Upload 11 files
Browse files- README.md +81 -0
- added_tokens.json +5 -0
- checkpoint-110235/config.json +61 -0
- checkpoint-110235/generation_config.json +6 -0
- checkpoint-110235/pytorch_model.bin +3 -0
- checkpoint-110235/trainer_state.json +1401 -0
- checkpoint-110235/training_args.bin +3 -0
- special_tokens_map.json +107 -0
- spiece.model +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +112 -0
README.md
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- generated_from_trainer
|
| 4 |
+
datasets:
|
| 5 |
+
- rebel-short
|
| 6 |
+
metrics:
|
| 7 |
+
- rouge
|
| 8 |
+
model-index:
|
| 9 |
+
- name: flan-t5-base-samsum
|
| 10 |
+
results:
|
| 11 |
+
- task:
|
| 12 |
+
name: Sequence-to-sequence Language Modeling
|
| 13 |
+
type: text2text-generation
|
| 14 |
+
dataset:
|
| 15 |
+
name: rebel-short
|
| 16 |
+
type: rebel-short
|
| 17 |
+
config: default
|
| 18 |
+
split: test
|
| 19 |
+
args: default
|
| 20 |
+
metrics:
|
| 21 |
+
- name: Rouge1
|
| 22 |
+
type: rouge
|
| 23 |
+
value: 51.5716
|
| 24 |
+
---
|
| 25 |
+
|
| 26 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 27 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 28 |
+
|
| 29 |
+
# flan-t5-base-samsum
|
| 30 |
+
|
| 31 |
+
This model is a fine-tuned version of [flan-t5-base-samsum\checkpoint-1658](https://huggingface.co/flan-t5-base-samsum\checkpoint-1658) on the rebel-short dataset.
|
| 32 |
+
It achieves the following results on the evaluation set:
|
| 33 |
+
- Loss: 0.1029
|
| 34 |
+
- Rouge1: 51.5716
|
| 35 |
+
- Rouge2: 40.2152
|
| 36 |
+
- Rougel: 49.9941
|
| 37 |
+
- Rougelsum: 49.9767
|
| 38 |
+
- Gen Len: 18.5898
|
| 39 |
+
|
| 40 |
+
## Model description
|
| 41 |
+
|
| 42 |
+
More information needed
|
| 43 |
+
|
| 44 |
+
## Intended uses & limitations
|
| 45 |
+
|
| 46 |
+
More information needed
|
| 47 |
+
|
| 48 |
+
## Training and evaluation data
|
| 49 |
+
|
| 50 |
+
More information needed
|
| 51 |
+
|
| 52 |
+
## Training procedure
|
| 53 |
+
|
| 54 |
+
### Training hyperparameters
|
| 55 |
+
|
| 56 |
+
The following hyperparameters were used during training:
|
| 57 |
+
- learning_rate: 5e-05
|
| 58 |
+
- train_batch_size: 4
|
| 59 |
+
- eval_batch_size: 4
|
| 60 |
+
- seed: 42
|
| 61 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 62 |
+
- lr_scheduler_type: linear
|
| 63 |
+
- num_epochs: 5
|
| 64 |
+
|
| 65 |
+
### Training results
|
| 66 |
+
|
| 67 |
+
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|
| 68 |
+
|:-------------:|:-----:|:------:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:|
|
| 69 |
+
| 0.1256 | 1.0 | 22047 | 0.1206 | 50.3892 | 38.2761 | 48.7657 | 48.7444 | 18.6112 |
|
| 70 |
+
| 0.1091 | 2.0 | 44094 | 0.1112 | 50.9615 | 39.2843 | 49.3865 | 49.3674 | 18.5447 |
|
| 71 |
+
| 0.0875 | 3.0 | 66141 | 0.1047 | 51.2045 | 39.7598 | 49.6483 | 49.6317 | 18.5763 |
|
| 72 |
+
| 0.0841 | 4.0 | 88188 | 0.1036 | 51.3543 | 39.9776 | 49.8528 | 49.8223 | 18.6178 |
|
| 73 |
+
| 0.0806 | 5.0 | 110235 | 0.1029 | 51.5716 | 40.2152 | 49.9941 | 49.9767 | 18.5898 |
|
| 74 |
+
|
| 75 |
+
|
| 76 |
+
### Framework versions
|
| 77 |
+
|
| 78 |
+
- Transformers 4.27.2
|
| 79 |
+
- Pytorch 1.13.1+cu117
|
| 80 |
+
- Datasets 2.10.1
|
| 81 |
+
- Tokenizers 0.12.1
|
added_tokens.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<obj>": 32100,
|
| 3 |
+
"<subj>": 32101,
|
| 4 |
+
"<triplet>": 32102
|
| 5 |
+
}
|
checkpoint-110235/config.json
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "google/flan-t5-base",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"T5ForConditionalGeneration"
|
| 5 |
+
],
|
| 6 |
+
"d_ff": 2048,
|
| 7 |
+
"d_kv": 64,
|
| 8 |
+
"d_model": 768,
|
| 9 |
+
"decoder_start_token_id": 0,
|
| 10 |
+
"dense_act_fn": "gelu_new",
|
| 11 |
+
"dropout_rate": 0.1,
|
| 12 |
+
"eos_token_id": 1,
|
| 13 |
+
"feed_forward_proj": "gated-gelu",
|
| 14 |
+
"initializer_factor": 1.0,
|
| 15 |
+
"is_encoder_decoder": true,
|
| 16 |
+
"is_gated_act": true,
|
| 17 |
+
"layer_norm_epsilon": 1e-06,
|
| 18 |
+
"model_type": "t5",
|
| 19 |
+
"n_positions": 512,
|
| 20 |
+
"num_decoder_layers": 12,
|
| 21 |
+
"num_heads": 12,
|
| 22 |
+
"num_layers": 12,
|
| 23 |
+
"output_past": true,
|
| 24 |
+
"pad_token_id": 0,
|
| 25 |
+
"relative_attention_max_distance": 128,
|
| 26 |
+
"relative_attention_num_buckets": 32,
|
| 27 |
+
"task_specific_params": {
|
| 28 |
+
"summarization": {
|
| 29 |
+
"early_stopping": true,
|
| 30 |
+
"length_penalty": 2.0,
|
| 31 |
+
"max_length": 200,
|
| 32 |
+
"min_length": 30,
|
| 33 |
+
"no_repeat_ngram_size": 3,
|
| 34 |
+
"num_beams": 4,
|
| 35 |
+
"prefix": "summarize: "
|
| 36 |
+
},
|
| 37 |
+
"translation_en_to_de": {
|
| 38 |
+
"early_stopping": true,
|
| 39 |
+
"max_length": 300,
|
| 40 |
+
"num_beams": 4,
|
| 41 |
+
"prefix": "translate English to German: "
|
| 42 |
+
},
|
| 43 |
+
"translation_en_to_fr": {
|
| 44 |
+
"early_stopping": true,
|
| 45 |
+
"max_length": 300,
|
| 46 |
+
"num_beams": 4,
|
| 47 |
+
"prefix": "translate English to French: "
|
| 48 |
+
},
|
| 49 |
+
"translation_en_to_ro": {
|
| 50 |
+
"early_stopping": true,
|
| 51 |
+
"max_length": 300,
|
| 52 |
+
"num_beams": 4,
|
| 53 |
+
"prefix": "translate English to Romanian: "
|
| 54 |
+
}
|
| 55 |
+
},
|
| 56 |
+
"tie_word_embeddings": false,
|
| 57 |
+
"torch_dtype": "float32",
|
| 58 |
+
"transformers_version": "4.27.2",
|
| 59 |
+
"use_cache": true,
|
| 60 |
+
"vocab_size": 32103
|
| 61 |
+
}
|
checkpoint-110235/generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"decoder_start_token_id": 0,
|
| 3 |
+
"eos_token_id": 1,
|
| 4 |
+
"pad_token_id": 0,
|
| 5 |
+
"transformers_version": "4.27.2"
|
| 6 |
+
}
|
checkpoint-110235/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:98848c42f62ed65cecf4d3f8011bef217974e1b61aad6d8ccd279885cead6aff
|
| 3 |
+
size 990255285
|
checkpoint-110235/trainer_state.json
ADDED
|
@@ -0,0 +1,1401 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.1043073907494545,
|
| 3 |
+
"best_model_checkpoint": "flan-t5-base-nlrebel\\checkpoint-110235",
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"global_step": 110235,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.02,
|
| 12 |
+
"learning_rate": 4.9773211774844656e-05,
|
| 13 |
+
"loss": 1.1883,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.05,
|
| 18 |
+
"learning_rate": 4.95464235496893e-05,
|
| 19 |
+
"loss": 0.4505,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.07,
|
| 24 |
+
"learning_rate": 4.9319635324533955e-05,
