Update from ec2-user
Browse files- README.md +72 -0
- all_results.json +15 -0
- config.json +60 -0
- eval_results.json +10 -0
- pytorch_model.bin +3 -0
- source.spm +0 -0
- special_tokens_map.json +1 -0
- target.spm +0 -0
- tokenizer_config.json +1 -0
- train_results.json +8 -0
- trainer_state.json +1253 -0
- training_args.bin +3 -0
- vocab.json +0 -0
README.md
ADDED
|
@@ -0,0 +1,72 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- en
|
| 4 |
+
- he
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
metrics:
|
| 8 |
+
- bleu
|
| 9 |
+
model-index:
|
| 10 |
+
- name: output_base
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# output_base
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [/home/ec2-user/SageMaker/marian_base](https://huggingface.co//home/ec2-user/SageMaker/marian_base) on the None dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 1.6852
|
| 22 |
+
- Bleu: 30.5903
|
| 23 |
+
- Gen Len: 64.8182
|
| 24 |
+
|
| 25 |
+
## Model description
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Intended uses & limitations
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training and evaluation data
|
| 34 |
+
|
| 35 |
+
More information needed
|
| 36 |
+
|
| 37 |
+
## Training procedure
|
| 38 |
+
|
| 39 |
+
### Training hyperparameters
|
| 40 |
+
|
| 41 |
+
The following hyperparameters were used during training:
|
| 42 |
+
- learning_rate: 5e-05
|
| 43 |
+
- train_batch_size: 48
|
| 44 |
+
- eval_batch_size: 16
|
| 45 |
+
- seed: 42
|
| 46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
+
- lr_scheduler_type: linear
|
| 48 |
+
- num_epochs: 10.0
|
| 49 |
+
- mixed_precision_training: Native AMP
|
| 50 |
+
|
| 51 |
+
### Training results
|
| 52 |
+
|
| 53 |
+
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|
| 54 |
+
|:-------------:|:-----:|:-------:|:---------------:|:-------:|:-------:|
|
| 55 |
+
| 1.9938 | 1.0 | 188563 | 2.0008 | 27.6169 | 66.0246 |
|
| 56 |
+
| 1.8171 | 2.0 | 377126 | 1.8753 | 28.4709 | 65.8859 |
|
| 57 |
+
| 1.7389 | 3.0 | 565689 | 1.8120 | 28.9724 | 65.8601 |
|
| 58 |
+
| 1.6893 | 4.0 | 754252 | 1.7690 | 29.5248 | 65.8846 |
|
| 59 |
+
| 1.6559 | 5.0 | 942815 | 1.7467 | 29.5757 | 65.8046 |
|
| 60 |
+
| 1.6279 | 6.0 | 1131378 | 1.7236 | 29.7512 | 66.0482 |
|
| 61 |
+
| 1.6053 | 7.0 | 1319941 | 1.7137 | 29.916 | 66.0031 |
|
| 62 |
+
| 1.5871 | 8.0 | 1508504 | 1.7007 | 30.1671 | 65.8853 |
|
| 63 |
+
| 1.5694 | 9.0 | 1697067 | 1.6921 | 30.3613 | 65.9506 |
|
| 64 |
+
| 1.5539 | 10.0 | 1885630 | 1.6852 | 30.4049 | 66.0487 |
|
| 65 |
+
|
| 66 |
+
|
| 67 |
+
### Framework versions
|
| 68 |
+
|
| 69 |
+
- Transformers 4.12.0.dev0
|
| 70 |
+
- Pytorch 1.9.1+cu102
|
| 71 |
+
- Datasets 1.12.1
|
| 72 |
+
- Tokenizers 0.10.3
|
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_bleu": 30.5903,
|
| 4 |
+
"eval_gen_len": 64.8182,
|
| 5 |
+
"eval_loss": 1.6851532459259033,
|
| 6 |
+
"eval_runtime": 901.837,
|
| 7 |
+
"eval_samples": 17165,
|
| 8 |
+
"eval_samples_per_second": 19.033,
|
| 9 |
+
"eval_steps_per_second": 1.19,
|
| 10 |
+
"train_loss": 1.743612184965204,
|
| 11 |
+
"train_runtime": 420652.6588,
|
| 12 |
+
"train_samples": 9051022,
|
| 13 |
+
"train_samples_per_second": 215.166,
|
| 14 |
+
"train_steps_per_second": 4.483
|
| 15 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/home/ec2-user/SageMaker/marian_base",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"activation_function": "swish",
|
| 5 |
+
"add_bias_logits": false,
|
| 6 |
+
"add_final_layer_norm": false,
|
| 7 |
+
"architectures": [
|
| 8 |
+
"MarianMTModel"
|
| 9 |
+
],
|
| 10 |
+
"attention_dropout": 0.0,
|
| 11 |
+
"bad_words_ids": [
|
| 12 |
+
[
|
| 13 |
+
62954
|
| 14 |
+
]
|
| 15 |
+
],
|
| 16 |
+
"bos_token_id": 0,
|
| 17 |
+
"classifier_dropout": 0.0,
|
| 18 |
+
"d_model": 512,
|
| 19 |
+
"decoder_attention_heads": 8,
|
| 20 |
+
"decoder_ffn_dim": 2048,
|
| 21 |
+
"decoder_layerdrop": 0.0,
|
| 22 |
+
"decoder_layers": 6,
|
| 23 |
+
"decoder_start_token_id": 62954,
|
| 24 |
+
"do_blenderbot_90_layernorm": false,
|
| 25 |
+
"dropout": 0.1,
|
| 26 |
+
"encoder_attention_heads": 8,
|
| 27 |
+
"encoder_ffn_dim": 2048,
|
| 28 |
+
"encoder_layerdrop": 0.0,
|
| 29 |
+
"encoder_layers": 6,
|
| 30 |
+
"eos_token_id": 0,
|
| 31 |
+
"extra_pos_embeddings": 2,
|
| 32 |
+
"force_bos_token_to_be_generated": false,
|
| 33 |
+
"forced_eos_token_id": 0,
|
| 34 |
+
"id2label": {
|
| 35 |
+
"0": "LABEL_0",
|
| 36 |
+
"1": "LABEL_1",
|
| 37 |
+
"2": "LABEL_2"
|
| 38 |
+
},
|
| 39 |
+
"init_std": 0.02,
|
| 40 |
+
"is_encoder_decoder": true,
|
| 41 |
+
"label2id": {
|
| 42 |
+
"LABEL_0": 0,
|
| 43 |
+
"LABEL_1": 1,
|
| 44 |
+
"LABEL_2": 2
|
| 45 |
+
},
|
| 46 |
+
"max_length": 512,
|
| 47 |
+
"max_position_embeddings": 512,
|
| 48 |
+
"model_type": "marian",
|
| 49 |
+
"normalize_before": false,
|
| 50 |
+
"normalize_embedding": false,
|
| 51 |
+
"num_beams": 6,
|
| 52 |
+
"num_hidden_layers": 6,
|
| 53 |
+
"pad_token_id": 62954,
|
| 54 |
+
"scale_embedding": true,
|
| 55 |
+
"static_position_embeddings": true,
|
| 56 |
+
"torch_dtype": "float32",
|
| 57 |
+
"transformers_version": "4.12.0.dev0",
|
| 58 |
+
"use_cache": true,
|
| 59 |
+
"vocab_size": 62955
|
| 60 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_bleu": 30.5903,
|
| 4 |
+
