrip": false,
|
| 281 |
+
"single_word": false
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"content": "<extra_id_59>",
|
| 285 |
+
"lstrip": true,
|
| 286 |
+
"normalized": true,
|
| 287 |
+
"rstrip": false,
|
| 288 |
+
"single_word": false
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"content": "<extra_id_58>",
|
| 292 |
+
"lstrip": true,
|
| 293 |
+
"normalized": true,
|
| 294 |
+
"rstrip": false,
|
| 295 |
+
"single_word": false
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"content": "<extra_id_57>",
|
| 299 |
+
"lstrip": true,
|
| 300 |
+
"normalized": true,
|
| 301 |
+
"rstrip": false,
|
| 302 |
+
"single_word": false
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"content": "<extra_id_56>",
|
| 306 |
+
"lstrip": true,
|
| 307 |
+
"normalized": true,
|
| 308 |
+
"rstrip": false,
|
| 309 |
+
"single_word": false
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"content": "<extra_id_55>",
|
| 313 |
+
"lstrip": true,
|
| 314 |
+
"normalized": true,
|
| 315 |
+
"rstrip": false,
|
| 316 |
+
"single_word": false
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"content": "<extra_id_54>",
|
| 320 |
+
"lstrip": true,
|
| 321 |
+
"normalized": true,
|
| 322 |
+
"rstrip": false,
|
| 323 |
+
"single_word": false
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"content": "<extra_id_53>",
|
| 327 |
+
"lstrip": true,
|
| 328 |
+
"normalized": true,
|
| 329 |
+
"rstrip": false,
|
| 330 |
+
"single_word": false
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"content": "<extra_id_52>",
|
| 334 |
+
"lstrip": true,
|
| 335 |
+
"normalized": true,
|
| 336 |
+
"rstrip": false,
|
| 337 |
+
"single_word": false
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"content": "<extra_id_51>",
|
| 341 |
+
"lstrip": true,
|
| 342 |
+
"normalized": true,
|
| 343 |
+
"rstrip": false,
|
| 344 |
+
"single_word": false
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"content": "<extra_id_50>",
|
| 348 |
+
"lstrip": true,
|
| 349 |
+
"normalized": true,
|
| 350 |
+
"rstrip": false,
|
| 351 |
+
"single_word": false
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"content": "<extra_id_49>",
|
| 355 |
+
"lstrip": true,
|
| 356 |
+
"normalized": true,
|
| 357 |
+
"rstrip": false,
|
| 358 |
+
"single_word": false
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"content": "<extra_id_48>",
|
| 362 |
+
"lstrip": true,
|
| 363 |
+
"normalized": true,
|
| 364 |
+
"rstrip": false,
|
| 365 |
+
"single_word": false
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"content": "<extra_id_47>",
|
| 369 |
+
"lstrip": true,
|
| 370 |
+
"normalized": true,
|
| 371 |
+
"rstrip": false,
|
| 372 |
+
"single_word": false
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"content": "<extra_id_46>",
|
| 376 |
+
"lstrip": true,
|
| 377 |
+
"normalized": true,
|
| 378 |
+
"rstrip": false,
|
| 379 |
+
"single_word": false
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"content": "<extra_id_45>",
|
| 383 |
+
"lstrip": true,
|
| 384 |
+
"normalized": true,
|
| 385 |
+
"rstrip": false,
|
| 386 |
+
"single_word": false
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"content": "<extra_id_44>",
|
| 390 |
+
"lstrip": true,
|
| 391 |
+
"normalized": true,
|
| 392 |
+
"rstrip": false,
|
| 393 |
+
"single_word": false
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"content": "<extra_id_43>",
|
| 397 |
+
"lstrip": true,
|
| 398 |
+
"normalized": true,
|
| 399 |
+
"rstrip": false,
|
| 400 |
+
"single_word": false
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"content": "<extra_id_42>",
|
| 404 |
+
"lstrip": true,
|
| 405 |
+
"normalized": true,
|
| 406 |
+
"rstrip": false,
|
| 407 |
+
"single_word": false
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"content": "<extra_id_41>",
|
| 411 |
+
"lstrip": true,
|
| 412 |
+
"normalized": true,
|
| 413 |
+
"rstrip": false,
|
| 414 |
+
"single_word": false
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"content": "<extra_id_40>",
|
| 418 |
+
"lstrip": true,
|
| 419 |
+
"normalized": true,
|
| 420 |
+
"rstrip": false,
|
| 421 |
+
"single_word": false
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"content": "<extra_id_39>",
|
| 425 |
+
"lstrip": true,
|
| 426 |
+
"normalized": true,
|
| 427 |
+
"rstrip": false,
|
| 428 |
+
"single_word": false
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"content": "<extra_id_38>",
|
| 432 |
+
"lstrip": true,
|
| 433 |
+
"normalized": true,
|
| 434 |
+
"rstrip": false,
|
| 435 |
+
"single_word": false
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"content": "<extra_id_37>",
|
| 439 |
+
"lstrip": true,
|
| 440 |
+
"normalized": true,
|
| 441 |
+
"rstrip": false,
|
| 442 |
+
"single_word": false
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"content": "<extra_id_36>",
|
| 446 |
+
"lstrip": true,
|
| 447 |
+
"normalized": true,
|
| 448 |
+
"rstrip": false,
|
| 449 |
+
"single_word": false
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"content": "<extra_id_35>",
|
| 453 |
+
"lstrip": true,
|
| 454 |
+
"normalized": true,
|
| 455 |
+
"rstrip": false,
|
| 456 |
+
"single_word": false
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"content": "<extra_id_34>",
|
| 460 |
+
"lstrip": true,
|
| 461 |
+
"normalized": true,
|
| 462 |
+
"rstrip": false,
|
| 463 |
+
"single_word": false
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"content": "<extra_id_33>",
|
| 467 |
+
"lstrip": true,
|
| 468 |
+
"normalized": true,
|
| 469 |
+
"rstrip": false,
|
| 470 |
+
"single_word": false
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"content": "<extra_id_32>",
|
| 474 |
+
"lstrip": true,
|
| 475 |
+
"normalized": true,
|
| 476 |
+
"rstrip": false,
|
| 477 |
+
"single_word": false
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"content": "<extra_id_31>",
|
| 481 |
+
"lstrip": true,
|
| 482 |
+
"normalized": true,
|
| 483 |
+
"rstrip": false,
|
| 484 |
+
"single_word": false
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"content": "<extra_id_30>",
|
| 488 |
+
"lstrip": true,
|
| 489 |
+
"normalized": true,
|
| 490 |
+
"rstrip": false,
|
| 491 |
+
"single_word": false
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"content": "<extra_id_29>",
|
| 495 |
+
"lstrip": true,
|
| 496 |
+
"normalized": true,
|
| 497 |
+
"rstrip": false,
|
| 498 |
+
"single_word": false
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"content": "<extra_id_28>",
|
| 502 |
+
"lstrip": true,
|
| 503 |
+
"normalized": true,
|
| 504 |
+
"rstrip": false,
|
| 505 |
+
"single_word": false
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"content": "<extra_id_27>",
|
| 509 |
+
"lstrip": true,
|
| 510 |
+
"normalized": true,
|
| 511 |
+
"rstrip": false,
|
| 512 |
+
"single_word": false
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"content": "<extra_id_26>",
|
| 516 |
+
"lstrip": true,
|
| 517 |
+
"normalized": true,
|
| 518 |
+
"rstrip": false,
|
| 519 |
+
"single_word": false
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"content": "<extra_id_25>",
|
| 523 |
+
"lstrip": true,
|
| 524 |
+
"normalized": true,
|
| 525 |
+
"rstrip": false,
|
| 526 |
+
"single_word": false
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"content": "<extra_id_24>",
|
| 530 |
+
"lstrip": true,
|
| 531 |
+
"normalized": true,
|
| 532 |
+
"rstrip": false,
|
| 533 |
+
"single_word": false
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"content": "<extra_id_23>",
|
| 537 |
+
"lstrip": true,
|
| 538 |
+
"normalized": true,
|
| 539 |
+
"rstrip": false,
|
| 540 |
+
"single_word": false
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"content": "<extra_id_22>",
|
| 544 |
+
"lstrip": true,
|
| 545 |
+
"normalized": true,
|
| 546 |
+
"rstrip": false,
|
| 547 |
+
"single_word": false
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"content": "<extra_id_21>",
|
| 551 |
+
"lstrip": true,
|
| 552 |
+
"normalized": true,
|
| 553 |
+
"rstrip": false,
|
| 554 |
+
"single_word": false
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"content": "<extra_id_20>",
|
| 558 |
+
"lstrip": true,
|
| 559 |
+
"normalized": true,
|
| 560 |
+
"rstrip": false,
|
| 561 |
+
"single_word": false
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"content": "<extra_id_19>",
|
| 565 |
+
"lstrip": true,
|
| 566 |
+
"normalized": true,
|
| 567 |
+
"rstrip": false,
|
| 568 |
+
"single_word": false
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"content": "<extra_id_18>",
|
| 572 |
+
"lstrip": true,
|
| 573 |
+
"normalized": true,
|
| 574 |
+
"rstrip": false,
|
| 575 |
+
"single_word": false
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"content": "<extra_id_17>",
|
| 579 |
+
"lstrip": true,
|
| 580 |
+
"normalized": true,
|
| 581 |
+
"rstrip": false,
|
| 582 |
+
"single_word": false
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"content": "<extra_id_16>",
|
| 586 |
+
"lstrip": true,
|
| 587 |
+
"normalized": true,
|
| 588 |
+
"rstrip": false,
|
| 589 |
+
"single_word": false
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"content": "<extra_id_15>",
|
| 593 |
+
"lstrip": true,
|
| 594 |
+
"normalized": true,
|
| 595 |
+
"rstrip": false,
|
| 596 |
+
"single_word": false
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"content": "<extra_id_14>",
|
| 600 |
+
"lstrip": true,
|
| 601 |
+
"normalized": true,
|
| 602 |
+
"rstrip": false,
|
| 603 |
+
"single_word": false
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"content": "<extra_id_13>",
|
| 607 |
+
"lstrip": true,
|
| 608 |
+
"normalized": true,
|
| 609 |
+
"rstrip": false,
|
| 610 |
+
"single_word": false
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"content": "<extra_id_12>",
|
| 614 |
+
"lstrip": true,
|
| 615 |
+
"normalized": true,
|
| 616 |
+
"rstrip": false,
|
| 617 |
+
"single_word": false
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"content": "<extra_id_11>",
|
| 621 |
+
"lstrip": true,
|
| 622 |
+
"normalized": true,
|
| 623 |
+
"rstrip": false,
|
| 624 |
+
"single_word": false
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"content": "<extra_id_10>",
|
| 628 |
+
"lstrip": true,
|
| 629 |
+
"normalized": true,
|
| 630 |
+
"rstrip": false,
|
| 631 |
+
"single_word": false
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"content": "<extra_id_9>",
|
| 635 |
+
"lstrip": true,
|
| 636 |
+
"normalized": true,
|
| 637 |
+
"rstrip": false,
|
| 638 |
+
"single_word": false
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"content": "<extra_id_8>",
|
| 642 |
+
"lstrip": true,
|
| 643 |
+
"normalized": true,
|
| 644 |
+
"rstrip": false,
|
| 645 |
+
"single_word": false
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"content": "<extra_id_7>",
|
| 649 |
+
"lstrip": true,
|
| 650 |
+
"normalized": true,
|
| 651 |
+
"rstrip": false,
|
| 652 |
+
"single_word": false
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"content": "<extra_id_6>",
|
| 656 |
+
"lstrip": true,
|
| 657 |
+
"normalized": true,
|
| 658 |
+
"rstrip": false,
|
| 659 |
+
"single_word": false
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"content": "<extra_id_5>",
|
| 663 |
+
"lstrip": true,
|
| 664 |
+
"normalized": true,
|
| 665 |
+
"rstrip": false,
|
| 666 |
+
"single_word": false
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"content": "<extra_id_4>",
|
| 670 |
+
"lstrip": true,
|
| 671 |
+
"normalized": true,
|
| 672 |
+
"rstrip": false,
|
| 673 |
+
"single_word": false
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"content": "<extra_id_3>",
|
| 677 |
+
"lstrip": true,
|
| 678 |
+
"normalized": true,
|
| 679 |
+
"rstrip": false,
|
| 680 |
+
"single_word": false
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"content": "<extra_id_2>",
|
| 684 |
+
"lstrip": true,
|
| 685 |
+
"normalized": true,
|
| 686 |
+
"rstrip": false,
|
| 687 |
+
"single_word": false
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"content": "<extra_id_1>",
|
| 691 |
+
"lstrip": true,
|
| 692 |
+
"normalized": true,
|
| 693 |
+
"rstrip": false,
|
| 694 |
+
"single_word": false
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"content": "<extra_id_0>",
|
| 698 |
+
"lstrip": true,
|
| 699 |
+
"normalized": true,
|
| 700 |
+
"rstrip": false,
|
| 701 |
+
"single_word": false
|
| 702 |
+
}
|
| 703 |
+
],
|
| 704 |
+
"bos_token": {
|
| 705 |
+
"content": "<s>",
|
| 706 |
+
"lstrip": false,
|
| 707 |
+
"normalized": true,
|
| 708 |
+
"rstrip": false,
|
| 709 |
+
"single_word": false
|
| 710 |
+
},
|
| 711 |
+
"cls_token": {
|
| 712 |
+
"content": "<s>",
|
| 713 |
+
"lstrip": false,
|
| 714 |
+
"normalized": true,
|
| 715 |
+
"rstrip": false,
|
| 716 |
+
"single_word": false
|
| 717 |
+
},
|
| 718 |
+
"eos_token": {
|
| 719 |
+
"content": "</s>",
|
| 720 |
+
"lstrip": false,
|
| 721 |
+
"normalized": true,
|
| 722 |
+
"rstrip": false,
|
| 723 |
+
"single_word": false
|
| 724 |
+
},
|
| 725 |
+
"mask_token": {
|
| 726 |
+
"content": "<mask>",
|
| 727 |
