Upload 9 files
Browse files- config.json +34 -0
- model.safetensors +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- sentencepiece.bpe.model +3 -0
- special_tokens_map.json +51 -0
- tokenizer_config.json +55 -0
- trainer_state.json +2471 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "BAAI/bge-reranker-base",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"XLMRobertaForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 768,
|
| 13 |
+
"id2label": {
|
| 14 |
+
"0": "LABEL_0"
|
| 15 |
+
},
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"intermediate_size": 3072,
|
| 18 |
+
"label2id": {
|
| 19 |
+
"LABEL_0": 0
|
| 20 |
+
},
|
| 21 |
+
"layer_norm_eps": 1e-05,
|
| 22 |
+
"max_position_embeddings": 514,
|
| 23 |
+
"model_type": "xlm-roberta",
|
| 24 |
+
"num_attention_heads": 12,
|
| 25 |
+
"num_hidden_layers": 12,
|
| 26 |
+
"output_past": true,
|
| 27 |
+
"pad_token_id": 1,
|
| 28 |
+
"position_embedding_type": "absolute",
|
| 29 |
+
"torch_dtype": "float32",
|
| 30 |
+
"transformers_version": "4.38.1",
|
| 31 |
+
"type_vocab_size": 1,
|
| 32 |
+
"use_cache": true,
|
| 33 |
+
"vocab_size": 250002
|
| 34 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:177292a33a59856b5f92569ad9cfdb2923a91743d222c00f7c9712d4a783dfc7
|
| 3 |
+
size 1112201932
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:75df776e428fdbf6972c671e86180fbff6ec019328075339a3a42f18aea0d6dc
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c0e7aadcd4ea1d9542b8f90b810e97e6ab2260fb93faaac5f1903d22603e7c2b
|
| 3 |
+
size 1064
|
sentencepiece.bpe.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
|
| 3 |
+
size 5069051
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"cls_token": {
|
| 10 |
+
"content": "<s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"eos_token": {
|
| 17 |
+
"content": "</s>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"mask_token": {
|
| 24 |
+
"content": "<mask>",
|
| 25 |
+
"lstrip": true,
|
| 26 |
+
"normalized": true,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"pad_token": {
|
| 31 |
+
"content": "<pad>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
},
|
| 37 |
+
"sep_token": {
|
| 38 |
+
"content": "</s>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false
|
| 43 |
+
},
|
| 44 |
+
"unk_token": {
|
| 45 |
+
"content": "<unk>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": false,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false
|
| 50 |
+
}
|
| 51 |
+
}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<s>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<pad>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "</s>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "<unk>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"250001": {
|
| 36 |
+
"content": "<mask>",
|
| 37 |
+
"lstrip": true,
|
| 38 |
+
"normalized": true,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"bos_token": "<s>",
|
| 45 |
+
"clean_up_tokenization_spaces": true,
|
| 46 |
+
"cls_token": "<s>",
|
| 47 |
+
"eos_token": "</s>",
|
| 48 |
+
"mask_token": "<mask>",
|
| 49 |
+
"model_max_length": 512,
|
| 50 |
+
"pad_token": "<pad>",
|
| 51 |
+
"sep_token": "</s>",
|
| 52 |
+
"sp_model_kwargs": {},
|
| 53 |
+
"tokenizer_class": "XLMRobertaTokenizer",
|
| 54 |
+
"unk_token": "<unk>"
|
| 55 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2471 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.5220125786163523,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 3500,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.01,
|
| 13 |
+
"grad_norm": 0.0008633775869384408,
|
| 14 |
+
"learning_rate": 5.987915407854985e-05,
|
| 15 |
+
"loss": 0.0029,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.02,
|
| 20 |
+
"grad_norm": 0.0003390114870853722,
|
| 21 |
+
"learning_rate": 5.97583081570997e-05,
|
| 22 |
+
"loss": 0.0,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.03,
|
| 27 |
+
"grad_norm": 6.441377627197653e-05,
|
| 28 |
+
"learning_rate": 5.963746223564955e-05,
|
| 29 |
+
"loss": 0.0,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.04,
|
| 34 |
+
"grad_norm": 1.1177830856468063e-06,
|
| 35 |
+
"learning_rate": 5.95166163141994e-05,
|
| 36 |
+
"loss": 0.0027,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.05,
|
| 41 |
+
"grad_norm": 0.017336083576083183,
|
| 42 |
+
"learning_rate": 5.9407854984894266e-05,
|
| 43 |
+
"loss": 0.0504,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.06,
