Upload folder using huggingface_hub
Browse files- .gitattributes +35 -35
- config.json +62 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +7 -0
- tokenizer.json +0 -0
- tokenizer_config.json +58 -0
- trainer_state.json +1666 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
.gitattributes
CHANGED
|
@@ -1,35 +1,35 @@
|
|
| 1 |
-
*.7z filter=lfs diff=lfs merge=lfs -text
|
| 2 |
-
*.arrow filter=lfs diff=lfs merge=lfs -text
|
| 3 |
-
*.bin filter=lfs diff=lfs merge=lfs -text
|
| 4 |
-
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
| 5 |
-
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
| 6 |
-
*.ftz filter=lfs diff=lfs merge=lfs -text
|
| 7 |
-
*.gz filter=lfs diff=lfs merge=lfs -text
|
| 8 |
-
*.h5 filter=lfs diff=lfs merge=lfs -text
|
| 9 |
-
*.joblib filter=lfs diff=lfs merge=lfs -text
|
| 10 |
-
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
| 11 |
-
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
| 12 |
-
*.model filter=lfs diff=lfs merge=lfs -text
|
| 13 |
-
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
| 14 |
-
*.npy filter=lfs diff=lfs merge=lfs -text
|
| 15 |
-
*.npz filter=lfs diff=lfs merge=lfs -text
|
| 16 |
-
*.onnx filter=lfs diff=lfs merge=lfs -text
|
| 17 |
-
*.ot filter=lfs diff=lfs merge=lfs -text
|
| 18 |
-
*.parquet filter=lfs diff=lfs merge=lfs -text
|
| 19 |
-
*.pb filter=lfs diff=lfs merge=lfs -text
|
| 20 |
-
*.pickle filter=lfs diff=lfs merge=lfs -text
|
| 21 |
-
*.pkl filter=lfs diff=lfs merge=lfs -text
|
| 22 |
-
*.pt filter=lfs diff=lfs merge=lfs -text
|
| 23 |
-
*.pth filter=lfs diff=lfs merge=lfs -text
|
| 24 |
-
*.rar filter=lfs diff=lfs merge=lfs -text
|
| 25 |
-
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
| 26 |
-
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
| 27 |
-
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
| 28 |
-
*.tar filter=lfs diff=lfs merge=lfs -text
|
| 29 |
-
*.tflite filter=lfs diff=lfs merge=lfs -text
|
| 30 |
-
*.tgz filter=lfs diff=lfs merge=lfs -text
|
| 31 |
-
*.wasm filter=lfs diff=lfs merge=lfs -text
|
| 32 |
-
*.xz filter=lfs diff=lfs merge=lfs -text
|
| 33 |
-
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
-
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
-
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
| 1 |
+
*.7z filter=lfs diff=lfs merge=lfs -text
|
| 2 |
+
*.arrow filter=lfs diff=lfs merge=lfs -text
|
| 3 |
+
*.bin filter=lfs diff=lfs merge=lfs -text
|
| 4 |
+
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
| 5 |
+
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
| 6 |
+
*.ftz filter=lfs diff=lfs merge=lfs -text
|
| 7 |
+
*.gz filter=lfs diff=lfs merge=lfs -text
|
| 8 |
+
*.h5 filter=lfs diff=lfs merge=lfs -text
|
| 9 |
+
*.joblib filter=lfs diff=lfs merge=lfs -text
|
| 10 |
+
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
| 11 |
+
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
| 12 |
+
*.model filter=lfs diff=lfs merge=lfs -text
|
| 13 |
+
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
| 14 |
+
*.npy filter=lfs diff=lfs merge=lfs -text
|
| 15 |
+
*.npz filter=lfs diff=lfs merge=lfs -text
|
| 16 |
+
*.onnx filter=lfs diff=lfs merge=lfs -text
|
| 17 |
+
*.ot filter=lfs diff=lfs merge=lfs -text
|
| 18 |
+
*.parquet filter=lfs diff=lfs merge=lfs -text
|
| 19 |
+
*.pb filter=lfs diff=lfs merge=lfs -text
|
| 20 |
+
*.pickle filter=lfs diff=lfs merge=lfs -text
|
| 21 |
+
*.pkl filter=lfs diff=lfs merge=lfs -text
|
| 22 |
+
*.pt filter=lfs diff=lfs merge=lfs -text
|
| 23 |
+
*.pth filter=lfs diff=lfs merge=lfs -text
|
| 24 |
+
*.rar filter=lfs diff=lfs merge=lfs -text
|
| 25 |
+
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
| 26 |
+
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
| 27 |
+
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
| 28 |
+
*.tar filter=lfs diff=lfs merge=lfs -text
|
| 29 |
+
*.tflite filter=lfs diff=lfs merge=lfs -text
|
| 30 |
+
*.tgz filter=lfs diff=lfs merge=lfs -text
|
| 31 |
+
*.wasm filter=lfs diff=lfs merge=lfs -text
|
| 32 |
+
*.xz filter=lfs diff=lfs merge=lfs -text
|
| 33 |
+
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
+
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
+
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
config.json
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"ElectraForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"dtype": "float32",
|
| 8 |
+
"embedding_size": 768,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "show_menu",
|
| 14 |
+
"1": "show_order",
|
| 15 |
+
"2": "show_basket",
|
| 16 |
+
"3": "do_order",
|
| 17 |
+
"4": "show_delivery",
|
| 18 |
+
"5": "cancel_order",
|
| 19 |
+
"6": "call_customer_service",
|
| 20 |
+
"7": "select_valentine_dinner",
|
| 21 |
+
"8": "select_french_dinner",
|
| 22 |
+
"9": "select_english_dinner",
|
| 23 |
+
"10": "select_champagne_feast_dinner",
|
| 24 |
+
"11": "select_style_simple",
|
| 25 |
+
"12": "select_style_grand",
|
| 26 |
+
"13": "select_style_deluxe"
|
| 27 |
+
},
|
| 28 |
+
"initializer_range": 0.02,
|
| 29 |
+
"intermediate_size": 3072,
|
| 30 |
+
"label2id": {
|
| 31 |
+
"call_customer_service": 6,
|
| 32 |
+
"cancel_order": 5,
|
| 33 |
+
"do_order": 3,
|
| 34 |
+
"select_champagne_feast_dinner": 10,
|
| 35 |
+
"select_english_dinner": 9,
|
| 36 |
+
"select_french_dinner": 8,
|
| 37 |
+
"select_style_deluxe": 13,
|
| 38 |
+
"select_style_grand": 12,
|
| 39 |
+
"select_style_simple": 11,
|
| 40 |
+
"select_valentine_dinner": 7,
|
| 41 |
+
"show_basket": 2,
|
| 42 |
+
"show_delivery": 4,
|
| 43 |
+
"show_menu": 0,
|
| 44 |
+
"show_order": 1
|
| 45 |
+
},
|
| 46 |
+
"layer_norm_eps": 1e-12,
|
| 47 |
+
"max_position_embeddings": 512,
|
| 48 |
+
"model_type": "electra",
|
| 49 |
+
"num_attention_heads": 12,
|
| 50 |
+
"num_hidden_layers": 12,
|
| 51 |
+
"pad_token_id": 0,
|
| 52 |
+
"position_embedding_type": "absolute",
|
| 53 |
+
"problem_type": "single_label_classification",
|
| 54 |
+
"summary_activation": "gelu",
|
| 55 |
+
"summary_last_dropout": 0.1,
|
| 56 |
+
"summary_type": "first",
|
| 57 |
+
"summary_use_proj": true,
|
| 58 |
+
"transformers_version": "4.56.1",
|
| 59 |
+
"type_vocab_size": 2,
|
| 60 |
+
"use_cache": true,
|
| 61 |
+
"vocab_size": 35000
|
| 62 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6fe8c0dd421487dfb023090cf8d6ee5e453330768148f9a034d1073f68562f82
|
| 3 |
+
size 451752584
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:166522d7372eb50192883e19b3251e1c8db651daad07a473e71377c2bed26491
|
| 3 |
+
size 903628171
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:316f2addcac6fe90224e7a2066a4d71c5f5b285562ddb363d329780bd4b31147
|
| 3 |
+
size 14645
|
scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:266c9878bf8dd608a94ec4c24d84a33825bbd82ec09fc32d27661b5cf35b1684
|
| 3 |
+
size 1383
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d4fb2533b16834d216796a0e534ef4c89bdcf195bd1371c2c7d6dd32c76deb33
|
| 3 |
+
size 1465
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
+
"mask_token": "[MASK]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"sep_token": "[SEP]",
|
| 6 |
+
"unk_token": "[UNK]"
|
| 7 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"4": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": true,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_basic_tokenize": true,
|
| 47 |
+
"do_lower_case": false,
|
| 48 |
+
"extra_special_tokens": {},
|
| 49 |
+
"mask_token": "[MASK]",
|
| 50 |
+
"model_max_length": 512,
|
| 51 |
+
"never_split": null,
|
| 52 |
+
"pad_token": "[PAD]",
|
| 53 |
+
"sep_token": "[SEP]",
|
| 54 |
+
"strip_accents": null,
|
| 55 |
+
"tokenize_chinese_chars": true,