|
| 25 |
+
"loss": 0.3265,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.09,
|
| 30 |
+
"learning_rate": 4.90928470993786e-05,
|
| 31 |
+
"loss": 0.2999,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.11,
|
| 36 |
+
"learning_rate": 4.8866058874223255e-05,
|
| 37 |
+
"loss": 0.282,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.14,
|
| 42 |
+
"learning_rate": 4.86392706490679e-05,
|
| 43 |
+
"loss": 0.2608,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.16,
|
| 48 |
+
"learning_rate": 4.8412482423912555e-05,
|
| 49 |
+
"loss": 0.2377,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.18,
|
| 54 |
+
"learning_rate": 4.81856941987572e-05,
|
| 55 |
+
"loss": 0.2276,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.2,
|
| 60 |
+
"learning_rate": 4.7958905973601855e-05,
|
| 61 |
+
"loss": 0.2064,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.23,
|
| 66 |
+
"learning_rate": 4.77321177484465e-05,
|
| 67 |
+
"loss": 0.2141,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.25,
|
| 72 |
+
"learning_rate": 4.7505329523291154e-05,
|
| 73 |
+
"loss": 0.193,
|
| 74 |
+
"step": 5500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.27,
|
| 78 |
+
"learning_rate": 4.72785412981358e-05,
|
| 79 |
+
"loss": 0.1955,
|
| 80 |
+
"step": 6000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.29,
|
| 84 |
+
"learning_rate": 4.7051753072980454e-05,
|
| 85 |
+
"loss": 0.1827,
|
| 86 |
+
"step": 6500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.32,
|
| 90 |
+
"learning_rate": 4.68249648478251e-05,
|
| 91 |
+
"loss": 0.183,
|
| 92 |
+
"step": 7000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.34,
|
| 96 |
+
"learning_rate": 4.6598176622669754e-05,
|
| 97 |
+
"loss": 0.1834,
|
| 98 |
+
"step": 7500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.36,
|
| 102 |
+
"learning_rate": 4.63713883975144e-05,
|
| 103 |
+
"loss": 0.1812,
|
| 104 |
+
"step": 8000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.39,
|
| 108 |
+
"learning_rate": 4.6144600172359054e-05,
|
| 109 |
+
"loss": 0.1808,
|
| 110 |
+
"step": 8500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.41,
|
| 114 |
+
"learning_rate": 4.591781194720371e-05,
|
| 115 |
+
"loss": 0.1779,
|
| 116 |
+
"step": 9000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.43,
|
| 120 |
+
"learning_rate": 4.569102372204835e-05,
|
| 121 |
+
"loss": 0.1567,
|
| 122 |
+
"step": 9500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.45,
|
| 126 |
+
"learning_rate": 4.5464235496893007e-05,
|
| 127 |
+
"loss": 0.1667,
|
| 128 |
+
"step": 10000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.48,
|
| 132 |
+
"learning_rate": 4.523744727173765e-05,
|
| 133 |
+
"loss": 0.1579,
|
| 134 |
+
"step": 10500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.5,
|
| 138 |
+
"learning_rate": 4.5010659046582306e-05,
|
| 139 |
+
"loss": 0.1621,
|
| 140 |
+
"step": 11000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.52,
|
| 144 |
+
"learning_rate": 4.478387082142695e-05,
|
| 145 |
+
"loss": 0.1631,
|
| 146 |
+
"step": 11500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.54,
|
| 150 |
+
"learning_rate": 4.4557082596271606e-05,
|
| 151 |
+
"loss": 0.1511,
|
| 152 |
+
"step": 12000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.57,
|
| 156 |
+
"learning_rate": 4.433029437111625e-05,
|
| 157 |
+
"loss": 0.1549,
|
| 158 |
+
"step": 12500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.59,
|
| 162 |
+
"learning_rate": 4.4103506145960906e-05,
|
| 163 |
+
"loss": 0.1528,
|
| 164 |
+
"step": 13000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.61,
|
| 168 |
+
"learning_rate": 4.387671792080555e-05,
|
| 169 |
+
"loss": 0.1518,
|
| 170 |
+
"step": 13500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.64,
|
| 174 |
+
"learning_rate": 4.3649929695650205e-05,
|
| 175 |
+
"loss": 0.15,
|
| 176 |
+
"step": 14000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.66,
|
| 180 |
+
"learning_rate": 4.342314147049485e-05,
|
| 181 |
+
"loss": 0.1457,
|
| 182 |
+
"step": 14500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.68,
|
| 186 |
+
"learning_rate": 4.3196353245339505e-05,
|
| 187 |
+
"loss": 0.1467,
|
| 188 |
+
"step": 15000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.7,
|
| 192 |
+
"learning_rate": 4.296956502018415e-05,
|
| 193 |
+
"loss": 0.1528,
|
| 194 |
+
"step": 15500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.73,
|
| 198 |
+
"learning_rate": 4.2742776795028805e-05,
|
| 199 |
+
"loss": 0.1459,
|
| 200 |
+
"step": 16000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.75,
|
| 204 |
+
"learning_rate": 4.251598856987346e-05,
|
| 205 |
+
"loss": 0.1369,
|
| 206 |
+
"step": 16500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.77,
|
| 210 |
+
"learning_rate": 4.2289200344718105e-05,
|
| 211 |
+
"loss": 0.144,
|
| 212 |
+
"step": 17000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.79,
|
| 216 |
+
"learning_rate": 4.206241211956276e-05,
|
| 217 |
+
"loss": 0.1387,
|
| 218 |
+
"step": 17500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.82,
|
| 222 |
+
"learning_rate": 4.1835623894407404e-05,
|
| 223 |
+
"loss": 0.1375,
|
| 224 |
+
"step": 18000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.84,
|
| 228 |
+
"learning_rate": 4.160883566925206e-05,
|
| 229 |
+
"loss": 0.1445,
|
| 230 |
+
"step": 18500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.86,
|
| 234 |
+
"learning_rate": 4.1382047444096704e-05,
|
| 235 |
+
"loss": 0.1436,
|
| 236 |
+
"step": 19000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.88,
|
| 240 |
+
"learning_rate": 4.115525921894136e-05,
|
| 241 |
+
"loss": 0.1373,
|
| 242 |
+
"step": 19500
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.91,
|
| 246 |
+
"learning_rate": 4.0928470993786004e-05,
|
| 247 |
+
"loss": 0.1466,
|
| 248 |
+
"step": 20000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.93,
|
| 252 |
+
"learning_rate": 4.070168276863066e-05,
|
| 253 |
+
"loss": 0.1342,
|
| 254 |
+
"step": 20500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.95,
|
| 258 |
+
"learning_rate": 4.0474894543475304e-05,
|
| 259 |
+
"loss": 0.1322,
|
| 260 |
+
"step": 21000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.98,
|
| 264 |
+
"learning_rate": 4.024810631831996e-05,
|
| 265 |
+
"loss": 0.1367,
|
| 266 |
+
"step": 21500
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 1.0,
|
| 270 |
+
"learning_rate": 4.00213180931646e-05,