"eval_gen_len": 64.8182,
|
| 5 |
+
"eval_loss": 1.6851532459259033,
|
| 6 |
+
"eval_runtime": 901.837,
|
| 7 |
+
"eval_samples": 17165,
|
| 8 |
+
"eval_samples_per_second": 19.033,
|
| 9 |
+
"eval_steps_per_second": 1.19
|
| 10 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dc4cbc6cfec26c88e60397f8e95fa4e1d36b41eb8cd25d1b0c118f9736d4fd02
|
| 3 |
+
size 305822085
|
source.spm
ADDED
|
Binary file (797 kB). View file
|
|
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>"}
|
target.spm
ADDED
|
Binary file (885 kB). View file
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"source_lang": null, "target_lang": null, "unk_token": "<unk>", "eos_token": "</s>", "pad_token": "<pad>", "model_max_length": 512, "sp_model_kwargs": {}, "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "/home/ec2-user/SageMaker/marian_base", "tokenizer_class": "MarianTokenizer"}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"train_loss": 1.743612184965204,
|
| 4 |
+
"train_runtime": 420652.6588,
|
| 5 |
+
"train_samples": 9051022,
|
| 6 |
+
"train_samples_per_second": 215.166,
|
| 7 |
+
"train_steps_per_second": 4.483
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1253 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"global_step": 1885630,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.05,
|
| 12 |
+
"learning_rate": 4.9734863149186215e-05,
|
| 13 |
+
"loss": 5.3726,
|
| 14 |
+
"step": 10000
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.11,
|
| 18 |
+
"learning_rate": 4.946977933104586e-05,
|
| 19 |
+
"loss": 3.517,
|
| 20 |
+
"step": 20000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.16,
|
| 24 |
+
"learning_rate": 4.920472202924222e-05,
|
| 25 |
+
"loss": 2.957,
|
| 26 |
+
"step": 30000
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.21,
|
| 30 |
+
"learning_rate": 4.893963821110186e-05,
|
| 31 |
+
"loss": 2.7004,
|
| 32 |
+
"step": 40000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.27,
|
| 36 |
+
"learning_rate": 4.867460742563494e-05,
|
| 37 |
+
"loss": 2.5453,
|
| 38 |
+
"step": 50000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.32,
|
| 42 |
+
"learning_rate": 4.840955012383129e-05,
|
| 43 |
+
"loss": 2.4342,
|
| 44 |
+
"step": 60000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.37,
|
| 48 |
+
"learning_rate": 4.814446630569094e-05,
|
| 49 |
+
"loss": 2.3507,
|
| 50 |
+
"step": 70000
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.42,
|
| 54 |
+
"learning_rate": 4.787943552022401e-05,
|
| 55 |
+
"loss": 2.2903,
|
| 56 |
+
"step": 80000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.48,
|
| 60 |
+
"learning_rate": 4.761437821842037e-05,
|
| 61 |
+
"loss": 2.2356,
|
| 62 |
+
"step": 90000
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.53,
|
| 66 |
+
"learning_rate": 4.734934743295344e-05,
|
| 67 |
+
"loss": 2.1933,
|
| 68 |
+
"step": 100000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.58,
|
| 72 |
+
"learning_rate": 4.70842901311498e-05,
|
| 73 |
+
"loss": 2.1565,
|
| 74 |
+
"step": 110000
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.64,
|
| 78 |
+
"learning_rate": 4.681923282934616e-05,
|
| 79 |
+
"loss": 2.1247,
|
| 80 |
+
"step": 120000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.69,
|
| 84 |
+
"learning_rate": 4.655417552754252e-05,
|
| 85 |
+
"loss": 2.0947,
|
| 86 |
+
"step": 130000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.74,
|
| 90 |
+
"learning_rate": 4.628909170940217e-05,
|
| 91 |
+
"loss": 2.0742,
|
| 92 |
+
"step": 140000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.8,
|
| 96 |
+
"learning_rate": 4.602403440759853e-05,
|
| 97 |
+
"loss": 2.048,
|
| 98 |
+
"step": 150000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.85,
|
| 102 |
+
"learning_rate": 4.575897710579488e-05,
|
| 103 |
+
"loss": 2.0292,
|
| 104 |
+
"step": 160000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.9,
|
| 108 |
+
"learning_rate": 4.5493893287654526e-05,
|
| 109 |
+
"loss": 2.0091,
|
| 110 |
+
"step": 170000
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.95,
|
| 114 |
+
"learning_rate": 4.5228862502187606e-05,
|
| 115 |
+
"loss": 1.9938,
|
| 116 |
+
"step": 180000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.0,
|
| 120 |
+
"eval_bleu": 27.6169,
|
| 121 |
+
"eval_gen_len": 66.0246,
|
| 122 |
+
"eval_loss": 2.0008087158203125,
|
| 123 |
+
"eval_runtime": 2487.0668,
|
| 124 |
+
"eval_samples_per_second": 6.902,
|
| 125 |
+
"eval_steps_per_second": 0.431,
|
| 126 |
+
"step": 188563
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 1.01,
|
| 130 |
+
"learning_rate": 4.496380520038396e-05,
|
| 131 |
+
"loss": 1.9756,
|
| 132 |
+
"step": 190000
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 1.06,
|
| 136 |
+
"learning_rate": 4.469874789858032e-05,
|
| 137 |
+
"loss": 1.9481,
|
| 138 |
+
"step": 200000
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 1.11,
|
| 142 |
+
"learning_rate": 4.443369059677668e-05,
|
| 143 |
+
"loss": 1.9385,
|
| 144 |
+
"step": 210000
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.17,
|
| 148 |
+
"learning_rate": 4.416865981130975e-05,
|
| 149 |
+
"loss": 1.9268,
|
| 150 |
+
"step": 220000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 1.22,
|
| 154 |
+
"learning_rate": 4.3903575993169396e-05,