|
| 56 |
+
"tokenizer_class": "ElectraTokenizer",
|
| 57 |
+
"unk_token": "[UNK]"
|
| 58 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1666 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 29.41176470588235,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 2000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.14705882352941177,
|
| 14 |
+
"grad_norm": 2.2869467735290527,
|
| 15 |
+
"learning_rate": 1.9911764705882353e-05,
|
| 16 |
+
"loss": 2.6394,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.29411764705882354,
|
| 21 |
+
"grad_norm": 1.5338459014892578,
|
| 22 |
+
"learning_rate": 1.981372549019608e-05,
|
| 23 |
+
"loss": 2.6312,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.4411764705882353,
|
| 28 |
+
"grad_norm": 2.0163090229034424,
|
| 29 |
+
"learning_rate": 1.9715686274509805e-05,
|
| 30 |
+
"loss": 2.6409,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.5882352941176471,
|
| 35 |
+
"grad_norm": 3.02846097946167,
|
| 36 |
+
"learning_rate": 1.961764705882353e-05,
|
| 37 |
+
"loss": 2.6167,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.7352941176470589,
|
| 42 |
+
"grad_norm": 4.507878303527832,
|
| 43 |
+
"learning_rate": 1.9519607843137257e-05,
|
| 44 |
+
"loss": 2.6231,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.8823529411764706,
|
| 49 |
+
"grad_norm": 2.433847188949585,
|
| 50 |
+
"learning_rate": 1.9421568627450982e-05,
|
| 51 |
+
"loss": 2.6153,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 1.0,
|
| 56 |
+
"eval_loss": 2.579890012741089,
|
| 57 |
+
"eval_runtime": 1.6326,
|
| 58 |
+
"eval_samples_per_second": 82.691,
|
| 59 |
+
"eval_steps_per_second": 10.413,
|
| 60 |
+
"step": 68
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.0294117647058822,
|
| 64 |
+
"grad_norm": 3.4059886932373047,
|
| 65 |
+
"learning_rate": 1.9323529411764706e-05,
|
| 66 |
+
"loss": 2.5863,
|
| 67 |
+
"step": 70
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.1764705882352942,
|
| 71 |
+
"grad_norm": 2.305633068084717,
|
| 72 |
+
"learning_rate": 1.9225490196078434e-05,
|
| 73 |
+
"loss": 2.5677,
|
| 74 |
+
"step": 80
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.3235294117647058,
|
| 78 |
+
"grad_norm": 2.493675470352173,
|
| 79 |
+
"learning_rate": 1.912745098039216e-05,
|
| 80 |
+
"loss": 2.5061,
|
| 81 |
+
"step": 90
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 1.4705882352941178,
|
| 85 |
+
"grad_norm": 2.6469180583953857,
|
| 86 |
+
"learning_rate": 1.9029411764705883e-05,
|
| 87 |
+
"loss": 2.4657,
|
| 88 |
+
"step": 100
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 1.6176470588235294,
|
| 92 |
+
"grad_norm": 3.5626657009124756,
|
| 93 |
+
"learning_rate": 1.8931372549019607e-05,
|
| 94 |
+
"loss": 2.4594,
|
| 95 |
+
"step": 110
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.7647058823529411,
|
| 99 |
+
"grad_norm": 2.9186179637908936,
|
| 100 |
+
"learning_rate": 1.8833333333333335e-05,
|
| 101 |
+
"loss": 2.4393,
|
| 102 |
+
"step": 120
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.9117647058823528,
|
| 106 |
+
"grad_norm": 4.3771772384643555,
|
| 107 |
+
"learning_rate": 1.873529411764706e-05,
|
| 108 |
+
"loss": 2.3884,
|
| 109 |
+
"step": 130
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 2.0,
|
| 113 |
+
"eval_loss": 2.311964750289917,
|
| 114 |
+
"eval_runtime": 1.6428,
|
| 115 |
+
"eval_samples_per_second": 82.178,
|
| 116 |
+
"eval_steps_per_second": 10.348,
|
| 117 |
+
"step": 136
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 2.0588235294117645,
|
| 121 |
+
"grad_norm": 2.706939935684204,
|
| 122 |
+
"learning_rate": 1.8637254901960787e-05,
|
| 123 |
+
"loss": 2.3395,
|
| 124 |
+
"step": 140
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 2.2058823529411766,
|
| 128 |
+
"grad_norm": 3.1243748664855957,
|
| 129 |
+
"learning_rate": 1.853921568627451e-05,
|
| 130 |
+
"loss": 2.2711,
|
| 131 |
+
"step": 150
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 2.3529411764705883,
|
| 135 |
+
"grad_norm": 2.9195902347564697,
|
| 136 |
+
"learning_rate": 1.844117647058824e-05,
|
| 137 |
+
"loss": 2.2252,
|
| 138 |
+
"step": 160
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 2.5,
|
| 142 |
+
"grad_norm": 5.713516712188721,
|
| 143 |
+
"learning_rate": 1.834313725490196e-05,
|
| 144 |
+
"loss": 2.2211,
|
| 145 |
+
"step": 170
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 2.6470588235294117,
|
| 149 |
+
"grad_norm": 2.779804229736328,
|
| 150 |
+
"learning_rate": 1.8245098039215688e-05,
|
| 151 |
+
"loss": 2.1251,
|
| 152 |
+
"step": 180
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 2.7941176470588234,
|
| 156 |
+
"grad_norm": 2.9840755462646484,
|
| 157 |
+
"learning_rate": 1.8147058823529412e-05,
|
| 158 |
+
"loss": 2.1162,
|
| 159 |
+
"step": 190
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 2.9411764705882355,
|
| 163 |
+
"grad_norm": 3.6137640476226807,
|
| 164 |
+
"learning_rate": 1.804901960784314e-05,
|
| 165 |
+
"loss": 2.0859,
|
| 166 |
+
"step": 200
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 3.0,
|
| 170 |
+
"eval_loss": 1.982031226158142,
|
| 171 |
+
"eval_runtime": 1.6147,
|
| 172 |
+
"eval_samples_per_second": 83.606,
|
| 173 |
+
"eval_steps_per_second": 10.528,
|
| 174 |
+
"step": 204
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 3.088235294117647,
|
| 178 |
+
"grad_norm": 2.8977713584899902,
|
| 179 |
+
"learning_rate": 1.7950980392156864e-05,
|
| 180 |
+
"loss": 2.0281,
|
| 181 |
+
"step": 210
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 3.235294117647059,
|
| 185 |
+
"grad_norm": 3.248961925506592,
|
| 186 |
+
"learning_rate": 1.7852941176470592e-05,
|
| 187 |
+
"loss": 1.9454,
|
| 188 |
+
"step": 220
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 3.3823529411764706,
|
| 192 |
+
"grad_norm": 3.525045394897461,
|
| 193 |
+
"learning_rate": 1.7754901960784313e-05,
|
| 194 |
+
"loss": 1.8824,
|
| 195 |
+
"step": 230
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 3.5294117647058822,
|
| 199 |
+
"grad_norm": 3.95782470703125,
|
| 200 |
+
"learning_rate": 1.765686274509804e-05,
|
| 201 |
+
"loss": 1.8538,
|
| 202 |
+
"step": 240
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 3.6764705882352944,
|
| 206 |
+
"grad_norm": 4.282369613647461,
|
| 207 |
+
"learning_rate": 1.7558823529411765e-05,
|
| 208 |
+
"loss": 1.823,
|
| 209 |
+
"step": 250
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 3.8235294117647056,
|
| 213 |
+
"grad_norm": 6.753427505493164,
|
| 214 |
+
"learning_rate": 1.7460784313725493e-05,
|
| 215 |
+
"loss": 1.8116,
|
| 216 |
+
"step": 260
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 3.9705882352941178,
|
| 220 |
+
"grad_norm": 4.552963733673096,
|
| 221 |
+
"learning_rate": 1.7362745098039217e-05,
|
| 222 |
+
"loss": 1.7387,
|
| 223 |
+
"step": 270
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 4.0,
|
| 227 |
+
"eval_loss": 1.6489075422286987,
|
| 228 |
+
"eval_runtime": 1.6483,
|
| 229 |
+
"eval_samples_per_second": 81.902,
|
| 230 |
+
"eval_steps_per_second": 10.314,
|
| 231 |
+
"step": 272
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 4.117647058823529,
|
| 235 |
+
"grad_norm": 3.1702029705047607,
|
| 236 |
+
"learning_rate": 1.7264705882352945e-05,
|
| 237 |
+
"loss": 1.6522,
|
| 238 |
+
"step": 280
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 4.264705882352941,
|
| 242 |
+
"grad_norm": 4.5589494705200195,
|
| 243 |
+
"learning_rate": 1.7166666666666666e-05,
|
| 244 |
+
"loss": 1.6078,
|
| 245 |
+
"step": 290
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 4.411764705882353,
|
| 249 |
+
