Commit
•
9ab896f
1
Parent(s):
9e59cb5
initial commit
Browse files- config.json +36 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +1 -0
- spiece.model +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +1 -0
- trainer_state.json +516 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "sarakolding/daT5-base",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"MT5ForConditionalGeneration"
|
| 5 |
+
],
|
| 6 |
+
"d_ff": 2048,
|
| 7 |
+
"d_kv": 64,
|
| 8 |
+
"d_model": 768,
|
| 9 |
+
"decoder_start_token_id": 0,
|
| 10 |
+
"dropout_rate": 0.01,
|
| 11 |
+
"early_stopping": true,
|
| 12 |
+
"eos_token_id": 1,
|
| 13 |
+
"feed_forward_proj": "gated-gelu",
|
| 14 |
+
"initializer_factor": 1.0,
|
| 15 |
+
"is_encoder_decoder": true,
|
| 16 |
+
"layer_norm_epsilon": 1e-06,
|
| 17 |
+
"length_penalty": 5,
|
| 18 |
+
"max_length": 128,
|
| 19 |
+
"min_length": 15,
|
| 20 |
+
"model_type": "mt5",
|
| 21 |
+
"no_repeat_ngram_size": 3,
|
| 22 |
+
"num_beams": 4,
|
| 23 |
+
"num_decoder_layers": 12,
|
| 24 |
+
"num_heads": 12,
|
| 25 |
+
"num_layers": 12,
|
| 26 |
+
"output_past": true,
|
| 27 |
+
"pad_token_id": 0,
|
| 28 |
+
"relative_attention_max_distance": 128,
|
| 29 |
+
"relative_attention_num_buckets": 32,
|
| 30 |
+
"tie_word_embeddings": false,
|
| 31 |
+
"tokenizer_class": "T5Tokenizer",
|
| 32 |
+
"torch_dtype": "float32",
|
| 33 |
+
"transformers_version": "4.20.0.dev0",
|
| 34 |
+
"use_cache": true,
|
| 35 |
+
"vocab_size": 30000
|
| 36 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fc91e2fa415ded95ce0b768fcf744fde086c771a2610643122ac4a93e015b460
|
| 3 |
+
size 1954640153
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a46dc9d540973454dc66d8964580d2bacf8336960c1d019241dbf2f5ac7dfa1d
|
| 3 |
+
size 977332173
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0362545844db8f9d3ba81e0d323c72d39cf5ff6a92b9f21b560faeaf428e6e6e
|
| 3 |
+
size 14503
|
scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bb514f3549501e7fbfb775146d40e9d3027ea20fa05036f069600337153c15ff
|
| 3 |
+
size 559
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5343e0993f724df030d39b5a8bed28c5148bbf4ad76c9e3fb3569506e763e049
|
| 3 |
+
size 623
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>"}
|
spiece.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dabe934d94d2182b6b15a66f895d7de8197a2c5e478c5a9e16746c51005e8b22
|
| 3 |
+
size 766659
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "extra_ids": 0, "additional_special_tokens": null, "sp_model_kwargs": {}, "special_tokens_map_file": "/home/ucloud/.cache/huggingface/transformers/90e4e02f17b0b14f97d96c94bd8e5652250a216e5950bb43c79afe24011e808c.294ebaa4cd17bb284635004c92d2c4d522ec488c828dcce0c2471b6f28e3fe82", "name_or_path": "sarakolding/daT5-base", "tokenizer_class": "T5Tokenizer"}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,516 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 2.132361888885498,
|
| 3 |
+
"best_model_checkpoint": "./26-125356_megasuperkanin/checkpoint-100000",
|
