Upload folder using huggingface_hub
Browse files- config.json +26 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +55 -0
- trainer_state.json +1888 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "intfloat/e5-base-v2",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertModel"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"gradient_checkpointing": false,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"layer_norm_eps": 1e-12,
|
| 15 |
+
"max_position_embeddings": 512,
|
| 16 |
+
"model_type": "bert",
|
| 17 |
+
"num_attention_heads": 12,
|
| 18 |
+
"num_hidden_layers": 12,
|
| 19 |
+
"pad_token_id": 0,
|
| 20 |
+
"position_embedding_type": "absolute",
|
| 21 |
+
"torch_dtype": "float32",
|
| 22 |
+
"transformers_version": "4.43.0",
|
| 23 |
+
"type_vocab_size": 2,
|
| 24 |
+
"use_cache": true,
|
| 25 |
+
"vocab_size": 30522
|
| 26 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5fd14e5a60a233a89b65f4e5e0214877284a315b7c91d095cccb98659375eec3
|
| 3 |
+
size 871298443
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e0128a3215d42508e2e197bcc28506531bdb9e60602387d7f2cd27d796f34617
|
| 3 |
+
size 437996607
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:53442c06a9fc645607ba081ae2825760c75829829d5f7ece4e608f24ffd5962a
|
| 3 |
+
size 14645
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2aed9d3908f9d60b0fc5e2ea967c2b39940266eed1a829376253b0feab8ea16c
|
| 3 |
+
size 1465
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"100": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"101": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"102": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"103": {
|
| 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_lower_case": true,
|
| 47 |
+
"mask_token": "[MASK]",
|
| 48 |
+
"model_max_length": 512,
|
| 49 |
+
"pad_token": "[PAD]",
|
| 50 |
+
"sep_token": "[SEP]",
|
| 51 |
+
"strip_accents": null,
|
| 52 |
+
"tokenize_chinese_chars": true,
|
| 53 |
+
"tokenizer_class": "BertTokenizer",
|
| 54 |
+
"unk_token": "[UNK]"
|
| 55 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1888 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 2658,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.007524454477050414,
|
| 13 |
+
"grad_norm": 63.752017974853516,
|
| 14 |
+
"learning_rate": 1e-05,
|
| 15 |
+
"loss": 3.3312,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.015048908954100828,
|
| 20 |
+
"grad_norm": 50.676944732666016,
|
| 21 |
+
"learning_rate": 1e-05,
|
| 22 |
+
"loss": 1.525,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.022573363431151242,
|
| 27 |
+
"grad_norm": 38.975914001464844,
|
| 28 |
+
"learning_rate": 1e-05,
|
| 29 |
+
"loss": 1.2213,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.030097817908201655,
|
| 34 |
+
"grad_norm": 30.093769073486328,
|
| 35 |
+
"learning_rate": 1e-05,
|
| 36 |
+
"loss": 0.8578,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.03762227238525207,
|
| 41 |
+
"grad_norm": 27.47739028930664,
|
| 42 |
+
"learning_rate": 1e-05,
|
| 43 |
+
"loss": 0.8364,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.045146726862302484,
|
| 48 |
+
"grad_norm": 26.15501594543457,
|
| 49 |
+
"learning_rate": 1e-05,
|
| 50 |
+
"loss": 0.8731,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.0526711813393529,
|
| 55 |
+
"grad_norm": 30.096651077270508,
|
| 56 |
+
"learning_rate": 1e-05,
|
| 57 |
+
"loss": 0.5887,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.06019563581640331,
|
| 62 |
+
"grad_norm": 21.479469299316406,
|
| 63 |
+
"learning_rate": 1e-05,
|
| 64 |
+
"loss": 0.6146,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.06772009029345373,
|
| 69 |
+
"grad_norm": 25.00172996520996,
|
| 70 |
+
"learning_rate": 1e-05,
|
| 71 |
+
"loss": 0.7224,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.07524454477050414,
|
| 76 |
+
"grad_norm": 19.167516708374023,
|
| 77 |
+
"learning_rate": 1e-05,
|
| 78 |
+
"loss": 0.6965,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.08276899924755456,
|
| 83 |
+
"grad_norm": 25.692691802978516,
|
| 84 |
+
"learning_rate": 1e-05,
|
| 85 |
+
"loss": 0.5539,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.09029345372460497,
|
| 90 |
+
"grad_norm": 19.874868392944336,
|
| 91 |
+
"learning_rate": 1e-05,
|
| 92 |
+
"loss": 0.5842,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.09781790820165538,
|
| 97 |
+
"grad_norm": 22.329219818115234,
|
| 98 |
+
"learning_rate": 1e-05,
|
| 99 |
+
"loss": 0.6003,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.1053423626787058,
|
| 104 |
+
"grad_norm": 14.971156120300293,
|
| 105 |
+
"learning_rate": 1e-05,
|
| 106 |
+
"loss": 0.5671,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.11286681715575621,
|
| 111 |
+
"grad_norm": 14.223251342773438,
|
| 112 |
+
"learning_rate": 1e-05,
|
| 113 |
+
"loss": 0.5096,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.12039127163280662,
|
| 118 |
+
"grad_norm": 19.30224609375,
|
| 119 |
+
"learning_rate": 1e-05,
|
| 120 |
+
"loss": 0.5067,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.12791572610985705,
|
| 125 |
+
"grad_norm": 14.27910327911377,
|
| 126 |
+
"learning_rate": 1e-05,
|
| 127 |
+
"loss": 0.5281,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.13544018058690746,
|
| 132 |
+
"grad_norm": 19.874217987060547,
|
| 133 |
+
"learning_rate": 1e-05,
|
| 134 |
+
"loss": 0.4569,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.14296463506395787,
|
| 139 |
+
"grad_norm": 16.16669273376465,
|
| 140 |
+
"learning_rate": 1e-05,
|
| 141 |
+
"loss": 0.5304,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.1504890895410083,
|
| 146 |
+
"grad_norm": 13.952829360961914,
|
| 147 |
+
"learning_rate": 1e-05,
|
| 148 |
+
"loss": 0.4415,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.1580135440180587,
|
| 153 |
+
"grad_norm": 16.3791446685791,
|
| 154 |
+
"learning_rate": 1e-05,
|
| 155 |
+
"loss": 0.5214,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.1655379984951091,
|
| 160 |
+
"grad_norm": 11.849374771118164,
|
| 161 |
+
"learning_rate": 1e-05,
|
| 162 |
+
"loss": 0.4079,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.17306245297215953,
|
| 167 |
+
"grad_norm": 17.892818450927734,
|
| 168 |
+
"learning_rate": 1e-05,
|
| 169 |
+
"loss": 0.562,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.18058690744920994,
|
| 174 |
+
"grad_norm": 11.97033977508545,
|
| 175 |
+
"learning_rate": 1e-05,
|
| 176 |
+
"loss": 0.3959,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.18811136192626035,
|
| 181 |
+
"grad_norm": 10.604959487915039,
|
| 182 |
+
"learning_rate": 1e-05,
|
| 183 |
+
"loss": 0.4519,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.19563581640331076,
|
| 188 |
+
"grad_norm": 10.842782974243164,
|
| 189 |
+
"learning_rate": 1e-05,
|
| 190 |
+
"loss": 0.4617,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.20316027088036118,
|
| 195 |
+
"grad_norm": 10.904434204101562,
|
| 196 |
+
"learning_rate": 1e-05,
|
| 197 |
+
"loss": 0.4858,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.2106847253574116,
|
| 202 |
+
"grad_norm": 9.698153495788574,