+
"lstrip": true,
|
| 728 |
+
"normalized": true,
|
| 729 |
+
"rstrip": false,
|
| 730 |
+
"single_word": false
|
| 731 |
+
},
|
| 732 |
+
"pad_token": {
|
| 733 |
+
"content": "<pad>",
|
| 734 |
+
"lstrip": false,
|
| 735 |
+
"normalized": true,
|
| 736 |
+
"rstrip": false,
|
| 737 |
+
"single_word": false
|
| 738 |
+
},
|
| 739 |
+
"sep_token": {
|
| 740 |
+
"content": "</s>",
|
| 741 |
+
"lstrip": false,
|
| 742 |
+
"normalized": true,
|
| 743 |
+
"rstrip": false,
|
| 744 |
+
"single_word": false
|
| 745 |
+
},
|
| 746 |
+
"unk_token": {
|
| 747 |
+
"content": "<unk>",
|
| 748 |
+
"lstrip": false,
|
| 749 |
+
"normalized": true,
|
| 750 |
+
"rstrip": false,
|
| 751 |
+
"single_word": false
|
| 752 |
+
}
|
| 753 |
+
}
|
test_results.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"test_loss": 0.20962746441364288,
|
| 3 |
+
"test_runtime": 564.3642,
|
| 4 |
+
"test_samples_per_second": 33.338,
|
| 5 |
+
"test_steps_per_second": 0.26
|
| 6 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,959 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
},
|
| 845 |
+
"additional_special_tokens": [
|
| 846 |
+
"<extra_id_99>",
|
| 847 |
+
"<extra_id_98>",
|
| 848 |
+
"<extra_id_97>",
|
| 849 |
+
"<extra_id_96>",
|
| 850 |
+
"<extra_id_95>",
|
| 851 |
+
"<extra_id_94>",
|
| 852 |
+
"<extra_id_93>",
|
| 853 |
+
"<extra_id_92>",
|
| 854 |
+
"<extra_id_91>",
|
| 855 |
+
"<extra_id_90>",
|
| 856 |
+
"<extra_id_89>",
|
| 857 |
+
"<extra_id_88>",
|
| 858 |
+
"<extra_id_87>",
|
| 859 |
+
"<extra_id_86>",
|
| 860 |
+
"<extra_id_85>",
|
| 861 |
+
"<extra_id_84>",
|
| 862 |
+
"<extra_id_83>",
|
| 863 |
+
"<extra_id_82>",
|
| 864 |
+
"<extra_id_81>",
|
| 865 |
+
"<extra_id_80>",
|
| 866 |
+
"<extra_id_79>",
|
| 867 |
+
"<extra_id_78>",
|
| 868 |
+
"<extra_id_77>",
|
| 869 |
+
"<extra_id_76>",
|
| 870 |
+
"<extra_id_75>",
|
| 871 |
+
"<extra_id_74>",
|
| 872 |
+
"<extra_id_73>",
|
| 873 |
+
"<extra_id_72>",
|
| 874 |
+
"<extra_id_71>",
|
| 875 |
+
"<extra_id_70>",
|
| 876 |
+
"<extra_id_69>",
|
| 877 |
+
"<extra_id_68>",
|
| 878 |
+
"<extra_id_67>",
|
| 879 |
+
"<extra_id_66>",
|
| 880 |
+
"<extra_id_65>",
|
| 881 |
+
"<extra_id_64>",
|
| 882 |
+
"<extra_id_63>",
|
| 883 |
+
"<extra_id_62>",
|
| 884 |
+
"<extra_id_61>",
|
| 885 |
+
"<extra_id_60>",
|
| 886 |
+
"<extra_id_59>",
|
| 887 |
+
"<extra_id_58>",
|
| 888 |
+
"<extra_id_57>",
|
| 889 |
+
"<extra_id_56>",
|
| 890 |
+
"<extra_id_55>",
|
| 891 |
+
"<extra_id_54>",
|
| 892 |
+
"<extra_id_53>",
|
| 893 |
+
"<extra_id_52>",
|
| 894 |
+
"<extra_id_51>",
|
| 895 |
+
"<extra_id_50>",
|
| 896 |
+
"<extra_id_49>",
|
| 897 |
+
"<extra_id_48>",
|
| 898 |
+
"<extra_id_47>",
|
| 899 |
+
"<extra_id_46>",
|
| 900 |
+
"<extra_id_45>",
|
| 901 |
+
"<extra_id_44>",
|
| 902 |
+
"<extra_id_43>",
|
| 903 |
+
"<extra_id_42>",
|
| 904 |
+
"<extra_id_41>",
|
| 905 |
+
"<extra_id_40>",
|
| 906 |
+
"<extra_id_39>",
|
| 907 |
+
"<extra_id_38>",
|
| 908 |
+
"<extra_id_37>",
|
| 909 |
+
"<extra_id_36>",
|
| 910 |
+
"<extra_id_35>",
|
| 911 |
+
"<extra_id_34>",
|
| 912 |
+
"<extra_id_33>",
|
| 913 |
+
"<extra_id_32>",
|
| 914 |
+
"<extra_id_31>",
|
| 915 |
+
"<extra_id_30>",
|
| 916 |
+
"<extra_id_29>",
|
| 917 |
+
"<extra_id_28>",
|
| 918 |
+
"<extra_id_27>",
|
| 919 |
+
"<extra_id_26>",
|
| 920 |
+
"<extra_id_25>",
|
| 921 |
+
"<extra_id_24>",
|
| 922 |
+
"<extra_id_23>",
|
| 923 |
+
"<extra_id_22>",
|
| 924 |
+
"<extra_id_21>",
|
| 925 |
+
"<extra_id_20>",
|
| 926 |
+
"<extra_id_19>",
|
| 927 |
+
"<extra_id_18>",
|
| 928 |
+
"<extra_id_17>",
|
| 929 |
+
"<extra_id_16>",
|
| 930 |
+
"<extra_id_15>",
|
| 931 |
+
"<extra_id_14>",
|
| 932 |
+
"<extra_id_13>",
|
| 933 |
+
"<extra_id_12>",
|
| 934 |
+
"<extra_id_11>",
|
| 935 |
+
"<extra_id_10>",
|
| 936 |
+
"<extra_id_9>",
|
| 937 |
+
"<extra_id_8>",
|
| 938 |
+
"<extra_id_7>",
|
| 939 |
+
"<extra_id_6>",
|
| 940 |
+
"<extra_id_5>",
|
| 941 |
+
"<extra_id_4>",
|
| 942 |
+
"<extra_id_3>",
|
| 943 |
+
"<extra_id_2>",
|
| 944 |
+
"<extra_id_1>",
|
| 945 |
+
"<extra_id_0>"
|
| 946 |
+
],
|
| 947 |
+
"bos_token": "<s>",
|
| 948 |
+
"clean_up_tokenization_spaces": true,
|
| 949 |
+
"cls_token": "<s>",
|
| 950 |
+
"eos_token": "</s>",
|
| 951 |
+
"errors": "replace",
|
| 952 |
+
"mask_token": "<mask>",
|
| 953 |
+
"model_max_length": 512,
|
| 954 |
+
"pad_token": "<pad>",
|
| 955 |
+
"sep_token": "</s>",
|
| 956 |
+
"tokenizer_class": "RobertaTokenizer",
|
| 957 |
+
"trim_offsets": true,
|
| 958 |
+
"unk_token": "<unk>"
|
| 959 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"train_loss": 0.2880888063569923,
|
| 4 |
+
"train_runtime": 54838.0441,
|
| 5 |
+
"train_samples_per_second": 109.795,
|
| 6 |
+
"train_steps_per_second": 1.715
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1666 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.20909027755260468,
|
| 3 |
+
"best_model_checkpoint": "/root/pretrain_utg4java_02/checkpoint-91689",
|
| 4 |
+
"epoch": 40.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 94040,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.21,
|
| 13 |
+
"grad_norm": 1.1109575033187866,
|
| 14 |
+
"learning_rate": 9.98e-05,
|
| 15 |
+
"loss": 0.4973,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.43,
|
| 20 |
+
"grad_norm": 1.1452946662902832,
|
| 21 |
+
"learning_rate": 0.0001998,
|
| 22 |
+
"loss": 0.4793,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.64,
|
| 27 |
+
"grad_norm": 0.9271925091743469,
|
| 28 |
+
"learning_rate": 0.0001989273430782459,
|
| 29 |
+
"loss": 0.4999,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.85,
|
| 34 |
+
"grad_norm": 0.9055812358856201,
|
| 35 |
+
"learning_rate": 0.0001978525365434222,
|
| 36 |
+
"loss": 0.4934,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 1.0,
|
| 41 |
+
"eval_loss": 0.4301895797252655,
|
| 42 |
+
"eval_runtime": 56.7204,
|
| 43 |
+
"eval_samples_per_second": 331.733,
|
| 44 |
+
"eval_steps_per_second": 2.592,
|
| 45 |
+
"step": 2351
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 1.06,
|
| 49 |
+
"grad_norm": 0.8988948464393616,
|
| 50 |
+
"learning_rate": 0.00019677773000859847,
|
| 51 |
+
"loss": 0.4818,
|
| 52 |
+
"step": 2500
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 1.28,
|
| 56 |
+
"grad_norm": 0.9011508822441101,
|
| 57 |
+
"learning_rate": 0.00019570292347377474,
|
| 58 |
+
"loss": 0.4651,
|
| 59 |
+
"step": 3000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.49,
|
| 63 |
+
"grad_norm": 0.905346691608429,
|
| 64 |
+
"learning_rate": 0.00019463026655202063,
|
| 65 |
+
"loss": 0.467,
|
| 66 |
+
"step": 3500
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.7,
|
| 70 |
+
"grad_norm": 0.7411425709724426,
|
| 71 |
+
"learning_rate": 0.0001935597592433362,
|
| 72 |
+
"loss": 0.4625,
|
| 73 |
+
"step": 4000
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.91,
|
| 77 |
+
"grad_norm": 0.8985588550567627,
|
| 78 |
+
"learning_rate": 0.00019248495270851249,
|
| 79 |
+
"loss": 0.4572,
|
| 80 |
+
"step": 4500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 2.0,
|
| 84 |
+
"eval_loss": 0.4058227837085724,
|
| 85 |
+
"eval_runtime": 56.7252,
|
| 86 |
+
"eval_samples_per_second": 331.705,
|
| 87 |
+
"eval_steps_per_second": 2.591,
|
| 88 |
+
"step": 4702
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 2.13,
|
| 92 |
+
"grad_norm": 0.827357292175293,
|
| 93 |
+
"learning_rate": 0.00019141014617368875,
|
| 94 |
+
"loss": 0.4472,
|
| 95 |
+
"step": 5000
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 2.34,
|
| 99 |
+
"grad_norm": 0.736473023891449,
|
| 100 |
+
"learning_rate": 0.00019033533963886502,
|
| 101 |
+
"loss": 0.4372,
|
| 102 |
+
"step": 5500
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 2.55,
|
| 106 |
+
"grad_norm": 0.7558779120445251,
|
| 107 |
+
"learning_rate": 0.00018926053310404129,
|
| 108 |
+
"loss": 0.4362,
|
| 109 |
+
"step": 6000
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 2.76,
|
| 113 |
+
"grad_norm": 0.7970576882362366,
|
| 114 |
+
"learning_rate": 0.00018818572656921755,
|
| 115 |
+
"loss": 0.4332,
|
| 116 |
+
"step": 6500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 2.98,
|
| 120 |
+
"grad_norm": 0.7692726850509644,
|
| 121 |
+
"learning_rate": 0.00018711092003439382,
|
| 122 |
+
"loss": 0.43,
|
| 123 |
+
"step": 7000
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 3.0,
|
| 127 |
+
"eval_loss": 0.38832882046699524,
|
| 128 |
+
"eval_runtime": 56.9476,
|
| 129 |
+
"eval_samples_per_second": 330.409,
|
| 130 |
+
"eval_steps_per_second": 2.581,
|
| 131 |
+
"step": 7053
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 3.19,
|
| 135 |
+
"grad_norm": 0.7109490633010864,
|
| 136 |
+
"learning_rate": 0.00018603611349957009,
|
| 137 |
+
"loss": 0.4217,
|
| 138 |
+
"step": 7500
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 3.4,
|
| 142 |
+
"grad_norm": 0.7125234603881836,
|
| 143 |
+
"learning_rate": 0.00018496130696474635,
|
| 144 |
+
"loss": 0.4141,
|
| 145 |
+
"step": 8000
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 3.62,
|
| 149 |
+
"grad_norm": 0.6914758682250977,
|
| 150 |
+
"learning_rate": 0.00018388650042992262,
|
| 151 |
+
"loss": 0.4169,
|
| 152 |
+
"step": 8500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 3.83,
|
| 156 |
+
"grad_norm": 0.7215930223464966,
|
| 157 |
+
"learning_rate": 0.00018281169389509889,
|
| 158 |
+
"loss": 0.4132,
|
| 159 |
+
"step": 9000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 4.0,
|
| 163 |
+
"eval_loss": 0.3813731372356415,
|
| 164 |
+
"eval_runtime": 56.8812,
|
| 165 |
+
"eval_samples_per_second": 330.794,
|
| 166 |
+
"eval_steps_per_second": 2.584,
|
| 167 |
+
"step": 9404
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 4.04,
|
| 171 |
+
"grad_norm": 0.7283121943473816,
|
| 172 |
+
"learning_rate": 0.00018173688736027515,
|
| 173 |
+
"loss": 0.4131,
|
| 174 |
+
"step": 9500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 4.25,
|
| 178 |
+
"grad_norm": 0.6947267055511475,
|
| 179 |
+
"learning_rate": 0.00018066208082545142,
|
| 180 |
+
"loss": 0.4059,
|
| 181 |
+
"step": 10000
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 4.47,
|
| 185 |
+
"grad_norm": 0.7544755339622498,
|
| 186 |
+
"learning_rate": 0.0001795872742906277,
|
| 187 |
+
"loss": 0.4023,
|
| 188 |
+
"step": 10500
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 4.68,
|
| 192 |
+
"grad_norm": 0.6204919815063477,
|
| 193 |
+
"learning_rate": 0.00017851246775580398,
|
| 194 |
+
"loss": 0.4038,
|
| 195 |
+
"step": 11000
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 4.89,
|
| 199 |
+
"grad_norm": 0.7036948800086975,
|
| 200 |
+
"learning_rate": 0.00017743766122098022,
|
| 201 |
+
"loss": 0.4017,
|
| 202 |
+
"step": 11500
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 5.0,
|
| 206 |
+
"eval_loss": 0.3638169765472412,
|
| 207 |
+
"eval_runtime": 56.8241,
|
| 208 |
+
"eval_samples_per_second": 331.127,
|
| 209 |
+
"eval_steps_per_second": 2.587,
|
| 210 |
+
"step": 11755
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 5.1,
|
| 214 |
+
"grad_norm": 0.62198805809021,
|
| 215 |
+
"learning_rate": 0.0001763628546861565,
|
| 216 |
+
"loss": 0.3921,
|
| 217 |
+
"step": 12000
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 5.32,
|
| 221 |
+
"grad_norm": 0.6095067858695984,
|
| 222 |
+
"learning_rate": 0.00017528804815133278,
|
| 223 |
+
"loss": 0.3901,
|
| 224 |
+
"step": 12500
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 5.53,
|
| 228 |
+
"grad_norm": 1.0518782138824463,
|
| 229 |
+
"learning_rate": 0.00017421324161650902,
|
| 230 |
+
"loss": 0.3888,
|
| 231 |
+
"step": 13000
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 5.74,
|
| 235 |
+
"grad_norm": 0.6801586151123047,
|
| 236 |
+
"learning_rate": 0.00017313843508168531,
|
| 237 |
+
"loss": 0.3859,
|
| 238 |
+
"step": 13500
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 5.95,
|
| 242 |
+
"grad_norm": 0.6489918828010559,
|
| 243 |
+
"learning_rate": 0.00017206362854686158,
|
| 244 |
+
"loss": 0.3839,
|
| 245 |
+
"step": 14000
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 6.0,
|
| 249 |
+
"eval_loss": 0.35733696818351746,
|
| 250 |
+
"eval_runtime": 56.779,
|
| 251 |
+
"eval_samples_per_second": 331.39,
|
| 252 |
+
"eval_steps_per_second": 2.589,
|
| 253 |
+
"step": 14106
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 6.17,
|
| 257 |
+
"grad_norm": 0.6174113154411316,
|
| 258 |
+
"learning_rate": 0.00017098882201203782,
|
| 259 |
+
"loss": 0.3782,
|
| 260 |
+
"step": 14500
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 6.38,
|
| 264 |
+
"grad_norm": 0.641315221786499,
|
| 265 |
+
"learning_rate": 0.00016991401547721411,
|
| 266 |
+
"loss": 0.3792,
|
| 267 |
+
"step": 15000
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 6.59,
|
| 271 |
+
"grad_norm": 0.6454831957817078,
|
| 272 |
+
"learning_rate": 0.00016883920894239038,
|
| 273 |
+
"loss": 0.3745,
|
| 274 |
+
"step": 15500
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 6.81,
