root
commited on
Commit
•
8b7f9b4
1
Parent(s):
6e732ae
12300 checkpoint
Browse files- config.json +29 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +1 -0
- spiece.model +3 -0
- tokenizer_config.json +1 -0
- trainer_state.json +631 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "kravchenko/uk-mt5-small",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"MT5ForConditionalGeneration"
|
| 5 |
+
],
|
| 6 |
+
"d_ff": 1024,
|
| 7 |
+
"d_kv": 64,
|
| 8 |
+
"d_model": 512,
|
| 9 |
+
"decoder_start_token_id": 0,
|
| 10 |
+
"dropout_rate": 0.1,
|
| 11 |
+
"eos_token_id": 1,
|
| 12 |
+
"feed_forward_proj": "gated-gelu",
|
| 13 |
+
"initializer_factor": 1.0,
|
| 14 |
+
"is_encoder_decoder": true,
|
| 15 |
+
"layer_norm_epsilon": 1e-06,
|
| 16 |
+
"model_type": "mt5",
|
| 17 |
+
"num_decoder_layers": 8,
|
| 18 |
+
"num_heads": 6,
|
| 19 |
+
"num_layers": 8,
|
| 20 |
+
"pad_token_id": 0,
|
| 21 |
+
"relative_attention_max_distance": 128,
|
| 22 |
+
"relative_attention_num_buckets": 32,
|
| 23 |
+
"tie_word_embeddings": false,
|
| 24 |
+
"tokenizer_class": "T5Tokenizer",
|
| 25 |
+
"torch_dtype": "float32",
|
| 26 |
+
"transformers_version": "4.19.2",
|
| 27 |
+
"use_cache": false,
|
| 28 |
+
"vocab_size": 30000
|
| 29 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3a5a4f28f9cce25fdf671b076b92b2827de6c6bdf6abe1c3a7ca6d1d5963a0c1
|
| 3 |
+
size 1117153
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a759b89fcc670eea9834044dba1adb6c388ab495f65a7f628af261e0d00aeb01
|
| 3 |
+
size 299192133
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b69a0c57d5a819ae48a351bc76bad781a68c37d773990ddce25eaebefb53f013
|
| 3 |
+
size 14503
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:593163c9a9abdfc9a621be41945657533e10a1d48dbb9d2062f0fa8a90856f47
|
| 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:fa8ea53d190098f2632bd1c44b62731a90a9257a96081183c1a361a24f7ea8ff
|
| 3 |
+
size 804017
|
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/jupyter/.cache/huggingface/transformers/7d115a6774b803276ce282b932e5802b0b6329001c4bc36e2b9bc26ec9276c0c.294ebaa4cd17bb284635004c92d2c4d522ec488c828dcce0c2471b6f28e3fe82", "name_or_path": "kravchenko/uk-mt5-small", "tokenizer_class": "T5Tokenizer"}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,631 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.1389758437871933,
|
| 3 |
+
"best_model_checkpoint": "uk-mt5-base-gec-synthetic/checkpoint-12300",
|
| 4 |
+
"epoch": 12.758942457231726,
|
| 5 |
+
"global_step": 12300,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.31,
|
| 12 |
+
"learning_rate": 0.000979253112033195,
|
| 13 |
+
"loss": 1.9255,
|
| 14 |
+
"step": 300
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.31,
|
| 18 |
+
"eval_google_bleu": 0.2703240096475649,
|
| 19 |
+
"eval_loss": 0.9424494504928589,
|
| 20 |
+
"eval_runtime": 347.3208,
|
| 21 |
+
"eval_samples_per_second": 44.429,
|
| 22 |
+
"eval_steps_per_second": 1.391,
|
| 23 |
+
"step": 300
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.62,
|
| 27 |
+
"learning_rate": 0.0009585062240663901,
|
| 28 |
+
"loss": 1.0814,
|
| 29 |
+
"step": 600
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.62,
|