|
| 203 |
+
"learning_rate": 1e-05,
|
| 204 |
+
"loss": 0.4126,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.218209179834462,
|
| 209 |
+
"grad_norm": 12.699883460998535,
|
| 210 |
+
"learning_rate": 1e-05,
|
| 211 |
+
"loss": 0.4756,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.22573363431151242,
|
| 216 |
+
"grad_norm": 14.62389850616455,
|
| 217 |
+
"learning_rate": 1e-05,
|
| 218 |
+
"loss": 0.3576,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.23325808878856283,
|
| 223 |
+
"grad_norm": 12.436488151550293,
|
| 224 |
+
"learning_rate": 1e-05,
|
| 225 |
+
"loss": 0.3493,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.24078254326561324,
|
| 230 |
+
"grad_norm": 17.488454818725586,
|
| 231 |
+
"learning_rate": 1e-05,
|
| 232 |
+
"loss": 0.4812,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.24830699774266365,
|
| 237 |
+
"grad_norm": 15.149370193481445,
|
| 238 |
+
"learning_rate": 1e-05,
|
| 239 |
+
"loss": 0.4218,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.2558314522197141,
|
| 244 |
+
"grad_norm": 11.76059341430664,
|
| 245 |
+
"learning_rate": 1e-05,
|
| 246 |
+
"loss": 0.3729,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.2633559066967645,
|
| 251 |
+
"grad_norm": 15.72620964050293,
|
| 252 |
+
"learning_rate": 1e-05,
|
| 253 |
+
"loss": 0.2586,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.2708803611738149,
|
| 258 |
+
"grad_norm": 16.726228713989258,
|
| 259 |
+
"learning_rate": 1e-05,
|
| 260 |
+
"loss": 0.4295,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.27840481565086533,
|
| 265 |
+
"grad_norm": 12.156024932861328,
|
| 266 |
+
"learning_rate": 1e-05,
|
| 267 |
+
"loss": 0.3179,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.28592927012791575,
|
| 272 |
+
"grad_norm": 12.32470417022705,
|
| 273 |
+
"learning_rate": 1e-05,
|
| 274 |
+
"loss": 0.3523,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.29345372460496616,
|
| 279 |
+
"grad_norm": 17.34354591369629,
|
| 280 |
+
"learning_rate": 1e-05,
|
| 281 |
+
"loss": 0.3724,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.3009781790820166,
|
| 286 |
+
"grad_norm": 9.95320987701416,
|
| 287 |
+
"learning_rate": 1e-05,
|
| 288 |
+
"loss": 0.4028,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.308502633559067,
|
| 293 |
+
"grad_norm": 10.40683650970459,
|
| 294 |
+
"learning_rate": 1e-05,
|
| 295 |
+
"loss": 0.5469,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.3160270880361174,
|
| 300 |
+
"grad_norm": 12.613582611083984,
|
| 301 |
+
"learning_rate": 1e-05,
|
| 302 |
+
"loss": 0.3672,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.3235515425131678,
|
| 307 |
+
"grad_norm": 13.326891899108887,
|
| 308 |
+
"learning_rate": 1e-05,
|
| 309 |
+
"loss": 0.3514,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.3310759969902182,
|
| 314 |
+
"grad_norm": 8.356232643127441,
|
| 315 |
+
"learning_rate": 1e-05,
|
| 316 |
+
"loss": 0.292,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.33860045146726864,
|
| 321 |
+
"grad_norm": 14.835829734802246,
|
| 322 |
+
"learning_rate": 1e-05,
|
| 323 |
+
"loss": 0.3765,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.34612490594431905,
|
| 328 |
+
"grad_norm": 7.975886821746826,
|
| 329 |
+
"learning_rate": 1e-05,
|
| 330 |
+
"loss": 0.3667,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.35364936042136946,
|
| 335 |
+
"grad_norm": 10.029479026794434,
|
| 336 |
+
"learning_rate": 1e-05,
|
| 337 |
+
"loss": 0.3788,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.3611738148984199,
|
| 342 |
+
"grad_norm": 11.4894437789917,
|
| 343 |
+
"learning_rate": 1e-05,
|
| 344 |
+
"loss": 0.309,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.3686982693754703,
|
| 349 |
+
"grad_norm": 12.190320014953613,
|
| 350 |
+
"learning_rate": 1e-05,
|
| 351 |
+
"loss": 0.3398,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.3762227238525207,
|
| 356 |
+
"grad_norm": 12.104024887084961,
|
| 357 |
+
"learning_rate": 1e-05,
|
| 358 |
+
"loss": 0.3907,
|
| 359 |
+
"step": 500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.3837471783295711,
|
| 363 |
+
"grad_norm": 11.915987014770508,
|
| 364 |
+
"learning_rate": 1e-05,
|
| 365 |
+
"loss": 0.3278,
|
| 366 |
+
"step": 510
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.3912716328066215,
|
| 370 |
+
"grad_norm": 16.552160263061523,
|
| 371 |
+
"learning_rate": 1e-05,
|
| 372 |
+
"loss": 0.3745,
|
| 373 |
+
"step": 520
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.39879608728367194,
|
| 377 |
+
"grad_norm": 11.056331634521484,
|
| 378 |
+
"learning_rate": 1e-05,
|
| 379 |
+
"loss": 0.4694,
|
| 380 |
+
"step": 530
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.40632054176072235,
|
| 384 |
+
"grad_norm": 10.76766586303711,
|
| 385 |
+
"learning_rate": 1e-05,
|
| 386 |
+
"loss": 0.3631,
|
| 387 |
+
"step": 540
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.41384499623777277,
|
| 391 |
+
"grad_norm": 10.77774715423584,
|
| 392 |
+
"learning_rate": 1e-05,
|
| 393 |
+
"loss": 0.3448,
|
| 394 |
+
"step": 550
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.4213694507148232,
|
| 398 |
+
"grad_norm": 11.11598014831543,
|
| 399 |
+
"learning_rate": 1e-05,
|
| 400 |
+
"loss": 0.339,
|
| 401 |
+
"step": 560
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.4288939051918736,
|
| 405 |
+
"grad_norm": 8.696084976196289,
|
| 406 |
+
"learning_rate": 1e-05,
|
| 407 |
+
"loss": 0.4023,
|
| 408 |
+
"step": 570
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.436418359668924,
|
| 412 |
+
"grad_norm": 15.626012802124023,
|
| 413 |
+
"learning_rate": 1e-05,
|
| 414 |
+
"loss": 0.4638,
|
| 415 |
+
"step": 580
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.4439428141459744,
|
| 419 |
+
"grad_norm": 14.812833786010742,
|
| 420 |
+
"learning_rate": 1e-05,
|
| 421 |
+
"loss": 0.468,
|
| 422 |
+
"step": 590
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.45146726862302483,
|
| 426 |
+
"grad_norm": 11.22861385345459,
|
| 427 |
+
"learning_rate": 1e-05,
|
| 428 |
+
"loss": 0.3772,
|
| 429 |
+
"step": 600
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.45899172310007524,
|
| 433 |
+
"grad_norm": 14.62263011932373,
|
| 434 |
+
"learning_rate": 1e-05,
|
| 435 |
+
"loss": 0.3682,
|
| 436 |
+
"step": 610
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.46651617757712566,
|
| 440 |
+
"grad_norm": 10.826017379760742,
|
| 441 |
+
"learning_rate": 1e-05,
|
| 442 |
+
"loss": 0.3671,
|
| 443 |
+
"step": 620
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.47404063205417607,
|
| 447 |
+
"grad_norm": 9.838117599487305,
|
| 448 |
+
"learning_rate": 1e-05,
|
| 449 |
+
"loss": 0.3459,
|
| 450 |
+
"step": 630
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.4815650865312265,
|
| 454 |
+
"grad_norm": 7.919167518615723,
|
| 455 |
+
"learning_rate": 1e-05,
|
| 456 |
+
"loss": 0.2914,
|
| 457 |
+
"step": 640
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.4890895410082769,
|
| 461 |
+
"grad_norm": 4.093368053436279,
|
| 462 |
+
"learning_rate": 1e-05,
|
| 463 |
+
"loss": 0.3329,
|
| 464 |