|
| 278 |
+
"grad_norm": 0.6489464640617371,
|
| 279 |
+
"learning_rate": 0.00016776440240756665,
|
| 280 |
+
"loss": 0.3729,
|
| 281 |
+
"step": 16000
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 7.0,
|
| 285 |
+
"eval_loss": 0.3489345610141754,
|
| 286 |
+
"eval_runtime": 57.0727,
|
| 287 |
+
"eval_samples_per_second": 329.685,
|
| 288 |
+
"eval_steps_per_second": 2.576,
|
| 289 |
+
"step": 16457
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 7.02,
|
| 293 |
+
"grad_norm": 0.6929277181625366,
|
| 294 |
+
"learning_rate": 0.0001666938950988822,
|
| 295 |
+
"loss": 0.3718,
|
| 296 |
+
"step": 16500
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 7.23,
|
| 300 |
+
"grad_norm": 0.5726625323295593,
|
| 301 |
+
"learning_rate": 0.00016562123817712813,
|
| 302 |
+
"loss": 0.3624,
|
| 303 |
+
"step": 17000
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 7.44,
|
| 307 |
+
"grad_norm": 0.603529155254364,
|
| 308 |
+
"learning_rate": 0.0001645464316423044,
|
| 309 |
+
"loss": 0.3664,
|
| 310 |
+
"step": 17500
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 7.66,
|
| 314 |
+
"grad_norm": 0.7436888813972473,
|
| 315 |
+
"learning_rate": 0.00016347162510748066,
|
| 316 |
+
"loss": 0.3614,
|
| 317 |
+
"step": 18000
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 7.87,
|
| 321 |
+
"grad_norm": 0.6405676007270813,
|
| 322 |
+
"learning_rate": 0.00016239681857265693,
|
| 323 |
+
"loss": 0.3646,
|
| 324 |
+
"step": 18500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 8.0,
|
| 328 |
+
"eval_loss": 0.3384884297847748,
|
| 329 |
+
"eval_runtime": 56.7342,
|
| 330 |
+
"eval_samples_per_second": 331.652,
|
| 331 |
+
"eval_steps_per_second": 2.591,
|
| 332 |
+
"step": 18808
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 8.08,
|
| 336 |
+
"grad_norm": 0.660517156124115,
|
| 337 |
+
"learning_rate": 0.0001613220120378332,
|
| 338 |
+
"loss": 0.3623,
|
| 339 |
+
"step": 19000
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 8.29,
|
| 343 |
+
"grad_norm": 0.6352570056915283,
|
| 344 |
+
"learning_rate": 0.0001602493551160791,
|
| 345 |
+
"loss": 0.3518,
|
| 346 |
+
"step": 19500
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 8.51,
|
| 350 |
+
"grad_norm": 0.6094992756843567,
|
| 351 |
+
"learning_rate": 0.0001591745485812554,
|
| 352 |
+
"loss": 0.3569,
|
| 353 |
+
"step": 20000
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 8.72,
|
| 357 |
+
"grad_norm": 0.6646167635917664,
|
| 358 |
+
"learning_rate": 0.00015809974204643166,
|
| 359 |
+
"loss": 0.3567,
|
| 360 |
+
"step": 20500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 8.93,
|
| 364 |
+
"grad_norm": 0.6240518093109131,
|
| 365 |
+
"learning_rate": 0.00015702493551160792,
|
| 366 |
+
"loss": 0.353,
|
| 367 |
+
"step": 21000
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 9.0,
|
| 371 |
+
"eval_loss": 0.3339434862136841,
|
| 372 |
+
"eval_runtime": 56.979,
|
| 373 |
+
"eval_samples_per_second": 330.227,
|
| 374 |
+
"eval_steps_per_second": 2.58,
|
| 375 |
+
"step": 21159
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 9.15,
|
| 379 |
+
"grad_norm": 0.6990819573402405,
|
| 380 |
+
"learning_rate": 0.0001559501289767842,
|
| 381 |
+
"loss": 0.3491,
|
| 382 |
+
"step": 21500
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 9.36,
|
| 386 |
+
"grad_norm": 0.6751831769943237,
|
| 387 |
+
"learning_rate": 0.00015487532244196046,
|
| 388 |
+
"loss": 0.346,
|
| 389 |
+
"step": 22000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 9.57,
|
| 393 |
+
"grad_norm": 0.615511417388916,
|
| 394 |
+
"learning_rate": 0.00015380051590713672,
|
| 395 |
+
"loss": 0.3488,
|
| 396 |
+
"step": 22500
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 9.78,
|
| 400 |
+
"grad_norm": 0.5277218222618103,
|
| 401 |
+
"learning_rate": 0.000152725709372313,
|
| 402 |
+
"loss": 0.3466,
|
| 403 |
+
"step": 23000
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 10.0,
|
| 407 |
+
"grad_norm": 0.5809108018875122,
|
| 408 |
+
"learning_rate": 0.00015165090283748926,
|
| 409 |
+
"loss": 0.3441,
|
| 410 |
+
"step": 23500
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 10.0,
|
| 414 |
+
"eval_loss": 0.324989378452301,
|
| 415 |
+
"eval_runtime": 56.9322,
|
| 416 |
+
"eval_samples_per_second": 330.499,
|
| 417 |
+
"eval_steps_per_second": 2.582,
|
| 418 |
+
"step": 23510
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 10.21,
|
| 422 |
+
"grad_norm": 0.5285485982894897,
|
| 423 |
+
"learning_rate": 0.00015058039552880484,
|
| 424 |
+
"loss": 0.3407,
|
| 425 |
+
"step": 24000
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 10.42,
|
| 429 |
+
"grad_norm": 0.6268051862716675,
|
| 430 |
+
"learning_rate": 0.00014950558899398108,
|
| 431 |
+
"loss": 0.3389,
|
| 432 |
+
"step": 24500
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 10.63,
|
| 436 |
+
"grad_norm": 0.5879684090614319,
|
| 437 |
+
"learning_rate": 0.00014843078245915735,
|
| 438 |
+
"loss": 0.3387,
|
| 439 |
+
"step": 25000
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 10.85,
|
| 443 |
+
"grad_norm": 0.5434576869010925,
|
| 444 |
+
"learning_rate": 0.00014735597592433364,
|
| 445 |
+
"loss": 0.3379,
|
| 446 |
+
"step": 25500
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 11.0,
|
| 450 |
+
"eval_loss": 0.31874018907546997,
|
| 451 |
+
"eval_runtime": 57.0459,
|
| 452 |
+
"eval_samples_per_second": 329.84,
|
| 453 |
+
"eval_steps_per_second": 2.577,
|
| 454 |
+
"step": 25861
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 11.06,
|
| 458 |
+
"grad_norm": 0.6547732949256897,
|
| 459 |
+
"learning_rate": 0.00014628116938950988,
|
| 460 |
+
"loss": 0.3369,
|
| 461 |
+
"step": 26000
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 11.27,
|
| 465 |
+
"grad_norm": 0.648601233959198,
|
| 466 |
+
"learning_rate": 0.00014520636285468615,
|
| 467 |
+
"loss": 0.3308,
|
| 468 |
+
"step": 26500
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 11.48,
|
| 472 |
+
"grad_norm": 0.6311826109886169,
|
| 473 |
+
"learning_rate": 0.00014413155631986244,
|
| 474 |
+
"loss": 0.3298,
|
| 475 |
+
"step": 27000
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 11.7,
|
| 479 |
+
"grad_norm": 0.5571895837783813,
|
| 480 |
+
"learning_rate": 0.00014305674978503868,
|
| 481 |
+
"loss": 0.3292,
|
| 482 |
+
"step": 27500
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 11.91,
|
| 486 |
+
"grad_norm": 0.5373964309692383,
|
| 487 |
+
"learning_rate": 0.00014198194325021498,
|
| 488 |
+
"loss": 0.3284,
|
| 489 |
+
"step": 28000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 12.0,
|
| 493 |
+
"eval_loss": 0.3123379051685333,
|
| 494 |
+
"eval_runtime": 56.7858,
|
| 495 |
+
"eval_samples_per_second": 331.351,
|
| 496 |
+
"eval_steps_per_second": 2.589,
|
| 497 |
+
"step": 28212
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 12.12,
|
| 501 |
+
"grad_norm": 0.5307437777519226,
|
| 502 |
+
"learning_rate": 0.00014090928632846088,
|
| 503 |
+
"loss": 0.3247,
|
| 504 |
+
"step": 28500
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 12.34,
|
| 508 |
+
"grad_norm": 0.6422705054283142,
|
| 509 |
+
"learning_rate": 0.00013983447979363714,
|
| 510 |
+
"loss": 0.3223,
|
| 511 |
+
"step": 29000
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 12.55,
|
| 515 |
+
"grad_norm": 0.6680347323417664,
|
| 516 |
+
"learning_rate": 0.00013875967325881344,
|
| 517 |
+
"loss": 0.3223,
|
| 518 |
+
"step": 29500
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 12.76,
|
| 522 |
+
"grad_norm": 0.5851155519485474,
|
| 523 |
+
"learning_rate": 0.00013768701633705933,
|
| 524 |
+
"loss": 0.3245,
|
| 525 |
+
"step": 30000
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 12.97,
|
| 529 |
+
"grad_norm": 0.544624924659729,
|
| 530 |
+
"learning_rate": 0.0001366122098022356,
|
| 531 |
+
"loss": 0.3252,
|
| 532 |
+
"step": 30500
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 13.0,
|
| 536 |
+
"eval_loss": 0.3076089322566986,
|
| 537 |
+
"eval_runtime": 57.0752,
|
| 538 |
+
"eval_samples_per_second": 329.67,
|
| 539 |
+
"eval_steps_per_second": 2.576,
|
| 540 |
+
"step": 30563
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 13.19,
|
| 544 |
+
"grad_norm": 0.5343388319015503,
|
| 545 |
+
"learning_rate": 0.00013553740326741187,
|
| 546 |
+
"loss": 0.3185,
|
| 547 |
+
"step": 31000
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 13.4,
|
| 551 |
+
"grad_norm": 0.6074999570846558,
|
| 552 |
+
"learning_rate": 0.00013446259673258814,
|
| 553 |
+
"loss": 0.3191,
|
| 554 |
+
"step": 31500
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 13.61,
|
| 558 |
+
"grad_norm": 0.6202041506767273,
|
| 559 |
+
"learning_rate": 0.00013338779019776443,
|
| 560 |
+
"loss": 0.3186,
|
| 561 |
+
"step": 32000
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 13.82,
|
| 565 |
+
"grad_norm": 0.6058717966079712,
|
| 566 |
+
"learning_rate": 0.00013231298366294067,
|
| 567 |
+
"loss": 0.3134,
|
| 568 |
+
"step": 32500
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 14.0,
|
| 572 |
+
"eval_loss": 0.3018127679824829,
|
| 573 |
+
"eval_runtime": 57.0557,
|
| 574 |
+
"eval_samples_per_second": 329.783,
|
| 575 |
+
"eval_steps_per_second": 2.576,
|
| 576 |
+
"step": 32914
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 14.04,
|
| 580 |
+
"grad_norm": 0.6439224481582642,
|
| 581 |
+
"learning_rate": 0.00013123817712811694,
|
| 582 |
+
"loss": 0.3161,
|
| 583 |
+
"step": 33000
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 14.25,
|
| 587 |
+
"grad_norm": 0.5888046622276306,
|
| 588 |
+
"learning_rate": 0.00013016337059329323,
|
| 589 |
+
"loss": 0.3094,
|
| 590 |
+
"step": 33500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 14.46,
|
| 594 |
+
"grad_norm": 0.6644863486289978,
|
| 595 |
+
"learning_rate": 0.00012908856405846947,
|
| 596 |
+
"loss": 0.3092,
|
| 597 |
+
"step": 34000
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 14.67,
|
| 601 |
+
"grad_norm": 0.5649863481521606,
|
| 602 |
+
"learning_rate": 0.00012801375752364576,
|
| 603 |
+
"loss": 0.3117,
|
| 604 |
+
"step": 34500
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 14.89,
|
| 608 |
+
"grad_norm": 0.4947267174720764,
|
| 609 |
+
"learning_rate": 0.00012693895098882203,
|
| 610 |
+
"loss": 0.3072,
|
| 611 |
+
"step": 35000
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 15.0,
|
| 615 |
+
"eval_loss": 0.29680636525154114,
|
| 616 |
+
"eval_runtime": 57.044,
|
| 617 |
+
"eval_samples_per_second": 329.851,
|
| 618 |
+
"eval_steps_per_second": 2.577,
|
| 619 |
+
"step": 35265
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 15.1,
|
| 623 |
+
"grad_norm": 0.6015808582305908,
|
| 624 |
+
"learning_rate": 0.00012586414445399827,
|
| 625 |
+
"loss": 0.3067,
|
| 626 |
+
"step": 35500
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 15.31,
|
| 630 |
+
"grad_norm": 0.6003320217132568,
|
| 631 |
+
"learning_rate": 0.00012478933791917456,
|
| 632 |
+
"loss": 0.3027,
|
| 633 |
+
"step": 36000
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 15.53,
|
| 637 |
+
"grad_norm": 0.5643883347511292,
|
| 638 |
+
"learning_rate": 0.00012371453138435083,
|
| 639 |
+
"loss": 0.3012,
|
| 640 |
+
"step": 36500
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 15.74,
|
| 644 |
+
"grad_norm": 0.5768193602561951,
|
| 645 |
+
"learning_rate": 0.00012264187446259673,
|
| 646 |
+
"loss": 0.303,
|
| 647 |
+
"step": 37000
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 15.95,
|
| 651 |
+
"grad_norm": 0.5599430799484253,
|
| 652 |
+
"learning_rate": 0.00012156921754084265,
|
| 653 |
+
"loss": 0.3034,
|
| 654 |
+
"step": 37500
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 16.0,
|
| 658 |
+
"eval_loss": 0.2938060164451599,
|
| 659 |
+
"eval_runtime": 56.7422,
|
| 660 |
+
"eval_samples_per_second": 331.605,
|
| 661 |
+
"eval_steps_per_second": 2.591,
|
| 662 |
+
"step": 37616
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 16.16,
|
| 666 |
+
"grad_norm": 0.6380453705787659,
|
| 667 |
+
"learning_rate": 0.00012049441100601892,
|
| 668 |
+
"loss": 0.2987,
|
| 669 |
+
"step": 38000
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 16.38,
|
| 673 |
+
"grad_norm": 0.5312320590019226,
|
| 674 |
+
"learning_rate": 0.0001194196044711952,
|
| 675 |
+
"loss": 0.2975,
|
| 676 |
+
"step": 38500
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 16.59,
|
| 680 |
+
"grad_norm": 0.7027547955513,
|
| 681 |
+
"learning_rate": 0.00011834479793637145,
|
| 682 |
+
"loss": 0.2934,
|
| 683 |
+
"step": 39000
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 16.8,
|
| 687 |
+
"grad_norm": 0.5144539475440979,
|
| 688 |
+
"learning_rate": 0.00011726999140154773,
|
| 689 |
+
"loss": 0.2959,
|
| 690 |
+
"step": 39500
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 17.0,
|
| 694 |
+
"eval_loss": 0.28754693269729614,
|
| 695 |
+
"eval_runtime": 56.9555,
|
| 696 |
+