|
| 271 |
+
"loss": 0.1309,
|
| 272 |
+
"step": 22000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 1.0,
|
| 276 |
+
"eval_gen_len": 18.65612244897959,
|
| 277 |
+
"eval_loss": 0.12737533450126648,
|
| 278 |
+
"eval_rouge1": 50.7462,
|
| 279 |
+
"eval_rouge2": 38.4276,
|
| 280 |
+
"eval_rougeL": 48.9889,
|
| 281 |
+
"eval_rougeLsum": 49.009,
|
| 282 |
+
"eval_runtime": 527.904,
|
| 283 |
+
"eval_samples_per_second": 9.282,
|
| 284 |
+
"eval_steps_per_second": 2.32,
|
| 285 |
+
"step": 22047
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 1.02,
|
| 289 |
+
"learning_rate": 3.979452986800926e-05,
|
| 290 |
+
"loss": 0.123,
|
| 291 |
+
"step": 22500
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 1.04,
|
| 295 |
+
"learning_rate": 3.95677416428539e-05,
|
| 296 |
+
"loss": 0.1248,
|
| 297 |
+
"step": 23000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.07,
|
| 301 |
+
"learning_rate": 3.9340953417698556e-05,
|
| 302 |
+
"loss": 0.1154,
|
| 303 |
+
"step": 23500
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 1.09,
|
| 307 |
+
"learning_rate": 3.91141651925432e-05,
|
| 308 |
+
"loss": 0.1254,
|
| 309 |
+
"step": 24000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.11,
|
| 313 |
+
"learning_rate": 3.8887376967387856e-05,
|
| 314 |
+
"loss": 0.1269,
|
| 315 |
+
"step": 24500
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 1.13,
|
| 319 |
+
"learning_rate": 3.866058874223251e-05,
|
| 320 |
+
"loss": 0.1198,
|
| 321 |
+
"step": 25000
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 1.16,
|
| 325 |
+
"learning_rate": 3.8433800517077156e-05,
|
| 326 |
+
"loss": 0.1186,
|
| 327 |
+
"step": 25500
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 1.18,
|
| 331 |
+
"learning_rate": 3.820701229192181e-05,
|
| 332 |
+
"loss": 0.1208,
|
| 333 |
+
"step": 26000
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 1.2,
|
| 337 |
+
"learning_rate": 3.7980224066766456e-05,
|
| 338 |
+
"loss": 0.1223,
|
| 339 |
+
"step": 26500
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.22,
|
| 343 |
+
"learning_rate": 3.775343584161111e-05,
|
| 344 |
+
"loss": 0.125,
|
| 345 |
+
"step": 27000
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 1.25,
|
| 349 |
+
"learning_rate": 3.7526647616455755e-05,
|
| 350 |
+
"loss": 0.1147,
|
| 351 |
+
"step": 27500
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 1.27,
|
| 355 |
+
"learning_rate": 3.729985939130041e-05,
|
| 356 |
+
"loss": 0.1266,
|
| 357 |
+
"step": 28000
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 1.29,
|
| 361 |
+
"learning_rate": 3.7073071166145055e-05,
|
| 362 |
+
"loss": 0.1197,
|
| 363 |
+
"step": 28500
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 1.32,
|
| 367 |
+
"learning_rate": 3.684628294098971e-05,
|
| 368 |
+
"loss": 0.1208,
|
| 369 |
+
"step": 29000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 1.34,
|
| 373 |
+
"learning_rate": 3.6619494715834355e-05,
|
| 374 |
+
"loss": 0.1166,
|
| 375 |
+
"step": 29500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 1.36,
|
| 379 |
+
"learning_rate": 3.639270649067901e-05,
|
| 380 |
+
"loss": 0.1195,
|
| 381 |
+
"step": 30000
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.38,
|
| 385 |
+
"learning_rate": 3.6165918265523655e-05,
|
| 386 |
+
"loss": 0.1153,
|
| 387 |
+
"step": 30500
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 1.41,
|
| 391 |
+
"learning_rate": 3.593913004036831e-05,
|
| 392 |
+
"loss": 0.1156,
|
| 393 |
+
"step": 31000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 1.43,
|
| 397 |
+
"learning_rate": 3.5712341815212954e-05,
|
| 398 |
+
"loss": 0.1215,
|
| 399 |
+
"step": 31500
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 1.45,
|
| 403 |
+
"learning_rate": 3.548555359005761e-05,
|
| 404 |
+
"loss": 0.1126,
|
| 405 |
+
"step": 32000
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 1.47,
|
| 409 |
+
"learning_rate": 3.525876536490226e-05,
|
| 410 |
+
"loss": 0.1176,
|
| 411 |
+
"step": 32500
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 1.5,
|
| 415 |
+
"learning_rate": 3.503197713974691e-05,
|
| 416 |
+
"loss": 0.1161,
|
| 417 |
+
"step": 33000
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 1.52,
|
| 421 |
+
"learning_rate": 3.480518891459156e-05,
|
| 422 |
+
"loss": 0.1131,
|
| 423 |
+
"step": 33500
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.54,
|
| 427 |
+
"learning_rate": 3.457840068943621e-05,
|
| 428 |
+
"loss": 0.1182,
|
| 429 |
+
"step": 34000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.56,
|
| 433 |
+
"learning_rate": 3.435161246428086e-05,
|
| 434 |
+
"loss": 0.1111,
|
| 435 |
+
"step": 34500
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 1.59,
|
| 439 |
+
"learning_rate": 3.412482423912551e-05,
|
| 440 |
+
"loss": 0.1166,
|
| 441 |
+
"step": 35000
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 1.61,
|
| 445 |
+
"learning_rate": 3.389803601397016e-05,
|
| 446 |
+
"loss": 0.1171,
|
| 447 |
+
"step": 35500
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 1.63,
|
| 451 |
+
"learning_rate": 3.3671247788814806e-05,
|
| 452 |
+
"loss": 0.1218,
|
| 453 |
+
"step": 36000
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 1.66,
|
| 457 |
+
"learning_rate": 3.344445956365946e-05,
|
| 458 |
+
"loss": 0.1088,
|
| 459 |
+
"step": 36500
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 1.68,
|
| 463 |
+
"learning_rate": 3.3217671338504106e-05,
|
| 464 |
+
"loss": 0.1139,
|
| 465 |
+
"step": 37000
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 1.7,
|
| 469 |
+
"learning_rate": 3.299088311334876e-05,
|
| 470 |
+
"loss": 0.1111,
|
| 471 |
+
"step": 37500
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.72,
|
| 475 |
+
"learning_rate": 3.2764094888193406e-05,
|
| 476 |
+
"loss": 0.1184,
|
| 477 |
+
"step": 38000
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 1.75,
|
| 481 |
+
"learning_rate": 3.253730666303806e-05,
|
| 482 |
+
"loss": 0.11,
|
| 483 |
+
"step": 38500
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 1.77,
|
| 487 |
+
"learning_rate": 3.2310518437882706e-05,
|
| 488 |
+
"loss": 0.1131,
|
| 489 |
+
"step": 39000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 1.79,
|
| 493 |
+
"learning_rate": 3.208373021272736e-05,
|
| 494 |
+
"loss": 0.111,
|
| 495 |
+
"step": 39500
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 1.81,
|
| 499 |
+
"learning_rate": 3.1856941987572005e-05,
|
| 500 |
+
"loss": 0.107,
|
| 501 |