+
"step": 650
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.4966139954853273,
|
| 468 |
+
"grad_norm": 12.66010856628418,
|
| 469 |
+
"learning_rate": 1e-05,
|
| 470 |
+
"loss": 0.4115,
|
| 471 |
+
"step": 660
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.5041384499623778,
|
| 475 |
+
"grad_norm": 11.424004554748535,
|
| 476 |
+
"learning_rate": 1e-05,
|
| 477 |
+
"loss": 0.4033,
|
| 478 |
+
"step": 670
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.5116629044394282,
|
| 482 |
+
"grad_norm": 9.730168342590332,
|
| 483 |
+
"learning_rate": 1e-05,
|
| 484 |
+
"loss": 0.3893,
|
| 485 |
+
"step": 680
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.5191873589164786,
|
| 489 |
+
"grad_norm": 9.054938316345215,
|
| 490 |
+
"learning_rate": 1e-05,
|
| 491 |
+
"loss": 0.3922,
|
| 492 |
+
"step": 690
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.526711813393529,
|
| 496 |
+
"grad_norm": 10.94675350189209,
|
| 497 |
+
"learning_rate": 1e-05,
|
| 498 |
+
"loss": 0.3291,
|
| 499 |
+
"step": 700
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.5342362678705794,
|
| 503 |
+
"grad_norm": 12.961570739746094,
|
| 504 |
+
"learning_rate": 1e-05,
|
| 505 |
+
"loss": 0.3225,
|
| 506 |
+
"step": 710
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.5417607223476298,
|
| 510 |
+
"grad_norm": 8.719619750976562,
|
| 511 |
+
"learning_rate": 1e-05,
|
| 512 |
+
"loss": 0.3267,
|
| 513 |
+
"step": 720
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.5492851768246803,
|
| 517 |
+
"grad_norm": 10.847646713256836,
|
| 518 |
+
"learning_rate": 1e-05,
|
| 519 |
+
"loss": 0.4268,
|
| 520 |
+
"step": 730
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.5568096313017307,
|
| 524 |
+
"grad_norm": 11.188985824584961,
|
| 525 |
+
"learning_rate": 1e-05,
|
| 526 |
+
"loss": 0.4236,
|
| 527 |
+
"step": 740
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.5643340857787811,
|
| 531 |
+
"grad_norm": 13.59192943572998,
|
| 532 |
+
"learning_rate": 1e-05,
|
| 533 |
+
"loss": 0.4197,
|
| 534 |
+
"step": 750
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.5718585402558315,
|
| 538 |
+
"grad_norm": 10.489006042480469,
|
| 539 |
+
"learning_rate": 1e-05,
|
| 540 |
+
"loss": 0.3314,
|
| 541 |
+
"step": 760
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.5793829947328819,
|
| 545 |
+
"grad_norm": 11.065324783325195,
|
| 546 |
+
"learning_rate": 1e-05,
|
| 547 |
+
"loss": 0.3667,
|
| 548 |
+
"step": 770
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.5869074492099323,
|
| 552 |
+
"grad_norm": 12.28297233581543,
|
| 553 |
+
"learning_rate": 1e-05,
|
| 554 |
+
"loss": 0.3189,
|
| 555 |
+
"step": 780
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.5944319036869827,
|
| 559 |
+
"grad_norm": 9.553642272949219,
|
| 560 |
+
"learning_rate": 1e-05,
|
| 561 |
+
"loss": 0.3991,
|
| 562 |
+
"step": 790
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.6019563581640331,
|
| 566 |
+
"grad_norm": 11.755203247070312,
|
| 567 |
+
"learning_rate": 1e-05,
|
| 568 |
+
"loss": 0.3528,
|
| 569 |
+
"step": 800
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.6094808126410836,
|
| 573 |
+
"grad_norm": 7.8607306480407715,
|
| 574 |
+
"learning_rate": 1e-05,
|
| 575 |
+
"loss": 0.3292,
|
| 576 |
+
"step": 810
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.617005267118134,
|
| 580 |
+
"grad_norm": 10.472386360168457,
|
| 581 |
+
"learning_rate": 1e-05,
|
| 582 |
+
"loss": 0.3549,
|
| 583 |
+
"step": 820
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.6245297215951844,
|
| 587 |
+
"grad_norm": 9.280732154846191,
|
| 588 |
+
"learning_rate": 1e-05,
|
| 589 |
+
"loss": 0.3183,
|
| 590 |
+
"step": 830
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.6320541760722348,
|
| 594 |
+
"grad_norm": 9.160599708557129,
|
| 595 |
+
"learning_rate": 1e-05,
|
| 596 |
+
"loss": 0.3686,
|
| 597 |
+
"step": 840
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.6395786305492852,
|
| 601 |
+
"grad_norm": 10.545658111572266,
|
| 602 |
+
"learning_rate": 1e-05,
|
| 603 |
+
"loss": 0.3517,
|
| 604 |
+
"step": 850
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.6471030850263356,
|
| 608 |
+
"grad_norm": 11.327434539794922,
|
| 609 |
+
"learning_rate": 1e-05,
|
| 610 |
+
"loss": 0.3465,
|
| 611 |
+
"step": 860
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.654627539503386,
|
| 615 |
+
"grad_norm": 12.003908157348633,
|
| 616 |
+
"learning_rate": 1e-05,
|
| 617 |
+
"loss": 0.2845,
|
| 618 |
+
"step": 870
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.6621519939804364,
|
| 622 |
+
"grad_norm": 9.960043907165527,
|
| 623 |
+
"learning_rate": 1e-05,
|
| 624 |
+
"loss": 0.3255,
|
| 625 |
+
"step": 880
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.6696764484574869,
|
| 629 |
+
"grad_norm": 11.36705207824707,
|
| 630 |
+
"learning_rate": 1e-05,
|
| 631 |
+
"loss": 0.3724,
|
| 632 |
+
"step": 890
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.6772009029345373,
|
| 636 |
+
"grad_norm": 9.673847198486328,
|
| 637 |
+
"learning_rate": 1e-05,
|
| 638 |
+
"loss": 0.3524,
|
| 639 |
+
"step": 900
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.6847253574115877,
|
| 643 |
+
"grad_norm": 10.644118309020996,
|
| 644 |
+
"learning_rate": 1e-05,
|
| 645 |
+
"loss": 0.3988,
|
| 646 |
+
"step": 910
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.6922498118886381,
|
| 650 |
+
"grad_norm": 11.484865188598633,
|
| 651 |
+
"learning_rate": 1e-05,
|
| 652 |
+
"loss": 0.3601,
|
| 653 |
+
"step": 920
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.6997742663656885,
|
| 657 |
+
"grad_norm": 7.940932750701904,
|
| 658 |
+
"learning_rate": 1e-05,
|
| 659 |
+
"loss": 0.2722,
|
| 660 |
+
"step": 930
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.7072987208427389,
|
| 664 |
+
"grad_norm": 9.51900863647461,
|
| 665 |
+
"learning_rate": 1e-05,
|
| 666 |
+
"loss": 0.2814,
|
| 667 |
+
"step": 940
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.7148231753197893,
|
| 671 |
+
"grad_norm": 14.423086166381836,
|
| 672 |
+
"learning_rate": 1e-05,
|
| 673 |
+
"loss": 0.4054,
|
| 674 |
+
"step": 950
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.7223476297968398,
|
| 678 |
+
"grad_norm": 12.655383110046387,
|
| 679 |
+
"learning_rate": 1e-05,
|
| 680 |
+
"loss": 0.3125,
|
| 681 |
+
"step": 960
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.7298720842738902,
|
| 685 |
+
"grad_norm": 13.050726890563965,
|
| 686 |
+
"learning_rate": 1e-05,
|
| 687 |
+
"loss": 0.3368,
|
| 688 |
+
"step": 970
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.7373965387509406,
|
| 692 |
+
"grad_norm": 8.44699764251709,
|
| 693 |
+
"learning_rate": 1e-05,
|
| 694 |
+
"loss": 0.3162,
|
| 695 |
+
"step": 980
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.744920993227991,
|
| 699 |
+
"grad_norm": 9.112492561340332,
|
| 700 |
+
"learning_rate": 1e-05,
|
| 701 |
+
"loss": 0.3428,
|
| 702 |
+
"step": 990
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.7524454477050414,
|
| 706 |
+
"grad_norm": 7.576210975646973,
|
| 707 |
+
"learning_rate": 1e-05,
