Commit ·
88d7dc4
1
Parent(s): 8f468b4
Upload 12 files
Browse files- config.json +59 -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
- tokenizer.json +0 -0
- tokenizer_config.json +1 -0
- trainer_state.json +882 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
config.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/mbart-large-cc25",
|
| 3 |
+
"_num_labels": 3,
|
| 4 |
+
"activation_dropout": 0.0,
|
| 5 |
+
"activation_function": "gelu",
|
| 6 |
+
"add_bias_logits": false,
|
| 7 |
+
"add_final_layer_norm": true,
|
| 8 |
+
"architectures": [
|
| 9 |
+
"MBartForConditionalGeneration"
|
| 10 |
+
],
|
| 11 |
+
"attention_dropout": 0.0,
|
| 12 |
+
"bos_token_id": 0,
|
| 13 |
+
"classif_dropout": 0.0,
|
| 14 |
+
"classifier_dropout": 0.0,
|
| 15 |
+
"d_model": 1024,
|
| 16 |
+
"decoder_attention_heads": 16,
|
| 17 |
+
"decoder_ffn_dim": 4096,
|
| 18 |
+
"decoder_layerdrop": 0.0,
|
| 19 |
+
"decoder_layers": 12,
|
| 20 |
+
"dropout": 0.1,
|
| 21 |
+
"encoder_attention_heads": 16,
|
| 22 |
+
"encoder_ffn_dim": 4096,
|
| 23 |
+
"encoder_layerdrop": 0.0,
|
| 24 |
+
"encoder_layers": 12,
|
| 25 |
+
"eos_token_id": 2,
|
| 26 |
+
"forced_eos_token_id": 2,
|
| 27 |
+
"id2label": {
|
| 28 |
+
"0": "LABEL_0",
|
| 29 |
+
"1": "LABEL_1",
|
| 30 |
+
"2": "LABEL_2"
|
| 31 |
+
},
|
| 32 |
+
"init_std": 0.02,
|
| 33 |
+
"is_encoder_decoder": true,
|
| 34 |
+
"label2id": {
|
| 35 |
+
"LABEL_0": 0,
|
| 36 |
+
"LABEL_1": 1,
|
| 37 |
+
"LABEL_2": 2
|
| 38 |
+
},
|
| 39 |
+
"max_length": 1024,
|
| 40 |
+
"max_position_embeddings": 1024,
|
| 41 |
+
"model_type": "mbart",
|
| 42 |
+
"normalize_before": true,
|
| 43 |
+
"normalize_embedding": true,
|
| 44 |
+
"num_beams": 5,
|
| 45 |
+
"num_hidden_layers": 12,
|
| 46 |
+
"output_past": true,
|
| 47 |
+
"pad_token_id": 1,
|
| 48 |
+
"scale_embedding": true,
|
| 49 |
+
"static_position_embeddings": false,
|
| 50 |
+
"task_specific_params": {
|
| 51 |
+
"translation_en_to_ro": {
|
| 52 |
+
"decoder_start_token_id": 250020
|
| 53 |
+
}
|
| 54 |
+
},
|
| 55 |
+
"torch_dtype": "float32",
|
| 56 |
+
"transformers_version": "4.18.0",
|
| 57 |
+
"use_cache": true,
|
| 58 |
+
"vocab_size": 250027
|
| 59 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:21a0a49860923c5dcc9b603aa1cb1e44043d47cf1ad045180afb2636923d2544
|
| 3 |
+
size 4887126731
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:41a0d27ca41de28c04930419a865b4d3018b44893cf7eed0d76d6e94779b8fd6
|
| 3 |
+
size 2444583325
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c9f2b47a7f87a15fe34b6d94effe621d9818f9ebda55550574e5f57d0c64d234
|
| 3 |
+
size 14575
|
scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6b5a7c407e2fba8639bdeb4db72c6d0f728aa3d3c20e3f7e1cd632c6f7514a27
|
| 3 |
+
size 557
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1f7c5642d3bc61b789a99d8f96ff4ce41b2d6a5bd5d01ad2353e0c03f9eaa641
|
| 3 |
+
size 627
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": false, "lowercase": false, "special_tokens_map_file": null, "name_or_path": "./dump", "do_basic_tokenize": true, "never_split": null, "tokenizer_class": "BertTokenizer"}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,882 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.98235965988409,
|
| 5 |
+
"global_step": 70500,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.02,
|
| 12 |
+
"learning_rate": 4.965241056446268e-05,
|
| 13 |
+