|
| 48 |
+
"grad_norm": 0.015523849986493587,
|
| 49 |
+
"learning_rate": 5.928700906344411e-05,
|
| 50 |
+
"loss": 0.0561,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.07,
|
| 55 |
+
"grad_norm": 0.974118709564209,
|
| 56 |
+
"learning_rate": 5.916616314199396e-05,
|
| 57 |
+
"loss": 0.0022,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.08,
|
| 62 |
+
"grad_norm": 0.051055677235126495,
|
| 63 |
+
"learning_rate": 5.904531722054381e-05,
|
| 64 |
+
"loss": 0.0006,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.09,
|
| 69 |
+
"grad_norm": 0.0001077930282917805,
|
| 70 |
+
"learning_rate": 5.892447129909366e-05,
|
| 71 |
+
"loss": 0.0,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.1,
|
| 76 |
+
"grad_norm": 1.7529870092403144e-05,
|
| 77 |
+
"learning_rate": 5.88036253776435e-05,
|
| 78 |
+
"loss": 0.0,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.11,
|
| 83 |
+
"grad_norm": 3.870090586133301e-05,
|
| 84 |
+
"learning_rate": 5.869486404833837e-05,
|
| 85 |
+
"loss": 0.0867,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.12,
|
| 90 |
+
"grad_norm": 0.0002746480458881706,
|
| 91 |
+
"learning_rate": 5.857401812688821e-05,
|
| 92 |
+
"loss": 0.0002,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.13,
|
| 97 |
+
"grad_norm": 0.30150529742240906,
|
| 98 |
+
"learning_rate": 5.8453172205438065e-05,
|
| 99 |
+
"loss": 0.0001,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.14,
|
| 104 |
+
"grad_norm": 0.07647743076086044,
|
| 105 |
+
"learning_rate": 5.833232628398792e-05,
|
| 106 |
+
"loss": 0.0019,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.15,
|
| 111 |
+
"grad_norm": 3.936241046176292e-05,
|
| 112 |
+
"learning_rate": 5.821148036253777e-05,
|
| 113 |
+
"loss": 0.0003,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.16,
|
| 118 |
+
"grad_norm": 0.019184544682502747,
|
| 119 |
+
"learning_rate": 5.809063444108762e-05,
|
| 120 |
+
"loss": 0.0001,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.17,
|
| 125 |
+
"grad_norm": 0.004856175277382135,
|
| 126 |
+
"learning_rate": 5.796978851963746e-05,
|
| 127 |
+
"loss": 0.0,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.18,
|
| 132 |
+
"grad_norm": 1.3400563148024958e-06,
|
| 133 |
+
"learning_rate": 5.7848942598187314e-05,
|
| 134 |
+
"loss": 0.0002,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.19,
|
| 139 |
+
"grad_norm": 0.7676287889480591,
|
| 140 |
+
"learning_rate": 5.772809667673716e-05,
|
| 141 |
+
"loss": 0.0218,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.2,
|
| 146 |
+
"grad_norm": 0.0006767328595742583,
|
| 147 |
+
"learning_rate": 5.760725075528701e-05,
|
| 148 |
+
"loss": 0.0,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.21,
|
| 153 |
+
"grad_norm": 0.4616200029850006,
|
| 154 |
+
"learning_rate": 5.749848942598187e-05,
|
| 155 |
+
"loss": 0.0479,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.22,
|
| 160 |
+
"grad_norm": 6.79653263092041,
|
| 161 |
+
"learning_rate": 5.737764350453173e-05,
|
| 162 |
+
"loss": 0.0514,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.23,
|
| 167 |
+
"grad_norm": 0.00680534215644002,
|
| 168 |
+
"learning_rate": 5.725679758308157e-05,
|
| 169 |
+
"loss": 0.0006,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.24,
|
| 174 |
+
"grad_norm": 8.962566789705306e-05,
|
| 175 |
+
"learning_rate": 5.7135951661631424e-05,
|
| 176 |
+
"loss": 0.0003,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.25,
|
| 181 |
+
"grad_norm": 0.016912635415792465,
|
| 182 |
+
"learning_rate": 5.701510574018127e-05,
|
| 183 |
+
"loss": 0.0044,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.26,
|
| 188 |
+
"grad_norm": 7.953944206237793,
|
| 189 |
+
"learning_rate": 5.689425981873112e-05,
|
| 190 |
+
"loss": 0.0081,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.27,
|
| 195 |
+
"grad_norm": 0.4362787902355194,
|
| 196 |
+
"learning_rate": 5.6773413897280964e-05,
|
| 197 |
+
"loss": 0.0556,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.28,
|
| 202 |
+
"grad_norm": 0.0008250115206465125,
|
| 203 |
+
"learning_rate": 5.6652567975830816e-05,
|
| 204 |
+
"loss": 0.0043,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.29,
|
| 209 |
+
"grad_norm": 0.056627318263053894,
|
| 210 |
+
"learning_rate": 5.653172205438066e-05,
|
| 211 |
+
"loss": 0.0021,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.3,
|
| 216 |