| 33 |
+
"eval_google_bleu": 0.2826040990979691,
|
| 34 |
+
"eval_loss": 0.7668491005897522,
|
| 35 |
+
"eval_runtime": 345.1153,
|
| 36 |
+
"eval_samples_per_second": 44.713,
|
| 37 |
+
"eval_steps_per_second": 1.4,
|
| 38 |
+
"step": 600
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.93,
|
| 42 |
+
"learning_rate": 0.000937759336099585,
|
| 43 |
+
"loss": 0.9355,
|
| 44 |
+
"step": 900
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.93,
|
| 48 |
+
"eval_google_bleu": 0.2865646045540358,
|
| 49 |
+
"eval_loss": 0.7197995185852051,
|
| 50 |
+
"eval_runtime": 342.8829,
|
| 51 |
+
"eval_samples_per_second": 45.004,
|
| 52 |
+
"eval_steps_per_second": 1.409,
|
| 53 |
+
"step": 900
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.24,
|
| 57 |
+
"learning_rate": 0.00091701244813278,
|
| 58 |
+
"loss": 0.8279,
|
| 59 |
+
"step": 1200
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.24,
|
| 63 |
+
"eval_google_bleu": 0.292388510379304,
|
| 64 |
+
"eval_loss": 0.6961821913719177,
|
| 65 |
+
"eval_runtime": 342.7478,
|
| 66 |
+
"eval_samples_per_second": 45.021,
|
| 67 |
+
"eval_steps_per_second": 1.409,
|
| 68 |
+
"step": 1200
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 1.56,
|
| 72 |
+
"learning_rate": 0.0008962655601659752,
|
| 73 |
+
"loss": 0.7798,
|
| 74 |
+
"step": 1500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.56,
|
| 78 |
+
"eval_google_bleu": 0.2966368793800289,
|
| 79 |
+
"eval_loss": 0.6216537952423096,
|
| 80 |
+
"eval_runtime": 341.7328,
|
| 81 |
+
"eval_samples_per_second": 45.155,
|
| 82 |
+
"eval_steps_per_second": 1.413,
|
| 83 |
+
"step": 1500
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 1.87,
|
| 87 |
+
"learning_rate": 0.0008755186721991702,
|
| 88 |
+
"loss": 0.7454,
|
| 89 |
+
"step": 1800
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 1.87,
|
| 93 |
+
"eval_google_bleu": 0.2970095360433254,
|
| 94 |
+
"eval_loss": 0.5832293629646301,
|
| 95 |
+
"eval_runtime": 342.7772,
|
| 96 |
+
"eval_samples_per_second": 45.018,
|
| 97 |
+
"eval_steps_per_second": 1.409,
|
| 98 |
+
"step": 1800
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 2.18,
|
| 102 |
+
"learning_rate": 0.0008547717842323651,
|
| 103 |
+
"loss": 0.6828,
|
| 104 |
+
"step": 2100
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.18,
|
| 108 |
+
"eval_google_bleu": 0.30616414078516274,
|
| 109 |
+
"eval_loss": 0.5547264814376831,
|
| 110 |
+
"eval_runtime": 342.5678,
|
| 111 |
+
"eval_samples_per_second": 45.045,
|
| 112 |
+
"eval_steps_per_second": 1.41,
|
| 113 |
+
"step": 2100
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 2.49,
|
| 117 |
+
"learning_rate": 0.0008340248962655602,
|
| 118 |
+
"loss": 0.6261,
|
| 119 |
+
"step": 2400
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 2.49,
|
| 123 |
+
"eval_google_bleu": 0.30804651852418713,
|
| 124 |
+
"eval_loss": 0.5183674693107605,
|
| 125 |
+
"eval_runtime": 342.3924,
|
| 126 |
+
"eval_samples_per_second": 45.068,
|
| 127 |
+
"eval_steps_per_second": 1.411,
|
| 128 |
+
"step": 2400
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 2.8,
|
| 132 |
+
"learning_rate": 0.0008132780082987552,
|
| 133 |
+
"loss": 0.6187,
|
| 134 |