"loss": 5.7812,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.04,
|
| 18 |
+
"learning_rate": 4.9299885781970474e-05,
|
| 19 |
+
"loss": 3.1232,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.06,
|
| 24 |
+
"learning_rate": 4.894736099947827e-05,
|
| 25 |
+
"loss": 2.7376,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.08,
|
| 30 |
+
"learning_rate": 4.859483621698606e-05,
|
| 31 |
+
"loss": 2.4571,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.11,
|
| 36 |
+
"learning_rate": 4.8242311434493845e-05,
|
| 37 |
+
"loss": 2.2177,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.13,
|
| 42 |
+
"learning_rate": 4.788978665200164e-05,
|
| 43 |
+
"loss": 1.9238,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.15,
|
| 48 |
+
"learning_rate": 4.753726186950943e-05,
|
| 49 |
+
"loss": 1.7636,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.17,
|
| 54 |
+
"learning_rate": 4.718473708701722e-05,
|
| 55 |
+
"loss": 1.5908,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.19,
|
| 60 |
+
"learning_rate": 4.683221230452501e-05,
|
| 61 |
+
"loss": 1.4306,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.21,
|
| 66 |
+
"learning_rate": 4.64796875220328e-05,
|
| 67 |
+
"loss": 1.2813,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.23,
|
| 72 |
+
"learning_rate": 4.6127162739540594e-05,
|
| 73 |
+
"loss": 1.2258,
|
| 74 |
+
"step": 5500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.25,
|
| 78 |
+
"learning_rate": 4.577463795704838e-05,
|
| 79 |
+
"loss": 1.1167,
|
| 80 |
+
"step": 6000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.27,
|
| 84 |
+
"learning_rate": 4.542211317455617e-05,
|
| 85 |
+
"loss": 1.0346,
|
| 86 |
+
"step": 6500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.3,
|
| 90 |
+
"learning_rate": 4.5069588392063965e-05,
|
| 91 |
+
"loss": 0.963,
|
| 92 |
+
"step": 7000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.32,
|
| 96 |
+
"learning_rate": 4.471706360957176e-05,
|
| 97 |
+
"loss": 0.9412,
|
| 98 |
+
"step": 7500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.34,
|
| 102 |
+
"learning_rate": 4.4364538827079544e-05,
|
| 103 |
+
"loss": 0.8301,
|
| 104 |
+
"step": 8000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.36,
|
| 108 |
+
"learning_rate": 4.4012014044587336e-05,
|
| 109 |
+
"loss": 0.8389,
|
| 110 |
+
"step": 8500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.38,
|
| 114 |
+
"learning_rate": 4.366019431166011e-05,
|
| 115 |
+
"loss": 0.8128,
|
| 116 |
+
"step": 9000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.4,
|
| 120 |
+
"learning_rate": 4.3307669529167903e-05,
|
| 121 |
+
"loss": 0.7062,
|
| 122 |
+
"step": 9500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.42,
|
| 126 |
+
"learning_rate": 4.295514474667569e-05,
|
| 127 |
+
"loss": 0.7305,
|
| 128 |
+
"step": 10000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.44,
|
| 132 |
+
"learning_rate": 4.260261996418348e-05,
|
| 133 |
+
"loss": 0.7048,
|
| 134 |
+
"step": 10500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.47,
|
| 138 |
+
"learning_rate": 4.2250095181691275e-05,
|
| 139 |
+
"loss": 0.6352,
|
| 140 |
+
"step": 11000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.49,
|
| 144 |
+
"learning_rate": 4.189757039919907e-05,
|
| 145 |
+
"loss": 0.6491,
|
| 146 |
+
"step": 11500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.51,
|
| 150 |
+
"learning_rate": 4.154504561670686e-05,
|
| 151 |