"grad_norm": 3.9019124507904053,
|
| 250 |
+
"learning_rate": 1.7068627450980394e-05,
|
| 251 |
+
"loss": 1.5599,
|
| 252 |
+
"step": 300
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 4.5588235294117645,
|
| 256 |
+
"grad_norm": 4.917637348175049,
|
| 257 |
+
"learning_rate": 1.6970588235294118e-05,
|
| 258 |
+
"loss": 1.5237,
|
| 259 |
+
"step": 310
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 4.705882352941177,
|
| 263 |
+
"grad_norm": 3.638333320617676,
|
| 264 |
+
"learning_rate": 1.6872549019607846e-05,
|
| 265 |
+
"loss": 1.4842,
|
| 266 |
+
"step": 320
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 4.852941176470588,
|
| 270 |
+
"grad_norm": 3.488672971725464,
|
| 271 |
+
"learning_rate": 1.677450980392157e-05,
|
| 272 |
+
"loss": 1.4391,
|
| 273 |
+
"step": 330
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 5.0,
|
| 277 |
+
"grad_norm": 8.806844711303711,
|
| 278 |
+
"learning_rate": 1.6676470588235295e-05,
|
| 279 |
+
"loss": 1.4165,
|
| 280 |
+
"step": 340
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 5.0,
|
| 284 |
+
"eval_loss": 1.3353697061538696,
|
| 285 |
+
"eval_runtime": 1.6321,
|
| 286 |
+
"eval_samples_per_second": 82.718,
|
| 287 |
+
"eval_steps_per_second": 10.416,
|
| 288 |
+
"step": 340
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 5.147058823529412,
|
| 292 |
+
"grad_norm": 5.466882705688477,
|
| 293 |
+
"learning_rate": 1.6578431372549022e-05,
|
| 294 |
+
"loss": 1.3357,
|
| 295 |
+
"step": 350
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 5.294117647058823,
|
| 299 |
+
"grad_norm": 3.95070481300354,
|
| 300 |
+
"learning_rate": 1.6480392156862747e-05,
|
| 301 |
+
"loss": 1.3317,
|
| 302 |
+
"step": 360
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 5.4411764705882355,
|
| 306 |
+
"grad_norm": 3.2761778831481934,
|
| 307 |
+
"learning_rate": 1.638235294117647e-05,
|
| 308 |
+
"loss": 1.1954,
|
| 309 |
+
"step": 370
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 5.588235294117647,
|
| 313 |
+
"grad_norm": 2.8257029056549072,
|
| 314 |
+
"learning_rate": 1.62843137254902e-05,
|
| 315 |
+
"loss": 1.1713,
|
| 316 |
+
"step": 380
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 5.735294117647059,
|
| 320 |
+
"grad_norm": 3.5745885372161865,
|
| 321 |
+
"learning_rate": 1.6186274509803923e-05,
|
| 322 |
+
"loss": 1.1767,
|
| 323 |
+
"step": 390
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 5.882352941176471,
|
| 327 |
+
"grad_norm": 3.625549077987671,
|
| 328 |
+
"learning_rate": 1.6088235294117648e-05,
|
| 329 |
+
"loss": 1.1028,
|
| 330 |
+
"step": 400
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 6.0,
|
| 334 |
+
"eval_loss": 1.0512332916259766,
|
| 335 |
+
"eval_runtime": 1.6283,
|
| 336 |
+
"eval_samples_per_second": 82.911,
|
| 337 |
+
"eval_steps_per_second": 10.441,
|
| 338 |
+
"step": 408
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 6.029411764705882,
|
| 342 |
+
"grad_norm": 3.7580277919769287,
|
| 343 |
+
"learning_rate": 1.5990196078431375e-05,
|
| 344 |
+
"loss": 1.0918,
|
| 345 |
+
"step": 410
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 6.176470588235294,
|
| 349 |
+
"grad_norm": 5.0767292976379395,
|
| 350 |
+
"learning_rate": 1.58921568627451e-05,
|
| 351 |
+
"loss": 1.0799,
|
| 352 |
+
"step": 420
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 6.323529411764706,
|
| 356 |
+
"grad_norm": 3.362245798110962,
|
| 357 |
+
"learning_rate": 1.5794117647058824e-05,
|
| 358 |
+
"loss": 1.0246,
|
| 359 |
+
"step": 430
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 6.470588235294118,
|
| 363 |
+
"grad_norm": 2.663010835647583,
|
| 364 |
+
"learning_rate": 1.569607843137255e-05,
|
| 365 |
+
"loss": 0.9335,
|
| 366 |
+
"step": 440
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 6.617647058823529,
|
| 370 |
+
"grad_norm": 2.7811036109924316,
|
| 371 |
+
"learning_rate": 1.5598039215686276e-05,
|
| 372 |
+
"loss": 0.9084,
|
| 373 |
+
"step": 450
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 6.764705882352941,
|
| 377 |
+
"grad_norm": 6.619537830352783,
|
| 378 |
+
"learning_rate": 1.55e-05,
|
| 379 |
+
"loss": 0.8652,
|
| 380 |
+
"step": 460
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 6.911764705882353,
|
| 384 |
+
"grad_norm": 4.524352073669434,
|
| 385 |
+
"learning_rate": 1.540196078431373e-05,
|
| 386 |
+
"loss": 0.8706,
|
| 387 |
+
"step": 470
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 7.0,
|
| 391 |
+
"eval_loss": 0.837912380695343,
|
| 392 |
+
"eval_runtime": 1.6194,
|
| 393 |
+
"eval_samples_per_second": 83.366,
|
| 394 |
+
"eval_steps_per_second": 10.498,
|
| 395 |
+
"step": 476
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 7.0588235294117645,
|
| 399 |
+
"grad_norm": 4.127895832061768,
|
| 400 |
+
"learning_rate": 1.5303921568627453e-05,
|
| 401 |
+
"loss": 0.7721,
|
| 402 |
+
"step": 480
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 7.205882352941177,
|
| 406 |
+
"grad_norm": 2.9023818969726562,
|
| 407 |
+
"learning_rate": 1.5205882352941177e-05,
|
| 408 |
+
"loss": 0.7268,
|
| 409 |
+
"step": 490
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 7.352941176470588,
|
| 413 |
+
"grad_norm": 3.2444231510162354,
|
| 414 |
+
"learning_rate": 1.5107843137254903e-05,
|
| 415 |
+
"loss": 0.7142,
|
| 416 |
+
"step": 500
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 7.5,
|
| 420 |
+
"grad_norm": 3.160628318786621,
|
| 421 |
+
"learning_rate": 1.500980392156863e-05,
|
| 422 |
+
"loss": 0.7019,
|
| 423 |
+
"step": 510
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 7.647058823529412,
|
| 427 |
+
"grad_norm": 2.913454294204712,
|
| 428 |
+
"learning_rate": 1.4911764705882354e-05,
|
| 429 |
+
"loss": 0.6911,
|
| 430 |
+
"step": 520
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 7.794117647058823,
|
| 434 |
+
"grad_norm": 2.9466731548309326,
|
| 435 |
+
"learning_rate": 1.481372549019608e-05,
|
| 436 |
+
"loss": 0.6639,
|
| 437 |
+
"step": 530
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 7.9411764705882355,
|
| 441 |
+
"grad_norm": 2.8029139041900635,
|
| 442 |
+
"learning_rate": 1.4715686274509806e-05,
|
| 443 |
+
"loss": 0.5554,
|
| 444 |
+
"step": 540
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 8.0,
|
| 448 |
+
"eval_loss": 0.5985532402992249,
|
| 449 |
+
"eval_runtime": 1.6867,
|
| 450 |
+
"eval_samples_per_second": 80.038,
|
| 451 |
+
"eval_steps_per_second": 10.079,
|
| 452 |
+
"step": 544
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 8.088235294117647,
|
| 456 |
+
"grad_norm": 2.267561912536621,
|
| 457 |
+
"learning_rate": 1.461764705882353e-05,
|
| 458 |
+
"loss": 0.5757,
|
| 459 |
+
"step": 550
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 8.235294117647058,
|
| 463 |
+
"grad_norm": 2.1232762336730957,
|
| 464 |
+
"learning_rate": 1.4519607843137256e-05,
|
| 465 |
+
"loss": 0.5091,
|
| 466 |
+
"step": 560
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 8.382352941176471,
|
| 470 |
+
"grad_norm": 3.035518169403076,
|
| 471 |
+
"learning_rate": 1.442156862745098e-05,
|
| 472 |
+
"loss": 0.5158,
|
| 473 |
+
"step": 570
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 8.529411764705882,
|
| 477 |
+
"grad_norm": 2.085606336593628,
|
| 478 |
+
"learning_rate": 1.4323529411764707e-05,
|
| 479 |
+
"loss": 0.4673,
|
| 480 |
+
"step": 580
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 8.676470588235293,
|
| 484 |
+
"grad_norm": 1.8189260959625244,
|
| 485 |
+
"learning_rate": 1.4225490196078433e-05,
|
| 486 |
+
"loss": 0.4604,
|
| 487 |
+
"step": 590
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 8.823529411764707,
|