|
| 708 |
+
"loss": 0.2674,
|
| 709 |
+
"step": 1000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.7599699021820918,
|
| 713 |
+
"grad_norm": 10.723271369934082,
|
| 714 |
+
"learning_rate": 1e-05,
|
| 715 |
+
"loss": 0.313,
|
| 716 |
+
"step": 1010
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.7674943566591422,
|
| 720 |
+
"grad_norm": 11.943977355957031,
|
| 721 |
+
"learning_rate": 1e-05,
|
| 722 |
+
"loss": 0.321,
|
| 723 |
+
"step": 1020
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.7750188111361926,
|
| 727 |
+
"grad_norm": 9.523961067199707,
|
| 728 |
+
"learning_rate": 1e-05,
|
| 729 |
+
"loss": 0.3475,
|
| 730 |
+
"step": 1030
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.782543265613243,
|
| 734 |
+
"grad_norm": 10.895538330078125,
|
| 735 |
+
"learning_rate": 1e-05,
|
| 736 |
+
"loss": 0.3763,
|
| 737 |
+
"step": 1040
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.7900677200902935,
|
| 741 |
+
"grad_norm": 6.124391078948975,
|
| 742 |
+
"learning_rate": 1e-05,
|
| 743 |
+
"loss": 0.251,
|
| 744 |
+
"step": 1050
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.7975921745673439,
|
| 748 |
+
"grad_norm": 9.036330223083496,
|
| 749 |
+
"learning_rate": 1e-05,
|
| 750 |
+
"loss": 0.3976,
|
| 751 |
+
"step": 1060
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.8051166290443943,
|
| 755 |
+
"grad_norm": 9.179152488708496,
|
| 756 |
+
"learning_rate": 1e-05,
|
| 757 |
+
"loss": 0.3052,
|
| 758 |
+
"step": 1070
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.8126410835214447,
|
| 762 |
+
"grad_norm": 6.6380157470703125,
|
| 763 |
+
"learning_rate": 1e-05,
|
| 764 |
+
"loss": 0.3397,
|
| 765 |
+
"step": 1080
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.8201655379984951,
|
| 769 |
+
"grad_norm": 9.615362167358398,
|
| 770 |
+
"learning_rate": 1e-05,
|
| 771 |
+
"loss": 0.3091,
|
| 772 |
+
"step": 1090
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.8276899924755455,
|
| 776 |
+
"grad_norm": 11.812026977539062,
|
| 777 |
+
"learning_rate": 1e-05,
|
| 778 |
+
"loss": 0.3156,
|
| 779 |
+
"step": 1100
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.835214446952596,
|
| 783 |
+
"grad_norm": 12.034916877746582,
|
| 784 |
+
"learning_rate": 1e-05,
|
| 785 |
+
"loss": 0.2479,
|
| 786 |
+
"step": 1110
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.8427389014296464,
|
| 790 |
+
"grad_norm": 13.908053398132324,
|
| 791 |
+
"learning_rate": 1e-05,
|
| 792 |
+
"loss": 0.256,
|
| 793 |
+
"step": 1120
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.8502633559066968,
|
| 797 |
+
"grad_norm": 11.581817626953125,
|
| 798 |
+
"learning_rate": 1e-05,
|
| 799 |
+
"loss": 0.3175,
|
| 800 |
+
"step": 1130
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.8577878103837472,
|
| 804 |
+
"grad_norm": 11.642997741699219,
|
| 805 |
+
"learning_rate": 1e-05,
|
| 806 |
+
"loss": 0.25,
|
| 807 |
+
"step": 1140
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.8653122648607976,
|
| 811 |
+
"grad_norm": 10.813202857971191,
|
| 812 |
+
"learning_rate": 1e-05,
|
| 813 |
+
"loss": 0.3413,
|
| 814 |
+
"step": 1150
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.872836719337848,
|
| 818 |
+
"grad_norm": 12.767478942871094,
|
| 819 |
+
"learning_rate": 1e-05,
|
| 820 |
+
"loss": 0.3094,
|
| 821 |
+
"step": 1160
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.8803611738148984,
|
| 825 |
+
"grad_norm": 9.270513534545898,
|
| 826 |
+
"learning_rate": 1e-05,
|
| 827 |
+
"loss": 0.2807,
|
| 828 |
+
"step": 1170
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.8878856282919488,
|
| 832 |
+
"grad_norm": 7.739561557769775,
|
| 833 |
+
"learning_rate": 1e-05,
|
| 834 |
+
"loss": 0.2789,
|
| 835 |
+
"step": 1180
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.8954100827689992,
|
| 839 |
+
"grad_norm": 7.118448734283447,
|
| 840 |
+
"learning_rate": 1e-05,
|
| 841 |
+
"loss": 0.2752,
|
| 842 |
+
"step": 1190
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.9029345372460497,
|
| 846 |
+
"grad_norm": 11.153009414672852,
|
| 847 |
+
"learning_rate": 1e-05,
|
| 848 |
+
"loss": 0.32,
|
| 849 |
+
"step": 1200
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.9104589917231001,
|
| 853 |
+
"grad_norm": 10.667895317077637,
|
| 854 |
+
"learning_rate": 1e-05,
|
| 855 |
+
"loss": 0.2285,
|
| 856 |
+
"step": 1210
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.9179834462001505,
|
| 860 |
+
"grad_norm": 12.081469535827637,
|
| 861 |
+
"learning_rate": 1e-05,
|
| 862 |
+
"loss": 0.3278,
|
| 863 |
+
"step": 1220
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.9255079006772009,
|
| 867 |
+
"grad_norm": 8.599586486816406,
|
| 868 |
+
"learning_rate": 1e-05,
|
| 869 |
+
"loss": 0.3049,
|
| 870 |
+
"step": 1230
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.9330323551542513,
|
| 874 |
+
"grad_norm": 10.062015533447266,
|
| 875 |
+
"learning_rate": 1e-05,
|
| 876 |
+
"loss": 0.3583,
|
| 877 |
+
"step": 1240
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.9405568096313017,
|
| 881 |
+
"grad_norm": 8.24731159210205,
|
| 882 |
+
"learning_rate": 1e-05,
|
| 883 |
+
"loss": 0.3162,
|
| 884 |
+
"step": 1250
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.9480812641083521,
|
| 888 |
+
"grad_norm": 8.026958465576172,
|
| 889 |
+
"learning_rate": 1e-05,
|
| 890 |
+
"loss": 0.3082,
|
| 891 |
+
"step": 1260
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.9556057185854026,
|
| 895 |
+
"grad_norm": 8.996962547302246,
|
| 896 |
+
"learning_rate": 1e-05,
|
| 897 |
+
"loss": 0.3969,
|
| 898 |
+
"step": 1270
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.963130173062453,
|
| 902 |
+
"grad_norm": 7.346575736999512,
|
| 903 |
+
"learning_rate": 1e-05,
|
| 904 |
+
"loss": 0.2905,
|
| 905 |
+
"step": 1280
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.9706546275395034,
|
| 909 |
+
"grad_norm": 7.704085826873779,
|
| 910 |
+
"learning_rate": 1e-05,
|
| 911 |
+
"loss": 0.2655,
|
| 912 |
+
"step": 1290
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.9781790820165538,
|
| 916 |
+
"grad_norm": 9.016671180725098,
|
| 917 |
+
"learning_rate": 1e-05,
|
| 918 |
+
"loss": 0.3071,
|
| 919 |
+
"step": 1300
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.9857035364936042,
|
| 923 |
+
"grad_norm": 10.028645515441895,
|
| 924 |
+
"learning_rate": 1e-05,
|
| 925 |
+
"loss": 0.3077,
|
| 926 |
+
"step": 1310
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.9932279909706546,
|
| 930 |
+
"grad_norm": 9.28890609741211,
|
| 931 |
+
"learning_rate": 1e-05,
|
| 932 |
+
"loss": 0.3317,
|
| 933 |
+
"step": 1320
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 1.000752445447705,
|
| 937 |
+
"grad_norm": 9.905320167541504,
|
| 938 |
+
"learning_rate": 1e-05,
|
| 939 |
+
"loss": 0.31,
|
| 940 |
+
"step": 1330
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 1.0082768999247556,
|
| 944 |
+
"grad_norm": 8.012602806091309,
|
| 945 |
+
"learning_rate": 1e-05,
|
| 946 |
+
"loss": 0.2343,
|
| 947 |
+
"step": 1340
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 1.0158013544018059,