+
"loss": 0.6416,
|
| 152 |
+
"step": 12000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.53,
|
| 156 |
+
"learning_rate": 4.1192520834214646e-05,
|
| 157 |
+
"loss": 0.618,
|
| 158 |
+
"step": 12500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.55,
|
| 162 |
+
"learning_rate": 4.083999605172244e-05,
|
| 163 |
+
"loss": 0.5734,
|
| 164 |
+
"step": 13000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.57,
|
| 168 |
+
"learning_rate": 4.048747126923023e-05,
|
| 169 |
+
"loss": 0.5764,
|
| 170 |
+
"step": 13500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.59,
|
| 174 |
+
"learning_rate": 4.0135651536303005e-05,
|
| 175 |
+
"loss": 0.6122,
|
| 176 |
+
"step": 14000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.61,
|
| 180 |
+
"learning_rate": 3.97831267538108e-05,
|
| 181 |
+
"loss": 0.4962,
|
| 182 |
+
"step": 14500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.63,
|
| 186 |
+
"learning_rate": 3.9430601971318584e-05,
|
| 187 |
+
"loss": 0.5173,
|
| 188 |
+
"step": 15000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.66,
|
| 192 |
+
"learning_rate": 3.907878223839136e-05,
|
| 193 |
+
"loss": 0.5122,
|
| 194 |
+
"step": 15500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.68,
|
| 198 |
+
"learning_rate": 3.872625745589915e-05,
|
| 199 |
+
"loss": 0.4673,
|
| 200 |
+
"step": 16000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.7,
|
| 204 |
+
"learning_rate": 3.8373732673406944e-05,
|
| 205 |
+
"loss": 0.4879,
|
| 206 |
+
"step": 16500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.72,
|
| 210 |
+
"learning_rate": 3.8021207890914736e-05,
|
| 211 |
+
"loss": 0.4461,
|
| 212 |
+
"step": 17000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.74,
|
| 216 |
+
"learning_rate": 3.766868310842253e-05,
|
| 217 |
+
"loss": 0.4864,
|
| 218 |
+
"step": 17500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.76,
|
| 222 |
+
"learning_rate": 3.7316158325930315e-05,
|
| 223 |
+
"loss": 0.4934,
|
| 224 |
+
"step": 18000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.78,
|
| 228 |
+
"learning_rate": 3.69636335434381e-05,
|
| 229 |
+
"loss": 0.465,
|
| 230 |
+
"step": 18500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.8,
|
| 234 |
+
"learning_rate": 3.661110876094589e-05,
|
| 235 |
+
"loss": 0.4444,
|
| 236 |
+
"step": 19000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.82,
|
| 240 |
+
"learning_rate": 3.6258583978453686e-05,
|
| 241 |
+
"loss": 0.4787,
|
| 242 |
+
"step": 19500
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.85,
|
| 246 |
+
"learning_rate": 3.590605919596148e-05,
|
| 247 |
+
"loss": 0.4358,
|
| 248 |
+
"step": 20000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.87,
|
| 252 |
+
"learning_rate": 3.555353441346927e-05,
|
| 253 |
+
"loss": 0.3978,
|
| 254 |
+
"step": 20500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.89,
|
| 258 |
+
"learning_rate": 3.5201009630977064e-05,
|
| 259 |
+
"loss": 0.4472,
|
| 260 |
+
"step": 21000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.91,
|
| 264 |
+
"learning_rate": 3.484918989804983e-05,
|
| 265 |
+
"loss": 0.4205,
|
| 266 |
+
"step": 21500
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.93,
|
| 270 |
+
"learning_rate": 3.4496665115557624e-05,
|
| 271 |
+
"loss": 0.394,
|
| 272 |
+
"step": 22000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.95,
|
| 276 |
+
"learning_rate": 3.4144845382630405e-05,