|
| 155 |
+
"loss": 1.9216,
|
| 156 |
+
"step": 230000
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 1.27,
|
| 160 |
+
"learning_rate": 4.363854520770247e-05,
|
| 161 |
+
"loss": 1.9097,
|
| 162 |
+
"step": 240000
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 1.33,
|
| 166 |
+
"learning_rate": 4.337348790589882e-05,
|
| 167 |
+
"loss": 1.8991,
|
| 168 |
+
"step": 250000
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 1.38,
|
| 172 |
+
"learning_rate": 4.31084571204319e-05,
|
| 173 |
+
"loss": 1.8921,
|
| 174 |
+
"step": 260000
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 1.43,
|
| 178 |
+
"learning_rate": 4.284339981862826e-05,
|
| 179 |
+
"loss": 1.8808,
|
| 180 |
+
"step": 270000
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 1.48,
|
| 184 |
+
"learning_rate": 4.2578342516824614e-05,
|
| 185 |
+
"loss": 1.8733,
|
| 186 |
+
"step": 280000
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 1.54,
|
| 190 |
+
"learning_rate": 4.231331173135769e-05,
|
| 191 |
+
"loss": 1.8671,
|
| 192 |
+
"step": 290000
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.59,
|
| 196 |
+
"learning_rate": 4.2048280945890766e-05,
|
| 197 |
+
"loss": 1.8612,
|
| 198 |
+
"step": 300000
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 1.64,
|
| 202 |
+
"learning_rate": 4.178325016042384e-05,
|
| 203 |
+
"loss": 1.8522,
|
| 204 |
+
"step": 310000
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 1.7,
|
| 208 |
+
"learning_rate": 4.151821937495691e-05,
|
| 209 |
+
"loss": 1.8443,
|
| 210 |
+
"step": 320000
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 1.75,
|
| 214 |
+
"learning_rate": 4.125313555681656e-05,
|
| 215 |
+
"loss": 1.8413,
|
| 216 |
+
"step": 330000
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 1.8,
|
| 220 |
+
"learning_rate": 4.098810477134963e-05,
|
| 221 |
+
"loss": 1.8361,
|
| 222 |
+
"step": 340000
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 1.86,
|
| 226 |
+
"learning_rate": 4.0723073985882704e-05,
|
| 227 |
+
"loss": 1.8276,
|
| 228 |
+
"step": 350000
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 1.91,
|
| 232 |
+
"learning_rate": 4.045801668407906e-05,
|
| 233 |
+
"loss": 1.824,
|
| 234 |
+
"step": 360000
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.96,
|
| 238 |
+
"learning_rate": 4.0192985898612136e-05,
|
| 239 |
+
"loss": 1.8171,
|
| 240 |
+
"step": 370000
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 2.0,
|
| 244 |
+
"eval_bleu": 28.4709,
|
| 245 |
+
"eval_gen_len": 65.8859,
|
| 246 |
+
"eval_loss": 1.8752530813217163,
|
| 247 |
+
"eval_runtime": 2537.4407,
|
| 248 |
+
"eval_samples_per_second": 6.765,
|
| 249 |
+
"eval_steps_per_second": 0.423,
|
| 250 |
+
"step": 377126
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 2.02,
|
| 254 |
+
"learning_rate": 3.9927928596808496e-05,
|
| 255 |
+
"loss": 1.8061,
|
| 256 |
+
"step": 380000
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 2.07,
|
| 260 |
+
"learning_rate": 3.966289781134157e-05,
|
| 261 |
+
"loss": 1.787,
|
| 262 |
+
"step": 390000
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 2.12,
|
| 266 |
+
"learning_rate": 3.939784050953793e-05,
|
| 267 |
+
"loss": 1.7839,
|
| 268 |
+
"step": 400000
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 2.17,
|
| 272 |
+
"learning_rate": 3.9132809724071e-05,
|
| 273 |
+
"loss": 1.7827,
|
| 274 |
+
"step": 410000
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 2.23,
|
| 278 |
+
"learning_rate": 3.886775242226736e-05,
|
| 279 |
+
"loss": 1.7787,
|
| 280 |
+
"step": 420000
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 2.28,
|
| 284 |
+
"learning_rate": 3.860269512046372e-05,
|
| 285 |
+
"loss": 1.776,
|
| 286 |
+
"step": 430000
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 2.33,
|
| 290 |
+
"learning_rate": 3.833763781866008e-05,
|
| 291 |
+
"loss": 1.773,
|
| 292 |
+
"step": 440000
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 2.39,
|
| 296 |
+
"learning_rate": 3.807258051685644e-05,
|
| 297 |
+
"loss": 1.7715,
|
| 298 |
+
"step": 450000
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 2.44,
|
| 302 |
+
"learning_rate": 3.78075232150528e-05,
|
| 303 |
+
"loss": 1.7672,
|
| 304 |
+
"step": 460000
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 2.49,
|
| 308 |
+
"learning_rate": 3.754246591324915e-05,
|
| 309 |
+
"loss": 1.764,
|
| 310 |
+
"step": 470000
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 2.55,
|
| 314 |
+
"learning_rate": 3.727740861144551e-05,
|
| 315 |
+
"loss": 1.7602,
|
| 316 |
+
"step": 480000
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 2.6,
|
| 320 |
+
"learning_rate": 3.701235130964188e-05,
|
| 321 |
+
"loss": 1.7608,
|
| 322 |
+
"step": 490000
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 2.65,
|
| 326 |
+
"learning_rate": 3.674729400783823e-05,
|
| 327 |
+
"loss": 1.7564,
|
| 328 |
+
"step": 500000
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 2.7,
|
| 332 |
+
"learning_rate": 3.648223670603459e-05,
|
| 333 |
+
"loss": 1.7532,
|
| 334 |
+
"step": 510000
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 2.76,
|
| 338 |
+
"learning_rate": 3.621720592056766e-05,
|
| 339 |
+
"loss": 1.7479,
|
| 340 |
+
"step": 520000
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 2.81,