| 491 |
+
"grad_norm": 6.74431037902832,
|
| 492 |
+
"learning_rate": 1.4127450980392159e-05,
|
| 493 |
+
"loss": 0.4651,
|
| 494 |
+
"step": 600
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 8.970588235294118,
|
| 498 |
+
"grad_norm": 1.56160306930542,
|
| 499 |
+
"learning_rate": 1.4029411764705883e-05,
|
| 500 |
+
"loss": 0.4017,
|
| 501 |
+
"step": 610
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 9.0,
|
| 505 |
+
"eval_loss": 0.40496423840522766,
|
| 506 |
+
"eval_runtime": 1.6727,
|
| 507 |
+
"eval_samples_per_second": 80.707,
|
| 508 |
+
"eval_steps_per_second": 10.163,
|
| 509 |
+
"step": 612
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 9.117647058823529,
|
| 513 |
+
"grad_norm": 1.9897135496139526,
|
| 514 |
+
"learning_rate": 1.3931372549019608e-05,
|
| 515 |
+
"loss": 0.3879,
|
| 516 |
+
"step": 620
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 9.264705882352942,
|
| 520 |
+
"grad_norm": 1.6671721935272217,
|
| 521 |
+
"learning_rate": 1.3833333333333334e-05,
|
| 522 |
+
"loss": 0.3206,
|
| 523 |
+
"step": 630
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 9.411764705882353,
|
| 527 |
+
"grad_norm": 1.8617277145385742,
|
| 528 |
+
"learning_rate": 1.373529411764706e-05,
|
| 529 |
+
"loss": 0.3565,
|
| 530 |
+
"step": 640
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 9.558823529411764,
|
| 534 |
+
"grad_norm": 1.64924955368042,
|
| 535 |
+
"learning_rate": 1.3637254901960786e-05,
|
| 536 |
+
"loss": 0.3457,
|
| 537 |
+
"step": 650
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 9.705882352941176,
|
| 541 |
+
"grad_norm": 1.5722205638885498,
|
| 542 |
+
"learning_rate": 1.3539215686274512e-05,
|
| 543 |
+
"loss": 0.3123,
|
| 544 |
+
"step": 660
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 9.852941176470589,
|
| 548 |
+
"grad_norm": 2.3067121505737305,
|
| 549 |
+
"learning_rate": 1.3441176470588238e-05,
|
| 550 |
+
"loss": 0.3137,
|
| 551 |
+
"step": 670
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 10.0,
|
| 555 |
+
"grad_norm": 4.091474533081055,
|
| 556 |
+
"learning_rate": 1.334313725490196e-05,
|
| 557 |
+
"loss": 0.2663,
|
| 558 |
+
"step": 680
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 10.0,
|
| 562 |
+
"eval_loss": 0.29727470874786377,
|
| 563 |
+
"eval_runtime": 1.6004,
|
| 564 |
+
"eval_samples_per_second": 84.356,
|
| 565 |
+
"eval_steps_per_second": 10.623,
|
| 566 |
+
"step": 680
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 10.147058823529411,
|
| 570 |
+
"grad_norm": 1.217851161956787,
|
| 571 |
+
"learning_rate": 1.3245098039215687e-05,
|
| 572 |
+
"loss": 0.2133,
|
| 573 |
+
"step": 690
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 10.294117647058824,
|
| 577 |
+
"grad_norm": 1.5889488458633423,
|
| 578 |
+
"learning_rate": 1.3147058823529413e-05,
|
| 579 |
+
"loss": 0.2299,
|
| 580 |
+
"step": 700
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 10.441176470588236,
|
| 584 |
+
"grad_norm": 1.2852109670639038,
|
| 585 |
+
"learning_rate": 1.3049019607843139e-05,
|
| 586 |
+
"loss": 0.1991,
|
| 587 |
+
"step": 710
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 10.588235294117647,
|
| 591 |
+
"grad_norm": 1.1250885725021362,
|
| 592 |
+
"learning_rate": 1.2950980392156865e-05,
|
| 593 |
+
"loss": 0.2625,
|
| 594 |
+
"step": 720
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 10.735294117647058,
|
| 598 |
+
"grad_norm": 0.8768565654754639,
|
| 599 |
+
"learning_rate": 1.285294117647059e-05,
|
| 600 |
+
"loss": 0.1657,
|
| 601 |
+
"step": 730
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 10.882352941176471,
|
| 605 |
+
"grad_norm": 1.0275033712387085,
|
| 606 |
+
"learning_rate": 1.2754901960784314e-05,
|
| 607 |
+
"loss": 0.1814,
|
| 608 |
+
"step": 740
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 11.0,
|
| 612 |
+
"eval_loss": 0.22045107185840607,
|
| 613 |
+
"eval_runtime": 1.6056,
|
| 614 |
+
"eval_samples_per_second": 84.083,
|
| 615 |
+
"eval_steps_per_second": 10.588,
|
| 616 |
+
"step": 748
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 11.029411764705882,
|
| 620 |
+
"grad_norm": 1.1377185583114624,
|
| 621 |
+
"learning_rate": 1.265686274509804e-05,
|
| 622 |
+
"loss": 0.2123,
|
| 623 |
+
"step": 750
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 11.176470588235293,
|
| 627 |
+
"grad_norm": 0.6125518679618835,
|
| 628 |
+
"learning_rate": 1.2558823529411766e-05,
|
| 629 |
+
"loss": 0.145,
|
| 630 |
+
"step": 760
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 11.323529411764707,
|
| 634 |
+
"grad_norm": 3.132908344268799,
|
| 635 |
+
"learning_rate": 1.2460784313725492e-05,
|
| 636 |
+
"loss": 0.1409,
|
| 637 |
+
"step": 770
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 11.470588235294118,
|
| 641 |
+
"grad_norm": 7.12354850769043,
|
| 642 |
+
"learning_rate": 1.2362745098039218e-05,
|
| 643 |
+
"loss": 0.171,
|
| 644 |
+
"step": 780
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 11.617647058823529,
|
| 648 |
+
"grad_norm": 0.9431573152542114,
|
| 649 |
+
"learning_rate": 1.2264705882352944e-05,
|
| 650 |
+
"loss": 0.1561,
|
| 651 |
+
"step": 790
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 11.764705882352942,
|
| 655 |
+
"grad_norm": 0.5725237131118774,
|
| 656 |
+
"learning_rate": 1.2166666666666667e-05,
|
| 657 |
+
"loss": 0.1012,
|
| 658 |
+
"step": 800
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 11.911764705882353,
|
| 662 |
+
"grad_norm": 0.6124898791313171,
|
| 663 |
+
"learning_rate": 1.2068627450980393e-05,
|
| 664 |
+
"loss": 0.1072,
|
| 665 |
+
"step": 810
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 12.0,
|
| 669 |
+
"eval_loss": 0.20114047825336456,
|
| 670 |
+
"eval_runtime": 1.61,
|
| 671 |
+
"eval_samples_per_second": 83.85,
|
| 672 |
+
"eval_steps_per_second": 10.559,
|
| 673 |
+
"step": 816
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 12.058823529411764,
|
| 677 |
+
"grad_norm": 0.6034418344497681,
|
| 678 |
+
"learning_rate": 1.1970588235294119e-05,
|
| 679 |
+
"loss": 0.1093,
|
| 680 |
+
"step": 820
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 12.205882352941176,
|
| 684 |
+
"grad_norm": 0.6741713881492615,
|
| 685 |
+
"learning_rate": 1.1872549019607845e-05,
|
| 686 |
+
"loss": 0.0901,
|
| 687 |
+
"step": 830
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 12.352941176470589,
|
| 691 |
+
"grad_norm": 0.7810522317886353,
|
| 692 |
+
"learning_rate": 1.177450980392157e-05,
|
| 693 |
+
"loss": 0.089,
|
| 694 |
+
"step": 840
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 12.5,
|
| 698 |
+
"grad_norm": 0.6442411541938782,
|
| 699 |
+
"learning_rate": 1.1676470588235295e-05,
|
| 700 |
+
"loss": 0.1219,
|
| 701 |
+
"step": 850
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 12.647058823529411,
|
| 705 |
+
"grad_norm": 3.0149757862091064,
|
| 706 |
+
"learning_rate": 1.1578431372549021e-05,
|
| 707 |
+
"loss": 0.0814,
|
| 708 |
+
"step": 860
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 12.794117647058824,
|
| 712 |
+
"grad_norm": 0.46087804436683655,
|
| 713 |
+
"learning_rate": 1.1480392156862746e-05,
|
| 714 |
+
"loss": 0.1077,
|
| 715 |
+
"step": 870
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 12.941176470588236,
|
| 719 |
+
"grad_norm": 0.56253981590271,
|
| 720 |
+
"learning_rate": 1.1382352941176472e-05,
|
| 721 |
+
"loss": 0.068,
|
| 722 |
+
"step": 880
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 13.0,
|
| 726 |
+
"eval_loss": 0.1641794890165329,
|
| 727 |
+
"eval_runtime": 1.6752,
|
| 728 |
+
"eval_samples_per_second": 80.589,
|
| 729 |
+
"eval_steps_per_second": 10.148,
|