+
"step": 40000
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 1.84,
|
| 505 |
+
"learning_rate": 3.163015376241666e-05,
|
| 506 |
+
"loss": 0.1122,
|
| 507 |
+
"step": 40500
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 1.86,
|
| 511 |
+
"learning_rate": 3.140336553726131e-05,
|
| 512 |
+
"loss": 0.1087,
|
| 513 |
+
"step": 41000
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.88,
|
| 517 |
+
"learning_rate": 3.117657731210596e-05,
|
| 518 |
+
"loss": 0.1149,
|
| 519 |
+
"step": 41500
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 1.91,
|
| 523 |
+
"learning_rate": 3.094978908695061e-05,
|
| 524 |
+
"loss": 0.1049,
|
| 525 |
+
"step": 42000
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 1.93,
|
| 529 |
+
"learning_rate": 3.072300086179526e-05,
|
| 530 |
+
"loss": 0.1127,
|
| 531 |
+
"step": 42500
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 1.95,
|
| 535 |
+
"learning_rate": 3.0496212636639908e-05,
|
| 536 |
+
"loss": 0.103,
|
| 537 |
+
"step": 43000
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 1.97,
|
| 541 |
+
"learning_rate": 3.0269424411484558e-05,
|
| 542 |
+
"loss": 0.1097,
|
| 543 |
+
"step": 43500
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 2.0,
|
| 547 |
+
"learning_rate": 3.0042636186329208e-05,
|
| 548 |
+
"loss": 0.113,
|
| 549 |
+
"step": 44000
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 2.0,
|
| 553 |
+
"eval_gen_len": 18.55,
|
| 554 |
+
"eval_loss": 0.11201398819684982,
|
| 555 |
+
"eval_rouge1": 51.3852,
|
| 556 |
+
"eval_rouge2": 39.4979,
|
| 557 |
+
"eval_rougeL": 49.6625,
|
| 558 |
+
"eval_rougeLsum": 49.6776,
|
| 559 |
+
"eval_runtime": 524.7178,
|
| 560 |
+
"eval_samples_per_second": 9.338,
|
| 561 |
+
"eval_steps_per_second": 2.335,
|
| 562 |
+
"step": 44094
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 2.02,
|
| 566 |
+
"learning_rate": 2.981584796117386e-05,
|
| 567 |
+
"loss": 0.0999,
|
| 568 |
+
"step": 44500
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 2.04,
|
| 572 |
+
"learning_rate": 2.958905973601851e-05,
|
| 573 |
+
"loss": 0.0997,
|
| 574 |
+
"step": 45000
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 2.06,
|
| 578 |
+
"learning_rate": 2.936227151086316e-05,
|
| 579 |
+
"loss": 0.102,
|
| 580 |
+
"step": 45500
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 2.09,
|
| 584 |
+
"learning_rate": 2.913548328570781e-05,
|
| 585 |
+
"loss": 0.1088,
|
| 586 |
+
"step": 46000
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 2.11,
|
| 590 |
+
"learning_rate": 2.890869506055246e-05,
|
| 591 |
+
"loss": 0.0993,
|
| 592 |
+
"step": 46500
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 2.13,
|
| 596 |
+
"learning_rate": 2.868190683539711e-05,
|
| 597 |
+
"loss": 0.1009,
|
| 598 |
+
"step": 47000
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 2.15,
|
| 602 |
+
"learning_rate": 2.845511861024176e-05,
|
| 603 |
+
"loss": 0.0922,
|
| 604 |
+
"step": 47500
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 2.18,
|
| 608 |
+
"learning_rate": 2.822833038508641e-05,
|
| 609 |
+
"loss": 0.0964,
|
| 610 |
+
"step": 48000
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 2.2,
|
| 614 |
+
"learning_rate": 2.800154215993106e-05,
|
| 615 |
+
"loss": 0.0985,
|
| 616 |
+
"step": 48500
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 2.22,
|
| 620 |
+
"learning_rate": 2.777475393477571e-05,
|
| 621 |
+
"loss": 0.099,
|
| 622 |
+
"step": 49000
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 2.25,
|
| 626 |
+
"learning_rate": 2.754796570962036e-05,
|
| 627 |
+
"loss": 0.0978,
|
| 628 |
+
"step": 49500
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 2.27,
|
| 632 |
+
"learning_rate": 2.732117748446501e-05,
|
| 633 |
+
"loss": 0.0977,
|
| 634 |
+
"step": 50000
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 2.29,
|
| 638 |
+
"learning_rate": 2.709438925930966e-05,
|
| 639 |
+
"loss": 0.1012,
|
| 640 |
+
"step": 50500
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 2.31,
|
| 644 |
+
"learning_rate": 2.686760103415431e-05,
|
| 645 |
+
"loss": 0.0967,
|
| 646 |
+
"step": 51000
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 2.34,
|
| 650 |
+
"learning_rate": 2.664081280899896e-05,
|
| 651 |
+
"loss": 0.101,
|
| 652 |
+
"step": 51500
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"epoch": 2.36,
|
| 656 |
+
"learning_rate": 2.641402458384361e-05,
|
| 657 |
+
"loss": 0.0961,
|
| 658 |
+
"step": 52000
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 2.38,
|
| 662 |
+
"learning_rate": 2.6187236358688262e-05,
|
| 663 |
+
"loss": 0.0939,
|
| 664 |
+
"step": 52500
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 2.4,
|
| 668 |
+
"learning_rate": 2.5960448133532912e-05,
|
| 669 |
+
"loss": 0.093,
|
| 670 |
+
"step": 53000
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 2.43,
|
| 674 |
+
"learning_rate": 2.5733659908377562e-05,
|
| 675 |
+
"loss": 0.0939,
|
| 676 |
+
"step": 53500
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 2.45,
|
| 680 |
+
"learning_rate": 2.5506871683222212e-05,
|
| 681 |
+
"loss": 0.0991,
|
| 682 |
+
"step": 54000
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 2.47,
|
| 686 |
+
"learning_rate": 2.5280083458066862e-05,
|
| 687 |
+
"loss": 0.0911,
|
| 688 |
+
"step": 54500
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 2.49,
|
| 692 |
+
"learning_rate": 2.505329523291151e-05,
|
| 693 |
+
"loss": 0.1011,
|
| 694 |
+
"step": 55000
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 2.52,
|
| 698 |
+
"learning_rate": 2.4826507007756158e-05,
|
| 699 |
+
"loss": 0.0989,
|
| 700 |
+
"step": 55500
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 2.54,
|
| 704 |
+
"learning_rate": 2.4599718782600808e-05,
|
| 705 |
+
"loss": 0.0978,
|
| 706 |
+
"step": 56000
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 2.56,
|
| 710 |
+
"learning_rate": 2.4372930557445458e-05,
|
| 711 |
+
"loss": 0.0994,
|
| 712 |
+
"step": 56500
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 2.59,
|
| 716 |
+
"learning_rate": 2.4146142332290108e-05,
|
| 717 |
+
"loss": 0.098,
|
| 718 |
+
"step": 57000
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 2.61,
|
| 722 |
+
"learning_rate": 2.3919354107134758e-05,
|
| 723 |
+
"loss": 0.0907,
|
| 724 |
+
"step": 57500
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 2.63,
|
| 728 |
+
"learning_rate": 2.3692565881979407e-05,
|
| 729 |
+
"loss": 0.0993,
|
| 730 |
+
"step": 58000