+
"step": 2700
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 2.8,
|
| 138 |
+
"eval_google_bleu": 0.3114351154903824,
|
| 139 |
+
"eval_loss": 0.4962822198867798,
|
| 140 |
+
"eval_runtime": 344.2093,
|
| 141 |
+
"eval_samples_per_second": 44.83,
|
| 142 |
+
"eval_steps_per_second": 1.403,
|
| 143 |
+
"step": 2700
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 3.11,
|
| 147 |
+
"learning_rate": 0.0007925311203319502,
|
| 148 |
+
"loss": 0.5809,
|
| 149 |
+
"step": 3000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 3.11,
|
| 153 |
+
"eval_google_bleu": 0.3130147936777771,
|
| 154 |
+
"eval_loss": 0.5069273114204407,
|
| 155 |
+
"eval_runtime": 343.9406,
|
| 156 |
+
"eval_samples_per_second": 44.865,
|
| 157 |
+
"eval_steps_per_second": 1.404,
|
| 158 |
+
"step": 3000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 3.42,
|
| 162 |
+
"learning_rate": 0.0007717842323651453,
|
| 163 |
+
"loss": 0.5366,
|
| 164 |
+
"step": 3300
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 3.42,
|
| 168 |
+
"eval_google_bleu": 0.31680302377876657,
|
| 169 |
+
"eval_loss": 0.46470168232917786,
|
| 170 |
+
"eval_runtime": 343.1588,
|
| 171 |
+
"eval_samples_per_second": 44.968,
|
| 172 |
+
"eval_steps_per_second": 1.408,
|
| 173 |
+
"step": 3300
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 3.73,
|
| 177 |
+
"learning_rate": 0.0007510373443983402,
|
| 178 |
+
"loss": 0.526,
|
| 179 |
+
"step": 3600
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 3.73,
|
| 183 |
+
"eval_google_bleu": 0.3175123454942687,
|
| 184 |
+
"eval_loss": 0.4724844694137573,
|
| 185 |
+
"eval_runtime": 344.2418,
|
| 186 |
+
"eval_samples_per_second": 44.826,
|
| 187 |
+
"eval_steps_per_second": 1.403,
|
| 188 |
+
"step": 3600
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 4.05,
|
| 192 |
+
"learning_rate": 0.0007302904564315352,
|
| 193 |
+
"loss": 0.5137,
|
| 194 |
+
"step": 3900
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 4.05,
|
| 198 |
+
"eval_google_bleu": 0.32288427448475915,
|
| 199 |
+
"eval_loss": 0.4414692223072052,
|
| 200 |
+
"eval_runtime": 344.0594,
|
| 201 |
+
"eval_samples_per_second": 44.85,
|
| 202 |
+
"eval_steps_per_second": 1.404,
|
| 203 |
+
"step": 3900
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 4.36,
|
| 207 |
+
"learning_rate": 0.0007095435684647303,
|
| 208 |
+
"loss": 0.4603,
|
| 209 |
+
"step": 4200
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 4.36,
|
| 213 |
+
"eval_google_bleu": 0.3268479743473163,
|
| 214 |
+
"eval_loss": 0.43008026480674744,
|
| 215 |
+
"eval_runtime": 342.6092,
|
| 216 |
+
"eval_samples_per_second": 45.04,
|
| 217 |
+
"eval_steps_per_second": 1.41,
|
| 218 |
+
"step": 4200
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 4.67,
|
| 222 |
+
"learning_rate": 0.0006887966804979254,
|
| 223 |
+
"loss": 0.4531,
|
| 224 |
+
"step": 4500
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 4.67,
|
| 228 |
+
"eval_google_bleu": 0.3273518151404788,
|
| 229 |
+
"eval_loss": 0.40794768929481506,
|
| 230 |
+
"eval_runtime": 342.7787,
|
| 231 |
+
"eval_samples_per_second": 45.017,
|
| 232 |
+
"eval_steps_per_second": 1.409,
|
| 233 |
+
"step": 4500