| 730 |
+
"step": 884
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 13.088235294117647,
|
| 734 |
+
"grad_norm": 0.6027768850326538,
|
| 735 |
+
"learning_rate": 1.1284313725490198e-05,
|
| 736 |
+
"loss": 0.0689,
|
| 737 |
+
"step": 890
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 13.235294117647058,
|
| 741 |
+
"grad_norm": 0.369243860244751,
|
| 742 |
+
"learning_rate": 1.1186274509803922e-05,
|
| 743 |
+
"loss": 0.0613,
|
| 744 |
+
"step": 900
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 13.382352941176471,
|
| 748 |
+
"grad_norm": 0.3756796419620514,
|
| 749 |
+
"learning_rate": 1.1088235294117648e-05,
|
| 750 |
+
"loss": 0.0599,
|
| 751 |
+
"step": 910
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 13.529411764705882,
|
| 755 |
+
"grad_norm": 0.42388486862182617,
|
| 756 |
+
"learning_rate": 1.0990196078431374e-05,
|
| 757 |
+
"loss": 0.0561,
|
| 758 |
+
"step": 920
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 13.676470588235293,
|
| 762 |
+
"grad_norm": 0.40439847111701965,
|
| 763 |
+
"learning_rate": 1.0892156862745099e-05,
|
| 764 |
+
"loss": 0.0542,
|
| 765 |
+
"step": 930
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 13.823529411764707,
|
| 769 |
+
"grad_norm": 0.39990735054016113,
|
| 770 |
+
"learning_rate": 1.0794117647058825e-05,
|
| 771 |
+
"loss": 0.0672,
|
| 772 |
+
"step": 940
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 13.970588235294118,
|
| 776 |
+
"grad_norm": 0.3704688251018524,
|
| 777 |
+
"learning_rate": 1.0696078431372549e-05,
|
| 778 |
+
"loss": 0.1051,
|
| 779 |
+
"step": 950
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 14.0,
|
| 783 |
+
"eval_loss": 0.14415012300014496,
|
| 784 |
+
"eval_runtime": 1.6471,
|
| 785 |
+
"eval_samples_per_second": 81.961,
|
| 786 |
+
"eval_steps_per_second": 10.321,
|
| 787 |
+
"step": 952
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 14.117647058823529,
|
| 791 |
+
"grad_norm": 0.33351826667785645,
|
| 792 |
+
"learning_rate": 1.0598039215686275e-05,
|
| 793 |
+
"loss": 0.0492,
|
| 794 |
+
"step": 960
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 14.264705882352942,
|
| 798 |
+
"grad_norm": 0.3827894926071167,
|
| 799 |
+
"learning_rate": 1.0500000000000001e-05,
|
| 800 |
+
"loss": 0.0484,
|
| 801 |
+
"step": 970
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 14.411764705882353,
|
| 805 |
+
"grad_norm": 0.4303022623062134,
|
| 806 |
+
"learning_rate": 1.0401960784313727e-05,
|
| 807 |
+
"loss": 0.0461,
|
| 808 |
+
"step": 980
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 14.558823529411764,
|
| 812 |
+
"grad_norm": 0.3589096665382385,
|
| 813 |
+
"learning_rate": 1.0303921568627452e-05,
|
| 814 |
+
"loss": 0.0468,
|
| 815 |
+
"step": 990
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 14.705882352941176,
|
| 819 |
+
"grad_norm": 0.30145418643951416,
|
| 820 |
+
"learning_rate": 1.0205882352941176e-05,
|
| 821 |
+
"loss": 0.0942,
|
| 822 |
+
"step": 1000
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 14.852941176470589,
|
| 826 |
+
"grad_norm": 0.3331129550933838,
|
| 827 |
+
"learning_rate": 1.0107843137254902e-05,
|
| 828 |
+
"loss": 0.0465,
|
| 829 |
+
"step": 1010
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 15.0,
|
| 833 |
+
"grad_norm": 0.790107011795044,
|
| 834 |
+
"learning_rate": 1.0009803921568628e-05,
|
| 835 |
+
"loss": 0.0431,
|
| 836 |
+
"step": 1020
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 15.0,
|
| 840 |
+
"eval_loss": 0.1515953093767166,
|
| 841 |
+
"eval_runtime": 1.5961,
|
| 842 |
+
"eval_samples_per_second": 84.583,
|
| 843 |
+
"eval_steps_per_second": 10.651,
|
| 844 |
+
"step": 1020
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 15.147058823529411,
|
| 848 |
+
"grad_norm": 0.4401082992553711,
|
| 849 |
+
"learning_rate": 9.911764705882354e-06,
|
| 850 |
+
"loss": 0.0511,
|
| 851 |
+
"step": 1030
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 15.294117647058824,
|
| 855 |
+
"grad_norm": 0.3193589150905609,
|
| 856 |
+
"learning_rate": 9.813725490196078e-06,
|
| 857 |
+
"loss": 0.0369,
|
| 858 |
+
"step": 1040
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 15.441176470588236,
|
| 862 |
+
"grad_norm": 0.33625075221061707,
|
| 863 |
+
"learning_rate": 9.715686274509805e-06,
|
| 864 |
+
"loss": 0.0363,
|
| 865 |
+
"step": 1050
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 15.588235294117647,
|
| 869 |
+
"grad_norm": 0.18927061557769775,
|
| 870 |
+
"learning_rate": 9.61764705882353e-06,
|
| 871 |
+
"loss": 0.0349,
|
| 872 |
+
"step": 1060
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 15.735294117647058,
|
| 876 |
+
"grad_norm": 0.28063905239105225,
|
| 877 |
+
"learning_rate": 9.519607843137255e-06,
|
| 878 |
+
"loss": 0.0355,
|
| 879 |
+
"step": 1070
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 15.882352941176471,
|
| 883 |
+
"grad_norm": 0.31923410296440125,
|
| 884 |
+
"learning_rate": 9.421568627450981e-06,
|
| 885 |
+
"loss": 0.0949,
|
| 886 |
+
"step": 1080
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 16.0,
|
| 890 |
+
"eval_loss": 0.16671152412891388,
|
| 891 |
+
"eval_runtime": 1.6889,
|
| 892 |
+
"eval_samples_per_second": 79.936,
|
| 893 |
+
"eval_steps_per_second": 10.066,
|
| 894 |
+
"step": 1088
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 16.029411764705884,
|
| 898 |
+
"grad_norm": 0.25026965141296387,
|
| 899 |
+
"learning_rate": 9.323529411764707e-06,
|
| 900 |
+
"loss": 0.0357,
|
| 901 |
+
"step": 1090
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 16.176470588235293,
|
| 905 |
+
"grad_norm": 0.39554500579833984,
|
| 906 |
+
"learning_rate": 9.225490196078433e-06,
|
| 907 |
+
"loss": 0.0946,
|
| 908 |
+
"step": 1100
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 16.323529411764707,
|
| 912 |
+
"grad_norm": 0.30123069882392883,
|
| 913 |
+
"learning_rate": 9.127450980392158e-06,
|
| 914 |
+
"loss": 0.0322,
|
| 915 |
+
"step": 1110
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 16.470588235294116,
|
| 919 |
+
"grad_norm": 0.25656214356422424,
|
| 920 |
+
"learning_rate": 9.029411764705884e-06,
|
| 921 |
+
"loss": 0.0291,
|
| 922 |
+
"step": 1120
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 16.61764705882353,
|
| 926 |
+
"grad_norm": 0.21169237792491913,
|
| 927 |
+
"learning_rate": 8.93137254901961e-06,
|
| 928 |
+
"loss": 0.029,
|
| 929 |
+
"step": 1130
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 16.764705882352942,
|
| 933 |
+
"grad_norm": 0.22442790865898132,
|
| 934 |
+
"learning_rate": 8.833333333333334e-06,
|
| 935 |
+
"loss": 0.0316,
|
| 936 |
+
"step": 1140
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 16.91176470588235,
|
| 940 |
+
"grad_norm": 0.20120474696159363,
|
| 941 |
+
"learning_rate": 8.73529411764706e-06,
|
| 942 |
+
"loss": 0.0287,
|
| 943 |
+
"step": 1150
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 17.0,
|
| 947 |
+
"eval_loss": 0.13797470927238464,
|
| 948 |
+
"eval_runtime": 1.6743,
|
| 949 |
+
"eval_samples_per_second": 80.63,
|
| 950 |
+
"eval_steps_per_second": 10.153,
|
| 951 |
+
"step": 1156
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 17.058823529411764,
|
| 955 |
+
"grad_norm": 0.20405986905097961,
|
| 956 |
+
"learning_rate": 8.637254901960786e-06,
|
| 957 |
+
"loss": 0.0321,
|
| 958 |
+
"step": 1160
|
| 959 |
+
},
|
| 960 |
+
{
|
| 961 |
+
"epoch": 17.205882352941178,
|
| 962 |
+
"grad_norm": 0.22055025398731232,
|
| 963 |
+
"learning_rate": 8.53921568627451e-06,
|
| 964 |
+
"loss": 0.027,
|
| 965 |
+
"step": 1170
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 17.352941176470587,
|
| 969 |
+