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 2.65,
|
| 734 |
+
"learning_rate": 2.3465777656824057e-05,
|
| 735 |
+
"loss": 0.0956,
|
| 736 |
+
"step": 58500
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 2.68,
|
| 740 |
+
"learning_rate": 2.323898943166871e-05,
|
| 741 |
+
"loss": 0.0997,
|
| 742 |
+
"step": 59000
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 2.7,
|
| 746 |
+
"learning_rate": 2.301220120651336e-05,
|
| 747 |
+
"loss": 0.0947,
|
| 748 |
+
"step": 59500
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 2.72,
|
| 752 |
+
"learning_rate": 2.278541298135801e-05,
|
| 753 |
+
"loss": 0.0969,
|
| 754 |
+
"step": 60000
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 2.74,
|
| 758 |
+
"learning_rate": 2.255862475620266e-05,
|
| 759 |
+
"loss": 0.0994,
|
| 760 |
+
"step": 60500
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 2.77,
|
| 764 |
+
"learning_rate": 2.233183653104731e-05,
|
| 765 |
+
"loss": 0.0927,
|
| 766 |
+
"step": 61000
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 2.79,
|
| 770 |
+
"learning_rate": 2.210504830589196e-05,
|
| 771 |
+
"loss": 0.0975,
|
| 772 |
+
"step": 61500
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 2.81,
|
| 776 |
+
"learning_rate": 2.187826008073661e-05,
|
| 777 |
+
"loss": 0.0953,
|
| 778 |
+
"step": 62000
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 2.83,
|
| 782 |
+
"learning_rate": 2.165147185558126e-05,
|
| 783 |
+
"loss": 0.0936,
|
| 784 |
+
"step": 62500
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 2.86,
|
| 788 |
+
"learning_rate": 2.142468363042591e-05,
|
| 789 |
+
"loss": 0.0944,
|
| 790 |
+
"step": 63000
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 2.88,
|
| 794 |
+
"learning_rate": 2.119789540527056e-05,
|
| 795 |
+
"loss": 0.1008,
|
| 796 |
+
"step": 63500
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 2.9,
|
| 800 |
+
"learning_rate": 2.097110718011521e-05,
|
| 801 |
+
"loss": 0.0945,
|
| 802 |
+
"step": 64000
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 2.93,
|
| 806 |
+
"learning_rate": 2.074431895495986e-05,
|
| 807 |
+
"loss": 0.0916,
|
| 808 |
+
"step": 64500
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 2.95,
|
| 812 |
+
"learning_rate": 2.051753072980451e-05,
|
| 813 |
+
"loss": 0.0998,
|
| 814 |
+
"step": 65000
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 2.97,
|
| 818 |
+
"learning_rate": 2.029074250464916e-05,
|
| 819 |
+
"loss": 0.0932,
|
| 820 |
+
"step": 65500
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 2.99,
|
| 824 |
+
"learning_rate": 2.006395427949381e-05,
|
| 825 |
+
"loss": 0.0925,
|
| 826 |
+
"step": 66000
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 3.0,
|
| 830 |
+
"eval_gen_len": 18.567551020408164,
|
| 831 |
+
"eval_loss": 0.10631391406059265,
|
| 832 |
+
"eval_rouge1": 51.5769,
|
| 833 |
+
"eval_rouge2": 39.9334,
|
| 834 |
+
"eval_rougeL": 49.9305,
|
| 835 |
+
"eval_rougeLsum": 49.9362,
|
| 836 |
+
"eval_runtime": 524.369,
|
| 837 |
+
"eval_samples_per_second": 9.345,
|
| 838 |
+
"eval_steps_per_second": 2.336,
|
| 839 |
+
"step": 66141
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 3.02,
|
| 843 |
+
"learning_rate": 1.983716605433846e-05,
|
| 844 |
+
"loss": 0.0874,
|
| 845 |
+
"step": 66500
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 3.04,
|
| 849 |
+
"learning_rate": 1.9610377829183112e-05,
|
| 850 |
+
"loss": 0.084,
|
| 851 |
+
"step": 67000
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 3.06,
|
| 855 |
+
"learning_rate": 1.9383589604027762e-05,
|
| 856 |
+
"loss": 0.0907,
|
| 857 |
+
"step": 67500
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 3.08,
|
| 861 |
+
"learning_rate": 1.915680137887241e-05,
|
| 862 |
+
"loss": 0.0955,
|
| 863 |
+
"step": 68000
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 3.11,
|
| 867 |
+
"learning_rate": 1.893001315371706e-05,
|
| 868 |
+
"loss": 0.0803,
|
| 869 |
+
"step": 68500
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 3.13,
|
| 873 |
+
"learning_rate": 1.870322492856171e-05,
|
| 874 |
+
"loss": 0.0962,
|
| 875 |
+
"step": 69000
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 3.15,
|
| 879 |
+
"learning_rate": 1.847643670340636e-05,
|
| 880 |
+
"loss": 0.0806,
|
| 881 |
+
"step": 69500
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 3.18,
|
| 885 |
+
"learning_rate": 1.824964847825101e-05,
|
| 886 |
+
"loss": 0.0887,
|
| 887 |
+
"step": 70000
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 3.2,
|
| 891 |
+
"learning_rate": 1.802286025309566e-05,
|
| 892 |
+
"loss": 0.0869,
|
| 893 |
+
"step": 70500
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 3.22,
|
| 897 |
+
"learning_rate": 1.779607202794031e-05,
|
| 898 |
+
"loss": 0.0908,
|
| 899 |
+
"step": 71000
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 3.24,
|
| 903 |
+
"learning_rate": 1.756928380278496e-05,
|
| 904 |
+
"loss": 0.0828,
|
| 905 |
+
"step": 71500
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 3.27,
|
| 909 |
+
"learning_rate": 1.734249557762961e-05,
|
| 910 |
+
"loss": 0.086,
|
| 911 |
+
"step": 72000
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 3.29,
|
| 915 |
+
"learning_rate": 1.711570735247426e-05,
|
| 916 |
+
"loss": 0.0827,
|
| 917 |
+
"step": 72500
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 3.31,
|
| 921 |
+
"learning_rate": 1.688891912731891e-05,
|
| 922 |
+
"loss": 0.085,
|
| 923 |
+
"step": 73000
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 3.33,
|
| 927 |
+
"learning_rate": 1.666213090216356e-05,
|
| 928 |
+
"loss": 0.0894,
|
| 929 |
+
"step": 73500
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 3.36,
|
| 933 |
+
"learning_rate": 1.643534267700821e-05,
|
| 934 |
+
"loss": 0.0834,
|
| 935 |
+
"step": 74000
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 3.38,
|
| 939 |
+
"learning_rate": 1.620855445185286e-05,
|
| 940 |
+
"loss": 0.0842,
|
| 941 |
+
"step": 74500
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 3.4,
|
| 945 |
+
"learning_rate": 1.5981766226697513e-05,
|
| 946 |
+
"loss": 0.0878,
|
| 947 |
+
"step": 75000
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 3.42,
|
| 951 |
+
"learning_rate": 1.5754978001542163e-05,
|
| 952 |
+
"loss": 0.0889,
|
| 953 |
+
"step": 75500
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 3.45,
|
| 957 |
+
"learning_rate": 1.5528189776386813e-05,
|
| 958 |
+
"loss": 0.0875,
|
| 959 |
+
"step": 76000