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 4.98,
|
| 237 |
+
"learning_rate": 0.0006680497925311203,
|
| 238 |
+
"loss": 0.4555,
|
| 239 |
+
"step": 4800
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 4.98,
|
| 243 |
+
"eval_google_bleu": 0.32905860892065497,
|
| 244 |
+
"eval_loss": 0.39621850848197937,
|
| 245 |
+
"eval_runtime": 343.6219,
|
| 246 |
+
"eval_samples_per_second": 44.907,
|
| 247 |
+
"eval_steps_per_second": 1.406,
|
| 248 |
+
"step": 4800
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 5.29,
|
| 252 |
+
"learning_rate": 0.0006473029045643154,
|
| 253 |
+
"loss": 0.392,
|
| 254 |
+
"step": 5100
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 5.29,
|
| 258 |
+
"eval_google_bleu": 0.3336905264492333,
|
| 259 |
+
"eval_loss": 0.3776128590106964,
|
| 260 |
+
"eval_runtime": 343.634,
|
| 261 |
+
"eval_samples_per_second": 44.905,
|
| 262 |
+
"eval_steps_per_second": 1.406,
|
| 263 |
+
"step": 5100
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 5.6,
|
| 267 |
+
"learning_rate": 0.0006265560165975104,
|
| 268 |
+
"loss": 0.402,
|
| 269 |
+
"step": 5400
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 5.6,
|
| 273 |
+
"eval_google_bleu": 0.3371762652161784,
|
| 274 |
+
"eval_loss": 0.3662883937358856,
|
| 275 |
+
"eval_runtime": 343.334,
|
| 276 |
+
"eval_samples_per_second": 44.945,
|
| 277 |
+
"eval_steps_per_second": 1.407,
|
| 278 |
+
"step": 5400
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 5.91,
|
| 282 |
+
"learning_rate": 0.0006058091286307054,
|
| 283 |
+
"loss": 0.3979,
|
| 284 |
+
"step": 5700
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 5.91,
|
| 288 |
+
"eval_google_bleu": 0.34215661656562757,
|
| 289 |
+
"eval_loss": 0.34634825587272644,
|
| 290 |
+
"eval_runtime": 343.5776,
|
| 291 |
+
"eval_samples_per_second": 44.913,
|
| 292 |
+
"eval_steps_per_second": 1.406,
|
| 293 |
+
"step": 5700
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 6.22,
|
| 297 |
+
"learning_rate": 0.0005850622406639005,
|
| 298 |
+
"loss": 0.3545,
|
| 299 |
+
"step": 6000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 6.22,
|
| 303 |
+
"eval_google_bleu": 0.339056903098911,
|
| 304 |
+
"eval_loss": 0.34026840329170227,
|
| 305 |
+
"eval_runtime": 344.1962,
|
| 306 |
+
"eval_samples_per_second": 44.832,
|
| 307 |
+
"eval_steps_per_second": 1.403,
|
| 308 |
+
"step": 6000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 6.53,
|
| 312 |
+
"learning_rate": 0.0005643153526970954,
|
| 313 |
+
"loss": 0.342,
|
| 314 |
+
"step": 6300
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 6.53,
|
| 318 |
+
"eval_google_bleu": 0.34660828418236983,
|
| 319 |
+
"eval_loss": 0.3360808789730072,
|
| 320 |
+
"eval_runtime": 344.3987,
|
| 321 |
+
"eval_samples_per_second": 44.806,
|
| 322 |
+
"eval_steps_per_second": 1.402,
|
| 323 |
+
"step": 6300
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 6.85,
|
| 327 |
+
"learning_rate": 0.0005435684647302904,
|
| 328 |
+
"loss": 0.3462,
|
| 329 |
+
"step": 6600
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 6.85,
|
| 333 |
+
"eval_google_bleu": 0.3519982269928361,
|
| 334 |
+
"eval_loss": 0.3193075954914093,
|
| 335 |
+
"eval_runtime": 343.4724,