|
| 277 |
+
"loss": 0.3727,
|
| 278 |
+
"step": 22500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.97,
|
| 282 |
+
"learning_rate": 3.379232060013819e-05,
|
| 283 |
+
"loss": 0.4089,
|
| 284 |
+
"step": 23000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.99,
|
| 288 |
+
"learning_rate": 3.343979581764598e-05,
|
| 289 |
+
"loss": 0.3507,
|
| 290 |
+
"step": 23500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.0,
|
| 294 |
+
"eval_bleu": 68.2888,
|
| 295 |
+
"eval_gen_len": 25.384,
|
| 296 |
+
"eval_loss": 0.06593693792819977,
|
| 297 |
+
"eval_runtime": 35.6722,
|
| 298 |
+
"eval_samples_per_second": 14.017,
|
| 299 |
+
"eval_steps_per_second": 3.504,
|
| 300 |
+
"step": 23639
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 1.02,
|
| 304 |
+
"learning_rate": 3.308727103515377e-05,
|
| 305 |
+
"loss": 0.2763,
|
| 306 |
+
"step": 24000
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 1.04,
|
| 310 |
+
"learning_rate": 3.273474625266156e-05,
|
| 311 |
+
"loss": 0.2929,
|
| 312 |
+
"step": 24500
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 1.06,
|
| 316 |
+
"learning_rate": 3.2383631569299325e-05,
|
| 317 |
+
"loss": 0.3075,
|
| 318 |
+
"step": 25000
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.08,
|
| 322 |
+
"learning_rate": 3.203110678680711e-05,
|
| 323 |
+
"loss": 0.262,
|
| 324 |
+
"step": 25500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.1,
|
| 328 |
+
"learning_rate": 3.1678582004314904e-05,
|
| 329 |
+
"loss": 0.2456,
|
| 330 |
+
"step": 26000
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 1.12,
|
| 334 |
+
"learning_rate": 3.1326057221822696e-05,
|
| 335 |
+
"loss": 0.2446,
|
| 336 |
+
"step": 26500
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 1.14,
|
| 340 |
+
"learning_rate": 3.097353243933049e-05,
|
| 341 |
+
"loss": 0.2347,
|
| 342 |
+
"step": 27000
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 1.16,
|
| 346 |
+
"learning_rate": 3.062100765683828e-05,
|
| 347 |
+
"loss": 0.2831,
|
| 348 |
+
"step": 27500
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 1.18,
|
| 352 |
+
"learning_rate": 3.0269187923911053e-05,
|
| 353 |
+
"loss": 0.2593,
|
| 354 |
+
"step": 28000
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 1.21,
|
| 358 |
+
"learning_rate": 2.9916663141418842e-05,
|
| 359 |
+
"loss": 0.2421,
|
| 360 |
+
"step": 28500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.23,
|
| 364 |
+
"learning_rate": 2.9564843408491616e-05,
|
| 365 |
+
"loss": 0.2507,
|
| 366 |
+
"step": 29000
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.25,
|
| 370 |
+
"learning_rate": 2.921231862599941e-05,
|
| 371 |
+
"loss": 0.2614,
|
| 372 |
+
"step": 29500
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 1.27,
|
| 376 |
+
"learning_rate": 2.8859793843507198e-05,
|
| 377 |
+
"loss": 0.2224,
|
| 378 |
+
"step": 30000
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 1.29,
|
| 382 |
+
"learning_rate": 2.850726906101499e-05,
|
| 383 |
+
"loss": 0.277,
|
| 384 |
+
"step": 30500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.31,
|
| 388 |
+
"learning_rate": 2.8154744278522784e-05,
|
| 389 |
+
"loss": 0.2291,
|
| 390 |
+
"step": 31000
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 1.33,
|
| 394 |
+
"learning_rate": 2.7802219496030573e-05,
|
| 395 |
+
"loss": 0.2347,
|
| 396 |
+
"step": 31500
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 1.35,