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 3.47,
|
| 963 |
+
"learning_rate": 1.5301401551231463e-05,
|
| 964 |
+
"loss": 0.0855,
|
| 965 |
+
"step": 76500
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 3.49,
|
| 969 |
+
"learning_rate": 1.5074613326076113e-05,
|
| 970 |
+
"loss": 0.0903,
|
| 971 |
+
"step": 77000
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 3.52,
|
| 975 |
+
"learning_rate": 1.4847825100920762e-05,
|
| 976 |
+
"loss": 0.0878,
|
| 977 |
+
"step": 77500
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 3.54,
|
| 981 |
+
"learning_rate": 1.4621036875765412e-05,
|
| 982 |
+
"loss": 0.0838,
|
| 983 |
+
"step": 78000
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 3.56,
|
| 987 |
+
"learning_rate": 1.4394248650610062e-05,
|
| 988 |
+
"loss": 0.0857,
|
| 989 |
+
"step": 78500
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 3.58,
|
| 993 |
+
"learning_rate": 1.4167460425454712e-05,
|
| 994 |
+
"loss": 0.0838,
|
| 995 |
+
"step": 79000
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 3.61,
|
| 999 |
+
"learning_rate": 1.3940672200299362e-05,
|
| 1000 |
+
"loss": 0.0894,
|
| 1001 |
+
"step": 79500
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 3.63,
|
| 1005 |
+
"learning_rate": 1.3713883975144012e-05,
|
| 1006 |
+
"loss": 0.0793,
|
| 1007 |
+
"step": 80000
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 3.65,
|
| 1011 |
+
"learning_rate": 1.3487095749988662e-05,
|
| 1012 |
+
"loss": 0.0859,
|
| 1013 |
+
"step": 80500
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 3.67,
|
| 1017 |
+
"learning_rate": 1.3260307524833313e-05,
|
| 1018 |
+
"loss": 0.0851,
|
| 1019 |
+
"step": 81000
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 3.7,
|
| 1023 |
+
"learning_rate": 1.3033519299677963e-05,
|
| 1024 |
+
"loss": 0.0847,
|
| 1025 |
+
"step": 81500
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 3.72,
|
| 1029 |
+
"learning_rate": 1.2806731074522613e-05,
|
| 1030 |
+
"loss": 0.0812,
|
| 1031 |
+
"step": 82000
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 3.74,
|
| 1035 |
+
"learning_rate": 1.2579942849367263e-05,
|
| 1036 |
+
"loss": 0.0878,
|
| 1037 |
+
"step": 82500
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 3.76,
|
| 1041 |
+
"learning_rate": 1.2353154624211911e-05,
|
| 1042 |
+
"loss": 0.0878,
|
| 1043 |
+
"step": 83000
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 3.79,
|
| 1047 |
+
"learning_rate": 1.2126366399056561e-05,
|
| 1048 |
+
"loss": 0.0872,
|
| 1049 |
+
"step": 83500
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 3.81,
|
| 1053 |
+
"learning_rate": 1.1899578173901212e-05,
|
| 1054 |
+
"loss": 0.082,
|
| 1055 |
+
"step": 84000
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 3.83,
|
| 1059 |
+
"learning_rate": 1.1672789948745862e-05,
|
| 1060 |
+
"loss": 0.0844,
|
| 1061 |
+
"step": 84500
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 3.86,
|
| 1065 |
+
"learning_rate": 1.1446001723590512e-05,
|
| 1066 |
+
"loss": 0.0872,
|
| 1067 |
+
"step": 85000
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 3.88,
|
| 1071 |
+
"learning_rate": 1.1219213498435162e-05,
|
| 1072 |
+
"loss": 0.0878,
|
| 1073 |
+
"step": 85500
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 3.9,
|
| 1077 |
+
"learning_rate": 1.0992425273279812e-05,
|
| 1078 |
+
"loss": 0.0854,
|
| 1079 |
+
"step": 86000
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 3.92,
|
| 1083 |
+
"learning_rate": 1.0765637048124462e-05,
|
| 1084 |
+
"loss": 0.0825,
|
| 1085 |
+
"step": 86500
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 3.95,
|
| 1089 |
+
"learning_rate": 1.0538848822969112e-05,
|
| 1090 |
+
"loss": 0.0817,
|
| 1091 |
+
"step": 87000
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 3.97,
|
| 1095 |
+
"learning_rate": 1.0312060597813762e-05,
|
| 1096 |
+
"loss": 0.0831,
|
| 1097 |
+
"step": 87500
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 3.99,
|
| 1101 |
+
"learning_rate": 1.0085272372658413e-05,
|
| 1102 |
+
"loss": 0.0878,
|
| 1103 |
+
"step": 88000
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 4.0,
|
| 1107 |
+
"eval_gen_len": 18.59469387755102,
|
| 1108 |
+
"eval_loss": 0.10489419847726822,
|
| 1109 |
+
"eval_rouge1": 51.7274,
|
| 1110 |
+
"eval_rouge2": 40.2559,
|
| 1111 |
+
"eval_rougeL": 50.122,
|
| 1112 |
+
"eval_rougeLsum": 50.147,
|
| 1113 |
+
"eval_runtime": 522.9353,
|
| 1114 |
+
"eval_samples_per_second": 9.37,
|
| 1115 |
+
"eval_steps_per_second": 2.343,
|
| 1116 |
+
"step": 88188
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 4.01,
|
| 1120 |
+
"learning_rate": 9.858484147503063e-06,
|
| 1121 |
+
"loss": 0.0782,
|
| 1122 |
+
"step": 88500
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 4.04,
|
| 1126 |
+
"learning_rate": 9.631695922347713e-06,
|
| 1127 |
+
"loss": 0.0791,
|
| 1128 |
+
"step": 89000
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 4.06,
|
| 1132 |
+
"learning_rate": 9.404907697192363e-06,
|
| 1133 |
+
"loss": 0.0764,
|
| 1134 |
+
"step": 89500
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 4.08,
|
| 1138 |
+
"learning_rate": 9.178119472037013e-06,
|
| 1139 |
+
"loss": 0.0786,
|
| 1140 |
+
"step": 90000
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 4.1,
|
| 1144 |
+
"learning_rate": 8.951331246881662e-06,
|
| 1145 |
+
"loss": 0.075,
|
| 1146 |
+
"step": 90500
|
| 1147 |
+
},
|
| 1148 |
+
{
|
| 1149 |
+
"epoch": 4.13,
|
| 1150 |
+
"learning_rate": 8.724543021726312e-06,
|
| 1151 |
+
"loss": 0.0819,
|
| 1152 |
+
"step": 91000
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 4.15,
|
| 1156 |
+
"learning_rate": 8.497754796570962e-06,
|
| 1157 |
+
"loss": 0.086,
|
| 1158 |
+
"step": 91500
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 4.17,
|
| 1162 |
+
"learning_rate": 8.270966571415614e-06,
|
| 1163 |
+
"loss": 0.0778,
|
| 1164 |
+
"step": 92000
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 4.2,
|
| 1168 |
+
"learning_rate": 8.044178346260264e-06,
|
| 1169 |
+
"loss": 0.0781,
|
| 1170 |
+
"step": 92500
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 4.22,
|
| 1174 |
+
"learning_rate": 7.817390121104913e-06,
|
| 1175 |
+
"loss": 0.0791,
|
| 1176 |
+
"step": 93000
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 4.24,
|
| 1180 |
+
"learning_rate": 7.590601895949563e-06,
|
| 1181 |
+
"loss": 0.0827,
|
| 1182 |
+
"step": 93500
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 4.26,
|
| 1186 |
+
"learning_rate": 7.363813670794213e-06,