+
"grad_norm": 0.24282953143119812,
|
| 217 |
+
"learning_rate": 5.641087613293052e-05,
|
| 218 |
+
"loss": 0.0019,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.31,
|
| 223 |
+
"grad_norm": 0.19482052326202393,
|
| 224 |
+
"learning_rate": 5.629003021148036e-05,
|
| 225 |
+
"loss": 0.0041,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.32,
|
| 230 |
+
"grad_norm": 6.74727198202163e-05,
|
| 231 |
+
"learning_rate": 5.6169184290030214e-05,
|
| 232 |
+
"loss": 0.0,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.33,
|
| 237 |
+
"grad_norm": 0.0013290628558024764,
|
| 238 |
+
"learning_rate": 5.604833836858006e-05,
|
| 239 |
+
"loss": 0.5076,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.34,
|
| 244 |
+
"grad_norm": 0.5521184206008911,
|
| 245 |
+
"learning_rate": 5.592749244712991e-05,
|
| 246 |
+
"loss": 0.0011,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.35,
|
| 251 |
+
"grad_norm": 0.0012500255834311247,
|
| 252 |
+
"learning_rate": 5.5806646525679755e-05,
|
| 253 |
+
"loss": 0.0049,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.36,
|
| 258 |
+
"grad_norm": 9.534443989878127e-08,
|
| 259 |
+
"learning_rate": 5.5685800604229606e-05,
|
| 260 |
+
"loss": 0.0014,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.37,
|
| 265 |
+
"grad_norm": 3.4361846701358445e-08,
|
| 266 |
+
"learning_rate": 5.5564954682779464e-05,
|
| 267 |
+
"loss": 0.0302,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.38,
|
| 272 |
+
"grad_norm": 1.528064785816241e-08,
|
| 273 |
+
"learning_rate": 5.544410876132931e-05,
|
| 274 |
+
"loss": 0.008,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.39,
|
| 279 |
+
"grad_norm": 8.260390131908935e-06,
|
| 280 |
+
"learning_rate": 5.532326283987916e-05,
|
| 281 |
+
"loss": 0.0,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.4,
|
| 286 |
+
"grad_norm": 3.2616500854492188,
|
| 287 |
+
"learning_rate": 5.5202416918429004e-05,
|
| 288 |
+
"loss": 0.0058,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.41,
|
| 293 |
+
"grad_norm": 6.549715294568159e-07,
|
| 294 |
+
"learning_rate": 5.5081570996978856e-05,
|
| 295 |
+
"loss": 0.0004,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.42,
|
| 300 |
+
"grad_norm": 1.4332237697090022e-05,
|
| 301 |
+
"learning_rate": 5.49607250755287e-05,
|
| 302 |
+
"loss": 0.0002,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.43,
|
| 307 |
+
"grad_norm": 0.00038707954809069633,
|
| 308 |
+
"learning_rate": 5.483987915407855e-05,
|
| 309 |
+
"loss": 0.0,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.44,
|
| 314 |
+
"grad_norm": 7.052252044559282e-07,
|
| 315 |
+
"learning_rate": 5.4719033232628396e-05,
|
| 316 |
+
"loss": 0.001,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.45,
|
| 321 |
+
"grad_norm": 2.283226336885491e-07,
|
| 322 |
+
"learning_rate": 5.4598187311178254e-05,
|
| 323 |
+
"loss": 0.0,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.46,
|
| 328 |
+
"grad_norm": 3.97150288335979e-05,
|
| 329 |
+
"learning_rate": 5.44773413897281e-05,
|
| 330 |
+
"loss": 0.0,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.47,
|
| 335 |
+
"grad_norm": 5.106105049890175e-08,
|
| 336 |
+
"learning_rate": 5.435649546827795e-05,
|
| 337 |
+
"loss": 0.0,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.48,
|
| 342 |
+
"grad_norm": 0.0003172747674398124,
|
| 343 |
+
"learning_rate": 5.4235649546827795e-05,
|
| 344 |
+
"loss": 0.0,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.49,
|
| 349 |
+
"grad_norm": 0.6184738278388977,
|
| 350 |
+
"learning_rate": 5.4114803625377646e-05,
|
| 351 |
+
"loss": 0.0002,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.5,
|
| 356 |
+
"grad_norm": 1.0878424205884585e-07,
|
| 357 |
+
"learning_rate": 5.399395770392749e-05,
|
| 358 |
+
"loss": 0.0,
|
| 359 |
+
"step": 500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.51,
|
| 363 |
+
"grad_norm": 8.74086580893163e-09,
|
| 364 |
+
"learning_rate": 5.387311178247734e-05,
|
| 365 |
+
"loss": 0.0,
|
| 366 |
+
"step": 510
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.52,
|
| 370 |
+
"grad_norm": 1.5521897012149566e-07,
|
| 371 |
+
"learning_rate": 5.375226586102719e-05,
|
| 372 |
+
"loss": 0.0,
|
| 373 |
+
"step": 520
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.53,
|
| 377 |
+
"grad_norm": 2.9636362341989297e-06,
|
| 378 |
+
"learning_rate": 5.3631419939577045e-05,
|
| 379 |
+
"loss": 0.0,
|
| 380 |