"grad_norm": 0.18829534947872162,
|
| 970 |
+
"learning_rate": 8.441176470588237e-06,
|
| 971 |
+
"loss": 0.0769,
|
| 972 |
+
"step": 1180
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 17.5,
|
| 976 |
+
"grad_norm": 0.7278969883918762,
|
| 977 |
+
"learning_rate": 8.343137254901961e-06,
|
| 978 |
+
"loss": 0.0276,
|
| 979 |
+
"step": 1190
|
| 980 |
+
},
|
| 981 |
+
{
|
| 982 |
+
"epoch": 17.647058823529413,
|
| 983 |
+
"grad_norm": 0.22244372963905334,
|
| 984 |
+
"learning_rate": 8.245098039215687e-06,
|
| 985 |
+
"loss": 0.0263,
|
| 986 |
+
"step": 1200
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 17.794117647058822,
|
| 990 |
+
"grad_norm": 0.1744055598974228,
|
| 991 |
+
"learning_rate": 8.147058823529413e-06,
|
| 992 |
+
"loss": 0.0241,
|
| 993 |
+
"step": 1210
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 17.941176470588236,
|
| 997 |
+
"grad_norm": 0.16366326808929443,
|
| 998 |
+
"learning_rate": 8.049019607843137e-06,
|
| 999 |
+
"loss": 0.0246,
|
| 1000 |
+
"step": 1220
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 18.0,
|
| 1004 |
+
"eval_loss": 0.15256382524967194,
|
| 1005 |
+
"eval_runtime": 1.6045,
|
| 1006 |
+
"eval_samples_per_second": 84.139,
|
| 1007 |
+
"eval_steps_per_second": 10.595,
|
| 1008 |
+
"step": 1224
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 18.08823529411765,
|
| 1012 |
+
"grad_norm": 0.15204133093357086,
|
| 1013 |
+
"learning_rate": 7.950980392156864e-06,
|
| 1014 |
+
"loss": 0.0254,
|
| 1015 |
+
"step": 1230
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 18.235294117647058,
|
| 1019 |
+
"grad_norm": 0.16595087945461273,
|
| 1020 |
+
"learning_rate": 7.85294117647059e-06,
|
| 1021 |
+
"loss": 0.0243,
|
| 1022 |
+
"step": 1240
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 18.38235294117647,
|
| 1026 |
+
"grad_norm": 0.2022247016429901,
|
| 1027 |
+
"learning_rate": 7.754901960784314e-06,
|
| 1028 |
+
"loss": 0.0242,
|
| 1029 |
+
"step": 1250
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 18.529411764705884,
|
| 1033 |
+
"grad_norm": 0.17330515384674072,
|
| 1034 |
+
"learning_rate": 7.65686274509804e-06,
|
| 1035 |
+
"loss": 0.024,
|
| 1036 |
+
"step": 1260
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 18.676470588235293,
|
| 1040 |
+
"grad_norm": 0.20826362073421478,
|
| 1041 |
+
"learning_rate": 7.558823529411765e-06,
|
| 1042 |
+
"loss": 0.0801,
|
| 1043 |
+
"step": 1270
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 18.823529411764707,
|
| 1047 |
+
"grad_norm": 0.2036994844675064,
|
| 1048 |
+
"learning_rate": 7.460784313725491e-06,
|
| 1049 |
+
"loss": 0.0232,
|
| 1050 |
+
"step": 1280
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 18.970588235294116,
|
| 1054 |
+
"grad_norm": 0.1467086523771286,
|
| 1055 |
+
"learning_rate": 7.3627450980392165e-06,
|
| 1056 |
+
"loss": 0.0217,
|
| 1057 |
+
"step": 1290
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 19.0,
|
| 1061 |
+
"eval_loss": 0.14522351324558258,
|
| 1062 |
+
"eval_runtime": 1.6184,
|
| 1063 |
+
"eval_samples_per_second": 83.417,
|
| 1064 |
+
"eval_steps_per_second": 10.504,
|
| 1065 |
+
"step": 1292
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 19.11764705882353,
|
| 1069 |
+
"grad_norm": 0.21091414988040924,
|
| 1070 |
+
"learning_rate": 7.264705882352942e-06,
|
| 1071 |
+
"loss": 0.0217,
|
| 1072 |
+
"step": 1300
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 19.264705882352942,
|
| 1076 |
+
"grad_norm": 0.16935530304908752,
|
| 1077 |
+
"learning_rate": 7.166666666666667e-06,
|
| 1078 |
+
"loss": 0.0742,
|
| 1079 |
+
"step": 1310
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 19.41176470588235,
|
| 1083 |
+
"grad_norm": 0.19531255960464478,
|
| 1084 |
+
"learning_rate": 7.068627450980393e-06,
|
| 1085 |
+
"loss": 0.0203,
|
| 1086 |
+
"step": 1320
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 19.558823529411764,
|
| 1090 |
+
"grad_norm": 0.18234974145889282,
|
| 1091 |
+
"learning_rate": 6.970588235294118e-06,
|
| 1092 |
+
"loss": 0.0539,
|
| 1093 |
+
"step": 1330
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 19.705882352941178,
|
| 1097 |
+
"grad_norm": 0.1523897647857666,
|
| 1098 |
+
"learning_rate": 6.8725490196078434e-06,
|
| 1099 |
+
"loss": 0.0207,
|
| 1100 |
+
"step": 1340
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 19.852941176470587,
|
| 1104 |
+
"grad_norm": 0.1887982189655304,
|
| 1105 |
+
"learning_rate": 6.7745098039215695e-06,
|
| 1106 |
+
"loss": 0.0205,
|
| 1107 |
+
"step": 1350
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 20.0,
|
| 1111 |
+
"grad_norm": 1.2818535566329956,
|
| 1112 |
+
"learning_rate": 6.676470588235294e-06,
|
| 1113 |
+
"loss": 0.0216,
|
| 1114 |
+
"step": 1360
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 20.0,
|
| 1118 |
+
"eval_loss": 0.15050244331359863,
|
| 1119 |
+
"eval_runtime": 1.6089,
|
| 1120 |
+
"eval_samples_per_second": 83.907,
|
| 1121 |
+
"eval_steps_per_second": 10.566,
|
| 1122 |
+
"step": 1360
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 20.147058823529413,
|
| 1126 |
+
"grad_norm": 0.1478956639766693,
|
| 1127 |
+
"learning_rate": 6.57843137254902e-06,
|
| 1128 |
+
"loss": 0.0181,
|
| 1129 |
+
"step": 1370
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 20.294117647058822,
|
| 1133 |
+
"grad_norm": 0.18312789499759674,
|
| 1134 |
+
"learning_rate": 6.480392156862746e-06,
|
| 1135 |
+
"loss": 0.1025,
|
| 1136 |
+
"step": 1380
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 20.441176470588236,
|
| 1140 |
+
"grad_norm": 0.15513214468955994,
|
| 1141 |
+
"learning_rate": 6.38235294117647e-06,
|
| 1142 |
+
"loss": 0.0209,
|
| 1143 |
+
"step": 1390
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 20.58823529411765,
|
| 1147 |
+
"grad_norm": 0.13837294280529022,
|
| 1148 |
+
"learning_rate": 6.2843137254901964e-06,
|
| 1149 |
+
"loss": 0.0199,
|
| 1150 |
+
"step": 1400
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 20.735294117647058,
|
| 1154 |
+
"grad_norm": 0.1465756893157959,
|
| 1155 |
+
"learning_rate": 6.1862745098039225e-06,
|
| 1156 |
+
"loss": 0.0192,
|
| 1157 |
+
"step": 1410
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 20.88235294117647,
|
| 1161 |
+
"grad_norm": 0.16790196299552917,
|
| 1162 |
+
"learning_rate": 6.088235294117647e-06,
|
| 1163 |
+
"loss": 0.0195,
|
| 1164 |
+
"step": 1420
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 21.0,
|
| 1168 |
+
"eval_loss": 0.1510893851518631,
|
| 1169 |
+
"eval_runtime": 1.6277,
|
| 1170 |
+
"eval_samples_per_second": 82.937,
|
| 1171 |
+
"eval_steps_per_second": 10.444,
|
| 1172 |
+
"step": 1428
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 21.029411764705884,
|
| 1176 |
+
"grad_norm": 0.29194769263267517,
|
| 1177 |
+
"learning_rate": 5.990196078431373e-06,
|
| 1178 |
+
"loss": 0.0181,
|
| 1179 |
+
"step": 1430
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 21.176470588235293,
|
| 1183 |
+
"grad_norm": 0.16755153238773346,
|
| 1184 |
+
"learning_rate": 5.892156862745099e-06,
|
| 1185 |
+
"loss": 0.0186,
|
| 1186 |
+
"step": 1440
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 21.323529411764707,
|
| 1190 |
+
"grad_norm": 0.19387616217136383,
|
| 1191 |
+
"learning_rate": 5.794117647058824e-06,
|
| 1192 |
+
"loss": 0.0184,
|
| 1193 |
+
"step": 1450
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 21.470588235294116,
|
| 1197 |
+
"grad_norm": 0.15613146126270294,
|
| 1198 |
+
"learning_rate": 5.6960784313725494e-06,
|
| 1199 |
+
"loss": 0.0196,
|
| 1200 |
+
"step": 1460
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 21.61764705882353,
|
| 1204 |
+