|
| 400 |
+
"learning_rate": 2.7450399763103347e-05,
|
| 401 |
+
"loss": 0.2425,
|
| 402 |
+
"step": 32000
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.37,
|
| 406 |
+
"learning_rate": 2.7098580030176125e-05,
|
| 407 |
+
"loss": 0.2481,
|
| 408 |
+
"step": 32500
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.4,
|
| 412 |
+
"learning_rate": 2.6746055247683914e-05,
|
| 413 |
+
"loss": 0.2437,
|
| 414 |
+
"step": 33000
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 1.42,
|
| 418 |
+
"learning_rate": 2.63935304651917e-05,
|
| 419 |
+
"loss": 0.2438,
|
| 420 |
+
"step": 33500
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 1.44,
|
| 424 |
+
"learning_rate": 2.6041005682699493e-05,
|
| 425 |
+
"loss": 0.2456,
|
| 426 |
+
"step": 34000
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 1.46,
|
| 430 |
+
"learning_rate": 2.5688480900207285e-05,
|
| 431 |
+
"loss": 0.2239,
|
| 432 |
+
"step": 34500
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 1.48,
|
| 436 |
+
"learning_rate": 2.5335956117715075e-05,
|
| 437 |
+
"loss": 0.2165,
|
| 438 |
+
"step": 35000
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 1.5,
|
| 442 |
+
"learning_rate": 2.4983431335222867e-05,
|
| 443 |
+
"loss": 0.1948,
|
| 444 |
+
"step": 35500
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.52,
|
| 448 |
+
"learning_rate": 2.463090655273066e-05,
|
| 449 |
+
"loss": 0.2161,
|
| 450 |
+
"step": 36000
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.54,
|
| 454 |
+
"learning_rate": 2.427838177023845e-05,
|
| 455 |
+
"loss": 0.2112,
|
| 456 |
+
"step": 36500
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 1.57,
|
| 460 |
+
"learning_rate": 2.392585698774624e-05,
|
| 461 |
+
"loss": 0.2025,
|
| 462 |
+
"step": 37000
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 1.59,
|
| 466 |
+
"learning_rate": 2.357333220525403e-05,
|
| 467 |
+
"loss": 0.2208,
|
| 468 |
+
"step": 37500
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 1.61,
|
| 472 |
+
"learning_rate": 2.322080742276182e-05,
|
| 473 |
+
"loss": 0.2175,
|
| 474 |
+
"step": 38000
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 1.63,
|
| 478 |
+
"learning_rate": 2.2868282640269613e-05,
|
| 479 |
+
"loss": 0.2222,
|
| 480 |
+
"step": 38500
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 1.65,
|
| 484 |
+
"learning_rate": 2.2515757857777402e-05,
|
| 485 |
+
"loss": 0.2122,
|
| 486 |
+
"step": 39000
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 1.67,
|
| 490 |
+
"learning_rate": 2.2163233075285195e-05,
|
| 491 |
+
"loss": 0.2034,
|
| 492 |
+
"step": 39500
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.69,
|
| 496 |
+
"learning_rate": 2.1810708292792984e-05,
|
| 497 |
+
"loss": 0.1905,
|
| 498 |
+
"step": 40000
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 1.71,
|
| 502 |
+
"learning_rate": 2.1458183510300777e-05,
|
| 503 |
+
"loss": 0.1869,
|
| 504 |
+
"step": 40500
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 1.73,
|
| 508 |
+
"learning_rate": 2.1105658727808566e-05,
|
| 509 |
+
"loss": 0.2258,
|
| 510 |
+
"step": 41000
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 1.76,
|
| 514 |
+
"learning_rate": 2.0753133945316355e-05,
|
| 515 |
+
"loss": 0.1938,
|
| 516 |
+
"step": 41500
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 1.78,
|
| 520 |
+
"learning_rate": 2.0400609162824148e-05,
|
| 521 |
+
"loss": 0.1924,