|
| 951 |
+
"grad_norm": 6.548900127410889,
|
| 952 |
+
"learning_rate": 1e-05,
|
| 953 |
+
"loss": 0.2686,
|
| 954 |
+
"step": 1350
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 1.0233258088788564,
|
| 958 |
+
"grad_norm": 9.645492553710938,
|
| 959 |
+
"learning_rate": 1e-05,
|
| 960 |
+
"loss": 0.3046,
|
| 961 |
+
"step": 1360
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 1.0308502633559067,
|
| 965 |
+
"grad_norm": 10.740938186645508,
|
| 966 |
+
"learning_rate": 1e-05,
|
| 967 |
+
"loss": 0.3873,
|
| 968 |
+
"step": 1370
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 1.0383747178329572,
|
| 972 |
+
"grad_norm": 11.067444801330566,
|
| 973 |
+
"learning_rate": 1e-05,
|
| 974 |
+
"loss": 0.3229,
|
| 975 |
+
"step": 1380
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 1.0458991723100075,
|
| 979 |
+
"grad_norm": 7.437419891357422,
|
| 980 |
+
"learning_rate": 1e-05,
|
| 981 |
+
"loss": 0.2907,
|
| 982 |
+
"step": 1390
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 1.053423626787058,
|
| 986 |
+
"grad_norm": 5.831209659576416,
|
| 987 |
+
"learning_rate": 1e-05,
|
| 988 |
+
"loss": 0.3107,
|
| 989 |
+
"step": 1400
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 1.0609480812641083,
|
| 993 |
+
"grad_norm": 8.783834457397461,
|
| 994 |
+
"learning_rate": 1e-05,
|
| 995 |
+
"loss": 0.3269,
|
| 996 |
+
"step": 1410
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 1.0684725357411589,
|
| 1000 |
+
"grad_norm": 12.042133331298828,
|
| 1001 |
+
"learning_rate": 1e-05,
|
| 1002 |
+
"loss": 0.2861,
|
| 1003 |
+
"step": 1420
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 1.0759969902182092,
|
| 1007 |
+
"grad_norm": 10.743906021118164,
|
| 1008 |
+
"learning_rate": 1e-05,
|
| 1009 |
+
"loss": 0.3,
|
| 1010 |
+
"step": 1430
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 1.0835214446952597,
|
| 1014 |
+
"grad_norm": 10.540002822875977,
|
| 1015 |
+
"learning_rate": 1e-05,
|
| 1016 |
+
"loss": 0.2708,
|
| 1017 |
+
"step": 1440
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 1.09104589917231,
|
| 1021 |
+
"grad_norm": 7.265504837036133,
|
| 1022 |
+
"learning_rate": 1e-05,
|
| 1023 |
+
"loss": 0.2723,
|
| 1024 |
+
"step": 1450
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 1.0985703536493605,
|
| 1028 |
+
"grad_norm": 5.650593280792236,
|
| 1029 |
+
"learning_rate": 1e-05,
|
| 1030 |
+
"loss": 0.3101,
|
| 1031 |
+
"step": 1460
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 1.1060948081264108,
|
| 1035 |
+
"grad_norm": 10.168730735778809,
|
| 1036 |
+
"learning_rate": 1e-05,
|
| 1037 |
+
"loss": 0.291,
|
| 1038 |
+
"step": 1470
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 1.1136192626034613,
|
| 1042 |
+
"grad_norm": 6.533019542694092,
|
| 1043 |
+
"learning_rate": 1e-05,
|
| 1044 |
+
"loss": 0.2925,
|
| 1045 |
+
"step": 1480
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 1.1211437170805116,
|
| 1049 |
+
"grad_norm": 9.97232437133789,
|
| 1050 |
+
"learning_rate": 1e-05,
|
| 1051 |
+
"loss": 0.2983,
|
| 1052 |
+
"step": 1490
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 1.1286681715575622,
|
| 1056 |
+
"grad_norm": 8.263399124145508,
|
| 1057 |
+
"learning_rate": 1e-05,
|
| 1058 |
+
"loss": 0.2818,
|
| 1059 |
+
"step": 1500
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 1.1361926260346125,
|
| 1063 |
+
"grad_norm": 8.396636962890625,
|
| 1064 |
+
"learning_rate": 1e-05,
|
| 1065 |
+
"loss": 0.3436,
|
| 1066 |
+
"step": 1510
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 1.143717080511663,
|
| 1070 |
+
"grad_norm": 13.860685348510742,
|
| 1071 |
+
"learning_rate": 1e-05,
|
| 1072 |
+
"loss": 0.4241,
|
| 1073 |
+
"step": 1520
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 1.1512415349887133,
|
| 1077 |
+
"grad_norm": 8.995695114135742,
|
| 1078 |
+
"learning_rate": 1e-05,
|
| 1079 |
+
"loss": 0.2742,
|
| 1080 |
+
"step": 1530
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 1.1587659894657638,
|
| 1084 |
+
"grad_norm": 12.496316909790039,
|
| 1085 |
+
"learning_rate": 1e-05,
|
| 1086 |
+
"loss": 0.3052,
|
| 1087 |
+
"step": 1540
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 1.1662904439428141,
|
| 1091 |
+
"grad_norm": 7.071567535400391,
|
| 1092 |
+
"learning_rate": 1e-05,
|
| 1093 |
+
"loss": 0.3227,
|
| 1094 |
+
"step": 1550
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 1.1738148984198646,
|
| 1098 |
+
"grad_norm": 9.216208457946777,
|
| 1099 |
+
"learning_rate": 1e-05,
|
| 1100 |
+
"loss": 0.2476,
|
| 1101 |
+
"step": 1560
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 1.181339352896915,
|
| 1105 |
+
"grad_norm": 7.963762283325195,
|
| 1106 |
+
"learning_rate": 1e-05,
|
| 1107 |
+
"loss": 0.2883,
|
| 1108 |
+
"step": 1570
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 1.1888638073739655,
|
| 1112 |
+
"grad_norm": 11.962204933166504,
|
| 1113 |
+
"learning_rate": 1e-05,
|
| 1114 |
+
"loss": 0.2792,
|
| 1115 |
+
"step": 1580
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 1.1963882618510158,
|
| 1119 |
+
"grad_norm": 11.451403617858887,
|
| 1120 |
+
"learning_rate": 1e-05,
|
| 1121 |
+
"loss": 0.3362,
|
| 1122 |
+
"step": 1590
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 1.2039127163280663,
|
| 1126 |
+
"grad_norm": 11.551766395568848,
|
| 1127 |
+
"learning_rate": 1e-05,
|
| 1128 |
+
"loss": 0.2431,
|
| 1129 |
+
"step": 1600
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 1.2114371708051166,
|
| 1133 |
+
"grad_norm": 5.913654327392578,
|
| 1134 |
+
"learning_rate": 1e-05,
|
| 1135 |
+
"loss": 0.277,
|
| 1136 |
+
"step": 1610
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 1.2189616252821671,
|
| 1140 |
+
"grad_norm": 13.468070030212402,
|
| 1141 |
+
"learning_rate": 1e-05,
|
| 1142 |
+
"loss": 0.3292,
|
| 1143 |
+
"step": 1620
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 1.2264860797592174,
|
| 1147 |
+
"grad_norm": 8.323406219482422,
|
| 1148 |
+
"learning_rate": 1e-05,
|
| 1149 |
+
"loss": 0.3182,
|
| 1150 |
+
"step": 1630
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 1.234010534236268,
|
| 1154 |
+
"grad_norm": 9.116568565368652,
|
| 1155 |
+
"learning_rate": 1e-05,
|
| 1156 |
+
"loss": 0.2328,
|
| 1157 |
+
"step": 1640
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 1.2415349887133182,
|
| 1161 |
+
"grad_norm": 8.88713264465332,
|
| 1162 |
+
"learning_rate": 1e-05,
|
| 1163 |
+
"loss": 0.2538,
|
| 1164 |
+
"step": 1650
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 1.2490594431903688,
|
| 1168 |
+
"grad_norm": 10.781469345092773,
|
| 1169 |
+
"learning_rate": 1e-05,
|
| 1170 |
+
"loss": 0.2389,
|
| 1171 |
+
"step": 1660
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 1.256583897667419,
|
| 1175 |
+
"grad_norm": 12.144160270690918,
|
| 1176 |
+
"learning_rate": 1e-05,
|
| 1177 |
+
"loss": 0.3006,
|
| 1178 |
+
"step": 1670
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 1.2641083521444696,
|
| 1182 |
+
"grad_norm": 7.866734027862549,
|
| 1183 |
+
"learning_rate": 1e-05,
|
| 1184 |
+
"loss": 0.345,
|
| 1185 |
+
"step": 1680