|
| 336 |
+
"eval_samples_per_second": 44.926,
|
| 337 |
+
"eval_steps_per_second": 1.406,
|
| 338 |
+
"step": 6600
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 7.16,
|
| 342 |
+
"learning_rate": 0.0005228215767634855,
|
| 343 |
+
"loss": 0.3097,
|
| 344 |
+
"step": 6900
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 7.16,
|
| 348 |
+
"eval_google_bleu": 0.35084156619797907,
|
| 349 |
+
"eval_loss": 0.3339296281337738,
|
| 350 |
+
"eval_runtime": 345.4283,
|
| 351 |
+
"eval_samples_per_second": 44.672,
|
| 352 |
+
"eval_steps_per_second": 1.398,
|
| 353 |
+
"step": 6900
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 7.47,
|
| 357 |
+
"learning_rate": 0.0005020746887966805,
|
| 358 |
+
"loss": 0.2937,
|
| 359 |
+
"step": 7200
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 7.47,
|
| 363 |
+
"eval_google_bleu": 0.35751958227664965,
|
| 364 |
+
"eval_loss": 0.29495447874069214,
|
| 365 |
+
"eval_runtime": 344.049,
|
| 366 |
+
"eval_samples_per_second": 44.851,
|
| 367 |
+
"eval_steps_per_second": 1.404,
|
| 368 |
+
"step": 7200
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 7.78,
|
| 372 |
+
"learning_rate": 0.00048132780082987554,
|
| 373 |
+
"loss": 0.2962,
|
| 374 |
+
"step": 7500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 7.78,
|
| 378 |
+
"eval_google_bleu": 0.3584187047483623,
|
| 379 |
+
"eval_loss": 0.27855053544044495,
|
| 380 |
+
"eval_runtime": 344.1164,
|
| 381 |
+
"eval_samples_per_second": 44.842,
|
| 382 |
+
"eval_steps_per_second": 1.404,
|
| 383 |
+
"step": 7500
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 8.09,
|
| 387 |
+
"learning_rate": 0.00046058091286307055,
|
| 388 |
+
"loss": 0.2795,
|
| 389 |
+
"step": 7800
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 8.09,
|
| 393 |
+
"eval_google_bleu": 0.3626892660498925,
|
| 394 |
+
"eval_loss": 0.273299902677536,
|
| 395 |
+
"eval_runtime": 344.7788,
|
| 396 |
+
"eval_samples_per_second": 44.756,
|
| 397 |
+
"eval_steps_per_second": 1.401,
|
| 398 |
+
"step": 7800
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 8.4,
|
| 402 |
+
"learning_rate": 0.00043983402489626555,
|
| 403 |
+
"loss": 0.2463,
|
| 404 |
+
"step": 8100
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 8.4,
|
| 408 |
+
"eval_google_bleu": 0.36640212432731034,
|
| 409 |
+
"eval_loss": 0.27249810099601746,
|
| 410 |
+
"eval_runtime": 344.7536,
|
| 411 |
+
"eval_samples_per_second": 44.76,
|
| 412 |
+
"eval_steps_per_second": 1.401,
|
| 413 |
+
"step": 8100
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 8.71,
|
| 417 |
+
"learning_rate": 0.0004190871369294606,
|
| 418 |
+
"loss": 0.2473,
|
| 419 |
+
"step": 8400
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 8.71,
|
| 423 |
+
"eval_google_bleu": 0.3680926990956975,
|
| 424 |
+
"eval_loss": 0.25599655508995056,
|
| 425 |
+
"eval_runtime": 343.8059,
|
| 426 |
+
"eval_samples_per_second": 44.883,
|
| 427 |
+
"eval_steps_per_second": 1.405,
|
| 428 |
+
"step": 8400
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 9.02,
|
| 432 |
+
"learning_rate": 0.00039834024896265557,
|
| 433 |
+
"loss": 0.2462,
|
| 434 |
+
"step": 8700
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 9.02,