| 4 |
+
"epoch": 0.9769822970807769,
|
| 5 |
+
"global_step": 100000,
|
| 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": 5e-05,
|
| 13 |
+
"loss": 2.6761,
|
| 14 |
+
"step": 2500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.05,
|
| 18 |
+
"learning_rate": 5e-05,
|
| 19 |
+
"loss": 2.551,
|
| 20 |
+
"step": 5000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.05,
|
| 24 |
+
"eval_gen_len": 28.4674,
|
| 25 |
+
"eval_loss": 2.423037052154541,
|
| 26 |
+
"eval_rouge1": 0.214,
|
| 27 |
+
"eval_rouge2": 0.0668,
|
| 28 |
+
"eval_rougeL": 0.1717,
|
| 29 |
+
"eval_rougeLsum": 0.1777,
|
| 30 |
+
"eval_runtime": 1015.6418,
|
| 31 |
+
"eval_samples_per_second": 2.265,
|
| 32 |
+
"eval_steps_per_second": 0.284,
|
| 33 |
+
"step": 5000
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.07,
|
| 37 |
+
"learning_rate": 5e-05,
|
| 38 |
+
"loss": 2.5186,
|
| 39 |
+
"step": 7500
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.1,
|
| 43 |
+
"learning_rate": 5e-05,
|
| 44 |
+
"loss": 2.4717,
|
| 45 |
+
"step": 10000
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.1,
|
| 49 |
+
"eval_gen_len": 25.6604,
|
| 50 |
+
"eval_loss": 2.3709843158721924,
|
| 51 |
+
"eval_rouge1": 0.2071,
|
| 52 |
+
"eval_rouge2": 0.0634,
|
| 53 |
+
"eval_rougeL": 0.1686,
|
| 54 |
+
"eval_rougeLsum": 0.1745,
|
| 55 |
+
"eval_runtime": 951.1096,
|
| 56 |
+
"eval_samples_per_second": 2.418,
|
| 57 |
+
"eval_steps_per_second": 0.303,
|
| 58 |
+
"step": 10000
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.12,
|
| 62 |
+
"learning_rate": 5e-05,
|
| 63 |
+
"loss": 2.4593,
|
| 64 |
+
"step": 12500
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.15,
|
| 68 |
+
"learning_rate": 5e-05,
|
| 69 |
+
"loss": 2.4281,
|
| 70 |
+
"step": 15000
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.15,
|
| 74 |
+
"eval_gen_len": 28.8296,
|
| 75 |
+
"eval_loss": 2.3228819370269775,
|
| 76 |
+
"eval_rouge1": 0.2137,
|
| 77 |
+
"eval_rouge2": 0.0662,
|
| 78 |
+
"eval_rougeL": 0.1711,
|
| 79 |
+
"eval_rougeLsum": 0.1768,
|
| 80 |
+
"eval_runtime": 1022.9494,
|
| 81 |
+
"eval_samples_per_second": 2.248,
|
| 82 |
+
"eval_steps_per_second": 0.282,
|
| 83 |
+
"step": 15000
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.17,
|
| 87 |
+
"learning_rate": 5e-05,
|
| 88 |
+
"loss": 2.4049,
|
| 89 |
+
"step": 17500
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.2,
|
| 93 |
+
"learning_rate": 5e-05,
|
| 94 |
+
"loss": 2.3735,
|
| 95 |
+
"step": 20000
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.2,
|
| 99 |
+
"eval_gen_len": 29.9183,
|
| 100 |
+
"eval_loss": 2.2881429195404053,
|
| 101 |
+
"eval_rouge1": 0.2164,
|
| 102 |
+
"eval_rouge2": 0.0668,
|
| 103 |
+
"eval_rougeL": 0.1735,
|
| 104 |
+
"eval_rougeLsum": 0.1808,
|
| 105 |
+
"eval_runtime": 1036.2984,
|
| 106 |
+
"eval_samples_per_second": 2.219,
|
| 107 |
+
"eval_steps_per_second": 0.278,
|
| 108 |
+
"step": 20000
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.22,
|
| 112 |
+
"learning_rate": 5e-05,