|
| 344 |
+
"learning_rate": 3.595214861876402e-05,
|
| 345 |
+
"loss": 1.744,
|
| 346 |
+
"step": 530000
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 2.86,
|
| 350 |
+
"learning_rate": 3.568706480062367e-05,
|
| 351 |
+
"loss": 1.744,
|
| 352 |
+
"step": 540000
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 2.92,
|
| 356 |
+
"learning_rate": 3.542203401515674e-05,
|
| 357 |
+
"loss": 1.7409,
|
| 358 |
+
"step": 550000
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 2.97,
|
| 362 |
+
"learning_rate": 3.5156950197016386e-05,
|
| 363 |
+
"loss": 1.7389,
|
| 364 |
+
"step": 560000
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 3.0,
|
| 368 |
+
"eval_bleu": 28.9724,
|
| 369 |
+
"eval_gen_len": 65.8601,
|
| 370 |
+
"eval_loss": 1.8119513988494873,
|
| 371 |
+
"eval_runtime": 2571.172,
|
| 372 |
+
"eval_samples_per_second": 6.676,
|
| 373 |
+
"eval_steps_per_second": 0.417,
|
| 374 |
+
"step": 565689
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 3.02,
|
| 378 |
+
"learning_rate": 3.489191941154946e-05,
|
| 379 |
+
"loss": 1.7262,
|
| 380 |
+
"step": 570000
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 3.08,
|
| 384 |
+
"learning_rate": 3.462686210974582e-05,
|
| 385 |
+
"loss": 1.711,
|
| 386 |
+
"step": 580000
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 3.13,
|
| 390 |
+
"learning_rate": 3.436183132427889e-05,
|
| 391 |
+
"loss": 1.7129,
|
| 392 |
+
"step": 590000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 3.18,
|
| 396 |
+
"learning_rate": 3.409674750613853e-05,
|
| 397 |
+
"loss": 1.7137,
|
| 398 |
+
"step": 600000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 3.23,
|
| 402 |
+
"learning_rate": 3.383166368799818e-05,
|
| 403 |
+
"loss": 1.7105,
|
| 404 |
+
"step": 610000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 3.29,
|
| 408 |
+
"learning_rate": 3.3566606386194536e-05,
|
| 409 |
+
"loss": 1.7114,
|
| 410 |
+
"step": 620000
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 3.34,
|
| 414 |
+
"learning_rate": 3.3301549084390896e-05,
|
| 415 |
+
"loss": 1.7078,
|
| 416 |
+
"step": 630000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 3.39,
|
| 420 |
+
"learning_rate": 3.3036491782587255e-05,
|
| 421 |
+
"loss": 1.7074,
|
| 422 |
+
"step": 640000
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 3.45,
|
| 426 |
+
"learning_rate": 3.2771434480783615e-05,
|
| 427 |
+
"loss": 1.7052,
|
| 428 |
+
"step": 650000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 3.5,
|
| 432 |
+
"learning_rate": 3.250637717897997e-05,
|
| 433 |
+
"loss": 1.7038,
|
| 434 |
+
"step": 660000
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 3.55,
|
| 438 |
+
"learning_rate": 3.224131987717633e-05,
|
| 439 |
+
"loss": 1.703,
|
| 440 |
+
"step": 670000
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 3.61,
|
| 444 |
+
"learning_rate": 3.1976262575372686e-05,
|
| 445 |
+
"loss": 1.7015,
|
| 446 |
+
"step": 680000
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 3.66,
|
| 450 |
+
"learning_rate": 3.171120527356905e-05,
|
| 451 |
+
"loss": 1.699,
|
| 452 |
+
"step": 690000
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 3.71,
|
| 456 |
+
"learning_rate": 3.144617448810212e-05,
|
| 457 |
+
"loss": 1.6991,
|
| 458 |
+
"step": 700000
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 3.77,
|
| 462 |
+
"learning_rate": 3.1181090669961765e-05,
|
| 463 |
+
"loss": 1.6956,
|
| 464 |
+
"step": 710000
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 3.82,
|
| 468 |
+
"learning_rate": 3.0916059884494844e-05,
|
| 469 |
+
"loss": 1.6951,
|
| 470 |
+
"step": 720000
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 3.87,
|
| 474 |
+
"learning_rate": 3.06510025826912e-05,
|
| 475 |
+
"loss": 1.6925,
|
| 476 |
+
"step": 730000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 3.92,
|
| 480 |
+
"learning_rate": 3.0385918764550843e-05,
|
| 481 |
+
"loss": 1.6925,
|
| 482 |
+
"step": 740000
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 3.98,
|
| 486 |
+
"learning_rate": 3.0120887979083916e-05,
|
| 487 |
+
"loss": 1.6893,
|
| 488 |
+
"step": 750000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 4.0,
|
| 492 |
+
"eval_bleu": 29.5248,
|
| 493 |
+
"eval_gen_len": 65.8846,
|
| 494 |
+
"eval_loss": 1.7690072059631348,
|
| 495 |
+
"eval_runtime": 2545.7813,
|
| 496 |
+
"eval_samples_per_second": 6.743,
|
| 497 |
+
"eval_steps_per_second": 0.421,
|
| 498 |
+
"step": 754252
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 4.03,
|
| 502 |
+
"learning_rate": 2.9855830677280272e-05,
|
| 503 |
+
"loss": 1.6732,
|
| 504 |
+
"step": 760000
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 4.08,
|
| 508 |
+
"learning_rate": 2.959077337547663e-05,
|
| 509 |
+
"loss": 1.6653,
|
| 510 |
+
"step": 770000
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 4.14,
|
| 514 |
+
"learning_rate": 2.9325716073672994e-05,
|
| 515 |
+
"loss": 1.6672,
|
| 516 |
+
"step": 780000
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 4.19,
|
| 520 |
+
"learning_rate": 2.9060658771869347e-05,
|
| 521 |
+
"loss": 1.667,
|
| 522 |
+
"step": 790000
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 4.24,
|
| 526 |
+
"learning_rate": 2.8795574953728993e-05,
|
| 527 |
+