"eval_samples_per_second": 330.363,
|
| 697 |
+
"eval_steps_per_second": 2.581,
|
| 698 |
+
"step": 39967
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 17.01,
|
| 702 |
+
"grad_norm": 0.5759513974189758,
|
| 703 |
+
"learning_rate": 0.00011619733447979365,
|
| 704 |
+
"loss": 0.2997,
|
| 705 |
+
"step": 40000
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 17.23,
|
| 709 |
+
"grad_norm": 0.593640148639679,
|
| 710 |
+
"learning_rate": 0.00011512467755803955,
|
| 711 |
+
"loss": 0.294,
|
| 712 |
+
"step": 40500
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 17.44,
|
| 716 |
+
"grad_norm": 0.6821351647377014,
|
| 717 |
+
"learning_rate": 0.00011404987102321583,
|
| 718 |
+
"loss": 0.2904,
|
| 719 |
+
"step": 41000
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 17.65,
|
| 723 |
+
"grad_norm": 0.5297681093215942,
|
| 724 |
+
"learning_rate": 0.0001129750644883921,
|
| 725 |
+
"loss": 0.2896,
|
| 726 |
+
"step": 41500
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 17.86,
|
| 730 |
+
"grad_norm": 0.5864290595054626,
|
| 731 |
+
"learning_rate": 0.00011190025795356836,
|
| 732 |
+
"loss": 0.2884,
|
| 733 |
+
"step": 42000
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 18.0,
|
| 737 |
+
"eval_loss": 0.28015270829200745,
|
| 738 |
+
"eval_runtime": 56.8879,
|
| 739 |
+
"eval_samples_per_second": 330.756,
|
| 740 |
+
"eval_steps_per_second": 2.584,
|
| 741 |
+
"step": 42318
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 18.08,
|
| 745 |
+
"grad_norm": 0.5764068961143494,
|
| 746 |
+
"learning_rate": 0.00011082545141874463,
|
| 747 |
+
"loss": 0.2874,
|
| 748 |
+
"step": 42500
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 18.29,
|
| 752 |
+
"grad_norm": 0.6073163747787476,
|
| 753 |
+
"learning_rate": 0.0001097506448839209,
|
| 754 |
+
"loss": 0.2828,
|
| 755 |
+
"step": 43000
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 18.5,
|
| 759 |
+
"grad_norm": 0.5691092610359192,
|
| 760 |
+
"learning_rate": 0.00010867583834909716,
|
| 761 |
+
"loss": 0.2848,
|
| 762 |
+
"step": 43500
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 18.72,
|
| 766 |
+
"grad_norm": 0.5399264097213745,
|
| 767 |
+
"learning_rate": 0.00010760103181427344,
|
| 768 |
+
"loss": 0.2861,
|
| 769 |
+
"step": 44000
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 18.93,
|
| 773 |
+
"grad_norm": 0.6221365928649902,
|
| 774 |
+
"learning_rate": 0.0001065262252794497,
|
| 775 |
+
"loss": 0.2839,
|
| 776 |
+
"step": 44500
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 19.0,
|
| 780 |
+
"eval_loss": 0.27618157863616943,
|
| 781 |
+
"eval_runtime": 56.9107,
|
| 782 |
+
"eval_samples_per_second": 330.623,
|
| 783 |
+
"eval_steps_per_second": 2.583,
|
| 784 |
+
"step": 44669
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 19.14,
|
| 788 |
+
"grad_norm": 0.5080223679542542,
|
| 789 |
+
"learning_rate": 0.00010545141874462596,
|
| 790 |
+
"loss": 0.2791,
|
| 791 |
+
"step": 45000
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 19.35,
|
| 795 |
+
"grad_norm": 0.567764163017273,
|
| 796 |
+
"learning_rate": 0.00010437661220980224,
|
| 797 |
+
"loss": 0.2792,
|
| 798 |
+
"step": 45500
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 19.57,
|
| 802 |
+
"grad_norm": 0.5838685035705566,
|
| 803 |
+
"learning_rate": 0.00010330180567497851,
|
| 804 |
+
"loss": 0.2768,
|
| 805 |
+
"step": 46000
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"epoch": 19.78,
|
| 809 |
+
"grad_norm": 0.5778998136520386,
|
| 810 |
+
"learning_rate": 0.00010222699914015476,
|
| 811 |
+
"loss": 0.2776,
|
| 812 |
+
"step": 46500
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 19.99,
|
| 816 |
+
"grad_norm": 0.615337073802948,
|
| 817 |
+
"learning_rate": 0.0001011543422184007,
|
| 818 |
+
"loss": 0.2785,
|
| 819 |
+
"step": 47000
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 20.0,
|
| 823 |
+
"eval_loss": 0.2724262773990631,
|
| 824 |
+
"eval_runtime": 56.852,
|
| 825 |
+
"eval_samples_per_second": 330.965,
|
| 826 |
+
"eval_steps_per_second": 2.586,
|
| 827 |
+
"step": 47020
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 20.2,
|
| 831 |
+
"grad_norm": 0.6153652667999268,
|
| 832 |
+
"learning_rate": 0.00010007953568357695,
|
| 833 |
+
"loss": 0.2712,
|
| 834 |
+
"step": 47500
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 20.42,
|
| 838 |
+
"grad_norm": 0.6126906275749207,
|
| 839 |
+
"learning_rate": 9.900472914875323e-05,
|
| 840 |
+
"loss": 0.2718,
|
| 841 |
+
"step": 48000
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 20.63,
|
| 845 |
+
"grad_norm": 0.5799471139907837,
|
| 846 |
+
"learning_rate": 9.792992261392949e-05,
|
| 847 |
+
"loss": 0.2701,
|
| 848 |
+
"step": 48500
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 20.84,
|
| 852 |
+
"grad_norm": 0.5679476261138916,
|
| 853 |
+
"learning_rate": 9.685511607910577e-05,
|
| 854 |
+
"loss": 0.2727,
|
| 855 |
+
"step": 49000
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 21.0,
|
| 859 |
+
"eval_loss": 0.26610955595970154,
|
| 860 |
+
"eval_runtime": 56.7263,
|
| 861 |
+
"eval_samples_per_second": 331.698,
|
| 862 |
+
"eval_steps_per_second": 2.591,
|
| 863 |
+
"step": 49371
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 21.05,
|
| 867 |
+
"grad_norm": 0.5923852324485779,
|
| 868 |
+
"learning_rate": 9.578245915735168e-05,
|
| 869 |
+
"loss": 0.2694,
|
| 870 |
+
"step": 49500
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 21.27,
|
| 874 |
+
"grad_norm": 0.6184679269790649,
|
| 875 |
+
"learning_rate": 9.470765262252796e-05,
|
| 876 |
+
"loss": 0.2649,
|
| 877 |
+
"step": 50000
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 21.48,
|
| 881 |
+
"grad_norm": 0.5939807295799255,
|
| 882 |
+
"learning_rate": 9.363284608770421e-05,
|
| 883 |
+
"loss": 0.2697,
|
| 884 |
+
"step": 50500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 21.69,
|
| 888 |
+
"grad_norm": 0.5925255417823792,
|
| 889 |
+
"learning_rate": 9.255803955288048e-05,
|
| 890 |
+
"loss": 0.2652,
|
| 891 |
+
"step": 51000
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 21.91,
|
| 895 |
+
"grad_norm": 0.5666438341140747,
|
| 896 |
+
"learning_rate": 9.148323301805676e-05,
|
| 897 |
+
"loss": 0.266,
|
| 898 |
+
"step": 51500
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 22.0,
|
| 902 |
+
"eval_loss": 0.266626238822937,
|
| 903 |
+
"eval_runtime": 56.6361,
|
| 904 |
+
"eval_samples_per_second": 332.226,
|
| 905 |
+
"eval_steps_per_second": 2.596,
|
| 906 |
+
"step": 51722
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 22.12,
|
| 910 |
+
"grad_norm": 0.5490565299987793,
|
| 911 |
+
"learning_rate": 9.040842648323303e-05,
|
| 912 |
+
"loss": 0.2613,
|
| 913 |
+
"step": 52000
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 22.33,
|
| 917 |
+
"grad_norm": 0.6534095406532288,
|
| 918 |
+
"learning_rate": 8.933361994840929e-05,
|
| 919 |
+
"loss": 0.2592,
|
| 920 |
+
"step": 52500
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 22.54,
|
| 924 |
+
"grad_norm": 0.568953812122345,
|
| 925 |
+
"learning_rate": 8.82609630266552e-05,
|
| 926 |
+
"loss": 0.2604,
|
| 927 |
+
"step": 53000
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 22.76,
|
| 931 |
+
"grad_norm": 0.5968701243400574,
|
| 932 |
+
"learning_rate": 8.718615649183147e-05,
|
| 933 |
+
"loss": 0.2602,
|
| 934 |
+
"step": 53500
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 22.97,
|
| 938 |
+
"grad_norm": 0.5995833277702332,
|
| 939 |
+
"learning_rate": 8.61134995700774e-05,
|
| 940 |
+
"loss": 0.2614,
|
| 941 |
+
"step": 54000
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 23.0,
|
| 945 |
+
"eval_loss": 0.26461803913116455,
|
| 946 |
+
"eval_runtime": 56.8736,
|
| 947 |
+
"eval_samples_per_second": 330.839,
|
| 948 |
+
"eval_steps_per_second": 2.585,
|
| 949 |
+
"step": 54073
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 23.18,
|
| 953 |
+
"grad_norm": 0.577324390411377,
|
| 954 |
+
"learning_rate": 8.503869303525366e-05,
|
| 955 |
+
"loss": 0.2502,
|
| 956 |
+
"step": 54500
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 23.39,
|
| 960 |
+
"grad_norm": 0.631363034248352,
|
| 961 |
+
"learning_rate": 8.396603611349956e-05,
|
| 962 |
+
"loss": 0.256,
|
| 963 |
+
"step": 55000
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 23.61,
|
| 967 |
+
"grad_norm": 0.5862709879875183,
|
| 968 |
+
"learning_rate": 8.289122957867584e-05,
|
| 969 |
+
"loss": 0.2559,
|
| 970 |
+
"step": 55500
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 23.82,
|
| 974 |
+
"grad_norm": 0.5620314478874207,
|
| 975 |
+
"learning_rate": 8.181857265692176e-05,
|
| 976 |
+
"loss": 0.2549,
|
| 977 |
+
"step": 56000
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 24.0,
|
| 981 |
+
"eval_loss": 0.2593855559825897,
|
| 982 |
+
"eval_runtime": 57.337,
|
| 983 |
+
"eval_samples_per_second": 328.165,
|
| 984 |
+
"eval_steps_per_second": 2.564,
|
| 985 |
+
"step": 56424
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 24.03,
|
| 989 |
+
"grad_norm": 0.5514592528343201,
|
| 990 |
+
"learning_rate": 8.074376612209804e-05,
|
| 991 |
+
"loss": 0.2542,
|
| 992 |
+
"step": 56500
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 24.25,
|
| 996 |
+
"grad_norm": 0.6351886987686157,
|
| 997 |
+
"learning_rate": 7.966895958727429e-05,
|
| 998 |
+
"loss": 0.2483,
|
| 999 |
+
"step": 57000
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 24.46,
|
| 1003 |
+
"grad_norm": 0.6075631380081177,
|
| 1004 |
+
"learning_rate": 7.859415305245056e-05,
|
| 1005 |
+
"loss": 0.2503,
|
| 1006 |
+
"step": 57500
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 24.67,
|
| 1010 |
+
"grad_norm": 0.5592435002326965,
|
| 1011 |
+
"learning_rate": 7.751934651762684e-05,
|
| 1012 |
+
"loss": 0.249,
|
| 1013 |
+
"step": 58000
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 24.88,
|
| 1017 |
+
"grad_norm": 0.5757043957710266,
|
| 1018 |
+
"learning_rate": 7.64445399828031e-05,
|
| 1019 |
+
"loss": 0.2497,
|
| 1020 |
+
"step": 58500
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 25.0,
|
| 1024 |
+
"eval_loss": 0.2535327076911926,
|
| 1025 |
+
"eval_runtime": 56.789,
|
| 1026 |
+
"eval_samples_per_second": 331.332,
|
| 1027 |
+
"eval_steps_per_second": 2.589,
|
| 1028 |
+
"step": 58775
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 25.1,
|
| 1032 |
+
"grad_norm": 0.5995878577232361,
|
| 1033 |
+
"learning_rate": 7.536973344797937e-05,
|
| 1034 |
+
"loss": 0.245,
|
| 1035 |
+
"step": 59000
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 25.31,
|
| 1039 |
+
"grad_norm": 0.6077148914337158,
|
| 1040 |
+
"learning_rate": 7.429492691315564e-05,
|
| 1041 |
+
"loss": 0.2461,
|
| 1042 |
+
"step": 59500
|
| 1043 |
+
},
|
| 1044 |
+
{
|
| 1045 |
+
"epoch": 25.52,
|
| 1046 |
+
"grad_norm": 0.5592058300971985,
|
| 1047 |
+
"learning_rate": 7.32201203783319e-05,
|
| 1048 |
+
"loss": 0.2433,
|
| 1049 |
+
"step": 60000
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 25.73,
|
| 1053 |
+
"grad_norm": 0.6243628859519958,
|
| 1054 |
+
"learning_rate": 7.214531384350817e-05,
|
| 1055 |
+
"loss": 0.2444,
|
| 1056 |
+
"step": 60500
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 25.95,
|
| 1060 |
+
"grad_norm": 0.6848371624946594,
|
| 1061 |
+
"learning_rate": 7.107050730868444e-05,
|
| 1062 |
+
"loss": 0.243,
|
| 1063 |
+
"step": 61000
|
| 1064 |
+
},
|
| 1065 |
+
{
|
| 1066 |
+
"epoch": 26.0,
|
| 1067 |
+
"eval_loss": 0.250180721282959,
|
| 1068 |
+
"eval_runtime": 56.9862,
|
| 1069 |
+
"eval_samples_per_second": 330.185,
|
| 1070 |
+
"eval_steps_per_second": 2.58,
|
| 1071 |
+
"step": 61126
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 26.16,
|
| 1075 |
+
"grad_norm": 0.5177180171012878,
|
| 1076 |
+
"learning_rate": 6.99957007738607e-05,
|
| 1077 |
+
"loss": 0.2382,
|
| 1078 |
+
"step": 61500
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 26.37,
|
| 1082 |
+
"grad_norm": 0.6144821047782898,
|
| 1083 |
+
"learning_rate": 6.892304385210663e-05,
|
| 1084 |
+
"loss": 0.2349,
|
| 1085 |
+
"step": 62000
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 26.58,
|
| 1089 |
+
"grad_norm": 0.5862034559249878,
|
| 1090 |
+
"learning_rate": 6.78482373172829e-05,
|
| 1091 |
+
"loss": 0.2385,
|
| 1092 |
+
"step": 62500
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 26.8,
|
| 1096 |
+
"grad_norm": 0.5799028277397156,
|
| 1097 |
+
"learning_rate": 6.677343078245916e-05,
|
| 1098 |
+
"loss": 0.2383,
|
| 1099 |
+
"step": 63000
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"epoch": 27.0,
|
| 1103 |
+
"eval_loss": 0.24638937413692474,
|
| 1104 |
+
"eval_runtime": 56.9276,
|
| 1105 |
+
"eval_samples_per_second": 330.525,
|
| 1106 |
+
"eval_steps_per_second": 2.582,
|
| 1107 |
+
"step": 63477
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 27.01,
|
| 1111 |
+
"grad_norm": 0.6466034054756165,
|
| 1112 |
+
"learning_rate": 6.569862424763543e-05,
|
| 1113 |