|
| 522 |
+
"step": 42000
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 1.8,
|
| 526 |
+
"learning_rate": 2.0048789429896922e-05,
|
| 527 |
+
"loss": 0.1651,
|
| 528 |
+
"step": 42500
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 1.82,
|
| 532 |
+
"learning_rate": 1.9696969696969697e-05,
|
| 533 |
+
"loss": 0.2262,
|
| 534 |
+
"step": 43000
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.84,
|
| 538 |
+
"learning_rate": 1.934444491447749e-05,
|
| 539 |
+
"loss": 0.1959,
|
| 540 |
+
"step": 43500
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 1.86,
|
| 544 |
+
"learning_rate": 1.899192013198528e-05,
|
| 545 |
+
"loss": 0.1973,
|
| 546 |
+
"step": 44000
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 1.88,
|
| 550 |
+
"learning_rate": 1.8639395349493068e-05,
|
| 551 |
+
"loss": 0.1683,
|
| 552 |
+
"step": 44500
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 1.9,
|
| 556 |
+
"learning_rate": 1.828687056700086e-05,
|
| 557 |
+
"loss": 0.2026,
|
| 558 |
+
"step": 45000
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 1.92,
|
| 562 |
+
"learning_rate": 1.7934345784508653e-05,
|
| 563 |
+
"loss": 0.1885,
|
| 564 |
+
"step": 45500
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 1.95,
|
| 568 |
+
"learning_rate": 1.7581821002016442e-05,
|
| 569 |
+
"loss": 0.1737,
|
| 570 |
+
"step": 46000
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 1.97,
|
| 574 |
+
"learning_rate": 1.722929621952423e-05,
|
| 575 |
+
"loss": 0.1895,
|
| 576 |
+
"step": 46500
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.99,
|
| 580 |
+
"learning_rate": 1.6878181536161995e-05,
|
| 581 |
+
"loss": 0.1645,
|
| 582 |
+
"step": 47000
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 2.0,
|
| 586 |
+
"eval_bleu": 0.0,
|
| 587 |
+
"eval_gen_len": 1.0,
|
| 588 |
+
"eval_loss": 0.00799082312732935,
|
| 589 |
+
"eval_runtime": 41.137,
|
| 590 |
+
"eval_samples_per_second": 12.155,
|
| 591 |
+
"eval_steps_per_second": 3.039,
|
| 592 |
+
"step": 47278
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 2.01,
|
| 596 |
+
"learning_rate": 1.6525656753669784e-05,
|
| 597 |
+
"loss": 0.1458,
|
| 598 |
+
"step": 47500
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 2.03,
|
| 602 |
+
"learning_rate": 1.6173131971177573e-05,
|
| 603 |
+
"loss": 0.088,
|
| 604 |
+
"step": 48000
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 2.05,
|
| 608 |
+
"learning_rate": 1.5820607188685366e-05,
|
| 609 |
+
"loss": 0.0819,
|
| 610 |
+
"step": 48500
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 2.07,
|
| 614 |
+
"learning_rate": 1.546808240619316e-05,
|
| 615 |
+
"loss": 0.0885,
|
| 616 |
+
"step": 49000
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 2.09,
|
| 620 |
+
"learning_rate": 1.5115557623700946e-05,
|
| 621 |
+
"loss": 0.099,
|
| 622 |
+
"step": 49500
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 2.12,
|
| 626 |
+
"learning_rate": 1.4763032841208737e-05,
|
| 627 |
+
"loss": 0.0994,
|
| 628 |
+
"step": 50000
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 2.14,
|
| 632 |
+
"learning_rate": 1.441050805871653e-05,
|
| 633 |
+
"loss": 0.0941,
|
| 634 |
+
"step": 50500
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 2.16,
|
| 638 |
+
"learning_rate": 1.4057983276224319e-05,
|
| 639 |
+
"loss": 0.0977,
|
| 640 |
+
"step": 51000
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 2.18,
|
| 644 |
+