|
| 113 |
+
"loss": 2.3732,
|
| 114 |
+
"step": 22500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.24,
|
| 118 |
+
"learning_rate": 5e-05,
|
| 119 |
+
"loss": 2.377,
|
| 120 |
+
"step": 25000
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.24,
|
| 124 |
+
"eval_gen_len": 29.5183,
|
| 125 |
+
"eval_loss": 2.2759358882904053,
|
| 126 |
+
"eval_rouge1": 0.2209,
|
| 127 |
+
"eval_rouge2": 0.0694,
|
| 128 |
+
"eval_rougeL": 0.1782,
|
| 129 |
+
"eval_rougeLsum": 0.1851,
|
| 130 |
+
"eval_runtime": 1036.1071,
|
| 131 |
+
"eval_samples_per_second": 2.22,
|
| 132 |
+
"eval_steps_per_second": 0.278,
|
| 133 |
+
"step": 25000
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 0.27,
|
| 137 |
+
"learning_rate": 5e-05,
|
| 138 |
+
"loss": 2.3513,
|
| 139 |
+
"step": 27500
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 0.29,
|
| 143 |
+
"learning_rate": 5e-05,
|
| 144 |
+
"loss": 2.3444,
|
| 145 |
+
"step": 30000
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.29,
|
| 149 |
+
"eval_gen_len": 29.3183,
|
| 150 |
+
"eval_loss": 2.2552034854888916,
|
| 151 |
+
"eval_rouge1": 0.2194,
|
| 152 |
+
"eval_rouge2": 0.0679,
|
| 153 |
+
"eval_rougeL": 0.1757,
|
| 154 |
+
"eval_rougeLsum": 0.1829,
|
| 155 |
+
"eval_runtime": 1037.4604,
|
| 156 |
+
"eval_samples_per_second": 2.217,
|
| 157 |
+
"eval_steps_per_second": 0.278,
|
| 158 |
+
"step": 30000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.32,
|
| 162 |
+
"learning_rate": 5e-05,
|
| 163 |
+
"loss": 2.3504,
|
| 164 |
+
"step": 32500
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.34,
|
| 168 |
+
"learning_rate": 5e-05,
|
| 169 |
+
"loss": 2.3203,
|
| 170 |
+
"step": 35000
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.34,
|
| 174 |
+
"eval_gen_len": 32.2061,
|
| 175 |
+
"eval_loss": 2.235518455505371,
|
| 176 |
+
"eval_rouge1": 0.2284,
|
| 177 |
+
"eval_rouge2": 0.0722,
|
| 178 |
+
"eval_rougeL": 0.1819,
|
| 179 |
+
"eval_rougeLsum": 0.1892,
|
| 180 |
+
"eval_runtime": 1121.1561,
|
| 181 |
+
"eval_samples_per_second": 2.051,
|
| 182 |
+
"eval_steps_per_second": 0.257,
|
| 183 |
+
"step": 35000
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 0.37,
|
| 187 |
+
"learning_rate": 5e-05,
|
| 188 |
+
"loss": 2.3087,
|
| 189 |
+
"step": 37500
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.39,
|
| 193 |
+
"learning_rate": 5e-05,
|
| 194 |
+
"loss": 2.3132,
|
| 195 |
+
"step": 40000
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.39,
|
| 199 |
+
"eval_gen_len": 29.5452,
|
| 200 |
+
"eval_loss": 2.2289836406707764,
|
| 201 |
+
"eval_rouge1": 0.2183,
|
| 202 |
+
"eval_rouge2": 0.0673,
|
| 203 |
+
"eval_rougeL": 0.1759,
|
| 204 |
+
"eval_rougeLsum": 0.1827,
|
| 205 |
+
"eval_runtime": 1055.2895,
|
| 206 |
+
"eval_samples_per_second": 2.179,
|
| 207 |
+
"eval_steps_per_second": 0.273,
|
| 208 |
+
"step": 40000
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.42,
|
| 212 |
+
"learning_rate": 5e-05,
|
| 213 |
+
"loss": 2.3063,
|
| 214 |
+
"step": 42500