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 1.27163280662152,
|
| 1189 |
+
"grad_norm": 7.459820747375488,
|
| 1190 |
+
"learning_rate": 1e-05,
|
| 1191 |
+
"loss": 0.2482,
|
| 1192 |
+
"step": 1690
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 1.2791572610985704,
|
| 1196 |
+
"grad_norm": 7.7605109214782715,
|
| 1197 |
+
"learning_rate": 1e-05,
|
| 1198 |
+
"loss": 0.214,
|
| 1199 |
+
"step": 1700
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 1.2866817155756207,
|
| 1203 |
+
"grad_norm": 9.145365715026855,
|
| 1204 |
+
"learning_rate": 1e-05,
|
| 1205 |
+
"loss": 0.3057,
|
| 1206 |
+
"step": 1710
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 1.2942061700526712,
|
| 1210 |
+
"grad_norm": 10.521879196166992,
|
| 1211 |
+
"learning_rate": 1e-05,
|
| 1212 |
+
"loss": 0.3376,
|
| 1213 |
+
"step": 1720
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 1.3017306245297215,
|
| 1217 |
+
"grad_norm": 5.132536888122559,
|
| 1218 |
+
"learning_rate": 1e-05,
|
| 1219 |
+
"loss": 0.2723,
|
| 1220 |
+
"step": 1730
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 1.309255079006772,
|
| 1224 |
+
"grad_norm": 10.422348022460938,
|
| 1225 |
+
"learning_rate": 1e-05,
|
| 1226 |
+
"loss": 0.2769,
|
| 1227 |
+
"step": 1740
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 1.3167795334838224,
|
| 1231 |
+
"grad_norm": 9.99517822265625,
|
| 1232 |
+
"learning_rate": 1e-05,
|
| 1233 |
+
"loss": 0.3455,
|
| 1234 |
+
"step": 1750
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 1.324303987960873,
|
| 1238 |
+
"grad_norm": 6.903396129608154,
|
| 1239 |
+
"learning_rate": 1e-05,
|
| 1240 |
+
"loss": 0.2813,
|
| 1241 |
+
"step": 1760
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 1.3318284424379232,
|
| 1245 |
+
"grad_norm": 5.721127986907959,
|
| 1246 |
+
"learning_rate": 1e-05,
|
| 1247 |
+
"loss": 0.2824,
|
| 1248 |
+
"step": 1770
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 1.3393528969149737,
|
| 1252 |
+
"grad_norm": 9.914773941040039,
|
| 1253 |
+
"learning_rate": 1e-05,
|
| 1254 |
+
"loss": 0.3353,
|
| 1255 |
+
"step": 1780
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 1.346877351392024,
|
| 1259 |
+
"grad_norm": 7.985681056976318,
|
| 1260 |
+
"learning_rate": 1e-05,
|
| 1261 |
+
"loss": 0.3295,
|
| 1262 |
+
"step": 1790
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 1.3544018058690745,
|
| 1266 |
+
"grad_norm": 10.242146492004395,
|
| 1267 |
+
"learning_rate": 1e-05,
|
| 1268 |
+
"loss": 0.3519,
|
| 1269 |
+
"step": 1800
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 1.3619262603461249,
|
| 1273 |
+
"grad_norm": 8.590888977050781,
|
| 1274 |
+
"learning_rate": 1e-05,
|
| 1275 |
+
"loss": 0.3171,
|
| 1276 |
+
"step": 1810
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 1.3694507148231754,
|
| 1280 |
+
"grad_norm": 9.34271240234375,
|
| 1281 |
+
"learning_rate": 1e-05,
|
| 1282 |
+
"loss": 0.3385,
|
| 1283 |
+
"step": 1820
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 1.3769751693002257,
|
| 1287 |
+
"grad_norm": 8.391048431396484,
|
| 1288 |
+
"learning_rate": 1e-05,
|
| 1289 |
+
"loss": 0.2975,
|
| 1290 |
+
"step": 1830
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 1.3844996237772762,
|
| 1294 |
+
"grad_norm": 11.190972328186035,
|
| 1295 |
+
"learning_rate": 1e-05,
|
| 1296 |
+
"loss": 0.3412,
|
| 1297 |
+
"step": 1840
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 1.3920240782543265,
|
| 1301 |
+
"grad_norm": 8.990412712097168,
|
| 1302 |
+
"learning_rate": 1e-05,
|
| 1303 |
+
"loss": 0.2985,
|
| 1304 |
+
"step": 1850
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 1.399548532731377,
|
| 1308 |
+
"grad_norm": 6.625011920928955,
|
| 1309 |
+
"learning_rate": 1e-05,
|
| 1310 |
+
"loss": 0.2485,
|
| 1311 |
+
"step": 1860
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 1.4070729872084273,
|
| 1315 |
+
"grad_norm": 8.91481876373291,
|
| 1316 |
+
"learning_rate": 1e-05,
|
| 1317 |
+
"loss": 0.3527,
|
| 1318 |
+
"step": 1870
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 1.4145974416854779,
|
| 1322 |
+
"grad_norm": 7.358391761779785,
|
| 1323 |
+
"learning_rate": 1e-05,
|
| 1324 |
+
"loss": 0.3243,
|
| 1325 |
+
"step": 1880
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 1.4221218961625282,
|
| 1329 |
+
"grad_norm": 6.641557216644287,
|
| 1330 |
+
"learning_rate": 1e-05,
|
| 1331 |
+
"loss": 0.242,
|
| 1332 |
+
"step": 1890
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 1.4296463506395787,
|
| 1336 |
+
"grad_norm": 8.88590145111084,
|
| 1337 |
+
"learning_rate": 1e-05,
|
| 1338 |
+
"loss": 0.2933,
|
| 1339 |
+
"step": 1900
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 1.437170805116629,
|
| 1343 |
+
"grad_norm": 9.170287132263184,
|
| 1344 |
+
"learning_rate": 1e-05,
|
| 1345 |
+
"loss": 0.2567,
|
| 1346 |
+
"step": 1910
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 1.4446952595936795,
|
| 1350 |
+
"grad_norm": 9.2968168258667,
|
| 1351 |
+
"learning_rate": 1e-05,
|
| 1352 |
+
"loss": 0.2649,
|
| 1353 |
+
"step": 1920
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 1.4522197140707298,
|
| 1357 |
+
"grad_norm": 8.246125221252441,
|
| 1358 |
+
"learning_rate": 1e-05,
|
| 1359 |
+
"loss": 0.2668,
|
| 1360 |
+
"step": 1930
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 1.4597441685477803,
|
| 1364 |
+
"grad_norm": 10.747627258300781,
|
| 1365 |
+
"learning_rate": 1e-05,
|
| 1366 |
+
"loss": 0.2905,
|
| 1367 |
+
"step": 1940
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 1.4672686230248306,
|
| 1371 |
+
"grad_norm": 7.501006603240967,
|
| 1372 |
+
"learning_rate": 1e-05,
|
| 1373 |
+
"loss": 0.3359,
|
| 1374 |
+
"step": 1950
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 1.4747930775018812,
|
| 1378 |
+
"grad_norm": 9.075075149536133,
|
| 1379 |
+
"learning_rate": 1e-05,
|
| 1380 |
+
"loss": 0.3039,
|
| 1381 |
+
"step": 1960
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 1.4823175319789315,
|
| 1385 |
+
"grad_norm": 9.139381408691406,
|
| 1386 |
+
"learning_rate": 1e-05,
|
| 1387 |
+
"loss": 0.3024,
|
| 1388 |
+
"step": 1970
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 1.489841986455982,
|
| 1392 |
+
"grad_norm": 10.483285903930664,
|
| 1393 |
+
"learning_rate": 1e-05,
|
| 1394 |
+
"loss": 0.2997,
|
| 1395 |
+
"step": 1980
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 1.4973664409330323,
|
| 1399 |
+
"grad_norm": 10.016528129577637,
|
| 1400 |
+
"learning_rate": 1e-05,
|
| 1401 |
+
"loss": 0.3105,
|
| 1402 |
+
"step": 1990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 1.5048908954100828,
|
| 1406 |
+
"grad_norm": 7.966375827789307,
|
| 1407 |
+
"learning_rate": 1e-05,
|
| 1408 |
+
"loss": 0.3121,
|
| 1409 |
+
"step": 2000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 1.5124153498871333,
|
| 1413 |
+
"grad_norm": 6.316532135009766,
|
| 1414 |
+
"learning_rate": 1e-05,
|
| 1415 |
+
"loss": 0.3148,
|
| 1416 |
+
"step": 2010
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 1.5199398043641836,
|
| 1420 |
+
"grad_norm": 9.16601276397705,
|
| 1421 |
+
"learning_rate": 1e-05,
|
| 1422 |
+
"loss": 0.2695,