+
"loss": 0.2375,
|
| 1114 |
+
"step": 63500
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 27.22,
|
| 1118 |
+
"grad_norm": 0.6161571741104126,
|
| 1119 |
+
"learning_rate": 6.46238177128117e-05,
|
| 1120 |
+
"loss": 0.2337,
|
| 1121 |
+
"step": 64000
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 27.44,
|
| 1125 |
+
"grad_norm": 0.6162907481193542,
|
| 1126 |
+
"learning_rate": 6.354901117798796e-05,
|
| 1127 |
+
"loss": 0.2328,
|
| 1128 |
+
"step": 64500
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 27.65,
|
| 1132 |
+
"grad_norm": 0.6643475890159607,
|
| 1133 |
+
"learning_rate": 6.247420464316423e-05,
|
| 1134 |
+
"loss": 0.2316,
|
| 1135 |
+
"step": 65000
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 27.86,
|
| 1139 |
+
"grad_norm": 0.5605382323265076,
|
| 1140 |
+
"learning_rate": 6.13993981083405e-05,
|
| 1141 |
+
"loss": 0.2325,
|
| 1142 |
+
"step": 65500
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 28.0,
|
| 1146 |
+
"eval_loss": 0.24465161561965942,
|
| 1147 |
+
"eval_runtime": 57.0032,
|
| 1148 |
+
"eval_samples_per_second": 330.087,
|
| 1149 |
+
"eval_steps_per_second": 2.579,
|
| 1150 |
+
"step": 65828
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 28.07,
|
| 1154 |
+
"grad_norm": 0.6108579635620117,
|
| 1155 |
+
"learning_rate": 6.032459157351676e-05,
|
| 1156 |
+
"loss": 0.231,
|
| 1157 |
+
"step": 66000
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 28.29,
|
| 1161 |
+
"grad_norm": 0.5990898013114929,
|
| 1162 |
+
"learning_rate": 5.924978503869304e-05,
|
| 1163 |
+
"loss": 0.2269,
|
| 1164 |
+
"step": 66500
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 28.5,
|
| 1168 |
+
"grad_norm": 0.5171389579772949,
|
| 1169 |
+
"learning_rate": 5.817712811693895e-05,
|
| 1170 |
+
"loss": 0.2281,
|
| 1171 |
+
"step": 67000
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 28.71,
|
| 1175 |
+
"grad_norm": 0.6478968262672424,
|
| 1176 |
+
"learning_rate": 5.710232158211523e-05,
|
| 1177 |
+
"loss": 0.23,
|
| 1178 |
+
"step": 67500
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 28.92,
|
| 1182 |
+
"grad_norm": 0.5454473495483398,
|
| 1183 |
+
"learning_rate": 5.602751504729149e-05,
|
| 1184 |
+
"loss": 0.224,
|
| 1185 |
+
"step": 68000
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 29.0,
|
| 1189 |
+
"eval_loss": 0.23835687339305878,
|
| 1190 |
+
"eval_runtime": 56.7685,
|
| 1191 |
+
"eval_samples_per_second": 331.452,
|
| 1192 |
+
"eval_steps_per_second": 2.589,
|
| 1193 |
+
"step": 68179
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 29.14,
|
| 1197 |
+
"grad_norm": 0.5909966826438904,
|
| 1198 |
+
"learning_rate": 5.4952708512467756e-05,
|
| 1199 |
+
"loss": 0.2237,
|
| 1200 |
+
"step": 68500
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 29.35,
|
| 1204 |
+
"grad_norm": 0.7489617466926575,
|
| 1205 |
+
"learning_rate": 5.387790197764403e-05,
|
| 1206 |
+
"loss": 0.2218,
|
| 1207 |
+
"step": 69000
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 29.56,
|
| 1211 |
+
"grad_norm": 0.5880895853042603,
|
| 1212 |
+
"learning_rate": 5.2803095442820296e-05,
|
| 1213 |
+
"loss": 0.2224,
|
| 1214 |
+
"step": 69500
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 29.77,
|
| 1218 |
+
"grad_norm": 0.6126420497894287,
|
| 1219 |
+
"learning_rate": 5.1728288907996556e-05,
|
| 1220 |
+
"loss": 0.2235,
|
| 1221 |
+
"step": 70000
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 29.99,
|
| 1225 |
+
"grad_norm": 0.7013407349586487,
|
| 1226 |
+
"learning_rate": 5.0653482373172836e-05,
|
| 1227 |
+
"loss": 0.2189,
|
| 1228 |
+
"step": 70500
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 30.0,
|
| 1232 |
+
"eval_loss": 0.23514553904533386,
|
| 1233 |
+
"eval_runtime": 56.6615,
|
| 1234 |
+
"eval_samples_per_second": 332.077,
|
| 1235 |
+
"eval_steps_per_second": 2.594,
|
| 1236 |
+
"step": 70530
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 30.2,
|
| 1240 |
+
"grad_norm": 0.6700057983398438,
|
| 1241 |
+
"learning_rate": 4.9578675838349096e-05,
|
| 1242 |
+
"loss": 0.2175,
|
| 1243 |
+
"step": 71000
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 30.41,
|
| 1247 |
+
"grad_norm": 0.5728330016136169,
|
| 1248 |
+
"learning_rate": 4.8506018916595015e-05,
|
| 1249 |
+
"loss": 0.2169,
|
| 1250 |
+
"step": 71500
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 30.63,
|
| 1254 |
+
"grad_norm": 0.5460176467895508,
|
| 1255 |
+
"learning_rate": 4.743121238177128e-05,
|
| 1256 |
+
"loss": 0.2159,
|
| 1257 |
+
"step": 72000
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 30.84,
|
| 1261 |
+
"grad_norm": 0.5714312791824341,
|
| 1262 |
+
"learning_rate": 4.63585554600172e-05,
|
| 1263 |
+
"loss": 0.2131,
|
| 1264 |
+
"step": 72500
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 31.0,
|
| 1268 |
+
"eval_loss": 0.23221717774868011,
|
| 1269 |
+
"eval_runtime": 56.8018,
|
| 1270 |
+
"eval_samples_per_second": 331.257,
|
| 1271 |
+
"eval_steps_per_second": 2.588,
|
| 1272 |
+
"step": 72881
|
| 1273 |
+
},
|
| 1274 |
+
{
|
| 1275 |
+
"epoch": 31.05,
|
| 1276 |
+
"grad_norm": 0.5165086984634399,
|
| 1277 |
+
"learning_rate": 4.528374892519347e-05,
|
| 1278 |
+
"loss": 0.2139,
|
| 1279 |
+
"step": 73000
|
| 1280 |
+
},
|
| 1281 |
+
{
|
| 1282 |
+
"epoch": 31.26,
|
| 1283 |
+
"grad_norm": 0.6647119522094727,
|
| 1284 |
+
"learning_rate": 4.4208942390369734e-05,
|
| 1285 |
+
"loss": 0.2098,
|
| 1286 |
+
"step": 73500
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 31.48,
|
| 1290 |
+
"grad_norm": 0.6132521629333496,
|
| 1291 |
+
"learning_rate": 4.3134135855546e-05,
|
| 1292 |
+
"loss": 0.213,
|
| 1293 |
+
"step": 74000
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 31.69,
|
| 1297 |
+
"grad_norm": 0.5501160025596619,
|
| 1298 |
+
"learning_rate": 4.2059329320722275e-05,
|
| 1299 |
+
"loss": 0.2111,
|
| 1300 |
+
"step": 74500
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 31.9,
|
| 1304 |
+
"grad_norm": 0.6153804659843445,
|
| 1305 |
+
"learning_rate": 4.098667239896819e-05,
|
| 1306 |
+
"loss": 0.2095,
|
| 1307 |
+
"step": 75000
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 32.0,
|
| 1311 |
+
"eval_loss": 0.22537025809288025,
|
| 1312 |
+
"eval_runtime": 56.6384,
|
| 1313 |
+
"eval_samples_per_second": 332.213,
|
| 1314 |
+
"eval_steps_per_second": 2.595,
|
| 1315 |
+
"step": 75232
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 32.11,
|
| 1319 |
+
"grad_norm": 0.5587669014930725,
|
| 1320 |
+
"learning_rate": 3.9911865864144454e-05,
|
| 1321 |
+
"loss": 0.2102,
|
| 1322 |
+
"step": 75500
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 32.33,
|
| 1326 |
+
"grad_norm": 0.6313532590866089,
|
| 1327 |
+
"learning_rate": 3.883705932932073e-05,
|
| 1328 |
+
"loss": 0.2066,
|
| 1329 |
+
"step": 76000
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 32.54,
|
| 1333 |
+
"grad_norm": 0.6371473670005798,
|
| 1334 |
+
"learning_rate": 3.7762252794496994e-05,
|
| 1335 |
+
"loss": 0.2067,
|
| 1336 |
+
"step": 76500
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 32.75,
|
| 1340 |
+
"grad_norm": 0.6398050785064697,
|
| 1341 |
+
"learning_rate": 3.668744625967326e-05,
|
| 1342 |
+
"loss": 0.2049,
|
| 1343 |
+
"step": 77000
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 32.96,
|
| 1347 |
+
"grad_norm": 0.6136685013771057,
|
| 1348 |
+
"learning_rate": 3.561263972484953e-05,
|
| 1349 |
+
"loss": 0.2072,
|
| 1350 |
+
"step": 77500
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 33.0,
|
| 1354 |
+
"eval_loss": 0.22324109077453613,
|
| 1355 |
+
"eval_runtime": 56.9404,
|
| 1356 |
+
"eval_samples_per_second": 330.451,
|
| 1357 |
+
"eval_steps_per_second": 2.582,
|
| 1358 |
+
"step": 77583
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 33.18,
|
| 1362 |
+
"grad_norm": 0.652153730392456,
|
| 1363 |
+
"learning_rate": 3.45378331900258e-05,
|
| 1364 |
+
"loss": 0.2024,
|
| 1365 |
+
"step": 78000
|
| 1366 |
+
},
|
| 1367 |
+
{
|
| 1368 |
+
"epoch": 33.39,
|
| 1369 |
+
"grad_norm": 0.5828467011451721,
|
| 1370 |
+
"learning_rate": 3.346302665520207e-05,
|
| 1371 |
+
"loss": 0.2008,
|
| 1372 |
+
"step": 78500
|
| 1373 |
+
},
|
| 1374 |
+
{
|
| 1375 |
+
"epoch": 33.6,
|
| 1376 |
+
"grad_norm": 0.5484930872917175,
|
| 1377 |
+
"learning_rate": 3.239036973344798e-05,
|
| 1378 |
+
"loss": 0.2032,
|
| 1379 |
+
"step": 79000
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 33.82,
|
| 1383 |
+
"grad_norm": 0.5094757676124573,
|
| 1384 |
+
"learning_rate": 3.1315563198624247e-05,
|
| 1385 |
+
"loss": 0.2012,
|
| 1386 |
+
"step": 79500
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"epoch": 34.0,
|
| 1390 |
+
"eval_loss": 0.22044067084789276,
|
| 1391 |
+
"eval_runtime": 56.7734,
|
| 1392 |
+
"eval_samples_per_second": 331.423,
|
| 1393 |
+
"eval_steps_per_second": 2.589,
|
| 1394 |
+
"step": 79934
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 34.03,
|
| 1398 |
+
"grad_norm": 0.6236255764961243,
|
| 1399 |
+
"learning_rate": 3.0240756663800517e-05,
|
| 1400 |
+
"loss": 0.1984,
|
| 1401 |
+
"step": 80000
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 34.24,
|
| 1405 |
+
"grad_norm": 0.6019966006278992,
|
| 1406 |
+
"learning_rate": 2.9165950128976787e-05,
|
| 1407 |
+
"loss": 0.1958,
|
| 1408 |
+
"step": 80500
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 34.45,
|
| 1412 |
+
"grad_norm": 0.6366199851036072,
|
| 1413 |
+
"learning_rate": 2.8093293207222702e-05,
|
| 1414 |
+
"loss": 0.1988,
|
| 1415 |
+
"step": 81000
|
| 1416 |
+
},
|
| 1417 |
+
{
|
| 1418 |
+
"epoch": 34.67,
|
| 1419 |
+
"grad_norm": 0.6756080389022827,
|
| 1420 |
+
"learning_rate": 2.701848667239897e-05,
|
| 1421 |
+
"loss": 0.1975,
|
| 1422 |
+
"step": 81500
|
| 1423 |
+
},
|
| 1424 |
+
{
|
| 1425 |
+
"epoch": 34.88,
|
| 1426 |
+
"grad_norm": 0.587624728679657,
|
| 1427 |
+
"learning_rate": 2.594368013757524e-05,
|
| 1428 |
+
"loss": 0.1938,
|
| 1429 |
+
"step": 82000
|
| 1430 |
+
},
|
| 1431 |
+
{
|
| 1432 |
+
"epoch": 35.0,
|
| 1433 |
+
"eval_loss": 0.21603567898273468,
|
| 1434 |
+
"eval_runtime": 56.8324,
|
| 1435 |
+
"eval_samples_per_second": 331.079,
|
| 1436 |
+
"eval_steps_per_second": 2.587,
|
| 1437 |
+
"step": 82285
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 35.09,
|
| 1441 |
+
"grad_norm": 0.5856329798698425,
|
| 1442 |
+
"learning_rate": 2.4868873602751506e-05,
|
| 1443 |
+
"loss": 0.193,
|
| 1444 |
+
"step": 82500
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 35.3,
|
| 1448 |
+
"grad_norm": 0.6045832633972168,
|
| 1449 |
+
"learning_rate": 2.3794067067927776e-05,
|
| 1450 |
+
"loss": 0.1936,
|
| 1451 |
+
"step": 83000
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 35.52,
|
| 1455 |
+
"grad_norm": 0.5652228593826294,
|
| 1456 |
+
"learning_rate": 2.2719260533104043e-05,
|
| 1457 |
+
"loss": 0.194,
|
| 1458 |
+
"step": 83500
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 35.73,
|
| 1462 |
+
"grad_norm": 0.6424867510795593,
|
| 1463 |
+
"learning_rate": 2.1644453998280313e-05,
|
| 1464 |
+
"loss": 0.1896,
|
| 1465 |
+
"step": 84000
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 35.94,
|
| 1469 |
+
"grad_norm": 0.5543941855430603,
|
| 1470 |
+
"learning_rate": 2.0571797076526225e-05,
|
| 1471 |
+
"loss": 0.1921,
|
| 1472 |
+
"step": 84500
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 36.0,
|
| 1476 |
+
"eval_loss": 0.21675606071949005,
|
| 1477 |
+
"eval_runtime": 56.9718,
|
| 1478 |
+
"eval_samples_per_second": 330.269,
|
| 1479 |
+
"eval_steps_per_second": 2.58,
|
| 1480 |
+
"step": 84636
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 36.15,
|
| 1484 |
+
"grad_norm": 0.6019295454025269,
|
| 1485 |
+
"learning_rate": 1.9496990541702495e-05,
|
| 1486 |
+
"loss": 0.1884,
|
| 1487 |
+
"step": 85000
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 36.37,
|
| 1491 |
+
"grad_norm": 0.645598292350769,
|
| 1492 |
+
"learning_rate": 1.8422184006878762e-05,
|
| 1493 |
+
"loss": 0.1888,
|
| 1494 |
+
"step": 85500
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 36.58,
|
| 1498 |
+
"grad_norm": 0.6292818188667297,
|
| 1499 |
+
"learning_rate": 1.7347377472055032e-05,
|
| 1500 |
+
"loss": 0.1886,
|
| 1501 |
+
"step": 86000
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 36.79,
|
| 1505 |
+
"grad_norm": 0.5169576406478882,
|
| 1506 |
+
"learning_rate": 1.62725709372313e-05,
|
| 1507 |
+
"loss": 0.1898,
|
| 1508 |
+
"step": 86500
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 37.0,
|
| 1512 |
+
"eval_loss": 0.21486429870128632,
|
| 1513 |
+
"eval_runtime": 56.8955,
|
| 1514 |
+
"eval_samples_per_second": 330.711,
|
| 1515 |
+
"eval_steps_per_second": 2.584,
|
| 1516 |
+
"step": 86987
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 37.01,
|
| 1520 |
+
"grad_norm": 0.680117130279541,
|
| 1521 |
+
"learning_rate": 1.5197764402407566e-05,
|
| 1522 |
+
"loss": 0.187,
|
| 1523 |
+
"step": 87000
|
| 1524 |
+
},
|
| 1525 |
+
{
|
| 1526 |