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.44,
|
| 218 |
+
"learning_rate": 5e-05,
|
| 219 |
+
"loss": 2.3116,
|
| 220 |
+
"step": 45000
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.44,
|
| 224 |
+
"eval_gen_len": 30.2935,
|
| 225 |
+
"eval_loss": 2.218207359313965,
|
| 226 |
+
"eval_rouge1": 0.2239,
|
| 227 |
+
"eval_rouge2": 0.07,
|
| 228 |
+
"eval_rougeL": 0.1798,
|
| 229 |
+
"eval_rougeLsum": 0.1879,
|
| 230 |
+
"eval_runtime": 1063.5185,
|
| 231 |
+
"eval_samples_per_second": 2.163,
|
| 232 |
+
"eval_steps_per_second": 0.271,
|
| 233 |
+
"step": 45000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.46,
|
| 237 |
+
"learning_rate": 5e-05,
|
| 238 |
+
"loss": 2.3014,
|
| 239 |
+
"step": 47500
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 0.49,
|
| 243 |
+
"learning_rate": 5e-05,
|
| 244 |
+
"loss": 2.2852,
|
| 245 |
+
"step": 50000
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 0.49,
|
| 249 |
+
"eval_gen_len": 28.6443,
|
| 250 |
+
"eval_loss": 2.2090706825256348,
|
| 251 |
+
"eval_rouge1": 0.2251,
|
| 252 |
+
"eval_rouge2": 0.0703,
|
| 253 |
+
"eval_rougeL": 0.1812,
|
| 254 |
+
"eval_rougeLsum": 0.1887,
|
| 255 |
+
"eval_runtime": 1045.7282,
|
| 256 |
+
"eval_samples_per_second": 2.199,
|
| 257 |
+
"eval_steps_per_second": 0.275,
|
| 258 |
+
"step": 50000
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 0.51,
|
| 262 |
+
"learning_rate": 5e-05,
|
| 263 |
+
"loss": 2.2963,
|
| 264 |
+
"step": 52500
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.54,
|
| 268 |
+
"learning_rate": 5e-05,
|
| 269 |
+
"loss": 2.2683,
|
| 270 |
+
"step": 55000
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 0.54,
|
| 274 |
+
"eval_gen_len": 29.9661,
|
| 275 |
+
"eval_loss": 2.1879115104675293,
|
| 276 |
+
"eval_rouge1": 0.2257,
|
| 277 |
+
"eval_rouge2": 0.0716,
|
| 278 |
+
"eval_rougeL": 0.1806,
|
| 279 |
+
"eval_rougeLsum": 0.1876,
|
| 280 |
+
"eval_runtime": 1061.3075,
|
| 281 |
+
"eval_samples_per_second": 2.167,
|
| 282 |
+
"eval_steps_per_second": 0.271,
|
| 283 |
+
"step": 55000
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.56,
|
| 287 |
+
"learning_rate": 5e-05,
|
| 288 |
+
"loss": 2.2735,
|
| 289 |
+
"step": 57500
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.59,
|
| 293 |
+
"learning_rate": 5e-05,
|
| 294 |
+
"loss": 2.2614,
|
| 295 |
+
"step": 60000
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.59,
|
| 299 |
+
"eval_gen_len": 30.4435,
|
| 300 |
+
"eval_loss": 2.1871089935302734,
|
| 301 |
+
"eval_rouge1": 0.2316,
|
| 302 |
+
"eval_rouge2": 0.075,
|
| 303 |
+
"eval_rougeL": 0.1863,
|
| 304 |
+
"eval_rougeLsum": 0.1936,
|
| 305 |
+
"eval_runtime": 1083.7377,
|
| 306 |
+
"eval_samples_per_second": 2.122,
|
| 307 |
+
"eval_steps_per_second": 0.266,
|
| 308 |
+
"step": 60000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.61,
|
| 312 |
+
"learning_rate": 5e-05,
|
| 313 |
+
"loss": 2.2735,
|
| 314 |
+
"step": 62500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.64,
|
| 318 |
+