"grad_norm": 0.12751713395118713,
|
| 1205 |
+
"learning_rate": 5.598039215686275e-06,
|
| 1206 |
+
"loss": 0.0178,
|
| 1207 |
+
"step": 1470
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 21.764705882352942,
|
| 1211 |
+
"grad_norm": 0.15414807200431824,
|
| 1212 |
+
"learning_rate": 5.500000000000001e-06,
|
| 1213 |
+
"loss": 0.0177,
|
| 1214 |
+
"step": 1480
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 21.91176470588235,
|
| 1218 |
+
"grad_norm": 0.12436957657337189,
|
| 1219 |
+
"learning_rate": 5.401960784313726e-06,
|
| 1220 |
+
"loss": 0.0749,
|
| 1221 |
+
"step": 1490
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 22.0,
|
| 1225 |
+
"eval_loss": 0.14262542128562927,
|
| 1226 |
+
"eval_runtime": 1.6896,
|
| 1227 |
+
"eval_samples_per_second": 79.901,
|
| 1228 |
+
"eval_steps_per_second": 10.062,
|
| 1229 |
+
"step": 1496
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 22.058823529411764,
|
| 1233 |
+
"grad_norm": 0.1797289252281189,
|
| 1234 |
+
"learning_rate": 5.303921568627451e-06,
|
| 1235 |
+
"loss": 0.0163,
|
| 1236 |
+
"step": 1500
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 22.205882352941178,
|
| 1240 |
+
"grad_norm": 0.15396994352340698,
|
| 1241 |
+
"learning_rate": 5.205882352941177e-06,
|
| 1242 |
+
"loss": 0.0164,
|
| 1243 |
+
"step": 1510
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 22.352941176470587,
|
| 1247 |
+
"grad_norm": 0.43706995248794556,
|
| 1248 |
+
"learning_rate": 5.107843137254902e-06,
|
| 1249 |
+
"loss": 0.0173,
|
| 1250 |
+
"step": 1520
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 22.5,
|
| 1254 |
+
"grad_norm": 0.13553638756275177,
|
| 1255 |
+
"learning_rate": 5.009803921568628e-06,
|
| 1256 |
+
"loss": 0.0701,
|
| 1257 |
+
"step": 1530
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 22.647058823529413,
|
| 1261 |
+
"grad_norm": 0.14506298303604126,
|
| 1262 |
+
"learning_rate": 4.911764705882353e-06,
|
| 1263 |
+
"loss": 0.0158,
|
| 1264 |
+
"step": 1540
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 22.794117647058822,
|
| 1268 |
+
"grad_norm": 0.25496020913124084,
|
| 1269 |
+
"learning_rate": 4.813725490196079e-06,
|
| 1270 |
+
"loss": 0.0152,
|
| 1271 |
+
"step": 1550
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 22.941176470588236,
|
| 1275 |
+
"grad_norm": 0.16556541621685028,
|
| 1276 |
+
"learning_rate": 4.715686274509804e-06,
|
| 1277 |
+
"loss": 0.0157,
|
| 1278 |
+
"step": 1560
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 23.0,
|
| 1282 |
+
"eval_loss": 0.1430669128894806,
|
| 1283 |
+
"eval_runtime": 1.6521,
|
| 1284 |
+
"eval_samples_per_second": 81.714,
|
| 1285 |
+
"eval_steps_per_second": 10.29,
|
| 1286 |
+
"step": 1564
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 23.08823529411765,
|
| 1290 |
+
"grad_norm": 0.12428230047225952,
|
| 1291 |
+
"learning_rate": 4.61764705882353e-06,
|
| 1292 |
+
"loss": 0.0159,
|
| 1293 |
+
"step": 1570
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 23.235294117647058,
|
| 1297 |
+
"grad_norm": 0.11920207738876343,
|
| 1298 |
+
"learning_rate": 4.519607843137255e-06,
|
| 1299 |
+
"loss": 0.0145,
|
| 1300 |
+
"step": 1580
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 23.38235294117647,
|
| 1304 |
+
"grad_norm": 0.1940208524465561,
|
| 1305 |
+
"learning_rate": 4.421568627450981e-06,
|
| 1306 |
+
"loss": 0.0162,
|
| 1307 |
+
"step": 1590
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 23.529411764705884,
|
| 1311 |
+
"grad_norm": 0.1417732536792755,
|
| 1312 |
+
"learning_rate": 4.323529411764707e-06,
|
| 1313 |
+
"loss": 0.0149,
|
| 1314 |
+
"step": 1600
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 23.676470588235293,
|
| 1318 |
+
"grad_norm": 0.13574868440628052,
|
| 1319 |
+
"learning_rate": 4.225490196078432e-06,
|
| 1320 |
+
"loss": 0.0152,
|
| 1321 |
+
"step": 1610
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 23.823529411764707,
|
| 1325 |
+
"grad_norm": 0.1438971906900406,
|
| 1326 |
+
"learning_rate": 4.127450980392157e-06,
|
| 1327 |
+
"loss": 0.0677,
|
| 1328 |
+
"step": 1620
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 23.970588235294116,
|
| 1332 |
+
"grad_norm": 0.12435536831617355,
|
| 1333 |
+
"learning_rate": 4.029411764705883e-06,
|
| 1334 |
+
"loss": 0.0152,
|
| 1335 |
+
"step": 1630
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 24.0,
|
| 1339 |
+
"eval_loss": 0.14459367096424103,
|
| 1340 |
+
"eval_runtime": 1.6409,
|
| 1341 |
+
"eval_samples_per_second": 82.271,
|
| 1342 |
+
"eval_steps_per_second": 10.36,
|
| 1343 |
+
"step": 1632
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 24.11764705882353,
|
| 1347 |
+
"grad_norm": 0.15185730159282684,
|
| 1348 |
+
"learning_rate": 3.931372549019608e-06,
|
| 1349 |
+
"loss": 0.0146,
|
| 1350 |
+
"step": 1640
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 24.264705882352942,
|
| 1354 |
+
"grad_norm": 0.14219020307064056,
|
| 1355 |
+
"learning_rate": 3.833333333333334e-06,
|
| 1356 |
+
"loss": 0.0141,
|
| 1357 |
+
"step": 1650
|
| 1358 |
+
},
|
| 1359 |
+
{
|
| 1360 |
+
"epoch": 24.41176470588235,
|
| 1361 |
+
"grad_norm": 0.1183333620429039,
|
| 1362 |
+
"learning_rate": 3.7352941176470593e-06,
|
| 1363 |
+
"loss": 0.0142,
|
| 1364 |
+
"step": 1660
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"epoch": 24.558823529411764,
|
| 1368 |
+
"grad_norm": 0.11505335569381714,
|
| 1369 |
+
"learning_rate": 3.6372549019607845e-06,
|
| 1370 |
+
"loss": 0.0601,
|
| 1371 |
+
"step": 1670
|
| 1372 |
+
},
|
| 1373 |
+
{
|
| 1374 |
+
"epoch": 24.705882352941178,
|
| 1375 |
+
"grad_norm": 0.1279689520597458,
|
| 1376 |
+
"learning_rate": 3.53921568627451e-06,
|
| 1377 |
+
"loss": 0.0149,
|
| 1378 |
+
"step": 1680
|
| 1379 |
+
},
|
| 1380 |
+
{
|
| 1381 |
+
"epoch": 24.852941176470587,
|
| 1382 |
+
"grad_norm": 0.09878543764352798,
|
| 1383 |
+
"learning_rate": 3.4411764705882358e-06,
|
| 1384 |
+
"loss": 0.0145,
|
| 1385 |
+
"step": 1690
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 25.0,
|
| 1389 |
+
"grad_norm": 0.2872982919216156,
|
| 1390 |
+
"learning_rate": 3.343137254901961e-06,
|
| 1391 |
+
"loss": 0.0142,
|
| 1392 |
+
"step": 1700
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 25.0,
|
| 1396 |
+
"eval_loss": 0.1434563398361206,
|
| 1397 |
+
"eval_runtime": 1.5906,
|
| 1398 |
+
"eval_samples_per_second": 84.875,
|
| 1399 |
+
"eval_steps_per_second": 10.688,
|
| 1400 |
+
"step": 1700
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 25.147058823529413,
|
| 1404 |
+
"grad_norm": 13.370408058166504,
|
| 1405 |
+
"learning_rate": 3.2450980392156866e-06,
|
| 1406 |
+
"loss": 0.0515,
|
| 1407 |
+
"step": 1710
|
| 1408 |
+
},
|
| 1409 |
+
{
|
| 1410 |
+
"epoch": 25.294117647058822,
|
| 1411 |
+
"grad_norm": 0.11658412218093872,
|
| 1412 |
+
"learning_rate": 3.147058823529412e-06,
|
| 1413 |
+
"loss": 0.0136,
|
| 1414 |
+
"step": 1720
|
| 1415 |
+
},
|
| 1416 |
+
{
|
| 1417 |
+
"epoch": 25.441176470588236,
|
| 1418 |
+
"grad_norm": 0.082937091588974,
|
| 1419 |
+
"learning_rate": 3.049019607843138e-06,
|
| 1420 |
+
"loss": 0.0145,
|
| 1421 |
+
"step": 1730
|
| 1422 |
+
},
|
| 1423 |
+
{
|
| 1424 |
+
"epoch": 25.58823529411765,
|
| 1425 |
+
"grad_norm": 0.09096076339483261,
|
| 1426 |
+
"learning_rate": 2.950980392156863e-06,
|
| 1427 |
+
"loss": 0.0132,
|
| 1428 |
+
"step": 1740
|
| 1429 |
+
},
|
| 1430 |
+
{
|
| 1431 |
+
"epoch": 25.735294117647058,
|
| 1432 |
+
"grad_norm": 0.12208091467618942,
|
| 1433 |
+
"learning_rate": 2.8529411764705883e-06,
|
| 1434 |
+
"loss": 0.0144,
|
| 1435 |
+
"step": 1750
|
| 1436 |
+
},
|
| 1437 |
+
{
|
| 1438 |
+
"epoch": 25.88235294117647,