"loss": 1.6665,
|
| 528 |
+
"step": 800000
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 4.3,
|
| 532 |
+
"learning_rate": 2.8530517651925353e-05,
|
| 533 |
+
"loss": 1.6661,
|
| 534 |
+
"step": 810000
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 4.35,
|
| 538 |
+
"learning_rate": 2.8265433833785e-05,
|
| 539 |
+
"loss": 1.6656,
|
| 540 |
+
"step": 820000
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 4.4,
|
| 544 |
+
"learning_rate": 2.8000403048318068e-05,
|
| 545 |
+
"loss": 1.666,
|
| 546 |
+
"step": 830000
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 4.45,
|
| 550 |
+
"learning_rate": 2.773534574651443e-05,
|
| 551 |
+
"loss": 1.6629,
|
| 552 |
+
"step": 840000
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 4.51,
|
| 556 |
+
"learning_rate": 2.747028844471079e-05,
|
| 557 |
+
"loss": 1.6648,
|
| 558 |
+
"step": 850000
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 4.56,
|
| 562 |
+
"learning_rate": 2.720520462657043e-05,
|
| 563 |
+
"loss": 1.6633,
|
| 564 |
+
"step": 860000
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 4.61,
|
| 568 |
+
"learning_rate": 2.694014732476679e-05,
|
| 569 |
+
"loss": 1.6631,
|
| 570 |
+
"step": 870000
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 4.67,
|
| 574 |
+
"learning_rate": 2.6675090022963152e-05,
|
| 575 |
+
"loss": 1.6616,
|
| 576 |
+
"step": 880000
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 4.72,
|
| 580 |
+
"learning_rate": 2.641000620482279e-05,
|
| 581 |
+
"loss": 1.6594,
|
| 582 |
+
"step": 890000
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 4.77,
|
| 586 |
+
"learning_rate": 2.614494890301915e-05,
|
| 587 |
+
"loss": 1.6588,
|
| 588 |
+
"step": 900000
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 4.83,
|
| 592 |
+
"learning_rate": 2.5879865084878797e-05,
|
| 593 |
+
"loss": 1.6582,
|
| 594 |
+
"step": 910000
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 4.88,
|
| 598 |
+
"learning_rate": 2.5614834299411866e-05,
|
| 599 |
+
"loss": 1.6569,
|
| 600 |
+
"step": 920000
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 4.93,
|
| 604 |
+
"learning_rate": 2.5349750481271512e-05,
|
| 605 |
+
"loss": 1.6539,
|
| 606 |
+
"step": 930000
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 4.99,
|
| 610 |
+
"learning_rate": 2.508471969580459e-05,
|
| 611 |
+
"loss": 1.6559,
|
| 612 |
+
"step": 940000
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 5.0,
|
| 616 |
+
"eval_bleu": 29.5757,
|
| 617 |
+
"eval_gen_len": 65.8046,
|
| 618 |
+
"eval_loss": 1.7467188835144043,
|
| 619 |
+
"eval_runtime": 2534.3035,
|
| 620 |
+
"eval_samples_per_second": 6.773,
|
| 621 |
+
"eval_steps_per_second": 0.423,
|
| 622 |
+
"step": 942815
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 5.04,
|
| 626 |
+
"learning_rate": 2.4819635877664228e-05,
|
| 627 |
+
"loss": 1.6372,
|
| 628 |
+
"step": 950000
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 5.09,
|
| 632 |
+
"learning_rate": 2.455457857586059e-05,
|
| 633 |
+
"loss": 1.6316,
|
| 634 |
+
"step": 960000
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 5.14,
|
| 638 |
+
"learning_rate": 2.4289494757720233e-05,
|
| 639 |
+
"loss": 1.6343,
|
| 640 |
+
"step": 970000
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 5.2,
|
| 644 |
+
"learning_rate": 2.4024463972253306e-05,
|
| 645 |
+
"loss": 1.6344,
|
| 646 |
+
"step": 980000
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 5.25,
|
| 650 |
+
"learning_rate": 2.3759406670449666e-05,
|
| 651 |
+
"loss": 1.6345,
|
| 652 |
+
"step": 990000
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"epoch": 5.3,
|
| 656 |
+
"learning_rate": 2.3494349368646025e-05,
|
| 657 |
+
"loss": 1.6343,
|
| 658 |
+
"step": 1000000
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 5.36,
|
| 662 |
+
"learning_rate": 2.322929206684238e-05,
|
| 663 |
+
"loss": 1.6353,
|
| 664 |
+
"step": 1010000
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 5.41,
|
| 668 |
+
"learning_rate": 2.296423476503874e-05,
|
| 669 |
+
"loss": 1.6329,
|
| 670 |
+
"step": 1020000
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 5.46,
|
| 674 |
+
"learning_rate": 2.26991774632351e-05,
|
| 675 |
+
"loss": 1.6338,
|
| 676 |
+
"step": 1030000
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 5.52,
|
| 680 |
+
"learning_rate": 2.2434093645094746e-05,
|
| 681 |
+
"loss": 1.6341,
|
| 682 |
+
"step": 1040000
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 5.57,
|
| 686 |
+
"learning_rate": 2.216906285962782e-05,
|
| 687 |
+
"loss": 1.633,
|
| 688 |
+
"step": 1050000
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 5.62,
|
| 692 |
+
"learning_rate": 2.1904005557824175e-05,
|
| 693 |
+
"loss": 1.6322,
|
| 694 |
+
"step": 1060000
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 5.67,
|
| 698 |
+
"learning_rate": 2.163892173968382e-05,
|
| 699 |
+
"loss": 1.6314,
|
| 700 |
+
"step": 1070000
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 5.73,
|
| 704 |
+
"learning_rate": 2.1373864437880177e-05,
|
| 705 |
+
"loss": 1.6301,
|
| 706 |
+
"step": 1080000
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 5.78,
|
| 710 |
+
"learning_rate": 2.110880713607654e-05,
|
| 711 |
+
"loss": 1.6313,
|
| 712 |
+
"step": 1090000
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 5.83,