"learning_rate": 5e-05,
|
| 319 |
+
"loss": 2.252,
|
| 320 |
+
"step": 65000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.64,
|
| 324 |
+
"eval_gen_len": 30.6239,
|
| 325 |
+
"eval_loss": 2.175469160079956,
|
| 326 |
+
"eval_rouge1": 0.226,
|
| 327 |
+
"eval_rouge2": 0.0729,
|
| 328 |
+
"eval_rougeL": 0.1834,
|
| 329 |
+
"eval_rougeLsum": 0.1914,
|
| 330 |
+
"eval_runtime": 1080.4009,
|
| 331 |
+
"eval_samples_per_second": 2.129,
|
| 332 |
+
"eval_steps_per_second": 0.267,
|
| 333 |
+
"step": 65000
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 0.66,
|
| 337 |
+
"learning_rate": 5e-05,
|
| 338 |
+
"loss": 2.2509,
|
| 339 |
+
"step": 67500
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.68,
|
| 343 |
+
"learning_rate": 5e-05,
|
| 344 |
+
"loss": 2.262,
|
| 345 |
+
"step": 70000
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.68,
|
| 349 |
+
"eval_gen_len": 30.9983,
|
| 350 |
+
"eval_loss": 2.16789174079895,
|
| 351 |
+
"eval_rouge1": 0.2256,
|
| 352 |
+
"eval_rouge2": 0.0716,
|
| 353 |
+
"eval_rougeL": 0.1815,
|
| 354 |
+
"eval_rougeLsum": 0.1889,
|
| 355 |
+
"eval_runtime": 1104.0224,
|
| 356 |
+
"eval_samples_per_second": 2.083,
|
| 357 |
+
"eval_steps_per_second": 0.261,
|
| 358 |
+
"step": 70000
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 0.71,
|
| 362 |
+
"learning_rate": 5e-05,
|
| 363 |
+
"loss": 2.2398,
|
| 364 |
+
"step": 72500
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 0.73,
|
| 368 |
+
"learning_rate": 5e-05,
|
| 369 |
+
"loss": 2.228,
|
| 370 |
+
"step": 75000
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 0.73,
|
| 374 |
+
"eval_gen_len": 29.9704,
|
| 375 |
+
"eval_loss": 2.1669178009033203,
|
| 376 |
+
"eval_rouge1": 0.2253,
|
| 377 |
+
"eval_rouge2": 0.0725,
|
| 378 |
+
"eval_rougeL": 0.1822,
|
| 379 |
+
"eval_rougeLsum": 0.1894,
|
| 380 |
+
"eval_runtime": 1052.7669,
|
| 381 |
+
"eval_samples_per_second": 2.185,
|
| 382 |
+
"eval_steps_per_second": 0.274,
|
| 383 |
+
"step": 75000
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 0.76,
|
| 387 |
+
"learning_rate": 5e-05,
|
| 388 |
+
"loss": 2.25,
|
| 389 |
+
"step": 77500
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.78,
|
| 393 |
+
"learning_rate": 5e-05,
|
| 394 |
+
"loss": 2.234,
|
| 395 |
+
"step": 80000
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.78,
|
| 399 |
+
"eval_gen_len": 29.4826,
|
| 400 |
+
"eval_loss": 2.1604671478271484,
|
| 401 |
+
"eval_rouge1": 0.2283,
|
| 402 |
+
"eval_rouge2": 0.0747,
|
| 403 |
+
"eval_rougeL": 0.1855,
|
| 404 |
+
"eval_rougeLsum": 0.1937,
|
| 405 |
+
"eval_runtime": 1075.8159,
|
| 406 |
+
"eval_samples_per_second": 2.138,
|
| 407 |
+
"eval_steps_per_second": 0.268,
|
| 408 |
+
"step": 80000
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.81,
|
| 412 |
+
"learning_rate": 5e-05,
|
| 413 |
+
"loss": 2.236,
|
| 414 |
+
"step": 82500
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.83,
|
| 418 |
+
"learning_rate": 5e-05,
|
| 419 |
+
"loss": 2.2289,
|
| 420 |
+
"step": 85000