+
"step": 530
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.54,
|
| 384 |
+
"grad_norm": 0.0023933914490044117,
|
| 385 |
+
"learning_rate": 5.351057401812689e-05,
|
| 386 |
+
"loss": 0.0004,
|
| 387 |
+
"step": 540
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.55,
|
| 391 |
+
"grad_norm": 0.0006123024504631758,
|
| 392 |
+
"learning_rate": 5.338972809667674e-05,
|
| 393 |
+
"loss": 0.0024,
|
| 394 |
+
"step": 550
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.56,
|
| 398 |
+
"grad_norm": 0.004097294993698597,
|
| 399 |
+
"learning_rate": 5.3268882175226585e-05,
|
| 400 |
+
"loss": 0.0673,
|
| 401 |
+
"step": 560
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.57,
|
| 405 |
+
"grad_norm": 2.348265013552009e-07,
|
| 406 |
+
"learning_rate": 5.3148036253776436e-05,
|
| 407 |
+
"loss": 0.0003,
|
| 408 |
+
"step": 570
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.58,
|
| 412 |
+
"grad_norm": 2.388081270510156e-07,
|
| 413 |
+
"learning_rate": 5.302719033232628e-05,
|
| 414 |
+
"loss": 0.0002,
|
| 415 |
+
"step": 580
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.59,
|
| 419 |
+
"grad_norm": 1.3560330671680276e-06,
|
| 420 |
+
"learning_rate": 5.290634441087613e-05,
|
| 421 |
+
"loss": 0.0,
|
| 422 |
+
"step": 590
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.6,
|
| 426 |
+
"grad_norm": 1.3283443678346885e-08,
|
| 427 |
+
"learning_rate": 5.2797583081571e-05,
|
| 428 |
+
"loss": 0.0986,
|
| 429 |
+
"step": 600
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.61,
|
| 433 |
+
"grad_norm": 0.00010959297651425004,
|
| 434 |
+
"learning_rate": 5.267673716012085e-05,
|
| 435 |
+
"loss": 0.0,
|
| 436 |
+
"step": 610
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.62,
|
| 440 |
+
"grad_norm": 0.00011195711704203859,
|
| 441 |
+
"learning_rate": 5.25558912386707e-05,
|
| 442 |
+
"loss": 0.0504,
|
| 443 |
+
"step": 620
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.63,
|
| 447 |
+
"grad_norm": 2.1371073722839355,
|
| 448 |
+
"learning_rate": 5.2435045317220546e-05,
|
| 449 |
+
"loss": 0.0036,
|
| 450 |
+
"step": 630
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.64,
|
| 454 |
+
"grad_norm": 0.001383950817398727,
|
| 455 |
+
"learning_rate": 5.23141993957704e-05,
|
| 456 |
+
"loss": 0.0052,
|
| 457 |
+
"step": 640
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.65,
|
| 461 |
+
"grad_norm": 0.001337925554253161,
|
| 462 |
+
"learning_rate": 5.219335347432024e-05,
|
| 463 |
+
"loss": 0.0001,
|
| 464 |
+
"step": 650
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.66,
|
| 468 |
+
"grad_norm": 2.282190052937949e-06,
|
| 469 |
+
"learning_rate": 5.207250755287009e-05,
|
| 470 |
+
"loss": 0.0001,
|
| 471 |
+
"step": 660
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.67,
|
| 475 |
+
"grad_norm": 0.0006644165841862559,
|
| 476 |
+
"learning_rate": 5.195166163141994e-05,
|
| 477 |
+
"loss": 0.0,
|
| 478 |
+
"step": 670
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.68,
|
| 482 |
+
"grad_norm": 0.00881275162100792,
|
| 483 |
+
"learning_rate": 5.1830815709969795e-05,
|
| 484 |
+
"loss": 0.0023,
|
| 485 |
+
"step": 680
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.69,
|
| 489 |
+
"grad_norm": 0.0035460491199046373,
|
| 490 |
+
"learning_rate": 5.170996978851964e-05,
|
| 491 |
+
"loss": 0.0,
|
| 492 |
+
"step": 690
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.7,
|
| 496 |
+
"grad_norm": 0.00038928634603507817,
|
| 497 |
+
"learning_rate": 5.158912386706949e-05,
|
| 498 |
+
"loss": 0.0008,
|
| 499 |
+
"step": 700
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.71,
|
| 503 |
+
"grad_norm": 5.376980993787583e-07,
|
| 504 |
+
"learning_rate": 5.1468277945619336e-05,
|
| 505 |
+
"loss": 0.0022,
|
| 506 |
+
"step": 710
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.72,
|
| 510 |
+
"grad_norm": 9.064051118912175e-05,
|
| 511 |
+
"learning_rate": 5.134743202416919e-05,
|
| 512 |
+
"loss": 0.0027,
|
| 513 |
+
"step": 720
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.73,
|
| 517 |
+
"grad_norm": 0.0011442841496318579,
|
| 518 |
+
"learning_rate": 5.122658610271903e-05,
|
| 519 |
+
"loss": 0.0003,
|
| 520 |
+
"step": 730
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.74,
|
| 524 |
+
"grad_norm": 0.004664595704525709,
|
| 525 |
+
"learning_rate": 5.110574018126888e-05,
|
| 526 |
+
"loss": 0.0022,
|
| 527 |
+
"step": 740