|
| 716 |
+
"learning_rate": 2.0843749834272896e-05,
|
| 717 |
+
"loss": 1.6313,
|
| 718 |
+
"step": 1100000
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 5.89,
|
| 722 |
+
"learning_rate": 2.057871904880597e-05,
|
| 723 |
+
"loss": 1.6291,
|
| 724 |
+
"step": 1110000
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 5.94,
|
| 728 |
+
"learning_rate": 2.0313635230665615e-05,
|
| 729 |
+
"loss": 1.627,
|
| 730 |
+
"step": 1120000
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 5.99,
|
| 734 |
+
"learning_rate": 2.0048604445198688e-05,
|
| 735 |
+
"loss": 1.6279,
|
| 736 |
+
"step": 1130000
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 6.0,
|
| 740 |
+
"eval_bleu": 29.7512,
|
| 741 |
+
"eval_gen_len": 66.0482,
|
| 742 |
+
"eval_loss": 1.7236360311508179,
|
| 743 |
+
"eval_runtime": 2786.837,
|
| 744 |
+
"eval_samples_per_second": 6.159,
|
| 745 |
+
"eval_steps_per_second": 0.385,
|
| 746 |
+
"step": 1131378
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 6.05,
|
| 750 |
+
"learning_rate": 1.9783520627058334e-05,
|
| 751 |
+
"loss": 1.6085,
|
| 752 |
+
"step": 1140000
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 6.1,
|
| 756 |
+
"learning_rate": 1.9518489841591407e-05,
|
| 757 |
+
"loss": 1.6066,
|
| 758 |
+
"step": 1150000
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 6.15,
|
| 762 |
+
"learning_rate": 1.9253432539787763e-05,
|
| 763 |
+
"loss": 1.6079,
|
| 764 |
+
"step": 1160000
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 6.2,
|
| 768 |
+
"learning_rate": 1.8988375237984123e-05,
|
| 769 |
+
"loss": 1.6108,
|
| 770 |
+
"step": 1170000
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 6.26,
|
| 774 |
+
"learning_rate": 1.8723317936180482e-05,
|
| 775 |
+
"loss": 1.6083,
|
| 776 |
+
"step": 1180000
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 6.31,
|
| 780 |
+
"learning_rate": 1.8458234118040125e-05,
|
| 781 |
+
"loss": 1.6082,
|
| 782 |
+
"step": 1190000
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 6.36,
|
| 786 |
+
"learning_rate": 1.8193176816236484e-05,
|
| 787 |
+
"loss": 1.61,
|
| 788 |
+
"step": 1200000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 6.42,
|
| 792 |
+
"learning_rate": 1.7928119514432844e-05,
|
| 793 |
+
"loss": 1.6092,
|
| 794 |
+
"step": 1210000
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 6.47,
|
| 798 |
+
"learning_rate": 1.7663062212629203e-05,
|
| 799 |
+
"loss": 1.6093,
|
| 800 |
+
"step": 1220000
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 6.52,
|
| 804 |
+
"learning_rate": 1.7397978394488846e-05,
|
| 805 |
+
"loss": 1.6106,
|
| 806 |
+
"step": 1230000
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 6.58,
|
| 810 |
+
"learning_rate": 1.7132921092685202e-05,
|
| 811 |
+
"loss": 1.6084,
|
| 812 |
+
"step": 1240000
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 6.63,
|
| 816 |
+
"learning_rate": 1.6867863790881565e-05,
|
| 817 |
+
"loss": 1.6094,
|
| 818 |
+
"step": 1250000
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 6.68,
|
| 822 |
+
"learning_rate": 1.6602779972741207e-05,
|
| 823 |
+
"loss": 1.6069,
|
| 824 |
+
"step": 1260000
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 6.74,
|
| 828 |
+
"learning_rate": 1.633774918727428e-05,
|
| 829 |
+
"loss": 1.6069,
|
| 830 |
+
"step": 1270000
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 6.79,
|
| 834 |
+
"learning_rate": 1.6072665369133923e-05,
|
| 835 |
+
"loss": 1.6075,
|
| 836 |
+
"step": 1280000
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 6.84,
|
| 840 |
+
"learning_rate": 1.5807608067330283e-05,
|
| 841 |
+
"loss": 1.6051,
|
| 842 |
+
"step": 1290000
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 6.89,
|
| 846 |
+
"learning_rate": 1.5542550765526642e-05,
|
| 847 |
+
"loss": 1.6044,
|
| 848 |
+
"step": 1300000
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 6.95,
|
| 852 |
+
"learning_rate": 1.5277493463723e-05,
|
| 853 |
+
"loss": 1.6053,
|
| 854 |
+
"step": 1310000
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 7.0,
|
| 858 |
+
"eval_bleu": 29.916,
|
| 859 |
+
"eval_gen_len": 66.0031,
|
| 860 |
+
"eval_loss": 1.7136626243591309,
|
| 861 |
+
"eval_runtime": 2659.0199,
|
| 862 |
+
"eval_samples_per_second": 6.455,
|
| 863 |
+
"eval_steps_per_second": 0.404,
|
| 864 |
+
"step": 1319941
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 7.0,
|
| 868 |
+
"learning_rate": 1.5012436161919361e-05,
|
| 869 |
+
"loss": 1.6068,
|
| 870 |
+
"step": 1320000
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 7.05,
|
| 874 |
+
"learning_rate": 1.4747378860115719e-05,
|
| 875 |
+
"loss": 1.5851,
|
| 876 |
+
"step": 1330000
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 7.11,
|
| 880 |
+
"learning_rate": 1.4482321558312076e-05,
|
| 881 |
+
"loss": 1.5864,
|
| 882 |
+
"step": 1340000
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 7.16,
|
| 886 |
+
"learning_rate": 1.4217264256508436e-05,
|
| 887 |
+
"loss": 1.5865,
|
| 888 |
+
"step": 1350000
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 7.21,
|
| 892 |
+
"learning_rate": 1.3952206954704794e-05,
|
| 893 |
+
"loss": 1.5866,
|
| 894 |
+
"step": 1360000
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 7.27,
|
| 898 |
+