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 0.83,
|
| 424 |
+
"eval_gen_len": 30.0213,
|
| 425 |
+
"eval_loss": 2.1517326831817627,
|
| 426 |
+
"eval_rouge1": 0.2226,
|
| 427 |
+
"eval_rouge2": 0.0705,
|
| 428 |
+
"eval_rougeL": 0.1801,
|
| 429 |
+
"eval_rougeLsum": 0.1873,
|
| 430 |
+
"eval_runtime": 1072.8178,
|
| 431 |
+
"eval_samples_per_second": 2.144,
|
| 432 |
+
"eval_steps_per_second": 0.268,
|
| 433 |
+
"step": 85000
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 0.85,
|
| 437 |
+
"learning_rate": 5e-05,
|
| 438 |
+
"loss": 2.2214,
|
| 439 |
+
"step": 87500
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 0.88,
|
| 443 |
+
"learning_rate": 5e-05,
|
| 444 |
+
"loss": 2.2043,
|
| 445 |
+
"step": 90000
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 0.88,
|
| 449 |
+
"eval_gen_len": 29.5361,
|
| 450 |
+
"eval_loss": 2.1455490589141846,
|
| 451 |
+
"eval_rouge1": 0.2265,
|
| 452 |
+
"eval_rouge2": 0.075,
|
| 453 |
+
"eval_rougeL": 0.1838,
|
| 454 |
+
"eval_rougeLsum": 0.1908,
|
| 455 |
+
"eval_runtime": 1058.731,
|
| 456 |
+
"eval_samples_per_second": 2.172,
|
| 457 |
+
"eval_steps_per_second": 0.272,
|
| 458 |
+
"step": 90000
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.9,
|
| 462 |
+
"learning_rate": 5e-05,
|
| 463 |
+
"loss": 2.2419,
|
| 464 |
+
"step": 92500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.93,
|
| 468 |
+
"learning_rate": 5e-05,
|
| 469 |
+
"loss": 2.2259,
|
| 470 |
+
"step": 95000
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.93,
|
| 474 |
+
"eval_gen_len": 29.6874,
|
| 475 |
+
"eval_loss": 2.1389129161834717,
|
| 476 |
+
"eval_rouge1": 0.2287,
|
| 477 |
+
"eval_rouge2": 0.0713,
|
| 478 |
+
"eval_rougeL": 0.1844,
|
| 479 |
+
"eval_rougeLsum": 0.1911,
|
| 480 |
+
"eval_runtime": 1069.2344,
|
| 481 |
+
"eval_samples_per_second": 2.151,
|
| 482 |
+
"eval_steps_per_second": 0.269,
|
| 483 |
+
"step": 95000
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 0.95,
|
| 487 |
+
"learning_rate": 5e-05,
|
| 488 |
+
"loss": 2.2202,
|
| 489 |
+
"step": 97500
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.98,
|
| 493 |
+
"learning_rate": 5e-05,
|
| 494 |
+
"loss": 2.2307,
|
| 495 |
+
"step": 100000
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 0.98,
|
| 499 |
+
"eval_gen_len": 30.7513,
|
| 500 |
+
"eval_loss": 2.132361888885498,
|
| 501 |
+
"eval_rouge1": 0.2293,
|
| 502 |
+
"eval_rouge2": 0.0741,
|
| 503 |
+
"eval_rougeL": 0.1845,
|
| 504 |
+
"eval_rougeLsum": 0.1924,
|
| 505 |
+
"eval_runtime": 1089.9927,
|
| 506 |
+
"eval_samples_per_second": 2.11,
|
| 507 |
+
"eval_steps_per_second": 0.264,
|
| 508 |
+
"step": 100000
|
| 509 |
+
}
|
| 510 |
+
],
|
| 511 |
+
"max_steps": 102356,
|
| 512 |
+
"num_train_epochs": 1,
|
| 513 |
+
"total_flos": 1.8696291573252096e+17,
|
| 514 |
+
"trial_name": null,
|
| 515 |
+
"trial_params": null
|
| 516 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:38a62961f75bc74d8cd1c3b0178206ed678b491c8842aa32921846d2fbbba66c
|
| 3 |
+
size 3375
|