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.75,
|
| 531 |
+
"grad_norm": 0.0058820717968046665,
|
| 532 |
+
"learning_rate": 5.098489425981873e-05,
|
| 533 |
+
"loss": 0.001,
|
| 534 |
+
"step": 750
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.76,
|
| 538 |
+
"grad_norm": 0.030588263645768166,
|
| 539 |
+
"learning_rate": 5.0864048338368586e-05,
|
| 540 |
+
"loss": 0.0005,
|
| 541 |
+
"step": 760
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.77,
|
| 545 |
+
"grad_norm": 0.002594970166683197,
|
| 546 |
+
"learning_rate": 5.074320241691843e-05,
|
| 547 |
+
"loss": 0.0161,
|
| 548 |
+
"step": 770
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.78,
|
| 552 |
+
"grad_norm": 0.0027177969459444284,
|
| 553 |
+
"learning_rate": 5.062235649546828e-05,
|
| 554 |
+
"loss": 0.0,
|
| 555 |
+
"step": 780
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.79,
|
| 559 |
+
"grad_norm": 7.966559678607155e-08,
|
| 560 |
+
"learning_rate": 5.0501510574018126e-05,
|
| 561 |
+
"loss": 0.0002,
|
| 562 |
+
"step": 790
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.81,
|
| 566 |
+
"grad_norm": 0.006196084897965193,
|
| 567 |
+
"learning_rate": 5.038066465256798e-05,
|
| 568 |
+
"loss": 0.0004,
|
| 569 |
+
"step": 800
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.82,
|
| 573 |
+
"grad_norm": 0.000184666394488886,
|
| 574 |
+
"learning_rate": 5.025981873111782e-05,
|
| 575 |
+
"loss": 0.0164,
|
| 576 |
+
"step": 810
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.83,
|
| 580 |
+
"grad_norm": 0.0,
|
| 581 |
+
"learning_rate": 5.013897280966767e-05,
|
| 582 |
+
"loss": 0.0071,
|
| 583 |
+
"step": 820
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.84,
|
| 587 |
+
"grad_norm": 2.8641309768318024e-07,
|
| 588 |
+
"learning_rate": 5.0018126888217525e-05,
|
| 589 |
+
"loss": 0.0003,
|
| 590 |
+
"step": 830
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.85,
|
| 594 |
+
"grad_norm": 0.0019273586804047227,
|
| 595 |
+
"learning_rate": 4.9897280966767376e-05,
|
| 596 |
+
"loss": 0.0,
|
| 597 |
+
"step": 840
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.86,
|
| 601 |
+
"grad_norm": 0.0644882395863533,
|
| 602 |
+
"learning_rate": 4.977643504531722e-05,
|
| 603 |
+
"loss": 0.0025,
|
| 604 |
+
"step": 850
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.87,
|
| 608 |
+
"grad_norm": 0.00033757061464712024,
|
| 609 |
+
"learning_rate": 4.965558912386707e-05,
|
| 610 |
+
"loss": 0.0004,
|
| 611 |
+
"step": 860
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.88,
|
| 615 |
+
"grad_norm": 0.0009230116847902536,
|
| 616 |
+
"learning_rate": 4.9534743202416916e-05,
|
| 617 |
+
"loss": 0.0156,
|
| 618 |
+
"step": 870
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.89,
|
| 622 |
+
"grad_norm": 2.1307731003616937e-05,
|
| 623 |
+
"learning_rate": 4.941389728096677e-05,
|
| 624 |
+
"loss": 0.0026,
|
| 625 |
+
"step": 880
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.9,
|
| 629 |
+
"grad_norm": 4.2529451071970925e-09,
|
| 630 |
+
"learning_rate": 4.929305135951661e-05,
|
| 631 |
+
"loss": 0.0,
|
| 632 |
+
"step": 890
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.91,
|
| 636 |
+
"grad_norm": 2.9835612735951145e-07,
|
| 637 |
+
"learning_rate": 4.9172205438066464e-05,
|
| 638 |
+
"loss": 0.0005,
|
| 639 |
+
"step": 900
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.92,
|
| 643 |
+
"grad_norm": 0.02193087339401245,
|
| 644 |
+
"learning_rate": 4.905135951661632e-05,
|
| 645 |
+
"loss": 0.0016,
|
| 646 |
+
"step": 910
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.93,
|
| 650 |
+
"grad_norm": 0.0027798530645668507,
|
| 651 |
+
"learning_rate": 4.8930513595166166e-05,
|
| 652 |
+
"loss": 0.0002,
|
| 653 |
+
"step": 920
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.94,
|
| 657 |
+
"grad_norm": 0.0008504379075020552,
|
| 658 |
+
"learning_rate": 4.880966767371602e-05,
|
| 659 |
+
"loss": 0.0007,
|
| 660 |
+
"step": 930
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.95,
|
| 664 |
+
"grad_norm": 4.395148067715127e-08,
|
| 665 |
+
"learning_rate": 4.868882175226586e-05,
|
| 666 |
+
"loss": 0.0,
|
| 667 |
+
"step": 940
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.96,
|
| 671 |
+
"grad_norm": 1.3475013474817388e-05,
|
| 672 |
+
"learning_rate": 4.8567975830815714e-05,
|
| 673 |
+
"loss": 0.011,
|
| 674 |
+
"step": 950
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.97,