+
"epoch": 37.22,
|
| 1527 |
+
"grad_norm": 0.5357288718223572,
|
| 1528 |
+
"learning_rate": 1.4122957867583836e-05,
|
| 1529 |
+
"loss": 0.1841,
|
| 1530 |
+
"step": 87500
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"epoch": 37.43,
|
| 1534 |
+
"grad_norm": 0.5753670334815979,
|
| 1535 |
+
"learning_rate": 1.3048151332760103e-05,
|
| 1536 |
+
"loss": 0.1829,
|
| 1537 |
+
"step": 88000
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 37.64,
|
| 1541 |
+
"grad_norm": 0.654707133769989,
|
| 1542 |
+
"learning_rate": 1.1975494411006018e-05,
|
| 1543 |
+
"loss": 0.1857,
|
| 1544 |
+
"step": 88500
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 37.86,
|
| 1548 |
+
"grad_norm": 0.5702338218688965,
|
| 1549 |
+
"learning_rate": 1.0900687876182287e-05,
|
| 1550 |
+
"loss": 0.1829,
|
| 1551 |
+
"step": 89000
|
| 1552 |
+
},
|
| 1553 |
+
{
|
| 1554 |
+
"epoch": 38.0,
|
| 1555 |
+
"eval_loss": 0.21089451014995575,
|
| 1556 |
+
"eval_runtime": 56.6335,
|
| 1557 |
+
"eval_samples_per_second": 332.241,
|
| 1558 |
+
"eval_steps_per_second": 2.596,
|
| 1559 |
+
"step": 89338
|
| 1560 |
+
},
|
| 1561 |
+
{
|
| 1562 |
+
"epoch": 38.07,
|
| 1563 |
+
"grad_norm": 0.6671079993247986,
|
| 1564 |
+
"learning_rate": 9.825881341358555e-06,
|
| 1565 |
+
"loss": 0.1846,
|
| 1566 |
+
"step": 89500
|
| 1567 |
+
},
|
| 1568 |
+
{
|
| 1569 |
+
"epoch": 38.28,
|
| 1570 |
+
"grad_norm": 0.5165778398513794,
|
| 1571 |
+
"learning_rate": 8.751074806534824e-06,
|
| 1572 |
+
"loss": 0.1809,
|
| 1573 |
+
"step": 90000
|
| 1574 |
+
},
|
| 1575 |
+
{
|
| 1576 |
+
"epoch": 38.49,
|
| 1577 |
+
"grad_norm": 0.6374748945236206,
|
| 1578 |
+
"learning_rate": 7.676268271711092e-06,
|
| 1579 |
+
"loss": 0.1793,
|
| 1580 |
+
"step": 90500
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 38.71,
|
| 1584 |
+
"grad_norm": 0.5679296851158142,
|
| 1585 |
+
"learning_rate": 6.60146173688736e-06,
|
| 1586 |
+
"loss": 0.1804,
|
| 1587 |
+
"step": 91000
|
| 1588 |
+
},
|
| 1589 |
+
{
|
| 1590 |
+
"epoch": 38.92,
|
| 1591 |
+
"grad_norm": 0.5970684885978699,
|
| 1592 |
+
"learning_rate": 5.526655202063629e-06,
|
| 1593 |
+
"loss": 0.1804,
|
| 1594 |
+
"step": 91500
|
| 1595 |
+
},
|
| 1596 |
+
{
|
| 1597 |
+
"epoch": 39.0,
|
| 1598 |
+
"eval_loss": 0.20909027755260468,
|
| 1599 |
+
"eval_runtime": 56.346,
|
| 1600 |
+
"eval_samples_per_second": 333.937,
|
| 1601 |
+
"eval_steps_per_second": 2.609,
|
| 1602 |
+
"step": 91689
|
| 1603 |
+
},
|
| 1604 |
+
{
|
| 1605 |
+
"epoch": 39.13,
|
| 1606 |
+
"grad_norm": 0.5444430112838745,
|
| 1607 |
+
"learning_rate": 4.4539982803095445e-06,
|
| 1608 |
+
"loss": 0.1804,
|
| 1609 |
+
"step": 92000
|
| 1610 |
+
},
|
| 1611 |
+
{
|
| 1612 |
+
"epoch": 39.34,
|
| 1613 |
+
"grad_norm": 0.6086540818214417,
|
| 1614 |
+
"learning_rate": 3.379191745485813e-06,
|
| 1615 |
+
"loss": 0.1786,
|
| 1616 |
+
"step": 92500
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 39.56,
|
| 1620 |
+
"grad_norm": 0.5558965802192688,
|
| 1621 |
+
"learning_rate": 2.304385210662081e-06,
|
| 1622 |
+
"loss": 0.1758,
|
| 1623 |
+
"step": 93000
|
| 1624 |
+
},
|
| 1625 |
+
{
|
| 1626 |
+
"epoch": 39.77,
|
| 1627 |
+
"grad_norm": 0.5531004667282104,
|
| 1628 |
+
"learning_rate": 1.2295786758383493e-06,
|
| 1629 |
+
"loss": 0.1788,
|
| 1630 |
+
"step": 93500
|
| 1631 |
+
},
|
| 1632 |
+
{
|
| 1633 |
+
"epoch": 39.98,
|
| 1634 |
+
"grad_norm": 0.5611669421195984,
|
| 1635 |
+
"learning_rate": 1.5477214101461737e-07,
|
| 1636 |
+
"loss": 0.1773,
|
| 1637 |
+
"step": 94000
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"epoch": 40.0,
|
| 1641 |
+
"eval_loss": 0.20976465940475464,
|
| 1642 |
+
"eval_runtime": 56.6405,
|
| 1643 |
+
"eval_samples_per_second": 332.2,
|
| 1644 |
+
"eval_steps_per_second": 2.595,
|
| 1645 |
+
"step": 94040
|
| 1646 |
+
},
|
| 1647 |
+
{
|
| 1648 |
+
"epoch": 40.0,
|
| 1649 |
+
"step": 94040,
|
| 1650 |
+
"total_flos": 3.6650496018087936e+18,
|
| 1651 |
+
"train_loss": 0.2880888063569923,
|
| 1652 |
+
"train_runtime": 54838.0441,
|
| 1653 |
+
"train_samples_per_second": 109.795,
|
| 1654 |
+
"train_steps_per_second": 1.715
|
| 1655 |
+
}
|
| 1656 |
+
],
|
| 1657 |
+
"logging_steps": 500,
|
| 1658 |
+
"max_steps": 94040,
|
| 1659 |
+
"num_input_tokens_seen": 0,
|
| 1660 |
+
"num_train_epochs": 40,
|
| 1661 |
+
"save_steps": 500,
|
| 1662 |
+
"total_flos": 3.6650496018087936e+18,
|
| 1663 |
+
"train_batch_size": 64,
|
| 1664 |
+
"trial_name": null,
|
| 1665 |
+
"trial_params": null
|
| 1666 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6b92941a1784c12a7e4b21c69dc4e4328ac5fee472111ccb5486de55e9d9adb0
|
| 3 |
+
size 5048
|
utg4java_pretrain_bleu_metrics.txt
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
TESTING RESULTS:
|
| 2 |
+
|
| 3 |
+
Elapsed time: 600.7056782245636 seconds
|
| 4 |
+
Elapsed time (formatted): 00:10:00
|
| 5 |
+
BLEU: 0.6220592260360718
|
| 6 |
+
CodeBLEU: 0.7619660425681312
|
utg4java_pretrain_log_history.txt
ADDED
|
@@ -0,0 +1,230 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
Log History:
|
| 2 |
+
{"loss": 0.4973, "grad_norm": 1.1109575033187866, "learning_rate": 9.98e-05, "epoch": 0.21, "step": 500}
|
| 3 |
+
{"loss": 0.4793, "grad_norm": 1.1452946662902832, "learning_rate": 0.0001998, "epoch": 0.43, "step": 1000}
|
| 4 |
+
{"loss": 0.4999, "grad_norm": 0.9271925091743469, "learning_rate": 0.0001989273430782459, "epoch": 0.64, "step": 1500}
|
| 5 |
+
{"loss": 0.4934, "grad_norm": 0.9055812358856201, "learning_rate": 0.0001978525365434222, "epoch": 0.85, "step": 2000}
|
| 6 |
+
{"eval_loss": 0.4301895797252655, "eval_runtime": 56.7204, "eval_samples_per_second": 331.733, "eval_steps_per_second": 2.592, "epoch": 1.0, "step": 2351}
|
| 7 |
+
{"loss": 0.4818, "grad_norm": 0.8988948464393616, "learning_rate": 0.00019677773000859847, "epoch": 1.06, "step": 2500}
|
| 8 |
+
{"loss": 0.4651, "grad_norm": 0.9011508822441101, "learning_rate": 0.00019570292347377474, "epoch": 1.28, "step": 3000}
|
| 9 |
+
{"loss": 0.467, "grad_norm": 0.905346691608429, "learning_rate": 0.00019463026655202063, "epoch": 1.49, "step": 3500}
|
| 10 |
+
{"loss": 0.4625, "grad_norm": 0.7411425709724426, "learning_rate": 0.0001935597592433362, "epoch": 1.7, "step": 4000}
|
| 11 |
+
{"loss": 0.4572, "grad_norm": 0.8985588550567627, "learning_rate": 0.00019248495270851249, "epoch": 1.91, "step": 4500}
|
| 12 |
+
{"eval_loss": 0.4058227837085724, "eval_runtime": 56.7252, "eval_samples_per_second": 331.705, "eval_steps_per_second": 2.591, "epoch": 2.0, "step": 4702}
|
| 13 |
+
{"loss": 0.4472, "grad_norm": 0.827357292175293, "learning_rate": 0.00019141014617368875, "epoch": 2.13, "step": 5000}
|
| 14 |
+
{"loss": 0.4372, "grad_norm": 0.736473023891449, "learning_rate": 0.00019033533963886502, "epoch": 2.34, "step": 5500}
|
| 15 |
+
{"loss": 0.4362, "grad_norm": 0.7558779120445251, "learning_rate": 0.00018926053310404129, "epoch": 2.55, "step": 6000}
|
| 16 |
+
{"loss": 0.4332, "grad_norm": 0.7970576882362366, "learning_rate": 0.00018818572656921755, "epoch": 2.76, "step": 6500}
|
| 17 |
+
{"loss": 0.43, "grad_norm": 0.7692726850509644, "learning_rate": 0.00018711092003439382, "epoch": 2.98, "step": 7000}
|
| 18 |
+
{"eval_loss": 0.38832882046699524, "eval_runtime": 56.9476, "eval_samples_per_second": 330.409, "eval_steps_per_second": 2.581, "epoch": 3.0, "step": 7053}
|
| 19 |
+
{"loss": 0.4217, "grad_norm": 0.7109490633010864, "learning_rate": 0.00018603611349957009, "epoch": 3.19, "step": 7500}
|
| 20 |
+
{"loss": 0.4141, "grad_norm": 0.7125234603881836, "learning_rate": 0.00018496130696474635, "epoch": 3.4, "step": 8000}
|
| 21 |
+
{"loss": 0.4169, "grad_norm": 0.6914758682250977, "learning_rate": 0.00018388650042992262, "epoch": 3.62, "step": 8500}
|
| 22 |
+
{"loss": 0.4132, "grad_norm": 0.7215930223464966, "learning_rate": 0.00018281169389509889, "epoch": 3.83, "step": 9000}
|
| 23 |
+
{"eval_loss": 0.3813731372356415, "eval_runtime": 56.8812, "eval_samples_per_second": 330.794, "eval_steps_per_second": 2.584, "epoch": 4.0, "step": 9404}
|
| 24 |
+
{"loss": 0.4131, "grad_norm": 0.7283121943473816, "learning_rate": 0.00018173688736027515, "epoch": 4.04, "step": 9500}
|
| 25 |
+
{"loss": 0.4059, "grad_norm": 0.6947267055511475, "learning_rate": 0.00018066208082545142, "epoch": 4.25, "step": 10000}
|
| 26 |
+
{"loss": 0.4023, "grad_norm": 0.7544755339622498, "learning_rate": 0.0001795872742906277, "epoch": 4.47, "step": 10500}
|
| 27 |
+
{"loss": 0.4038, "grad_norm": 0.6204919815063477, "learning_rate": 0.00017851246775580398, "epoch": 4.68, "step": 11000}
|
| 28 |
+
{"loss": 0.4017, "grad_norm": 0.7036948800086975, "learning_rate": 0.00017743766122098022, "epoch": 4.89, "step": 11500}
|
| 29 |
+
{"eval_loss": 0.3638169765472412, "eval_runtime": 56.8241, "eval_samples_per_second": 331.127, "eval_steps_per_second": 2.587, "epoch": 5.0, "step": 11755}
|
| 30 |
+
{"loss": 0.3921, "grad_norm": 0.62198805809021, "learning_rate": 0.0001763628546861565, "epoch": 5.1, "step": 12000}
|
| 31 |
+
{"loss": 0.3901, "grad_norm": 0.6095067858695984, "learning_rate": 0.00017528804815133278, "epoch": 5.32, "step": 12500}
|
| 32 |
+
{"loss": 0.3888, "grad_norm": 1.0518782138824463, "learning_rate": 0.00017421324161650902, "epoch": 5.53, "step": 13000}
|
| 33 |
+
{"loss": 0.3859, "grad_norm": 0.6801586151123047, "learning_rate": 0.00017313843508168531, "epoch": 5.74, "step": 13500}
|
| 34 |
+
{"loss": 0.3839, "grad_norm": 0.6489918828010559, "learning_rate": 0.00017206362854686158, "epoch": 5.95, "step": 14000}
|
| 35 |
+
{"eval_loss": 0.35733696818351746, "eval_runtime": 56.779, "eval_samples_per_second": 331.39, "eval_steps_per_second": 2.589, "epoch": 6.0, "step": 14106}
|
| 36 |
+
{"loss": 0.3782, "grad_norm": 0.6174113154411316, "learning_rate": 0.00017098882201203782, "epoch": 6.17, "step": 14500}
|
| 37 |
+
{"loss": 0.3792, "grad_norm": 0.641315221786499, "learning_rate": 0.00016991401547721411, "epoch": 6.38, "step": 15000}
|
| 38 |
+
{"loss": 0.3745, "grad_norm": 0.6454831957817078, "learning_rate": 0.00016883920894239038, "epoch": 6.59, "step": 15500}
|
| 39 |
+
{"loss": 0.3729, "grad_norm": 0.6489464640617371, "learning_rate": 0.00016776440240756665, "epoch": 6.81, "step": 16000}
|
| 40 |
+
{"eval_loss": 0.3489345610141754, "eval_runtime": 57.0727, "eval_samples_per_second": 329.685, "eval_steps_per_second": 2.576, "epoch": 7.0, "step": 16457}
|
| 41 |
+
{"loss": 0.3718, "grad_norm": 0.6929277181625366, "learning_rate": 0.0001666938950988822, "epoch": 7.02, "step": 16500}
|
| 42 |
+
{"loss": 0.3624, "grad_norm": 0.5726625323295593, "learning_rate": 0.00016562123817712813, "epoch": 7.23, "step": 17000}
|
| 43 |
+
{"loss": 0.3664, "grad_norm": 0.603529155254364, "learning_rate": 0.0001645464316423044, "epoch": 7.44, "step": 17500}
|
| 44 |
+
{"loss": 0.3614, "grad_norm": 0.7436888813972473, "learning_rate": 0.00016347162510748066, "epoch": 7.66, "step": 18000}
|
| 45 |
+
{"loss": 0.3646, "grad_norm": 0.6405676007270813, "learning_rate": 0.00016239681857265693, "epoch": 7.87, "step": 18500}
|
| 46 |
+
{"eval_loss": 0.3384884297847748, "eval_runtime": 56.7342, "eval_samples_per_second": 331.652, "eval_steps_per_second": 2.591, "epoch": 8.0, "step": 18808}
|
| 47 |
+
{"loss": 0.3623, "grad_norm": 0.660517156124115, "learning_rate": 0.0001613220120378332, "epoch": 8.08, "step": 19000}
|
| 48 |
+
{"loss": 0.3518, "grad_norm": 0.6352570056915283, "learning_rate": 0.0001602493551160791, "epoch": 8.29, "step": 19500}
|
| 49 |
+
{"loss": 0.3569, "grad_norm": 0.6094992756843567, "learning_rate": 0.0001591745485812554, "epoch": 8.51, "step": 20000}
|
| 50 |
+
{"loss": 0.3567, "grad_norm": 0.6646167635917664, "learning_rate": 0.00015809974204643166, "epoch": 8.72, "step": 20500}
|
| 51 |
+
{"loss": 0.353, "grad_norm": 0.6240518093109131, "learning_rate": 0.00015702493551160792, "epoch": 8.93, "step": 21000}
|
| 52 |
+
{"eval_loss": 0.3339434862136841, "eval_runtime": 56.979, "eval_samples_per_second": 330.227, "eval_steps_per_second": 2.58, "epoch": 9.0, "step": 21159}
|
| 53 |
+
{"loss": 0.3491, "grad_norm": 0.6990819573402405, "learning_rate": 0.0001559501289767842, "epoch": 9.15, "step": 21500}
|
| 54 |
+
{"loss": 0.346, "grad_norm": 0.6751831769943237, "learning_rate": 0.00015487532244196046, "epoch": 9.36, "step": 22000}
|
| 55 |
+
{"loss": 0.3488, "grad_norm": 0.615511417388916, "learning_rate": 0.00015380051590713672, "epoch": 9.57, "step": 22500}
|
| 56 |
+
{"loss": 0.3466, "grad_norm": 0.5277218222618103, "learning_rate": 0.000152725709372313, "epoch": 9.78, "step": 23000}
|
| 57 |
+
{"loss": 0.3441, "grad_norm": 0.5809108018875122, "learning_rate": 0.00015165090283748926, "epoch": 10.0, "step": 23500}
|
| 58 |
+
{"eval_loss": 0.324989378452301, "eval_runtime": 56.9322, "eval_samples_per_second": 330.499, "eval_steps_per_second": 2.582, "epoch": 10.0, "step": 23510}
|
| 59 |
+
{"loss": 0.3407, "grad_norm": 0.5285485982894897, "learning_rate": 0.00015058039552880484, "epoch": 10.21, "step": 24000}