"learning_rate": 1.370545849373211e-05,
|
| 645 |
+
"loss": 0.0892,
|
| 646 |
+
"step": 51500
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 2.2,
|
| 650 |
+
"learning_rate": 1.33529337112399e-05,
|
| 651 |
+
"loss": 0.0935,
|
| 652 |
+
"step": 52000
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"epoch": 2.22,
|
| 656 |
+
"learning_rate": 1.3000408928747693e-05,
|
| 657 |
+
"loss": 0.1018,
|
| 658 |
+
"step": 52500
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 2.24,
|
| 662 |
+
"learning_rate": 1.264788414625548e-05,
|
| 663 |
+
"loss": 0.078,
|
| 664 |
+
"step": 53000
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 2.26,
|
| 668 |
+
"learning_rate": 1.2295359363763273e-05,
|
| 669 |
+
"loss": 0.0866,
|
| 670 |
+
"step": 53500
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 2.28,
|
| 674 |
+
"learning_rate": 1.1942834581271064e-05,
|
| 675 |
+
"loss": 0.0956,
|
| 676 |
+
"step": 54000
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 2.31,
|
| 680 |
+
"learning_rate": 1.1590309798778855e-05,
|
| 681 |
+
"loss": 0.0853,
|
| 682 |
+
"step": 54500
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 2.33,
|
| 686 |
+
"learning_rate": 1.1237785016286644e-05,
|
| 687 |
+
"loss": 0.0882,
|
| 688 |
+
"step": 55000
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 2.35,
|
| 692 |
+
"learning_rate": 1.0885260233794437e-05,
|
| 693 |
+
"loss": 0.0994,
|
| 694 |
+
"step": 55500
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 2.37,
|
| 698 |
+
"learning_rate": 1.0534145550432197e-05,
|
| 699 |
+
"loss": 0.087,
|
| 700 |
+
"step": 56000
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 2.39,
|
| 704 |
+
"learning_rate": 1.0181620767939986e-05,
|
| 705 |
+
"loss": 0.0754,
|
| 706 |
+
"step": 56500
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 2.41,
|
| 710 |
+
"learning_rate": 9.829095985447777e-06,
|
| 711 |
+
"loss": 0.1035,
|
| 712 |
+
"step": 57000
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 2.43,
|
| 716 |
+
"learning_rate": 9.476571202955568e-06,
|
| 717 |
+
"loss": 0.0855,
|
| 718 |
+
"step": 57500
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 2.45,
|
| 722 |
+
"learning_rate": 9.124046420463359e-06,
|
| 723 |
+
"loss": 0.0868,
|
| 724 |
+
"step": 58000
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 2.47,
|
| 728 |
+
"learning_rate": 8.772226687536133e-06,
|
| 729 |
+
"loss": 0.0873,
|
| 730 |
+
"step": 58500
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 2.5,
|
| 734 |
+
"learning_rate": 8.419701905043926e-06,
|
| 735 |
+
"loss": 0.0849,
|
| 736 |
+
"step": 59000
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 2.52,
|
| 740 |
+
"learning_rate": 8.0678821721167e-06,
|
| 741 |
+
"loss": 0.0863,
|
| 742 |
+
"step": 59500
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 2.54,
|
| 746 |
+
"learning_rate": 7.71535738962449e-06,
|
| 747 |
+
"loss": 0.0833,
|
| 748 |
+
"step": 60000
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 2.56,
|
| 752 |
+
"learning_rate": 7.362832607132282e-06,
|
| 753 |
+
"loss": 0.0988,
|
| 754 |
+
"step": 60500
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 2.58,
|
| 758 |
+
"learning_rate": 7.010307824640072e-06,
|
| 759 |
+
"loss": 0.0712,
|
| 760 |
+
"step": 61000
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 2.6,
|
| 764 |
+
"learning_rate": 6.657783042147864e-06,
|
| 765 |
+
"loss": 0.0784,
|
| 766 |
+