|
| 438 |
+
"eval_google_bleu": 0.3726668030252467,
|
| 439 |
+
"eval_loss": 0.24074989557266235,
|
| 440 |
+
"eval_runtime": 343.4854,
|
| 441 |
+
"eval_samples_per_second": 44.925,
|
| 442 |
+
"eval_steps_per_second": 1.406,
|
| 443 |
+
"step": 8700
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 9.34,
|
| 447 |
+
"learning_rate": 0.00037759336099585063,
|
| 448 |
+
"loss": 0.2003,
|
| 449 |
+
"step": 9000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 9.34,
|
| 453 |
+
"eval_google_bleu": 0.37575691510452164,
|
| 454 |
+
"eval_loss": 0.23459354043006897,
|
| 455 |
+
"eval_runtime": 343.9506,
|
| 456 |
+
"eval_samples_per_second": 44.864,
|
| 457 |
+
"eval_steps_per_second": 1.404,
|
| 458 |
+
"step": 9000
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 9.65,
|
| 462 |
+
"learning_rate": 0.00035684647302904564,
|
| 463 |
+
"loss": 0.2096,
|
| 464 |
+
"step": 9300
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 9.65,
|
| 468 |
+
"eval_google_bleu": 0.3810405269728534,
|
| 469 |
+
"eval_loss": 0.23272807896137238,
|
| 470 |
+
"eval_runtime": 345.3533,
|
| 471 |
+
"eval_samples_per_second": 44.682,
|
| 472 |
+
"eval_steps_per_second": 1.399,
|
| 473 |
+
"step": 9300
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 9.96,
|
| 477 |
+
"learning_rate": 0.0003360995850622407,
|
| 478 |
+
"loss": 0.2109,
|
| 479 |
+
"step": 9600
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 9.96,
|
| 483 |
+
"eval_google_bleu": 0.3860127254969551,
|
| 484 |
+
"eval_loss": 0.2076478898525238,
|
| 485 |
+
"eval_runtime": 345.3314,
|
| 486 |
+
"eval_samples_per_second": 44.685,
|
| 487 |
+
"eval_steps_per_second": 1.399,
|
| 488 |
+
"step": 9600
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 10.27,
|
| 492 |
+
"learning_rate": 0.00031535269709543565,
|
| 493 |
+
"loss": 0.1673,
|
| 494 |
+
"step": 9900
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 10.27,
|
| 498 |
+
"eval_google_bleu": 0.39099070617947457,
|
| 499 |
+
"eval_loss": 0.2026001662015915,
|
| 500 |
+
"eval_runtime": 344.3009,
|
| 501 |
+
"eval_samples_per_second": 44.818,
|
| 502 |
+
"eval_steps_per_second": 1.403,
|
| 503 |
+
"step": 9900
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 10.58,
|
| 507 |
+
"learning_rate": 0.0002946058091286307,
|
| 508 |
+
"loss": 0.1722,
|
| 509 |
+
"step": 10200
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 10.58,
|
| 513 |
+
"eval_google_bleu": 0.39423925473127097,
|
| 514 |
+
"eval_loss": 0.19409596920013428,
|
| 515 |
+
"eval_runtime": 345.022,
|
| 516 |
+
"eval_samples_per_second": 44.725,
|
| 517 |
+
"eval_steps_per_second": 1.4,
|
| 518 |
+
"step": 10200
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 10.89,
|
| 522 |
+
"learning_rate": 0.0002738589211618258,
|
| 523 |
+
"loss": 0.1692,
|
| 524 |
+
"step": 10500
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 10.89,
|
| 528 |
+
"eval_google_bleu": 0.39808409620231944,
|
| 529 |
+
"eval_loss": 0.18527193367481232,
|
| 530 |
+
"eval_runtime": 344.9054,
|
| 531 |
+
"eval_samples_per_second": 44.74,
|
| 532 |
+
"eval_steps_per_second": 1.4,
|
| 533 |
+
"step": 10500
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 11.2,