|
| 678 |
+
"grad_norm": 4.958523192044595e-08,
|
| 679 |
+
"learning_rate": 4.844712990936556e-05,
|
| 680 |
+
"loss": 0.0001,
|
| 681 |
+
"step": 960
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.98,
|
| 685 |
+
"grad_norm": 0.00013180672249291092,
|
| 686 |
+
"learning_rate": 4.832628398791541e-05,
|
| 687 |
+
"loss": 0.0,
|
| 688 |
+
"step": 970
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.99,
|
| 692 |
+
"grad_norm": 0.000387900770874694,
|
| 693 |
+
"learning_rate": 4.8205438066465254e-05,
|
| 694 |
+
"loss": 0.0515,
|
| 695 |
+
"step": 980
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.0,
|
| 699 |
+
"grad_norm": 0.024301297962665558,
|
| 700 |
+
"learning_rate": 4.808459214501511e-05,
|
| 701 |
+
"loss": 0.0002,
|
| 702 |
+
"step": 990
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.01,
|
| 706 |
+
"grad_norm": 0.054384149610996246,
|
| 707 |
+
"learning_rate": 4.796374622356496e-05,
|
| 708 |
+
"loss": 0.0002,
|
| 709 |
+
"step": 1000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 1.02,
|
| 713 |
+
"grad_norm": 0.0062861149199306965,
|
| 714 |
+
"learning_rate": 4.784290030211481e-05,
|
| 715 |
+
"loss": 0.0008,
|
| 716 |
+
"step": 1010
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 1.03,
|
| 720 |
+
"grad_norm": 0.00028014209237881005,
|
| 721 |
+
"learning_rate": 4.772205438066465e-05,
|
| 722 |
+
"loss": 0.0298,
|
| 723 |
+
"step": 1020
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 1.04,
|
| 727 |
+
"grad_norm": 8.147145854309201e-05,
|
| 728 |
+
"learning_rate": 4.7601208459214504e-05,
|
| 729 |
+
"loss": 0.0001,
|
| 730 |
+
"step": 1030
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 1.05,
|
| 734 |
+
"grad_norm": 0.0004538837820291519,
|
| 735 |
+
"learning_rate": 4.748036253776435e-05,
|
| 736 |
+
"loss": 0.0,
|
| 737 |
+
"step": 1040
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.06,
|
| 741 |
+
"grad_norm": 3.3908912882907316e-05,
|
| 742 |
+
"learning_rate": 4.73595166163142e-05,
|
| 743 |
+
"loss": 0.0,
|
| 744 |
+
"step": 1050
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.07,
|
| 748 |
+
"grad_norm": 15.499226570129395,
|
| 749 |
+
"learning_rate": 4.723867069486405e-05,
|
| 750 |
+
"loss": 0.0171,
|
| 751 |
+
"step": 1060
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 1.08,
|
| 755 |
+
"grad_norm": 6.438665390014648,
|
| 756 |
+
"learning_rate": 4.71178247734139e-05,
|
| 757 |
+
"loss": 0.0035,
|
| 758 |
+
"step": 1070
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 1.09,
|
| 762 |
+
"grad_norm": 0.00015530579548794776,
|
| 763 |
+
"learning_rate": 4.699697885196375e-05,
|
| 764 |
+
"loss": 0.0001,
|
| 765 |
+
"step": 1080
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 1.1,
|
| 769 |
+
"grad_norm": 5.526881068362854e-07,
|
| 770 |
+
"learning_rate": 4.68761329305136e-05,
|
| 771 |
+
"loss": 0.0,
|
| 772 |
+
"step": 1090
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 1.11,
|
| 776 |
+
"grad_norm": 3.55797128577251e-05,
|
| 777 |
+
"learning_rate": 4.675528700906344e-05,
|
| 778 |
+
"loss": 0.007,
|
| 779 |
+
"step": 1100
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.12,
|
| 783 |
+
"grad_norm": 1.0427943379909266e-05,
|
| 784 |
+
"learning_rate": 4.6634441087613294e-05,
|
| 785 |
+
"loss": 0.0,
|
| 786 |
+
"step": 1110
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 1.13,
|
| 790 |
+
"grad_norm": 2.200431765686517e-07,
|
| 791 |
+
"learning_rate": 4.651359516616314e-05,
|
| 792 |
+
"loss": 0.0009,
|
| 793 |
+
"step": 1120
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 1.14,
|
| 797 |
+
"grad_norm": 2.6252129828208126e-05,
|
| 798 |
+
"learning_rate": 4.639274924471299e-05,
|
| 799 |
+
"loss": 0.0001,
|
| 800 |
+
"step": 1130
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 1.15,
|
| 804 |
+
"grad_norm": 3.5910196061195165e-07,
|
| 805 |
+
"learning_rate": 4.627190332326284e-05,
|
| 806 |
+
"loss": 0.0,
|
| 807 |
+
"step": 1140
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 1.16,
|
| 811 |
+
"grad_norm": 7.039247407192306e-09,
|
| 812 |
+
"learning_rate": 4.615105740181269e-05,
|
| 813 |
+
"loss": 0.0,
|
| 814 |
+
"step": 1150
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 1.17,
|
| 818 |
+
"grad_norm": 6.980984380788868e-06,
|
| 819 |
+
"learning_rate": 4.603021148036254e-05,
|
| 820 |
+
"loss": 0.0014,
|
| 821 |
+
"step": 1160
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 1.18,