|
| 60 |
+
{"loss": 0.3389, "grad_norm": 0.6268051862716675, "learning_rate": 0.00014950558899398108, "epoch": 10.42, "step": 24500}
|
| 61 |
+
{"loss": 0.3387, "grad_norm": 0.5879684090614319, "learning_rate": 0.00014843078245915735, "epoch": 10.63, "step": 25000}
|
| 62 |
+
{"loss": 0.3379, "grad_norm": 0.5434576869010925, "learning_rate": 0.00014735597592433364, "epoch": 10.85, "step": 25500}
|
| 63 |
+
{"eval_loss": 0.31874018907546997, "eval_runtime": 57.0459, "eval_samples_per_second": 329.84, "eval_steps_per_second": 2.577, "epoch": 11.0, "step": 25861}
|
| 64 |
+
{"loss": 0.3369, "grad_norm": 0.6547732949256897, "learning_rate": 0.00014628116938950988, "epoch": 11.06, "step": 26000}
|
| 65 |
+
{"loss": 0.3308, "grad_norm": 0.648601233959198, "learning_rate": 0.00014520636285468615, "epoch": 11.27, "step": 26500}
|
| 66 |
+
{"loss": 0.3298, "grad_norm": 0.6311826109886169, "learning_rate": 0.00014413155631986244, "epoch": 11.48, "step": 27000}
|
| 67 |
+
{"loss": 0.3292, "grad_norm": 0.5571895837783813, "learning_rate": 0.00014305674978503868, "epoch": 11.7, "step": 27500}
|
| 68 |
+
{"loss": 0.3284, "grad_norm": 0.5373964309692383, "learning_rate": 0.00014198194325021498, "epoch": 11.91, "step": 28000}
|
| 69 |
+
{"eval_loss": 0.3123379051685333, "eval_runtime": 56.7858, "eval_samples_per_second": 331.351, "eval_steps_per_second": 2.589, "epoch": 12.0, "step": 28212}
|
| 70 |
+
{"loss": 0.3247, "grad_norm": 0.5307437777519226, "learning_rate": 0.00014090928632846088, "epoch": 12.12, "step": 28500}
|
| 71 |
+
{"loss": 0.3223, "grad_norm": 0.6422705054283142, "learning_rate": 0.00013983447979363714, "epoch": 12.34, "step": 29000}
|
| 72 |
+
{"loss": 0.3223, "grad_norm": 0.6680347323417664, "learning_rate": 0.00013875967325881344, "epoch": 12.55, "step": 29500}
|
| 73 |
+
{"loss": 0.3245, "grad_norm": 0.5851155519485474, "learning_rate": 0.00013768701633705933, "epoch": 12.76, "step": 30000}
|
| 74 |
+
{"loss": 0.3252, "grad_norm": 0.544624924659729, "learning_rate": 0.0001366122098022356, "epoch": 12.97, "step": 30500}
|
| 75 |
+
{"eval_loss": 0.3076089322566986, "eval_runtime": 57.0752, "eval_samples_per_second": 329.67, "eval_steps_per_second": 2.576, "epoch": 13.0, "step": 30563}
|
| 76 |
+
{"loss": 0.3185, "grad_norm": 0.5343388319015503, "learning_rate": 0.00013553740326741187, "epoch": 13.19, "step": 31000}
|
| 77 |
+
{"loss": 0.3191, "grad_norm": 0.6074999570846558, "learning_rate": 0.00013446259673258814, "epoch": 13.4, "step": 31500}
|
| 78 |
+
{"loss": 0.3186, "grad_norm": 0.6202041506767273, "learning_rate": 0.00013338779019776443, "epoch": 13.61, "step": 32000}
|
| 79 |
+
{"loss": 0.3134, "grad_norm": 0.6058717966079712, "learning_rate": 0.00013231298366294067, "epoch": 13.82, "step": 32500}
|
| 80 |
+
{"eval_loss": 0.3018127679824829, "eval_runtime": 57.0557, "eval_samples_per_second": 329.783, "eval_steps_per_second": 2.576, "epoch": 14.0, "step": 32914}
|
| 81 |
+
{"loss": 0.3161, "grad_norm": 0.6439224481582642, "learning_rate": 0.00013123817712811694, "epoch": 14.04, "step": 33000}
|
| 82 |
+
{"loss": 0.3094, "grad_norm": 0.5888046622276306, "learning_rate": 0.00013016337059329323, "epoch": 14.25, "step": 33500}
|
| 83 |
+
{"loss": 0.3092, "grad_norm": 0.6644863486289978, "learning_rate": 0.00012908856405846947, "epoch": 14.46, "step": 34000}
|
| 84 |
+
{"loss": 0.3117, "grad_norm": 0.5649863481521606, "learning_rate": 0.00012801375752364576, "epoch": 14.67, "step": 34500}
|
| 85 |
+
{"loss": 0.3072, "grad_norm": 0.4947267174720764, "learning_rate": 0.00012693895098882203, "epoch": 14.89, "step": 35000}
|
| 86 |
+
{"eval_loss": 0.29680636525154114, "eval_runtime": 57.044, "eval_samples_per_second": 329.851, "eval_steps_per_second": 2.577, "epoch": 15.0, "step": 35265}
|
| 87 |
+
{"loss": 0.3067, "grad_norm": 0.6015808582305908, "learning_rate": 0.00012586414445399827, "epoch": 15.1, "step": 35500}
|
| 88 |
+
{"loss": 0.3027, "grad_norm": 0.6003320217132568, "learning_rate": 0.00012478933791917456, "epoch": 15.31, "step": 36000}
|
| 89 |
+
{"loss": 0.3012, "grad_norm": 0.5643883347511292, "learning_rate": 0.00012371453138435083, "epoch": 15.53, "step": 36500}
|
| 90 |
+
{"loss": 0.303, "grad_norm": 0.5768193602561951, "learning_rate": 0.00012264187446259673, "epoch": 15.74, "step": 37000}
|
| 91 |
+
{"loss": 0.3034, "grad_norm": 0.5599430799484253, "learning_rate": 0.00012156921754084265, "epoch": 15.95, "step": 37500}
|
| 92 |
+
{"eval_loss": 0.2938060164451599, "eval_runtime": 56.7422, "eval_samples_per_second": 331.605, "eval_steps_per_second": 2.591, "epoch": 16.0, "step": 37616}
|
| 93 |
+
{"loss": 0.2987, "grad_norm": 0.6380453705787659, "learning_rate": 0.00012049441100601892, "epoch": 16.16, "step": 38000}
|
| 94 |
+
{"loss": 0.2975, "grad_norm": 0.5312320590019226, "learning_rate": 0.0001194196044711952, "epoch": 16.38, "step": 38500}
|
| 95 |
+
{"loss": 0.2934, "grad_norm": 0.7027547955513, "learning_rate": 0.00011834479793637145, "epoch": 16.59, "step": 39000}
|
| 96 |
+
{"loss": 0.2959, "grad_norm": 0.5144539475440979, "learning_rate": 0.00011726999140154773, "epoch": 16.8, "step": 39500}
|
| 97 |
+
{"eval_loss": 0.28754693269729614, "eval_runtime": 56.9555, "eval_samples_per_second": 330.363, "eval_steps_per_second": 2.581, "epoch": 17.0, "step": 39967}
|
| 98 |
+
{"loss": 0.2997, "grad_norm": 0.5759513974189758, "learning_rate": 0.00011619733447979365, "epoch": 17.01, "step": 40000}
|
| 99 |
+
{"loss": 0.294, "grad_norm": 0.593640148639679, "learning_rate": 0.00011512467755803955, "epoch": 17.23, "step": 40500}
|
| 100 |
+
{"loss": 0.2904, "grad_norm": 0.6821351647377014, "learning_rate": 0.00011404987102321583, "epoch": 17.44, "step": 41000}
|
| 101 |
+
{"loss": 0.2896, "grad_norm": 0.5297681093215942, "learning_rate": 0.0001129750644883921, "epoch": 17.65, "step": 41500}
|
| 102 |
+
{"loss": 0.2884, "grad_norm": 0.5864290595054626, "learning_rate": 0.00011190025795356836, "epoch": 17.86, "step": 42000}
|
| 103 |
+
{"eval_loss": 0.28015270829200745, "eval_runtime": 56.8879, "eval_samples_per_second": 330.756, "eval_steps_per_second": 2.584, "epoch": 18.0, "step": 42318}
|
| 104 |
+
{"loss": 0.2874, "grad_norm": 0.5764068961143494, "learning_rate": 0.00011082545141874463, "epoch": 18.08, "step": 42500}
|
| 105 |
+
{"loss": 0.2828, "grad_norm": 0.6073163747787476, "learning_rate": 0.0001097506448839209, "epoch": 18.29, "step": 43000}
|
| 106 |
+
{"loss": 0.2848, "grad_norm": 0.5691092610359192, "learning_rate": 0.00010867583834909716, "epoch": 18.5, "step": 43500}
|
| 107 |
+
{"loss": 0.2861, "grad_norm": 0.5399264097213745, "learning_rate": 0.00010760103181427344, "epoch": 18.72, "step": 44000}
|
| 108 |
+
{"loss": 0.2839, "grad_norm": 0.6221365928649902, "learning_rate": 0.0001065262252794497, "epoch": 18.93, "step": 44500}
|
| 109 |
+
{"eval_loss": 0.27618157863616943, "eval_runtime": 56.9107, "eval_samples_per_second": 330.623, "eval_steps_per_second": 2.583, "epoch": 19.0, "step": 44669}
|
| 110 |
+
{"loss": 0.2791, "grad_norm": 0.5080223679542542, "learning_rate": 0.00010545141874462596, "epoch": 19.14, "step": 45000}
|
| 111 |
+
{"loss": 0.2792, "grad_norm": 0.567764163017273, "learning_rate": 0.00010437661220980224, "epoch": 19.35, "step": 45500}
|
| 112 |
+
{"loss": 0.2768, "grad_norm": 0.5838685035705566, "learning_rate": 0.00010330180567497851, "epoch": 19.57, "step": 46000}
|
| 113 |
+
{"loss": 0.2776, "grad_norm": 0.5778998136520386, "learning_rate": 0.00010222699914015476, "epoch": 19.78, "step": 46500}
|
| 114 |
+
{"loss": 0.2785, "grad_norm": 0.615337073802948, "learning_rate": 0.0001011543422184007, "epoch": 19.99, "step": 47000}
|
| 115 |
+
{"eval_loss": 0.2724262773990631, "eval_runtime": 56.852, "eval_samples_per_second": 330.965, "eval_steps_per_second": 2.586, "epoch": 20.0, "step": 47020}
|
| 116 |
+
{"loss": 0.2712, "grad_norm": 0.6153652667999268, "learning_rate": 0.00010007953568357695, "epoch": 20.2, "step": 47500}
|
| 117 |
+
{"loss": 0.2718, "grad_norm": 0.6126906275749207, "learning_rate": 9.900472914875323e-05, "epoch": 20.42, "step": 48000}
|
| 118 |
+
{"loss": 0.2701, "grad_norm": 0.5799471139907837, "learning_rate": 9.792992261392949e-05, "epoch": 20.63, "step": 48500}
|
| 119 |
+
{"loss": 0.2727, "grad_norm": 0.5679476261138916, "learning_rate": 9.685511607910577e-05, "epoch": 20.84, "step": 49000}
|
| 120 |
+
{"eval_loss": 0.26610955595970154, "eval_runtime": 56.7263, "eval_samples_per_second": 331.698, "eval_steps_per_second": 2.591, "epoch": 21.0, "step": 49371}
|
| 121 |
+
{"loss": 0.2694, "grad_norm": 0.5923852324485779, "learning_rate": 9.578245915735168e-05, "epoch": 21.05, "step": 49500}
|
| 122 |
+
{"loss": 0.2649, "grad_norm": 0.6184679269790649, "learning_rate": 9.470765262252796e-05, "epoch": 21.27, "step": 50000}
|
| 123 |
+
{"loss": 0.2697, "grad_norm": 0.5939807295799255, "learning_rate": 9.363284608770421e-05, "epoch": 21.48, "step": 50500}
|
| 124 |
+
{"loss": 0.2652, "grad_norm": 0.5925255417823792, "learning_rate": 9.255803955288048e-05, "epoch": 21.69, "step": 51000}
|
| 125 |
+
{"loss": 0.266, "grad_norm": 0.5666438341140747, "learning_rate": 9.148323301805676e-05, "epoch": 21.91, "step": 51500}
|
| 126 |
+
{"eval_loss": 0.266626238822937, "eval_runtime": 56.6361, "eval_samples_per_second": 332.226, "eval_steps_per_second": 2.596, "epoch": 22.0, "step": 51722}
|
| 127 |
+
{"loss": 0.2613, "grad_norm": 0.5490565299987793, "learning_rate": 9.040842648323303e-05, "epoch": 22.12, "step": 52000}
|
| 128 |
+
{"loss": 0.2592, "grad_norm": 0.6534095406532288, "learning_rate": 8.933361994840929e-05, "epoch": 22.33, "step": 52500}
|
| 129 |
+
{"loss": 0.2604, "grad_norm": 0.568953812122345, "learning_rate": 8.82609630266552e-05, "epoch": 22.54, "step": 53000}
|
| 130 |
+
{"loss": 0.2602, "grad_norm": 0.5968701243400574, "learning_rate": 8.718615649183147e-05, "epoch": 22.76, "step": 53500}
|
| 131 |
+
{"loss": 0.2614, "grad_norm": 0.5995833277702332, "learning_rate": 8.61134995700774e-05, "epoch": 22.97, "step": 54000}
|
| 132 |
+
{"eval_loss": 0.26461803913116455, "eval_runtime": 56.8736, "eval_samples_per_second": 330.839, "eval_steps_per_second": 2.585, "epoch": 23.0, "step": 54073}
|
| 133 |
+
{"loss": 0.2502, "grad_norm": 0.577324390411377, "learning_rate": 8.503869303525366e-05, "epoch": 23.18, "step": 54500}
|
| 134 |
+
{"loss": 0.256, "grad_norm": 0.631363034248352, "learning_rate": 8.396603611349956e-05, "epoch": 23.39, "step": 55000}
|
| 135 |
+
{"loss": 0.2559, "grad_norm": 0.5862709879875183, "learning_rate": 8.289122957867584e-05, "epoch": 23.61, "step": 55500}
|
| 136 |
+
{"loss": 0.2549, "grad_norm": 0.5620314478874207, "learning_rate": 8.181857265692176e-05, "epoch": 23.82, "step": 56000}
|
| 137 |
+
{"eval_loss": 0.2593855559825897, "eval_runtime": 57.337, "eval_samples_per_second": 328.165, "eval_steps_per_second": 2.564, "epoch": 24.0, "step": 56424}
|
| 138 |
+
{"loss": 0.2542, "grad_norm": 0.5514592528343201, "learning_rate": 8.074376612209804e-05, "epoch": 24.03, "step": 56500}
|
| 139 |
+
{"loss": 0.2483, "grad_norm": 0.6351886987686157, "learning_rate": 7.966895958727429e-05, "epoch": 24.25, "step": 57000}
|
| 140 |
+
{"loss": 0.2503, "grad_norm": 0.6075631380081177, "learning_rate": 7.859415305245056e-05, "epoch": 24.46, "step": 57500}
|
| 141 |
+
{"loss": 0.249, "grad_norm": 0.5592435002326965, "learning_rate": 7.751934651762684e-05, "epoch": 24.67, "step": 58000}
|
| 142 |
+
{"loss": 0.2497, "grad_norm": 0.5757043957710266, "learning_rate": 7.64445399828031e-05, "epoch": 24.88, "step": 58500}
|
| 143 |
+
{"eval_loss": 0.2535327076911926, "eval_runtime": 56.789, "eval_samples_per_second": 331.332, "eval_steps_per_second": 2.589, "epoch": 25.0, "step": 58775}
|
| 144 |
+
{"loss": 0.245, "grad_norm": 0.5995878577232361, "learning_rate": 7.536973344797937e-05, "epoch": 25.1, "step": 59000}
|
| 145 |
+
{"loss": 0.2461, "grad_norm": 0.6077148914337158, "learning_rate": 7.429492691315564e-05, "epoch": 25.31, "step": 59500}
|
| 146 |
+
{"loss": 0.2433, "grad_norm": 0.5592058300971985, "learning_rate": 7.32201203783319e-05, "epoch": 25.52, "step": 60000}
|
| 147 |
+
{"loss": 0.2444, "grad_norm": 0.6243628859519958, "learning_rate": 7.214531384350817e-05, "epoch": 25.73, "step": 60500}
|
| 148 |
+
{"loss": 0.243, "grad_norm": 0.6848371624946594, "learning_rate": 7.107050730868444e-05, "epoch": 25.95, "step": 61000}
|
| 149 |
+
{"eval_loss": 0.250180721282959, "eval_runtime": 56.9862, "eval_samples_per_second": 330.185, "eval_steps_per_second": 2.58, "epoch": 26.0, "step": 61126}
|
| 150 |
+
{"loss": 0.2382, "grad_norm": 0.5177180171012878, "learning_rate": 6.99957007738607e-05, "epoch": 26.16, "step": 61500}
|
| 151 |
+
{"loss": 0.2349, "grad_norm": 0.6144821047782898, "learning_rate": 6.892304385210663e-05, "epoch": 26.37, "step": 62000}
|
| 152 |
+