"step": 61500
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 2.62,
|
| 770 |
+
"learning_rate": 6.305258259655654e-06,
|
| 771 |
+
"loss": 0.0871,
|
| 772 |
+
"step": 62000
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 2.64,
|
| 776 |
+
"learning_rate": 5.952733477163445e-06,
|
| 777 |
+
"loss": 0.0697,
|
| 778 |
+
"step": 62500
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 2.67,
|
| 782 |
+
"learning_rate": 5.600208694671235e-06,
|
| 783 |
+
"loss": 0.0681,
|
| 784 |
+
"step": 63000
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 2.69,
|
| 788 |
+
"learning_rate": 5.247683912179026e-06,
|
| 789 |
+
"loss": 0.0763,
|
| 790 |
+
"step": 63500
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 2.71,
|
| 794 |
+
"learning_rate": 4.895159129686817e-06,
|
| 795 |
+
"loss": 0.0844,
|
| 796 |
+
"step": 64000
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 2.73,
|
| 800 |
+
"learning_rate": 4.542634347194608e-06,
|
| 801 |
+
"loss": 0.0803,
|
| 802 |
+
"step": 64500
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 2.75,
|
| 806 |
+
"learning_rate": 4.190109564702399e-06,
|
| 807 |
+
"loss": 0.0743,
|
| 808 |
+
"step": 65000
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 2.77,
|
| 812 |
+
"learning_rate": 3.837584782210189e-06,
|
| 813 |
+
"loss": 0.079,
|
| 814 |
+
"step": 65500
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 2.79,
|
| 818 |
+
"learning_rate": 3.48505999971798e-06,
|
| 819 |
+
"loss": 0.0857,
|
| 820 |
+
"step": 66000
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 2.81,
|
| 824 |
+
"learning_rate": 3.1332402667907553e-06,
|
| 825 |
+
"loss": 0.0706,
|
| 826 |
+
"step": 66500
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 2.83,
|
| 830 |
+
"learning_rate": 2.7807154842985467e-06,
|
| 831 |
+
"loss": 0.0666,
|
| 832 |
+
"step": 67000
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 2.86,
|
| 836 |
+
"learning_rate": 2.428190701806337e-06,
|
| 837 |
+
"loss": 0.0659,
|
| 838 |
+
"step": 67500
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 2.88,
|
| 842 |
+
"learning_rate": 2.0756659193141277e-06,
|
| 843 |
+
"loss": 0.0793,
|
| 844 |
+
"step": 68000
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 2.9,
|
| 848 |
+
"learning_rate": 1.723846186386903e-06,
|
| 849 |
+
"loss": 0.0578,
|
| 850 |
+
"step": 68500
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 2.92,
|
| 854 |
+
"learning_rate": 1.3713214038946937e-06,
|
| 855 |
+
"loss": 0.0788,
|
| 856 |
+
"step": 69000
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 2.94,
|
| 860 |
+
"learning_rate": 1.0187966214024847e-06,
|
| 861 |
+
"loss": 0.0711,
|
| 862 |
+
"step": 69500
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 2.96,
|
| 866 |
+
"learning_rate": 6.662718389102755e-07,
|
| 867 |
+
"loss": 0.0622,
|
| 868 |
+
"step": 70000
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 2.98,
|
| 872 |
+
"learning_rate": 3.1445210598305064e-07,
|
| 873 |
+
"loss": 0.0716,
|
| 874 |
+
"step": 70500
|
| 875 |
+
}
|
| 876 |
+
],
|
| 877 |
+
"max_steps": 70917,
|
| 878 |
+
"num_train_epochs": 3,
|
| 879 |
+
"total_flos": 1.3984054543269888e+16,
|
| 880 |
+
"trial_name": null,
|
| 881 |
+
"trial_params": null
|
| 882 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:17bd0431fff19a2651e362543069188af49380e108a731e9f789a638407dfd7e
|
| 3 |
+
size 3195
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|