|
| 1423 |
+
"step": 2020
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 1.527464258841234,
|
| 1427 |
+
"grad_norm": 4.997910499572754,
|
| 1428 |
+
"learning_rate": 1e-05,
|
| 1429 |
+
"loss": 0.2746,
|
| 1430 |
+
"step": 2030
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 1.5349887133182845,
|
| 1434 |
+
"grad_norm": 9.328558921813965,
|
| 1435 |
+
"learning_rate": 1e-05,
|
| 1436 |
+
"loss": 0.2873,
|
| 1437 |
+
"step": 2040
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 1.542513167795335,
|
| 1441 |
+
"grad_norm": 7.824413299560547,
|
| 1442 |
+
"learning_rate": 1e-05,
|
| 1443 |
+
"loss": 0.2925,
|
| 1444 |
+
"step": 2050
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 1.5500376222723853,
|
| 1448 |
+
"grad_norm": 8.306281089782715,
|
| 1449 |
+
"learning_rate": 1e-05,
|
| 1450 |
+
"loss": 0.3059,
|
| 1451 |
+
"step": 2060
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 1.5575620767494356,
|
| 1455 |
+
"grad_norm": 11.394743919372559,
|
| 1456 |
+
"learning_rate": 1e-05,
|
| 1457 |
+
"loss": 0.2995,
|
| 1458 |
+
"step": 2070
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 1.565086531226486,
|
| 1462 |
+
"grad_norm": 8.414088249206543,
|
| 1463 |
+
"learning_rate": 1e-05,
|
| 1464 |
+
"loss": 0.2914,
|
| 1465 |
+
"step": 2080
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 1.5726109857035366,
|
| 1469 |
+
"grad_norm": 10.023848533630371,
|
| 1470 |
+
"learning_rate": 1e-05,
|
| 1471 |
+
"loss": 0.239,
|
| 1472 |
+
"step": 2090
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 1.580135440180587,
|
| 1476 |
+
"grad_norm": 8.992568016052246,
|
| 1477 |
+
"learning_rate": 1e-05,
|
| 1478 |
+
"loss": 0.2613,
|
| 1479 |
+
"step": 2100
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 1.5876598946576372,
|
| 1483 |
+
"grad_norm": 9.712190628051758,
|
| 1484 |
+
"learning_rate": 1e-05,
|
| 1485 |
+
"loss": 0.2558,
|
| 1486 |
+
"step": 2110
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 1.5951843491346878,
|
| 1490 |
+
"grad_norm": 7.813495635986328,
|
| 1491 |
+
"learning_rate": 1e-05,
|
| 1492 |
+
"loss": 0.218,
|
| 1493 |
+
"step": 2120
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 1.6027088036117383,
|
| 1497 |
+
"grad_norm": 6.5997443199157715,
|
| 1498 |
+
"learning_rate": 1e-05,
|
| 1499 |
+
"loss": 0.285,
|
| 1500 |
+
"step": 2130
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 1.6102332580887886,
|
| 1504 |
+
"grad_norm": 11.630515098571777,
|
| 1505 |
+
"learning_rate": 1e-05,
|
| 1506 |
+
"loss": 0.2677,
|
| 1507 |
+
"step": 2140
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 1.617757712565839,
|
| 1511 |
+
"grad_norm": 7.1673359870910645,
|
| 1512 |
+
"learning_rate": 1e-05,
|
| 1513 |
+
"loss": 0.2845,
|
| 1514 |
+
"step": 2150
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 1.6252821670428894,
|
| 1518 |
+
"grad_norm": 10.230573654174805,
|
| 1519 |
+
"learning_rate": 1e-05,
|
| 1520 |
+
"loss": 0.2695,
|
| 1521 |
+
"step": 2160
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 1.63280662151994,
|
| 1525 |
+
"grad_norm": 7.908997535705566,
|
| 1526 |
+
"learning_rate": 1e-05,
|
| 1527 |
+
"loss": 0.2732,
|
| 1528 |
+
"step": 2170
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 1.6403310759969902,
|
| 1532 |
+
"grad_norm": 6.134283542633057,
|
| 1533 |
+
"learning_rate": 1e-05,
|
| 1534 |
+
"loss": 0.2613,
|
| 1535 |
+
"step": 2180
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 1.6478555304740405,
|
| 1539 |
+
"grad_norm": 9.178680419921875,
|
| 1540 |
+
"learning_rate": 1e-05,
|
| 1541 |
+
"loss": 0.3006,
|
| 1542 |
+
"step": 2190
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 1.655379984951091,
|
| 1546 |
+
"grad_norm": 8.385205268859863,
|
| 1547 |
+
"learning_rate": 1e-05,
|
| 1548 |
+
"loss": 0.2391,
|
| 1549 |
+
"step": 2200
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 1.6629044394281416,
|
| 1553 |
+
"grad_norm": 8.024893760681152,
|
| 1554 |
+
"learning_rate": 1e-05,
|
| 1555 |
+
"loss": 0.3149,
|
| 1556 |
+
"step": 2210
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 1.670428893905192,
|
| 1560 |
+
"grad_norm": 7.160342216491699,
|
| 1561 |
+
"learning_rate": 1e-05,
|
| 1562 |
+
"loss": 0.2321,
|
| 1563 |
+
"step": 2220
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 1.6779533483822422,
|
| 1567 |
+
"grad_norm": 9.467365264892578,
|
| 1568 |
+
"learning_rate": 1e-05,
|
| 1569 |
+
"loss": 0.3183,
|
| 1570 |
+
"step": 2230
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 1.6854778028592927,
|
| 1574 |
+
"grad_norm": 8.656280517578125,
|
| 1575 |
+
"learning_rate": 1e-05,
|
| 1576 |
+
"loss": 0.3128,
|
| 1577 |
+
"step": 2240
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 1.6930022573363432,
|
| 1581 |
+
"grad_norm": 5.692852973937988,
|
| 1582 |
+
"learning_rate": 1e-05,
|
| 1583 |
+
"loss": 0.3139,
|
| 1584 |
+
"step": 2250
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 1.7005267118133935,
|
| 1588 |
+
"grad_norm": 7.600724220275879,
|
| 1589 |
+
"learning_rate": 1e-05,
|
| 1590 |
+
"loss": 0.249,
|
| 1591 |
+
"step": 2260
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 1.7080511662904438,
|
| 1595 |
+
"grad_norm": 7.849809646606445,
|
| 1596 |
+
"learning_rate": 1e-05,
|
| 1597 |
+
"loss": 0.2587,
|
| 1598 |
+
"step": 2270
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 1.7155756207674944,
|
| 1602 |
+
"grad_norm": 8.293899536132812,
|
| 1603 |
+
"learning_rate": 1e-05,
|
| 1604 |
+
"loss": 0.2805,
|
| 1605 |
+
"step": 2280
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 1.723100075244545,
|
| 1609 |
+
"grad_norm": 5.557303428649902,
|
| 1610 |
+
"learning_rate": 1e-05,
|
| 1611 |
+
"loss": 0.2789,
|
| 1612 |
+
"step": 2290
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 1.7306245297215952,
|
| 1616 |
+
"grad_norm": 10.24751091003418,
|
| 1617 |
+
"learning_rate": 1e-05,
|
| 1618 |
+
"loss": 0.3244,
|
| 1619 |
+
"step": 2300
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 1.7381489841986455,
|
| 1623 |
+
"grad_norm": 7.778582572937012,
|
| 1624 |
+
"learning_rate": 1e-05,
|
| 1625 |
+
"loss": 0.3005,
|
| 1626 |
+
"step": 2310
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 1.745673438675696,
|
| 1630 |
+
"grad_norm": 7.266477584838867,
|
| 1631 |
+
"learning_rate": 1e-05,
|
| 1632 |
+
"loss": 0.2207,
|
| 1633 |
+
"step": 2320
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 1.7531978931527465,
|
| 1637 |
+
"grad_norm": 10.221525192260742,
|
| 1638 |
+
"learning_rate": 1e-05,
|
| 1639 |
+
"loss": 0.2742,
|
| 1640 |
+
"step": 2330
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 1.7607223476297968,
|
| 1644 |
+
"grad_norm": 8.85750961303711,
|
| 1645 |
+
"learning_rate": 1e-05,
|
| 1646 |
+
"loss": 0.265,
|
| 1647 |
+
"step": 2340
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 1.7682468021068471,
|
| 1651 |
+
"grad_norm": 8.104692459106445,
|
| 1652 |
+
"learning_rate": 1e-05,
|
| 1653 |
+
"loss": 0.277,
|
| 1654 |
+
"step": 2350
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 1.7757712565838977,
|
| 1658 |
+
"grad_norm": 8.834745407104492,
|