|
| 1187 |
+
"loss": 0.0812,
|
| 1188 |
+
"step": 94000
|
| 1189 |
+
},
|
| 1190 |
+
{
|
| 1191 |
+
"epoch": 4.29,
|
| 1192 |
+
"learning_rate": 7.137025445638863e-06,
|
| 1193 |
+
"loss": 0.0755,
|
| 1194 |
+
"step": 94500
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 4.31,
|
| 1198 |
+
"learning_rate": 6.910237220483514e-06,
|
| 1199 |
+
"loss": 0.0807,
|
| 1200 |
+
"step": 95000
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 4.33,
|
| 1204 |
+
"learning_rate": 6.683448995328164e-06,
|
| 1205 |
+
"loss": 0.0822,
|
| 1206 |
+
"step": 95500
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 4.35,
|
| 1210 |
+
"learning_rate": 6.4566607701728135e-06,
|
| 1211 |
+
"loss": 0.0786,
|
| 1212 |
+
"step": 96000
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"epoch": 4.38,
|
| 1216 |
+
"learning_rate": 6.2298725450174625e-06,
|
| 1217 |
+
"loss": 0.0833,
|
| 1218 |
+
"step": 96500
|
| 1219 |
+
},
|
| 1220 |
+
{
|
| 1221 |
+
"epoch": 4.4,
|
| 1222 |
+
"learning_rate": 6.003084319862113e-06,
|
| 1223 |
+
"loss": 0.078,
|
| 1224 |
+
"step": 97000
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 4.42,
|
| 1228 |
+
"learning_rate": 5.776296094706763e-06,
|
| 1229 |
+
"loss": 0.079,
|
| 1230 |
+
"step": 97500
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"epoch": 4.45,
|
| 1234 |
+
"learning_rate": 5.549507869551413e-06,
|
| 1235 |
+
"loss": 0.0769,
|
| 1236 |
+
"step": 98000
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 4.47,
|
| 1240 |
+
"learning_rate": 5.322719644396063e-06,
|
| 1241 |
+
"loss": 0.077,
|
| 1242 |
+
"step": 98500
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 4.49,
|
| 1246 |
+
"learning_rate": 5.0959314192407136e-06,
|
| 1247 |
+
"loss": 0.0839,
|
| 1248 |
+
"step": 99000
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 4.51,
|
| 1252 |
+
"learning_rate": 4.8691431940853634e-06,
|
| 1253 |
+
"loss": 0.0808,
|
| 1254 |
+
"step": 99500
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 4.54,
|
| 1258 |
+
"learning_rate": 4.642354968930013e-06,
|
| 1259 |
+
"loss": 0.081,
|
| 1260 |
+
"step": 100000
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"epoch": 4.56,
|
| 1264 |
+
"learning_rate": 4.415566743774663e-06,
|
| 1265 |
+
"loss": 0.0763,
|
| 1266 |
+
"step": 100500
|
| 1267 |
+
},
|
| 1268 |
+
{
|
| 1269 |
+
"epoch": 4.58,
|
| 1270 |
+
"learning_rate": 4.188778518619314e-06,
|
| 1271 |
+
"loss": 0.0785,
|
| 1272 |
+
"step": 101000
|
| 1273 |
+
},
|
| 1274 |
+
{
|
| 1275 |
+
"epoch": 4.6,
|
| 1276 |
+
"learning_rate": 3.961990293463964e-06,
|
| 1277 |
+
"loss": 0.0756,
|
| 1278 |
+
"step": 101500
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 4.63,
|
| 1282 |
+
"learning_rate": 3.7352020683086136e-06,
|
| 1283 |
+
"loss": 0.0779,
|
| 1284 |
+
"step": 102000
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 4.65,
|
| 1288 |
+
"learning_rate": 3.508413843153264e-06,
|
| 1289 |
+
"loss": 0.0801,
|
| 1290 |
+
"step": 102500
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 4.67,
|
| 1294 |
+
"learning_rate": 3.2816256179979138e-06,
|
| 1295 |
+
"loss": 0.0782,
|
| 1296 |
+
"step": 103000
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 4.69,
|
| 1300 |
+
"learning_rate": 3.0548373928425636e-06,
|
| 1301 |
+
"loss": 0.0826,
|
| 1302 |
+
"step": 103500
|
| 1303 |
+
},
|
| 1304 |
+
{
|
| 1305 |
+
"epoch": 4.72,
|
| 1306 |
+
"learning_rate": 2.828049167687214e-06,
|
| 1307 |
+
"loss": 0.088,
|
| 1308 |
+
"step": 104000
|
| 1309 |
+
},
|
| 1310 |
+
{
|
| 1311 |
+
"epoch": 4.74,
|
| 1312 |
+
"learning_rate": 2.601260942531864e-06,
|
| 1313 |
+
"loss": 0.078,
|
| 1314 |
+
"step": 104500
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 4.76,
|
| 1318 |
+
"learning_rate": 2.374472717376514e-06,
|
| 1319 |
+
"loss": 0.0761,
|
| 1320 |
+
"step": 105000
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 4.79,
|
| 1324 |
+
"learning_rate": 2.147684492221164e-06,
|
| 1325 |
+
"loss": 0.0772,
|
| 1326 |
+
"step": 105500
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 4.81,
|
| 1330 |
+
"learning_rate": 1.9208962670658143e-06,
|
| 1331 |
+
"loss": 0.0777,
|
| 1332 |
+
"step": 106000
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 4.83,
|
| 1336 |
+
"learning_rate": 1.6941080419104641e-06,
|
| 1337 |
+
"loss": 0.0788,
|
| 1338 |
+
"step": 106500
|
| 1339 |
+
},
|
| 1340 |
+
{
|
| 1341 |
+
"epoch": 4.85,
|
| 1342 |
+
"learning_rate": 1.4673198167551142e-06,
|
| 1343 |
+
"loss": 0.079,
|
| 1344 |
+
"step": 107000
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 4.88,
|
| 1348 |
+
"learning_rate": 1.240531591599764e-06,
|
| 1349 |
+
"loss": 0.0799,
|
| 1350 |
+
"step": 107500
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 4.9,
|
| 1354 |
+
"learning_rate": 1.0137433664444142e-06,
|
| 1355 |
+
"loss": 0.0765,
|
| 1356 |
+
"step": 108000
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 4.92,
|
| 1360 |
+
"learning_rate": 7.869551412890642e-07,
|
| 1361 |
+
"loss": 0.081,
|
| 1362 |
+
"step": 108500
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"epoch": 4.94,
|
| 1366 |
+
"learning_rate": 5.601669161337143e-07,
|
| 1367 |
+
"loss": 0.0819,
|
| 1368 |
+
"step": 109000
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 4.97,
|
| 1372 |
+
"learning_rate": 3.333786909783644e-07,
|
| 1373 |
+
"loss": 0.0801,
|
| 1374 |
+
"step": 109500
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 4.99,
|
| 1378 |
+
"learning_rate": 1.0659046582301448e-07,
|
| 1379 |
+
"loss": 0.0813,
|
| 1380 |
+
"step": 110000
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 5.0,
|
| 1384 |
+
"eval_gen_len": 18.58265306122449,
|
| 1385 |
+
"eval_loss": 0.1043073907494545,
|
| 1386 |
+
"eval_rouge1": 51.9039,
|
| 1387 |
+
"eval_rouge2": 40.5297,
|
| 1388 |
+
"eval_rougeL": 50.2805,
|
| 1389 |
+
"eval_rougeLsum": 50.2968,
|
| 1390 |
+
"eval_runtime": 522.6723,
|
| 1391 |
+
"eval_samples_per_second": 9.375,
|
| 1392 |
+
"eval_steps_per_second": 2.344,
|
| 1393 |
+
"step": 110235
|
| 1394 |
+
}
|
| 1395 |
+
],
|
| 1396 |
+
"max_steps": 110235,
|
| 1397 |
+
"num_train_epochs": 5,
|
| 1398 |
+
"total_flos": 1.509503144067072e+17,
|
| 1399 |
+
"trial_name": null,
|
| 1400 |
+
"trial_params": null
|
| 1401 |
+
}
|
checkpoint-110235/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9eba9c339578a61f86733bafa9ac4b5fc2a7688c706e37321d6867e424f5bdd3
|
| 3 |
+
size 3707
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,107 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<extra_id_0>",