|
| 825 |
+
"grad_norm": 7.229643728123847e-08,
|
| 826 |
+
"learning_rate": 4.590936555891239e-05,
|
| 827 |
+
"loss": 0.0653,
|
| 828 |
+
"step": 1170
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 1.19,
|
| 832 |
+
"grad_norm": 0.03556942939758301,
|
| 833 |
+
"learning_rate": 4.578851963746223e-05,
|
| 834 |
+
"loss": 0.0002,
|
| 835 |
+
"step": 1180
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 1.2,
|
| 839 |
+
"grad_norm": 0.022921539843082428,
|
| 840 |
+
"learning_rate": 4.5667673716012084e-05,
|
| 841 |
+
"loss": 0.0019,
|
| 842 |
+
"step": 1190
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.21,
|
| 846 |
+
"grad_norm": 1.2204433005535975e-05,
|
| 847 |
+
"learning_rate": 4.554682779456193e-05,
|
| 848 |
+
"loss": 0.0,
|
| 849 |
+
"step": 1200
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.22,
|
| 853 |
+
"grad_norm": 0.003104632254689932,
|
| 854 |
+
"learning_rate": 4.542598187311179e-05,
|
| 855 |
+
"loss": 0.0004,
|
| 856 |
+
"step": 1210
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.23,
|
| 860 |
+
"grad_norm": 0.007898018695414066,
|
| 861 |
+
"learning_rate": 4.530513595166163e-05,
|
| 862 |
+
"loss": 0.0012,
|
| 863 |
+
"step": 1220
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 1.24,
|
| 867 |
+
"grad_norm": 0.0009976280853152275,
|
| 868 |
+
"learning_rate": 4.518429003021148e-05,
|
| 869 |
+
"loss": 0.0002,
|
| 870 |
+
"step": 1230
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 1.25,
|
| 874 |
+
"grad_norm": 0.03468003869056702,
|
| 875 |
+
"learning_rate": 4.5063444108761334e-05,
|
| 876 |
+
"loss": 0.0,
|
| 877 |
+
"step": 1240
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 1.26,
|
| 881 |
+
"grad_norm": 1.0441084441481507e-06,
|
| 882 |
+
"learning_rate": 4.494259818731118e-05,
|
| 883 |
+
"loss": 0.0001,
|
| 884 |
+
"step": 1250
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 1.27,
|
| 888 |
+
"grad_norm": 3.78264850553478e-08,
|
| 889 |
+
"learning_rate": 4.482175226586103e-05,
|
| 890 |
+
"loss": 0.0003,
|
| 891 |
+
"step": 1260
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 1.28,
|
| 895 |
+
"grad_norm": 6.2697478320217215e-09,
|
| 896 |
+
"learning_rate": 4.4700906344410875e-05,
|
| 897 |
+
"loss": 0.0,
|
| 898 |
+
"step": 1270
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 1.29,
|
| 902 |
+
"grad_norm": 0.0001720537693472579,
|
| 903 |
+
"learning_rate": 4.4580060422960726e-05,
|
| 904 |
+
"loss": 0.0,
|
| 905 |
+
"step": 1280
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 1.3,
|
| 909 |
+
"grad_norm": 6.12790336163016e-06,
|
| 910 |
+
"learning_rate": 4.445921450151058e-05,
|
| 911 |
+
"loss": 0.0,
|
| 912 |
+
"step": 1290
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 1.31,
|
| 916 |
+
"grad_norm": 2.106332302093506,
|
| 917 |
+
"learning_rate": 4.433836858006043e-05,
|
| 918 |
+
"loss": 0.0033,
|
| 919 |
+
"step": 1300
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 1.32,
|
| 923 |
+
"grad_norm": 0.000930789508856833,
|
| 924 |
+
"learning_rate": 4.421752265861027e-05,
|
| 925 |
+
"loss": 0.0,
|
| 926 |
+
"step": 1310
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 1.33,
|
| 930 |
+
"grad_norm": 1.7019704046106199e-06,
|
| 931 |
+
"learning_rate": 4.4096676737160125e-05,
|
| 932 |
+
"loss": 0.0001,
|
| 933 |
+
"step": 1320
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 1.34,
|
| 937 |
+
"grad_norm": 0.01025537308305502,
|
| 938 |
+
"learning_rate": 4.397583081570997e-05,
|
| 939 |
+
"loss": 0.0449,
|
| 940 |
+
"step": 1330
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 1.35,
|
| 944 |
+
"grad_norm": 0.0023111889604479074,
|
| 945 |
+
"learning_rate": 4.385498489425982e-05,
|
| 946 |
+
"loss": 0.0,
|
| 947 |
+
"step": 1340
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 1.36,
|
| 951 |
+
"grad_norm": 0.0005836488562636077,
|
| 952 |
+
"learning_rate": 4.3734138972809665e-05,
|
| 953 |
+
"loss": 0.0015,
|
| 954 |
+
"step": 1350
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 1.37,
|
| 958 |
+
"grad_norm": 0.001271137036383152,
|
| 959 |
+
"learning_rate": 4.3613293051359516e-05,
|
| 960 |
+
"loss": 0.0,
|
| 961 |
+
"step": 1360
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 1.38,
|
| 965 |
+
"grad_norm": 0.01227193046361208,
|
| 966 |
+
"learning_rate": 4.349244712990937e-05,
|
| 967 |
+
"loss": 0.0,
|
| 968 |
+
"step": 1370