{"loss": 0.2385, "grad_norm": 0.5862034559249878, "learning_rate": 6.78482373172829e-05, "epoch": 26.58, "step": 62500}
|
| 153 |
+
{"loss": 0.2383, "grad_norm": 0.5799028277397156, "learning_rate": 6.677343078245916e-05, "epoch": 26.8, "step": 63000}
|
| 154 |
+
{"eval_loss": 0.24638937413692474, "eval_runtime": 56.9276, "eval_samples_per_second": 330.525, "eval_steps_per_second": 2.582, "epoch": 27.0, "step": 63477}
|
| 155 |
+
{"loss": 0.2375, "grad_norm": 0.6466034054756165, "learning_rate": 6.569862424763543e-05, "epoch": 27.01, "step": 63500}
|
| 156 |
+
{"loss": 0.2337, "grad_norm": 0.6161571741104126, "learning_rate": 6.46238177128117e-05, "epoch": 27.22, "step": 64000}
|
| 157 |
+
{"loss": 0.2328, "grad_norm": 0.6162907481193542, "learning_rate": 6.354901117798796e-05, "epoch": 27.44, "step": 64500}
|
| 158 |
+
{"loss": 0.2316, "grad_norm": 0.6643475890159607, "learning_rate": 6.247420464316423e-05, "epoch": 27.65, "step": 65000}
|
| 159 |
+
{"loss": 0.2325, "grad_norm": 0.5605382323265076, "learning_rate": 6.13993981083405e-05, "epoch": 27.86, "step": 65500}
|
| 160 |
+
{"eval_loss": 0.24465161561965942, "eval_runtime": 57.0032, "eval_samples_per_second": 330.087, "eval_steps_per_second": 2.579, "epoch": 28.0, "step": 65828}
|
| 161 |
+
{"loss": 0.231, "grad_norm": 0.6108579635620117, "learning_rate": 6.032459157351676e-05, "epoch": 28.07, "step": 66000}
|
| 162 |
+
{"loss": 0.2269, "grad_norm": 0.5990898013114929, "learning_rate": 5.924978503869304e-05, "epoch": 28.29, "step": 66500}
|
| 163 |
+
{"loss": 0.2281, "grad_norm": 0.5171389579772949, "learning_rate": 5.817712811693895e-05, "epoch": 28.5, "step": 67000}
|
| 164 |
+
{"loss": 0.23, "grad_norm": 0.6478968262672424, "learning_rate": 5.710232158211523e-05, "epoch": 28.71, "step": 67500}
|
| 165 |
+
{"loss": 0.224, "grad_norm": 0.5454473495483398, "learning_rate": 5.602751504729149e-05, "epoch": 28.92, "step": 68000}
|
| 166 |
+
{"eval_loss": 0.23835687339305878, "eval_runtime": 56.7685, "eval_samples_per_second": 331.452, "eval_steps_per_second": 2.589, "epoch": 29.0, "step": 68179}
|
| 167 |
+
{"loss": 0.2237, "grad_norm": 0.5909966826438904, "learning_rate": 5.4952708512467756e-05, "epoch": 29.14, "step": 68500}
|
| 168 |
+
{"loss": 0.2218, "grad_norm": 0.7489617466926575, "learning_rate": 5.387790197764403e-05, "epoch": 29.35, "step": 69000}
|
| 169 |
+
{"loss": 0.2224, "grad_norm": 0.5880895853042603, "learning_rate": 5.2803095442820296e-05, "epoch": 29.56, "step": 69500}
|
| 170 |
+
{"loss": 0.2235, "grad_norm": 0.6126420497894287, "learning_rate": 5.1728288907996556e-05, "epoch": 29.77, "step": 70000}
|
| 171 |
+
{"loss": 0.2189, "grad_norm": 0.7013407349586487, "learning_rate": 5.0653482373172836e-05, "epoch": 29.99, "step": 70500}
|
| 172 |
+
{"eval_loss": 0.23514553904533386, "eval_runtime": 56.6615, "eval_samples_per_second": 332.077, "eval_steps_per_second": 2.594, "epoch": 30.0, "step": 70530}
|
| 173 |
+
{"loss": 0.2175, "grad_norm": 0.6700057983398438, "learning_rate": 4.9578675838349096e-05, "epoch": 30.2, "step": 71000}
|
| 174 |
+
{"loss": 0.2169, "grad_norm": 0.5728330016136169, "learning_rate": 4.8506018916595015e-05, "epoch": 30.41, "step": 71500}
|
| 175 |
+
{"loss": 0.2159, "grad_norm": 0.5460176467895508, "learning_rate": 4.743121238177128e-05, "epoch": 30.63, "step": 72000}
|
| 176 |
+
{"loss": 0.2131, "grad_norm": 0.5714312791824341, "learning_rate": 4.63585554600172e-05, "epoch": 30.84, "step": 72500}
|
| 177 |
+
{"eval_loss": 0.23221717774868011, "eval_runtime": 56.8018, "eval_samples_per_second": 331.257, "eval_steps_per_second": 2.588, "epoch": 31.0, "step": 72881}
|
| 178 |
+
{"loss": 0.2139, "grad_norm": 0.5165086984634399, "learning_rate": 4.528374892519347e-05, "epoch": 31.05, "step": 73000}
|
| 179 |
+
{"loss": 0.2098, "grad_norm": 0.6647119522094727, "learning_rate": 4.4208942390369734e-05, "epoch": 31.26, "step": 73500}
|
| 180 |
+
{"loss": 0.213, "grad_norm": 0.6132521629333496, "learning_rate": 4.3134135855546e-05, "epoch": 31.48, "step": 74000}
|
| 181 |
+
{"loss": 0.2111, "grad_norm": 0.5501160025596619, "learning_rate": 4.2059329320722275e-05, "epoch": 31.69, "step": 74500}
|
| 182 |
+
{"loss": 0.2095, "grad_norm": 0.6153804659843445, "learning_rate": 4.098667239896819e-05, "epoch": 31.9, "step": 75000}
|
| 183 |
+
{"eval_loss": 0.22537025809288025, "eval_runtime": 56.6384, "eval_samples_per_second": 332.213, "eval_steps_per_second": 2.595, "epoch": 32.0, "step": 75232}
|
| 184 |
+
{"loss": 0.2102, "grad_norm": 0.5587669014930725, "learning_rate": 3.9911865864144454e-05, "epoch": 32.11, "step": 75500}
|
| 185 |
+
{"loss": 0.2066, "grad_norm": 0.6313532590866089, "learning_rate": 3.883705932932073e-05, "epoch": 32.33, "step": 76000}
|
| 186 |
+
{"loss": 0.2067, "grad_norm": 0.6371473670005798, "learning_rate": 3.7762252794496994e-05, "epoch": 32.54, "step": 76500}
|
| 187 |
+
{"loss": 0.2049, "grad_norm": 0.6398050785064697, "learning_rate": 3.668744625967326e-05, "epoch": 32.75, "step": 77000}
|
| 188 |
+
{"loss": 0.2072, "grad_norm": 0.6136685013771057, "learning_rate": 3.561263972484953e-05, "epoch": 32.96, "step": 77500}
|
| 189 |
+
{"eval_loss": 0.22324109077453613, "eval_runtime": 56.9404, "eval_samples_per_second": 330.451, "eval_steps_per_second": 2.582, "epoch": 33.0, "step": 77583}
|
| 190 |
+
{"loss": 0.2024, "grad_norm": 0.652153730392456, "learning_rate": 3.45378331900258e-05, "epoch": 33.18, "step": 78000}
|
| 191 |
+
{"loss": 0.2008, "grad_norm": 0.5828467011451721, "learning_rate": 3.346302665520207e-05, "epoch": 33.39, "step": 78500}
|
| 192 |
+
{"loss": 0.2032, "grad_norm": 0.5484930872917175, "learning_rate": 3.239036973344798e-05, "epoch": 33.6, "step": 79000}
|
| 193 |
+
{"loss": 0.2012, "grad_norm": 0.5094757676124573, "learning_rate": 3.1315563198624247e-05, "epoch": 33.82, "step": 79500}
|
| 194 |
+
{"eval_loss": 0.22044067084789276, "eval_runtime": 56.7734, "eval_samples_per_second": 331.423, "eval_steps_per_second": 2.589, "epoch": 34.0, "step": 79934}
|
| 195 |
+
{"loss": 0.1984, "grad_norm": 0.6236255764961243, "learning_rate": 3.0240756663800517e-05, "epoch": 34.03, "step": 80000}
|
| 196 |
+
{"loss": 0.1958, "grad_norm": 0.6019966006278992, "learning_rate": 2.9165950128976787e-05, "epoch": 34.24, "step": 80500}
|
| 197 |
+
{"loss": 0.1988, "grad_norm": 0.6366199851036072, "learning_rate": 2.8093293207222702e-05, "epoch": 34.45, "step": 81000}
|
| 198 |
+
{"loss": 0.1975, "grad_norm": 0.6756080389022827, "learning_rate": 2.701848667239897e-05, "epoch": 34.67, "step": 81500}
|
| 199 |
+
{"loss": 0.1938, "grad_norm": 0.587624728679657, "learning_rate": 2.594368013757524e-05, "epoch": 34.88, "step": 82000}
|
| 200 |
+
{"eval_loss": 0.21603567898273468, "eval_runtime": 56.8324, "eval_samples_per_second": 331.079, "eval_steps_per_second": 2.587, "epoch": 35.0, "step": 82285}
|
| 201 |
+
{"loss": 0.193, "grad_norm": 0.5856329798698425, "learning_rate": 2.4868873602751506e-05, "epoch": 35.09, "step": 82500}
|
| 202 |
+
{"loss": 0.1936, "grad_norm": 0.6045832633972168, "learning_rate": 2.3794067067927776e-05, "epoch": 35.3, "step": 83000}
|
| 203 |
+
{"loss": 0.194, "grad_norm": 0.5652228593826294, "learning_rate": 2.2719260533104043e-05, "epoch": 35.52, "step": 83500}
|
| 204 |
+
{"loss": 0.1896, "grad_norm": 0.6424867510795593, "learning_rate": 2.1644453998280313e-05, "epoch": 35.73, "step": 84000}
|
| 205 |
+
{"loss": 0.1921, "grad_norm": 0.5543941855430603, "learning_rate": 2.0571797076526225e-05, "epoch": 35.94, "step": 84500}
|
| 206 |
+
{"eval_loss": 0.21675606071949005, "eval_runtime": 56.9718, "eval_samples_per_second": 330.269, "eval_steps_per_second": 2.58, "epoch": 36.0, "step": 84636}
|
| 207 |
+
{"loss": 0.1884, "grad_norm": 0.6019295454025269, "learning_rate": 1.9496990541702495e-05, "epoch": 36.15, "step": 85000}
|
| 208 |
+
{"loss": 0.1888, "grad_norm": 0.645598292350769, "learning_rate": 1.8422184006878762e-05, "epoch": 36.37, "step": 85500}
|
| 209 |
+
{"loss": 0.1886, "grad_norm": 0.6292818188667297, "learning_rate": 1.7347377472055032e-05, "epoch": 36.58, "step": 86000}
|
| 210 |
+
{"loss": 0.1898, "grad_norm": 0.5169576406478882, "learning_rate": 1.62725709372313e-05, "epoch": 36.79, "step": 86500}
|
| 211 |
+
{"eval_loss": 0.21486429870128632, "eval_runtime": 56.8955, "eval_samples_per_second": 330.711, "eval_steps_per_second": 2.584, "epoch": 37.0, "step": 86987}
|
| 212 |
+
{"loss": 0.187, "grad_norm": 0.680117130279541, "learning_rate": 1.5197764402407566e-05, "epoch": 37.01, "step": 87000}
|
| 213 |
+
{"loss": 0.1841, "grad_norm": 0.5357288718223572, "learning_rate": 1.4122957867583836e-05, "epoch": 37.22, "step": 87500}
|
| 214 |
+
{"loss": 0.1829, "grad_norm": 0.5753670334815979, "learning_rate": 1.3048151332760103e-05, "epoch": 37.43, "step": 88000}
|
| 215 |
+
{"loss": 0.1857, "grad_norm": 0.654707133769989, "learning_rate": 1.1975494411006018e-05, "epoch": 37.64, "step": 88500}
|
| 216 |
+
{"loss": 0.1829, "grad_norm": 0.5702338218688965, "learning_rate": 1.0900687876182287e-05, "epoch": 37.86, "step": 89000}
|
| 217 |
+
{"eval_loss": 0.21089451014995575, "eval_runtime": 56.6335, "eval_samples_per_second": 332.241, "eval_steps_per_second": 2.596, "epoch": 38.0, "step": 89338}
|
| 218 |
+
{"loss": 0.1846, "grad_norm": 0.6671079993247986, "learning_rate": 9.825881341358555e-06, "epoch": 38.07, "step": 89500}
|
| 219 |
+
{"loss": 0.1809, "grad_norm": 0.5165778398513794, "learning_rate": 8.751074806534824e-06, "epoch": 38.28, "step": 90000}
|
| 220 |
+
{"loss": 0.1793, "grad_norm": 0.6374748945236206, "learning_rate": 7.676268271711092e-06, "epoch": 38.49, "step": 90500}
|
| 221 |
+
{"loss": 0.1804, "grad_norm": 0.5679296851158142, "learning_rate": 6.60146173688736e-06, "epoch": 38.71, "step": 91000}
|
| 222 |
+
{"loss": 0.1804, "grad_norm": 0.5970684885978699, "learning_rate": 5.526655202063629e-06, "epoch": 38.92, "step": 91500}
|
| 223 |
+
{"eval_loss": 0.20909027755260468, "eval_runtime": 56.346, "eval_samples_per_second": 333.937, "eval_steps_per_second": 2.609, "epoch": 39.0, "step": 91689}
|
| 224 |
+
{"loss": 0.1804, "grad_norm": 0.5444430112838745, "learning_rate": 4.4539982803095445e-06, "epoch": 39.13, "step": 92000}
|
| 225 |
+
{"loss": 0.1786, "grad_norm": 0.6086540818214417, "learning_rate": 3.379191745485813e-06, "epoch": 39.34, "step": 92500}
|
| 226 |
+
{"loss": 0.1758, "grad_norm": 0.5558965802192688, "learning_rate": 2.304385210662081e-06, "epoch": 39.56, "step": 93000}
|
| 227 |
+
{"loss": 0.1788, "grad_norm": 0.5531004667282104, "learning_rate": 1.2295786758383493e-06, "epoch": 39.77, "step": 93500}
|
| 228 |
+
{"loss": 0.1773, "grad_norm": 0.5611669421195984, "learning_rate": 1.5477214101461737e-07, "epoch": 39.98, "step": 94000}
|
| 229 |
+
{"eval_loss": 0.20976465940475464, "eval_runtime": 56.6405, "eval_samples_per_second": 332.2, "eval_steps_per_second": 2.595, "epoch": 40.0, "step": 94040}
|
| 230 |
+
{"train_runtime": 54838.0441, "train_samples_per_second": 109.795, "train_steps_per_second": 1.715, "total_flos": 3.6650496018087936e+18, "train_loss": 0.2880888063569923, "epoch": 40.0, "step": 94040}
|
utg4java_pretrain_results.txt
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
SUMMARY MODEL:
|
| 2 |
+
|
| 3 |
+
Model Params Size: 222882048
|
| 4 |
+
Model Params Size Formatted: 222.88 M
|
| 5 |
+
Model Disk Size: 891648255
|
| 6 |
+
Model Disk Size Formatted: 891.65 MB
|
| 7 |
+
|
| 8 |
+
|
| 9 |
+
TRAINING AND VALIDATION RESULTS:
|
| 10 |
+
|
| 11 |
+
Training batch size: 64
|
| 12 |
+
Validation batch size: 128
|
| 13 |
+
Total expected epochs: 40
|
| 14 |
+
Total expected trainig steps: 94040
|
| 15 |
+
Total expected trainig steps 2: 94040
|
| 16 |
+
Total trained epochs: 40.0
|
| 17 |
+
Total trained steps: 94040
|
| 18 |
+
Elapsed time: 54841.370337963104 seconds
|
| 19 |
+
Elapsed time (formatted): 15:14:01
|
| 20 |
+
Total flos: 3.6650496018087936e+18
|
| 21 |
+
Total flos (formatted): 3.665050e+18
|
| 22 |
+
Best epoch val_loss: 0.20909027755260468
|
| 23 |
+
Best model checkpoint: /root/pretrain_utg4java_02/checkpoint-91689
|
| 24 |
+
|
| 25 |
+
|
| 26 |
+
|
| 27 |
+
SUMMARY DATASETS:
|
| 28 |
+
|
| 29 |
+
Loaded Dataset:
|
| 30 |
+
DatasetDict({
|
| 31 |
+
train: Dataset({
|
| 32 |
+
features: ['text'],
|
| 33 |
+
num_rows: 150523
|
| 34 |
+
})
|
| 35 |
+
valid: Dataset({
|
| 36 |
+
features: ['text'],
|
| 37 |
+
num_rows: 18816
|
| 38 |
+
})
|
| 39 |
+
test: Dataset({
|
| 40 |
+
features: ['text'],
|
| 41 |
+
num_rows: 18815
|
| 42 |
+
})
|
| 43 |
+
})
|
| 44 |
+
|
| 45 |
+
Tokenized Dataset:
|
| 46 |
+
DatasetDict({
|
| 47 |
+
train: Dataset({
|
| 48 |
+
features: ['input_ids'],
|
| 49 |
+
num_rows: 150523
|
| 50 |
+
})
|
| 51 |
+
valid: Dataset({
|
| 52 |
+
features: ['input_ids'],
|
| 53 |
+
num_rows: 18816
|
| 54 |
+
})
|
| 55 |
+
test: Dataset({
|
| 56 |
+
features: ['input_ids'],
|
| 57 |
+
num_rows: 18815
|
| 58 |
+
})
|
| 59 |
+
})
|
| 60 |
+
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|