"learning_rate": 1.3687149652901151e-05,
|
| 899 |
+
"loss": 1.589,
|
| 900 |
+
"step": 1370000
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 7.32,
|
| 904 |
+
"learning_rate": 1.3422092351097513e-05,
|
| 905 |
+
"loss": 1.5885,
|
| 906 |
+
"step": 1380000
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 7.37,
|
| 910 |
+
"learning_rate": 1.315703504929387e-05,
|
| 911 |
+
"loss": 1.589,
|
| 912 |
+
"step": 1390000
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 7.42,
|
| 916 |
+
"learning_rate": 1.2891951231153515e-05,
|
| 917 |
+
"loss": 1.5882,
|
| 918 |
+
"step": 1400000
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 7.48,
|
| 922 |
+
"learning_rate": 1.2626893929349873e-05,
|
| 923 |
+
"loss": 1.5897,
|
| 924 |
+
"step": 1410000
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 7.53,
|
| 928 |
+
"learning_rate": 1.2361836627546232e-05,
|
| 929 |
+
"loss": 1.5868,
|
| 930 |
+
"step": 1420000
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 7.58,
|
| 934 |
+
"learning_rate": 1.2096805842079307e-05,
|
| 935 |
+
"loss": 1.5884,
|
| 936 |
+
"step": 1430000
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 7.64,
|
| 940 |
+
"learning_rate": 1.1831748540275664e-05,
|
| 941 |
+
"loss": 1.5896,
|
| 942 |
+
"step": 1440000
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 7.69,
|
| 946 |
+
"learning_rate": 1.1566664722135307e-05,
|
| 947 |
+
"loss": 1.5874,
|
| 948 |
+
"step": 1450000
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 7.74,
|
| 952 |
+
"learning_rate": 1.1301633936668382e-05,
|
| 953 |
+
"loss": 1.5874,
|
| 954 |
+
"step": 1460000
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 7.8,
|
| 958 |
+
"learning_rate": 1.1036576634864741e-05,
|
| 959 |
+
"loss": 1.5878,
|
| 960 |
+
"step": 1470000
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 7.85,
|
| 964 |
+
"learning_rate": 1.0771492816724385e-05,
|
| 965 |
+
"loss": 1.5864,
|
| 966 |
+
"step": 1480000
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 7.9,
|
| 970 |
+
"learning_rate": 1.0506408998584028e-05,
|
| 971 |
+
"loss": 1.5848,
|
| 972 |
+
"step": 1490000
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 7.95,
|
| 976 |
+
"learning_rate": 1.0241351696780386e-05,
|
| 977 |
+
"loss": 1.5871,
|
| 978 |
+
"step": 1500000
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 8.0,
|
| 982 |
+
"eval_bleu": 30.1671,
|
| 983 |
+
"eval_gen_len": 65.8853,
|
| 984 |
+
"eval_loss": 1.7006529569625854,
|
| 985 |
+
"eval_runtime": 2553.3513,
|
| 986 |
+
"eval_samples_per_second": 6.723,
|
| 987 |
+
"eval_steps_per_second": 0.42,
|
| 988 |
+
"step": 1508504
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 8.01,
|
| 992 |
+
"learning_rate": 9.97626787864003e-06,
|
| 993 |
+
"loss": 1.582,
|
| 994 |
+
"step": 1510000
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 8.06,
|
| 998 |
+
"learning_rate": 9.711237093173105e-06,
|
| 999 |
+
"loss": 1.5691,
|
| 1000 |
+
"step": 1520000
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 8.11,
|
| 1004 |
+
"learning_rate": 9.446153275032749e-06,
|
| 1005 |
+
"loss": 1.5695,
|
| 1006 |
+
"step": 1530000
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 8.17,
|
| 1010 |
+
"learning_rate": 9.181095973229107e-06,
|
| 1011 |
+
"loss": 1.5702,
|
| 1012 |
+
"step": 1540000
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 8.22,
|
| 1016 |
+
"learning_rate": 8.91601215508875e-06,
|
| 1017 |
+
"loss": 1.5701,
|
| 1018 |
+
"step": 1550000
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 8.27,
|
| 1022 |
+
"learning_rate": 8.650954853285109e-06,
|
| 1023 |
+
"loss": 1.5711,
|
| 1024 |
+
"step": 1560000
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 8.33,
|
| 1028 |
+
"learning_rate": 8.385897551481469e-06,
|
| 1029 |
+
"loss": 1.5693,
|
| 1030 |
+
"step": 1570000
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 8.38,
|
| 1034 |
+
"learning_rate": 8.120840249677828e-06,
|
| 1035 |
+
"loss": 1.5703,
|
| 1036 |
+
"step": 1580000
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 8.43,
|
| 1040 |
+
"learning_rate": 7.855809464210901e-06,
|
| 1041 |
+
"loss": 1.571,
|
| 1042 |
+
"step": 1590000
|
| 1043 |
+
},
|
| 1044 |
+
{
|
| 1045 |
+
"epoch": 8.49,
|
| 1046 |
+
"learning_rate": 7.5907521624072595e-06,
|
| 1047 |
+
"loss": 1.5709,
|
| 1048 |
+
"step": 1600000
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 8.54,
|
| 1052 |
+
"learning_rate": 7.325694860603618e-06,
|
| 1053 |
+
"loss": 1.5698,
|
| 1054 |
+
"step": 1610000
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 8.59,
|
| 1058 |
+
"learning_rate": 7.0606110424632625e-06,
|
| 1059 |
+
"loss": 1.5716,
|
| 1060 |
+
"step": 1620000
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 8.64,
|
| 1064 |
+
"learning_rate": 6.79555374065962e-06,
|
| 1065 |
+
"loss": 1.5702,
|
| 1066 |
+
"step": 1630000
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 8.7,
|
| 1070 |
+
"learning_rate": 6.530496438855979e-06,
|
| 1071 |
+
"loss": 1.5698,
|
| 1072 |
+
"step": 1640000
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 8.75,
|
| 1076 |
+
"learning_rate": 6.265439137052338e-06,
|
| 1077 |
+
"loss": 1.5699,
|
| 1078 |
+
"step": 1650000
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 8.8,
|
| 1082 |
+
"learning_rate": 6.000381835248697e-06,