|
| 4 |
+
"<extra_id_1>",
|
| 5 |
+
"<extra_id_2>",
|
| 6 |
+
"<extra_id_3>",
|
| 7 |
+
"<extra_id_4>",
|
| 8 |
+
"<extra_id_5>",
|
| 9 |
+
"<extra_id_6>",
|
| 10 |
+
"<extra_id_7>",
|
| 11 |
+
"<extra_id_8>",
|
| 12 |
+
"<extra_id_9>",
|
| 13 |
+
"<extra_id_10>",
|
| 14 |
+
"<extra_id_11>",
|
| 15 |
+
"<extra_id_12>",
|
| 16 |
+
"<extra_id_13>",
|
| 17 |
+
"<extra_id_14>",
|
| 18 |
+
"<extra_id_15>",
|
| 19 |
+
"<extra_id_16>",
|
| 20 |
+
"<extra_id_17>",
|
| 21 |
+
"<extra_id_18>",
|
| 22 |
+
"<extra_id_19>",
|
| 23 |
+
"<extra_id_20>",
|
| 24 |
+
"<extra_id_21>",
|
| 25 |
+
"<extra_id_22>",
|
| 26 |
+
"<extra_id_23>",
|
| 27 |
+
"<extra_id_24>",
|
| 28 |
+
"<extra_id_25>",
|
| 29 |
+
"<extra_id_26>",
|
| 30 |
+
"<extra_id_27>",
|
| 31 |
+
"<extra_id_28>",
|
| 32 |
+
"<extra_id_29>",
|
| 33 |
+
"<extra_id_30>",
|
| 34 |
+
"<extra_id_31>",
|
| 35 |
+
"<extra_id_32>",
|
| 36 |
+
"<extra_id_33>",
|
| 37 |
+
"<extra_id_34>",
|
| 38 |
+
"<extra_id_35>",
|
| 39 |
+
"<extra_id_36>",
|
| 40 |
+
"<extra_id_37>",
|
| 41 |
+
"<extra_id_38>",
|
| 42 |
+
"<extra_id_39>",
|
| 43 |
+
"<extra_id_40>",
|
| 44 |
+
"<extra_id_41>",
|
| 45 |
+
"<extra_id_42>",
|
| 46 |
+
"<extra_id_43>",
|
| 47 |
+
"<extra_id_44>",
|
| 48 |
+
"<extra_id_45>",
|
| 49 |
+
"<extra_id_46>",
|
| 50 |
+
"<extra_id_47>",
|
| 51 |
+
"<extra_id_48>",
|
| 52 |
+
"<extra_id_49>",
|
| 53 |
+
"<extra_id_50>",
|
| 54 |
+
"<extra_id_51>",
|
| 55 |
+
"<extra_id_52>",
|
| 56 |
+
"<extra_id_53>",
|
| 57 |
+
"<extra_id_54>",
|
| 58 |
+
"<extra_id_55>",
|
| 59 |
+
"<extra_id_56>",
|
| 60 |
+
"<extra_id_57>",
|
| 61 |
+
"<extra_id_58>",
|
| 62 |
+
"<extra_id_59>",
|
| 63 |
+
"<extra_id_60>",
|
| 64 |
+
"<extra_id_61>",
|
| 65 |
+
"<extra_id_62>",
|
| 66 |
+
"<extra_id_63>",
|
| 67 |
+
"<extra_id_64>",
|
| 68 |
+
"<extra_id_65>",
|
| 69 |
+
"<extra_id_66>",
|
| 70 |
+
"<extra_id_67>",
|
| 71 |
+
"<extra_id_68>",
|
| 72 |
+
"<extra_id_69>",
|
| 73 |
+
"<extra_id_70>",
|
| 74 |
+
"<extra_id_71>",
|
| 75 |
+
"<extra_id_72>",
|
| 76 |
+
"<extra_id_73>",
|
| 77 |
+
"<extra_id_74>",
|
| 78 |
+
"<extra_id_75>",
|
| 79 |
+
"<extra_id_76>",
|
| 80 |
+
"<extra_id_77>",
|
| 81 |
+
"<extra_id_78>",
|
| 82 |
+
"<extra_id_79>",
|
| 83 |
+
"<extra_id_80>",
|
| 84 |
+
"<extra_id_81>",
|
| 85 |
+
"<extra_id_82>",
|
| 86 |
+
"<extra_id_83>",
|
| 87 |
+
"<extra_id_84>",
|
| 88 |
+
"<extra_id_85>",
|
| 89 |
+
"<extra_id_86>",
|
| 90 |
+
"<extra_id_87>",
|
| 91 |
+
"<extra_id_88>",
|
| 92 |
+
"<extra_id_89>",
|
| 93 |
+
"<extra_id_90>",
|
| 94 |
+
"<extra_id_91>",
|
| 95 |
+
"<extra_id_92>",
|
| 96 |
+
"<extra_id_93>",
|
| 97 |
+
"<extra_id_94>",
|
| 98 |
+
"<extra_id_95>",
|
| 99 |
+
"<extra_id_96>",
|
| 100 |
+
"<extra_id_97>",
|
| 101 |
+
"<extra_id_98>",
|
| 102 |
+
"<extra_id_99>"
|
| 103 |
+
],
|
| 104 |
+
"eos_token": "</s>",
|
| 105 |
+
"pad_token": "<pad>",
|
| 106 |
+
"unk_token": "<unk>"
|
| 107 |
+
}
|
spiece.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d60acb128cf7b7f2536e8f38a5b18a05535c9e14c7a355904270e15b0945ea86
|
| 3 |
+
size 791656
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,112 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<extra_id_0>",
|
| 4 |
+
"<extra_id_1>",
|
| 5 |
+
"<extra_id_2>",
|
| 6 |
+
"<extra_id_3>",
|
| 7 |
+
"<extra_id_4>",
|
| 8 |
+
"<extra_id_5>",
|
| 9 |
+
"<extra_id_6>",
|
| 10 |
+
"<extra_id_7>",
|
| 11 |
+
"<extra_id_8>",
|
| 12 |
+
"<extra_id_9>",
|
| 13 |
+
"<extra_id_10>",
|
| 14 |
+
"<extra_id_11>",
|
| 15 |
+
"<extra_id_12>",
|
| 16 |
+
"<extra_id_13>",
|
| 17 |
+
"<extra_id_14>",
|
| 18 |
+
"<extra_id_15>",
|
| 19 |
+
"<extra_id_16>",
|
| 20 |
+
"<extra_id_17>",
|
| 21 |
+
"<extra_id_18>",
|
| 22 |
+
"<extra_id_19>",
|
| 23 |
+
"<extra_id_20>",
|
| 24 |
+
"<extra_id_21>",
|
| 25 |
+
"<extra_id_22>",
|
| 26 |
+
"<extra_id_23>",
|
| 27 |
+
"<extra_id_24>",
|
| 28 |
+
"<extra_id_25>",
|
| 29 |
+
"<extra_id_26>",
|
| 30 |
+
"<extra_id_27>",
|
| 31 |
+
"<extra_id_28>",
|
| 32 |
+
"<extra_id_29>",
|
| 33 |
+
"<extra_id_30>",
|
| 34 |
+
"<extra_id_31>",
|
| 35 |
+
"<extra_id_32>",
|
| 36 |
+
"<extra_id_33>",
|
| 37 |
+
"<extra_id_34>",
|
| 38 |
+
"<extra_id_35>",
|
| 39 |
+
"<extra_id_36>",
|
| 40 |
+
"<extra_id_37>",
|
| 41 |
+
"<extra_id_38>",
|
| 42 |
+
"<extra_id_39>",
|
| 43 |
+
"<extra_id_40>",
|
| 44 |
+
"<extra_id_41>",
|
| 45 |
+
"<extra_id_42>",
|
| 46 |
+
"<extra_id_43>",
|
| 47 |
+
"<extra_id_44>",
|
| 48 |
+
"<extra_id_45>",
|
| 49 |
+
"<extra_id_46>",
|
| 50 |
+
"<extra_id_47>",
|
| 51 |
+
"<extra_id_48>",
|
| 52 |
+
"<extra_id_49>",
|
| 53 |
+
"<extra_id_50>",
|
| 54 |
+
"<extra_id_51>",
|
| 55 |
+
"<extra_id_52>",
|
| 56 |
+
"<extra_id_53>",
|
| 57 |
+
"<extra_id_54>",
|
| 58 |
+
"<extra_id_55>",
|
| 59 |
+
"<extra_id_56>",
|
| 60 |
+
"<extra_id_57>",
|
| 61 |
+
"<extra_id_58>",
|
| 62 |
+
"<extra_id_59>",
|
| 63 |
+
"<extra_id_60>",
|
| 64 |
+
"<extra_id_61>",
|
| 65 |
+
"<extra_id_62>",
|
| 66 |
+
"<extra_id_63>",
|
| 67 |
+
"<extra_id_64>",
|
| 68 |
+
"<extra_id_65>",
|
| 69 |
+
"<extra_id_66>",
|
| 70 |
+
"<extra_id_67>",
|
| 71 |
+
"<extra_id_68>",
|
| 72 |
+
"<extra_id_69>",
|
| 73 |
+
"<extra_id_70>",
|
| 74 |
+
"<extra_id_71>",
|
| 75 |
+
"<extra_id_72>",
|
| 76 |
+
"<extra_id_73>",
|
| 77 |
+
"<extra_id_74>",
|
| 78 |
+
"<extra_id_75>",
|
| 79 |
+
"<extra_id_76>",
|
| 80 |
+
"<extra_id_77>",
|
| 81 |
+
"<extra_id_78>",
|
| 82 |
+
"<extra_id_79>",
|
| 83 |
+
"<extra_id_80>",
|
| 84 |
+
"<extra_id_81>",
|
| 85 |
+
"<extra_id_82>",
|
| 86 |
+
"<extra_id_83>",
|
| 87 |
+
"<extra_id_84>",
|
| 88 |
+
"<extra_id_85>",
|
| 89 |
+
"<extra_id_86>",
|
| 90 |
+
"<extra_id_87>",
|
| 91 |
+
"<extra_id_88>",
|
| 92 |
+
"<extra_id_89>",
|
| 93 |
+
"<extra_id_90>",
|
| 94 |
+
"<extra_id_91>",
|
| 95 |
+
"<extra_id_92>",
|
| 96 |
+
"<extra_id_93>",
|
| 97 |
+
"<extra_id_94>",
|
| 98 |
+
"<extra_id_95>",
|
| 99 |
+
"<extra_id_96>",
|
| 100 |
+
"<extra_id_97>",
|
| 101 |
+
"<extra_id_98>",
|
| 102 |
+
"<extra_id_99>"
|
| 103 |
+
],
|
| 104 |
+
"eos_token": "</s>",
|
| 105 |
+
"extra_ids": 100,
|
| 106 |
+
"model_max_length": 512,
|
| 107 |
+
"pad_token": "<pad>",
|
| 108 |
+
"sp_model_kwargs": {},
|
| 109 |
+
"special_tokens_map_file": "/home/younes_huggingface_co/.cache/huggingface/hub/models--google--t5-v1_1-base/snapshots/650d7745bf1e502d6949b22cc19155cd656d3d4e/special_tokens_map.json",
|
| 110 |
+
"tokenizer_class": "T5Tokenizer",
|
| 111 |
+
"unk_token": "<unk>"
|
| 112 |
+
}
|