| 1659 |
+
"learning_rate": 1e-05,
|
| 1660 |
+
"loss": 0.2315,
|
| 1661 |
+
"step": 2360
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 1.7832957110609482,
|
| 1665 |
+
"grad_norm": 10.258501052856445,
|
| 1666 |
+
"learning_rate": 1e-05,
|
| 1667 |
+
"loss": 0.2921,
|
| 1668 |
+
"step": 2370
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 1.7908201655379985,
|
| 1672 |
+
"grad_norm": 8.523322105407715,
|
| 1673 |
+
"learning_rate": 1e-05,
|
| 1674 |
+
"loss": 0.2043,
|
| 1675 |
+
"step": 2380
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 1.7983446200150488,
|
| 1679 |
+
"grad_norm": 9.75324821472168,
|
| 1680 |
+
"learning_rate": 1e-05,
|
| 1681 |
+
"loss": 0.2187,
|
| 1682 |
+
"step": 2390
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 1.8058690744920993,
|
| 1686 |
+
"grad_norm": 6.369287490844727,
|
| 1687 |
+
"learning_rate": 1e-05,
|
| 1688 |
+
"loss": 0.2365,
|
| 1689 |
+
"step": 2400
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 1.8133935289691498,
|
| 1693 |
+
"grad_norm": 6.650455951690674,
|
| 1694 |
+
"learning_rate": 1e-05,
|
| 1695 |
+
"loss": 0.3027,
|
| 1696 |
+
"step": 2410
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 1.8209179834462002,
|
| 1700 |
+
"grad_norm": 7.7705397605896,
|
| 1701 |
+
"learning_rate": 1e-05,
|
| 1702 |
+
"loss": 0.2942,
|
| 1703 |
+
"step": 2420
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 1.8284424379232505,
|
| 1707 |
+
"grad_norm": 8.137877464294434,
|
| 1708 |
+
"learning_rate": 1e-05,
|
| 1709 |
+
"loss": 0.3317,
|
| 1710 |
+
"step": 2430
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 1.835966892400301,
|
| 1714 |
+
"grad_norm": 8.068604469299316,
|
| 1715 |
+
"learning_rate": 1e-05,
|
| 1716 |
+
"loss": 0.2657,
|
| 1717 |
+
"step": 2440
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 1.8434913468773515,
|
| 1721 |
+
"grad_norm": 5.773308753967285,
|
| 1722 |
+
"learning_rate": 1e-05,
|
| 1723 |
+
"loss": 0.2706,
|
| 1724 |
+
"step": 2450
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 1.8510158013544018,
|
| 1728 |
+
"grad_norm": 6.239875793457031,
|
| 1729 |
+
"learning_rate": 1e-05,
|
| 1730 |
+
"loss": 0.2696,
|
| 1731 |
+
"step": 2460
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 1.858540255831452,
|
| 1735 |
+
"grad_norm": 11.373642921447754,
|
| 1736 |
+
"learning_rate": 1e-05,
|
| 1737 |
+
"loss": 0.2964,
|
| 1738 |
+
"step": 2470
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 1.8660647103085026,
|
| 1742 |
+
"grad_norm": 10.659271240234375,
|
| 1743 |
+
"learning_rate": 1e-05,
|
| 1744 |
+
"loss": 0.2839,
|
| 1745 |
+
"step": 2480
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 1.8735891647855532,
|
| 1749 |
+
"grad_norm": 7.298862934112549,
|
| 1750 |
+
"learning_rate": 1e-05,
|
| 1751 |
+
"loss": 0.2748,
|
| 1752 |
+
"step": 2490
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 1.8811136192626035,
|
| 1756 |
+
"grad_norm": 12.347573280334473,
|
| 1757 |
+
"learning_rate": 1e-05,
|
| 1758 |
+
"loss": 0.2654,
|
| 1759 |
+
"step": 2500
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 1.8886380737396538,
|
| 1763 |
+
"grad_norm": 6.894641399383545,
|
| 1764 |
+
"learning_rate": 1e-05,
|
| 1765 |
+
"loss": 0.2522,
|
| 1766 |
+
"step": 2510
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 1.8961625282167043,
|
| 1770 |
+
"grad_norm": 7.589210510253906,
|
| 1771 |
+
"learning_rate": 1e-05,
|
| 1772 |
+
"loss": 0.3053,
|
| 1773 |
+
"step": 2520
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 1.9036869826937548,
|
| 1777 |
+
"grad_norm": 6.738051891326904,
|
| 1778 |
+
"learning_rate": 1e-05,
|
| 1779 |
+
"loss": 0.224,
|
| 1780 |
+
"step": 2530
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 1.911211437170805,
|
| 1784 |
+
"grad_norm": 7.8747239112854,
|
| 1785 |
+
"learning_rate": 1e-05,
|
| 1786 |
+
"loss": 0.2751,
|
| 1787 |
+
"step": 2540
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 1.9187358916478554,
|
| 1791 |
+
"grad_norm": 6.456340789794922,
|
| 1792 |
+
"learning_rate": 1e-05,
|
| 1793 |
+
"loss": 0.1926,
|
| 1794 |
+
"step": 2550
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 1.926260346124906,
|
| 1798 |
+
"grad_norm": 9.186247825622559,
|
| 1799 |
+
"learning_rate": 1e-05,
|
| 1800 |
+
"loss": 0.3004,
|
| 1801 |
+
"step": 2560
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 1.9337848006019565,
|
| 1805 |
+
"grad_norm": 11.75734806060791,
|
| 1806 |
+
"learning_rate": 1e-05,
|
| 1807 |
+
"loss": 0.2749,
|
| 1808 |
+
"step": 2570
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 1.9413092550790068,
|
| 1812 |
+
"grad_norm": 10.06619644165039,
|
| 1813 |
+
"learning_rate": 1e-05,
|
| 1814 |
+
"loss": 0.3011,
|
| 1815 |
+
"step": 2580
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 1.948833709556057,
|
| 1819 |
+
"grad_norm": 11.47065544128418,
|
| 1820 |
+
"learning_rate": 1e-05,
|
| 1821 |
+
"loss": 0.3124,
|
| 1822 |
+
"step": 2590
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 1.9563581640331076,
|
| 1826 |
+
"grad_norm": 9.72806167602539,
|
| 1827 |
+
"learning_rate": 1e-05,
|
| 1828 |
+
"loss": 0.298,
|
| 1829 |
+
"step": 2600
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 1.963882618510158,
|
| 1833 |
+
"grad_norm": 9.43126106262207,
|
| 1834 |
+
"learning_rate": 1e-05,
|
| 1835 |
+
"loss": 0.2617,
|
| 1836 |
+
"step": 2610
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 1.9714070729872084,
|
| 1840 |
+
"grad_norm": 12.323174476623535,
|
| 1841 |
+
"learning_rate": 1e-05,
|
| 1842 |
+
"loss": 0.3397,
|
| 1843 |
+
"step": 2620
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 1.9789315274642587,
|
| 1847 |
+
"grad_norm": 7.464193820953369,
|
| 1848 |
+
"learning_rate": 1e-05,
|
| 1849 |
+
"loss": 0.2187,
|
| 1850 |
+
"step": 2630
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 1.9864559819413092,
|
| 1854 |
+
"grad_norm": 6.475297927856445,
|
| 1855 |
+
"learning_rate": 1e-05,
|
| 1856 |
+
"loss": 0.2879,
|
| 1857 |
+
"step": 2640
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 1.9939804364183598,
|
| 1861 |
+
"grad_norm": 8.84854793548584,
|
| 1862 |
+
"learning_rate": 1e-05,
|
| 1863 |
+
"loss": 0.2828,
|
| 1864 |
+
"step": 2650
|
| 1865 |
+
}
|
| 1866 |
+
],
|
| 1867 |
+
"logging_steps": 10,
|
| 1868 |
+
"max_steps": 39870,
|
| 1869 |
+
"num_input_tokens_seen": 0,
|
| 1870 |
+
"num_train_epochs": 30,
|
| 1871 |
+
"save_steps": 500,
|
| 1872 |
+
"stateful_callbacks": {
|
| 1873 |
+
"TrainerControl": {
|
| 1874 |
+
"args": {
|
| 1875 |
+
"should_epoch_stop": false,
|
| 1876 |
+
"should_evaluate": false,
|
| 1877 |
+
"should_log": false,
|
| 1878 |
+
"should_save": true,
|
| 1879 |
+
"should_training_stop": false
|
| 1880 |
+
},
|
| 1881 |
+
"attributes": {}
|
| 1882 |
+
}
|
| 1883 |
+
},
|
| 1884 |
+
"total_flos": 0.0,
|
| 1885 |
+
"train_batch_size": 64,
|
| 1886 |
+
"trial_name": null,
|
| 1887 |
+
"trial_params": null
|
| 1888 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a2c1821ca0f9d93bc486c36647133fda0fce87fbb23384115135978f2c56fb07
|
| 3 |
+
size 5713
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|