|
| 1083 |
+
"loss": 1.5683,
|
| 1084 |
+
"step": 1660000
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 8.86,
|
| 1088 |
+
"learning_rate": 5.735351049781771e-06,
|
| 1089 |
+
"loss": 1.5683,
|
| 1090 |
+
"step": 1670000
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 8.91,
|
| 1094 |
+
"learning_rate": 5.470267231641415e-06,
|
| 1095 |
+
"loss": 1.5701,
|
| 1096 |
+
"step": 1680000
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 8.96,
|
| 1100 |
+
"learning_rate": 5.205236446174489e-06,
|
| 1101 |
+
"loss": 1.5694,
|
| 1102 |
+
"step": 1690000
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 9.0,
|
| 1106 |
+
"eval_bleu": 30.3613,
|
| 1107 |
+
"eval_gen_len": 65.9506,
|
| 1108 |
+
"eval_loss": 1.692053198814392,
|
| 1109 |
+
"eval_runtime": 2749.5168,
|
| 1110 |
+
"eval_samples_per_second": 6.243,
|
| 1111 |
+
"eval_steps_per_second": 0.39,
|
| 1112 |
+
"step": 1697067
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 9.02,
|
| 1116 |
+
"learning_rate": 4.940126111697417e-06,
|
| 1117 |
+
"loss": 1.5631,
|
| 1118 |
+
"step": 1700000
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 9.07,
|
| 1122 |
+
"learning_rate": 4.675095326230491e-06,
|
| 1123 |
+
"loss": 1.5545,
|
| 1124 |
+
"step": 1710000
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 9.12,
|
| 1128 |
+
"learning_rate": 4.4100380244268495e-06,
|
| 1129 |
+
"loss": 1.554,
|
| 1130 |
+
"step": 1720000
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 9.17,
|
| 1134 |
+
"learning_rate": 4.144980722623208e-06,
|
| 1135 |
+
"loss": 1.5572,
|
| 1136 |
+
"step": 1730000
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 9.23,
|
| 1140 |
+
"learning_rate": 3.879896904482852e-06,
|
| 1141 |
+
"loss": 1.5576,
|
| 1142 |
+
"step": 1740000
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 9.28,
|
| 1146 |
+
"learning_rate": 3.614866119015926e-06,
|
| 1147 |
+
"loss": 1.5542,
|
| 1148 |
+
"step": 1750000
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 9.33,
|
| 1152 |
+
"learning_rate": 3.3497823008755693e-06,
|
| 1153 |
+
"loss": 1.5546,
|
| 1154 |
+
"step": 1760000
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 9.39,
|
| 1158 |
+
"learning_rate": 3.0847249990719284e-06,
|
| 1159 |
+
"loss": 1.5557,
|
| 1160 |
+
"step": 1770000
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 9.44,
|
| 1164 |
+
"learning_rate": 2.819667697268287e-06,
|
| 1165 |
+
"loss": 1.5561,
|
| 1166 |
+
"step": 1780000
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 9.49,
|
| 1170 |
+
"learning_rate": 2.554583879127931e-06,
|
| 1171 |
+
"loss": 1.5553,
|
| 1172 |
+
"step": 1790000
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 9.55,
|
| 1176 |
+
"learning_rate": 2.2895530936610047e-06,
|
| 1177 |
+
"loss": 1.5558,
|
| 1178 |
+
"step": 1800000
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 9.6,
|
| 1182 |
+
"learning_rate": 2.0244957918573637e-06,
|
| 1183 |
+
"loss": 1.554,
|
| 1184 |
+
"step": 1810000
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 9.65,
|
| 1188 |
+
"learning_rate": 1.759411973717007e-06,
|
| 1189 |
+
"loss": 1.5543,
|
| 1190 |
+
"step": 1820000
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 9.7,
|
| 1194 |
+
"learning_rate": 1.494381188250081e-06,
|
| 1195 |
+
"loss": 1.5553,
|
| 1196 |
+
"step": 1830000
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 9.76,
|
| 1200 |
+
"learning_rate": 1.2292973701097247e-06,
|
| 1201 |
+
"loss": 1.5543,
|
| 1202 |
+
"step": 1840000
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 9.81,
|
| 1206 |
+
"learning_rate": 9.642665846427986e-07,
|
| 1207 |
+
"loss": 1.5546,
|
| 1208 |
+
"step": 1850000
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 9.86,
|
| 1212 |
+
"learning_rate": 6.992092828391572e-07,
|
| 1213 |
+
"loss": 1.5534,
|
| 1214 |
+
"step": 1860000
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 9.92,
|
| 1218 |
+
"learning_rate": 4.3415198103551603e-07,
|
| 1219 |
+
"loss": 1.5536,
|
| 1220 |
+
"step": 1870000
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 9.97,
|
| 1224 |
+
"learning_rate": 1.690681628951597e-07,
|
| 1225 |
+
"loss": 1.5539,
|
| 1226 |
+
"step": 1880000
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 10.0,
|
| 1230 |
+
"eval_bleu": 30.4049,
|
| 1231 |
+
"eval_gen_len": 66.0487,
|
| 1232 |
+
"eval_loss": 1.6851532459259033,
|
| 1233 |
+
"eval_runtime": 2548.9872,
|
| 1234 |
+
"eval_samples_per_second": 6.734,
|
| 1235 |
+
"eval_steps_per_second": 0.421,
|
| 1236 |
+
"step": 1885630
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 10.0,
|
| 1240 |
+
"step": 1885630,
|
| 1241 |
+
"total_flos": 5.018247011089908e+18,
|
| 1242 |
+
"train_loss": 1.743612184965204,
|
| 1243 |
+
"train_runtime": 420652.6588,
|
| 1244 |
+
"train_samples_per_second": 215.166,
|
| 1245 |
+
"train_steps_per_second": 4.483
|
| 1246 |
+
}
|
| 1247 |
+
],
|
| 1248 |
+
"max_steps": 1885630,
|
| 1249 |
+
"num_train_epochs": 10,
|
| 1250 |
+
"total_flos": 5.018247011089908e+18,
|
| 1251 |
+
"trial_name": null,
|
| 1252 |
+
"trial_params": null
|
| 1253 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:16c94757d1777e2cea000476da865ed5e54d61b84197eea5f402bbddd3f7735a
|
| 3 |
+
size 2991
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|