|
| 1439 |
+
"grad_norm": 0.11157109588384628,
|
| 1440 |
+
"learning_rate": 2.754901960784314e-06,
|
| 1441 |
+
"loss": 0.0134,
|
| 1442 |
+
"step": 1760
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 26.0,
|
| 1446 |
+
"eval_loss": 0.14469991624355316,
|
| 1447 |
+
"eval_runtime": 1.6847,
|
| 1448 |
+
"eval_samples_per_second": 80.133,
|
| 1449 |
+
"eval_steps_per_second": 10.091,
|
| 1450 |
+
"step": 1768
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"epoch": 26.029411764705884,
|
| 1454 |
+
"grad_norm": 0.13175475597381592,
|
| 1455 |
+
"learning_rate": 2.6568627450980396e-06,
|
| 1456 |
+
"loss": 0.0139,
|
| 1457 |
+
"step": 1770
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 26.176470588235293,
|
| 1461 |
+
"grad_norm": 0.12328551709651947,
|
| 1462 |
+
"learning_rate": 2.558823529411765e-06,
|
| 1463 |
+
"loss": 0.0127,
|
| 1464 |
+
"step": 1780
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 26.323529411764707,
|
| 1468 |
+
"grad_norm": 0.08687178790569305,
|
| 1469 |
+
"learning_rate": 2.4607843137254905e-06,
|
| 1470 |
+
"loss": 0.0731,
|
| 1471 |
+
"step": 1790
|
| 1472 |
+
},
|
| 1473 |
+
{
|
| 1474 |
+
"epoch": 26.470588235294116,
|
| 1475 |
+
"grad_norm": 0.11597796529531479,
|
| 1476 |
+
"learning_rate": 2.3627450980392157e-06,
|
| 1477 |
+
"loss": 0.0135,
|
| 1478 |
+
"step": 1800
|
| 1479 |
+
},
|
| 1480 |
+
{
|
| 1481 |
+
"epoch": 26.61764705882353,
|
| 1482 |
+
"grad_norm": 0.09727944433689117,
|
| 1483 |
+
"learning_rate": 2.2647058823529413e-06,
|
| 1484 |
+
"loss": 0.014,
|
| 1485 |
+
"step": 1810
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"epoch": 26.764705882352942,
|
| 1489 |
+
"grad_norm": 0.1099948063492775,
|
| 1490 |
+
"learning_rate": 2.166666666666667e-06,
|
| 1491 |
+
"loss": 0.0139,
|
| 1492 |
+
"step": 1820
|
| 1493 |
+
},
|
| 1494 |
+
{
|
| 1495 |
+
"epoch": 26.91176470588235,
|
| 1496 |
+
"grad_norm": 0.1616813987493515,
|
| 1497 |
+
"learning_rate": 2.068627450980392e-06,
|
| 1498 |
+
"loss": 0.0137,
|
| 1499 |
+
"step": 1830
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 27.0,
|
| 1503 |
+
"eval_loss": 0.15348146855831146,
|
| 1504 |
+
"eval_runtime": 1.6268,
|
| 1505 |
+
"eval_samples_per_second": 82.982,
|
| 1506 |
+
"eval_steps_per_second": 10.45,
|
| 1507 |
+
"step": 1836
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 27.058823529411764,
|
| 1511 |
+
"grad_norm": 0.09702517092227936,
|
| 1512 |
+
"learning_rate": 1.970588235294118e-06,
|
| 1513 |
+
"loss": 0.013,
|
| 1514 |
+
"step": 1840
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 27.205882352941178,
|
| 1518 |
+
"grad_norm": 0.4694797992706299,
|
| 1519 |
+
"learning_rate": 1.8725490196078432e-06,
|
| 1520 |
+
"loss": 0.014,
|
| 1521 |
+
"step": 1850
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 27.352941176470587,
|
| 1525 |
+
"grad_norm": 0.11010562628507614,
|
| 1526 |
+
"learning_rate": 1.7745098039215689e-06,
|
| 1527 |
+
"loss": 0.0126,
|
| 1528 |
+
"step": 1860
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 27.5,
|
| 1532 |
+
"grad_norm": 0.11507220566272736,
|
| 1533 |
+
"learning_rate": 1.676470588235294e-06,
|
| 1534 |
+
"loss": 0.0134,
|
| 1535 |
+
"step": 1870
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 27.647058823529413,
|
| 1539 |
+
"grad_norm": 0.10667148232460022,
|
| 1540 |
+
"learning_rate": 1.5784313725490197e-06,
|
| 1541 |
+
"loss": 0.0703,
|
| 1542 |
+
"step": 1880
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 27.794117647058822,
|
| 1546 |
+
"grad_norm": 0.09035525470972061,
|
| 1547 |
+
"learning_rate": 1.4803921568627452e-06,
|
| 1548 |
+
"loss": 0.0127,
|
| 1549 |
+
"step": 1890
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 27.941176470588236,
|
| 1553 |
+
"grad_norm": 0.13801293075084686,
|
| 1554 |
+
"learning_rate": 1.3823529411764708e-06,
|
| 1555 |
+
"loss": 0.0127,
|
| 1556 |
+
"step": 1900
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 28.0,
|
| 1560 |
+
"eval_loss": 0.151557058095932,
|
| 1561 |
+
"eval_runtime": 1.6327,
|
| 1562 |
+
"eval_samples_per_second": 82.687,
|
| 1563 |
+
"eval_steps_per_second": 10.412,
|
| 1564 |
+
"step": 1904
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 28.08823529411765,
|
| 1568 |
+
"grad_norm": 0.12456930428743362,
|
| 1569 |
+
"learning_rate": 1.284313725490196e-06,
|
| 1570 |
+
"loss": 0.0121,
|
| 1571 |
+
"step": 1910
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 28.235294117647058,
|
| 1575 |
+
"grad_norm": 0.1714206337928772,
|
| 1576 |
+
"learning_rate": 1.1862745098039217e-06,
|
| 1577 |
+
"loss": 0.0143,
|
| 1578 |
+
"step": 1920
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 28.38235294117647,
|
| 1582 |
+
"grad_norm": 0.09358106553554535,
|
| 1583 |
+
"learning_rate": 1.088235294117647e-06,
|
| 1584 |
+
"loss": 0.0124,
|
| 1585 |
+
"step": 1930
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 28.529411764705884,
|
| 1589 |
+
"grad_norm": 0.12065358459949493,
|
| 1590 |
+
"learning_rate": 9.901960784313725e-07,
|
| 1591 |
+
"loss": 0.013,
|
| 1592 |
+
"step": 1940
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 28.676470588235293,
|
| 1596 |
+
"grad_norm": 0.10697101801633835,
|
| 1597 |
+
"learning_rate": 8.921568627450982e-07,
|
| 1598 |
+
"loss": 0.0128,
|
| 1599 |
+
"step": 1950
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 28.823529411764707,
|
| 1603 |
+
"grad_norm": 0.11727521568536758,
|
| 1604 |
+
"learning_rate": 7.941176470588236e-07,
|
| 1605 |
+
"loss": 0.0121,
|
| 1606 |
+
"step": 1960
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 28.970588235294116,
|
| 1610 |
+
"grad_norm": 0.10474637895822525,
|
| 1611 |
+
"learning_rate": 6.960784313725491e-07,
|
| 1612 |
+
"loss": 0.0548,
|
| 1613 |
+
"step": 1970
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 29.0,
|
| 1617 |
+
"eval_loss": 0.15194356441497803,
|
| 1618 |
+
"eval_runtime": 1.6187,
|
| 1619 |
+
"eval_samples_per_second": 83.399,
|
| 1620 |
+
"eval_steps_per_second": 10.502,
|
| 1621 |
+
"step": 1972
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 29.11764705882353,
|
| 1625 |
+
"grad_norm": 0.11856939643621445,
|
| 1626 |
+
"learning_rate": 5.980392156862745e-07,
|
| 1627 |
+
"loss": 0.0121,
|
| 1628 |
+
"step": 1980
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 29.264705882352942,
|
| 1632 |
+
"grad_norm": 0.12045503407716751,
|
| 1633 |
+
"learning_rate": 5.000000000000001e-07,
|
| 1634 |
+
"loss": 0.0128,
|
| 1635 |
+
"step": 1990
|
| 1636 |
+
},
|
| 1637 |
+
{
|
| 1638 |
+
"epoch": 29.41176470588235,
|
| 1639 |
+
"grad_norm": 0.11332525312900543,
|
| 1640 |
+
"learning_rate": 4.0196078431372556e-07,
|
| 1641 |
+
"loss": 0.0119,
|
| 1642 |
+
"step": 2000
|
| 1643 |
+
}
|
| 1644 |
+
],
|
| 1645 |
+
"logging_steps": 10,
|
| 1646 |
+
"max_steps": 2040,
|
| 1647 |
+
"num_input_tokens_seen": 0,
|
| 1648 |
+
"num_train_epochs": 30,
|
| 1649 |
+
"save_steps": 500,
|
| 1650 |
+
"stateful_callbacks": {
|
| 1651 |
+
"TrainerControl": {
|
| 1652 |
+
"args": {
|
| 1653 |
+
"should_epoch_stop": false,
|
| 1654 |
+
"should_evaluate": false,
|
| 1655 |
+
"should_log": false,
|
| 1656 |
+
"should_save": true,
|
| 1657 |
+
"should_training_stop": false
|
| 1658 |
+
},
|
| 1659 |
+
"attributes": {}
|
| 1660 |
+
}
|
| 1661 |
+
},
|
| 1662 |
+
"total_flos": 4156813161449472.0,
|
| 1663 |
+
"train_batch_size": 8,
|
| 1664 |
+
"trial_name": null,
|
| 1665 |
+
"trial_params": null
|
| 1666 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0ac7204b325961b954225fd6c5ea6cbc6125f315fac2247932e723704e953da7
|
| 3 |
+
size 5713
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|