|
| 537 |
+
"learning_rate": 0.00025311203319502073,
|
| 538 |
+
"loss": 0.1414,
|
| 539 |
+
"step": 10800
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 11.2,
|
| 543 |
+
"eval_google_bleu": 0.3988704921153549,
|
| 544 |
+
"eval_loss": 0.18296599388122559,
|
| 545 |
+
"eval_runtime": 344.3914,
|
| 546 |
+
"eval_samples_per_second": 44.807,
|
| 547 |
+
"eval_steps_per_second": 1.402,
|
| 548 |
+
"step": 10800
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 11.51,
|
| 552 |
+
"learning_rate": 0.00023236514522821577,
|
| 553 |
+
"loss": 0.1357,
|
| 554 |
+
"step": 11100
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 11.51,
|
| 558 |
+
"eval_google_bleu": 0.40408064479411665,
|
| 559 |
+
"eval_loss": 0.17169393599033356,
|
| 560 |
+
"eval_runtime": 345.2605,
|
| 561 |
+
"eval_samples_per_second": 44.694,
|
| 562 |
+
"eval_steps_per_second": 1.399,
|
| 563 |
+
"step": 11100
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 11.83,
|
| 567 |
+
"learning_rate": 0.0002116182572614108,
|
| 568 |
+
"loss": 0.1335,
|
| 569 |
+
"step": 11400
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 11.83,
|
| 573 |
+
"eval_google_bleu": 0.4105103547726813,
|
| 574 |
+
"eval_loss": 0.15817226469516754,
|
| 575 |
+
"eval_runtime": 345.0386,
|
| 576 |
+
"eval_samples_per_second": 44.723,
|
| 577 |
+
"eval_steps_per_second": 1.4,
|
| 578 |
+
"step": 11400
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 12.14,
|
| 582 |
+
"learning_rate": 0.0001908713692946058,
|
| 583 |
+
"loss": 0.119,
|
| 584 |
+
"step": 11700
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 12.14,
|
| 588 |
+
"eval_google_bleu": 0.4141623418407881,
|
| 589 |
+
"eval_loss": 0.1521884799003601,
|
| 590 |
+
"eval_runtime": 344.7777,
|
| 591 |
+
"eval_samples_per_second": 44.756,
|
| 592 |
+
"eval_steps_per_second": 1.401,
|
| 593 |
+
"step": 11700
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 12.45,
|
| 597 |
+
"learning_rate": 0.00017012448132780084,
|
| 598 |
+
"loss": 0.102,
|
| 599 |
+
"step": 12000
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 12.45,
|
| 603 |
+
"eval_google_bleu": 0.4170916396229331,
|
| 604 |
+
"eval_loss": 0.150096133351326,
|
| 605 |
+
"eval_runtime": 344.5413,
|
| 606 |
+
"eval_samples_per_second": 44.787,
|
| 607 |
+
"eval_steps_per_second": 1.402,
|
| 608 |
+
"step": 12000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 12.76,
|
| 612 |
+
"learning_rate": 0.00014937759336099585,
|
| 613 |
+
"loss": 0.1014,
|
| 614 |
+
"step": 12300
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 12.76,
|
| 618 |
+
"eval_google_bleu": 0.4215356451396477,
|
| 619 |
+
"eval_loss": 0.1389758437871933,
|
| 620 |
+
"eval_runtime": 343.8578,
|
| 621 |
+
"eval_samples_per_second": 44.876,
|
| 622 |
+
"eval_steps_per_second": 1.405,
|
| 623 |
+
"step": 12300
|
| 624 |
+
}
|
| 625 |
+
],
|
| 626 |
+
"max_steps": 14460,
|
| 627 |
+
"num_train_epochs": 15,
|
| 628 |
+
"total_flos": 2.3750619195838464e+16,
|
| 629 |
+
"trial_name": null,
|
| 630 |
+
"trial_params": null
|
| 631 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:03c9c8d6bdffdcd4d3c4206e3f365d795d7752c39d5a48be64e7ecb9cbe28956
|
| 3 |
+
size 3247
|