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 1.39,
|
| 972 |
+
"grad_norm": 3.68997419286643e-08,
|
| 973 |
+
"learning_rate": 4.337160120845922e-05,
|
| 974 |
+
"loss": 0.0009,
|
| 975 |
+
"step": 1380
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 1.4,
|
| 979 |
+
"grad_norm": 1.0907036340768173e-07,
|
| 980 |
+
"learning_rate": 4.3250755287009064e-05,
|
| 981 |
+
"loss": 0.0,
|
| 982 |
+
"step": 1390
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 1.41,
|
| 986 |
+
"grad_norm": 6.332713383017108e-05,
|
| 987 |
+
"learning_rate": 4.3129909365558915e-05,
|
| 988 |
+
"loss": 0.0,
|
| 989 |
+
"step": 1400
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 1.42,
|
| 993 |
+
"grad_norm": 2.5950125603912966e-08,
|
| 994 |
+
"learning_rate": 4.300906344410876e-05,
|
| 995 |
+
"loss": 0.0,
|
| 996 |
+
"step": 1410
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 1.43,
|
| 1000 |
+
"grad_norm": 0.6216311454772949,
|
| 1001 |
+
"learning_rate": 4.288821752265861e-05,
|
| 1002 |
+
"loss": 0.0001,
|
| 1003 |
+
"step": 1420
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 1.44,
|
| 1007 |
+
"grad_norm": 3.573224205410952e-08,
|
| 1008 |
+
"learning_rate": 4.2767371601208455e-05,
|
| 1009 |
+
"loss": 0.0,
|
| 1010 |
+
"step": 1430
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 1.45,
|
| 1014 |
+
"grad_norm": 2.763231634617114e-07,
|
| 1015 |
+
"learning_rate": 4.2646525679758313e-05,
|
| 1016 |
+
"loss": 0.0,
|
| 1017 |
+
"step": 1440
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 1.46,
|
| 1021 |
+
"grad_norm": 2.6494699767454222e-08,
|
| 1022 |
+
"learning_rate": 4.252567975830816e-05,
|
| 1023 |
+
"loss": 0.015,
|
| 1024 |
+
"step": 1450
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 1.47,
|
| 1028 |
+
"grad_norm": 1.0802775340934545e-09,
|
| 1029 |
+
"learning_rate": 4.240483383685801e-05,
|
| 1030 |
+
"loss": 0.0,
|
| 1031 |
+
"step": 1460
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 1.48,
|
| 1035 |
+
"grad_norm": 2.746639538031559e-08,
|
| 1036 |
+
"learning_rate": 4.2283987915407854e-05,
|
| 1037 |
+
"loss": 0.0,
|
| 1038 |
+
"step": 1470
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 1.49,
|
| 1042 |
+
"grad_norm": 5.9745045291492715e-05,
|
| 1043 |
+
"learning_rate": 4.2163141993957705e-05,
|
| 1044 |
+
"loss": 0.0,
|
| 1045 |
+
"step": 1480
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 1.5,
|
| 1049 |
+
"grad_norm": 3.736948031018983e-07,
|
| 1050 |
+
"learning_rate": 4.204229607250755e-05,
|
| 1051 |
+
"loss": 0.0,
|
| 1052 |
+
"step": 1490
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 1.51,
|
| 1056 |
+
"grad_norm": 5.944487213582761e-08,
|
| 1057 |
+
"learning_rate": 4.19214501510574e-05,
|
| 1058 |
+
"loss": 0.0,
|
| 1059 |
+
"step": 1500
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 1.52,
|
| 1063 |
+
"grad_norm": 9.614995910567359e-09,
|
| 1064 |
+
"learning_rate": 4.1800604229607246e-05,
|
| 1065 |
+
"loss": 0.0,
|
| 1066 |
+
"step": 1510
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 1.53,
|
| 1070 |
+
"grad_norm": 9.067125716910596e-08,
|
| 1071 |
+
"learning_rate": 4.1679758308157104e-05,
|
| 1072 |
+
"loss": 0.0,
|
| 1073 |
+
"step": 1520
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 1.54,
|
| 1077 |
+
"grad_norm": 3.790767255651417e-09,
|
| 1078 |
+
"learning_rate": 4.155891238670695e-05,
|
| 1079 |
+
"loss": 0.0018,
|
| 1080 |
+
"step": 1530
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 1.55,
|
| 1084 |
+
"grad_norm": 0.0014521937118843198,
|
| 1085 |
+
"learning_rate": 4.14380664652568e-05,
|
| 1086 |
+
"loss": 0.0141,
|
| 1087 |
+
"step": 1540
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 1.56,
|
| 1091 |
+
"grad_norm": 1.1887091204698663e-06,
|
| 1092 |
+
"learning_rate": 4.131722054380665e-05,
|
| 1093 |
+
"loss": 0.0001,
|
| 1094 |
+
"step": 1550
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 1.57,
|
| 1098 |
+
"grad_norm": 0.21191056072711945,
|
| 1099 |
+
"learning_rate": 4.1196374622356495e-05,
|
| 1100 |
+
"loss": 0.0001,
|
| 1101 |
+
"step": 1560
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 1.58,
|
| 1105 |
+
"grad_norm": 1.7035688415489858e-09,
|
| 1106 |
+
"learning_rate": 4.107552870090635e-05,
|
| 1107 |
+
"loss": 0.103,
|
| 1108 |
+
"step": 1570
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 1.59,
|
| 1112 |
+
"grad_norm": 4.13519810535945e-05,
|
| 1113 |
+
"learning_rate": 4.095468277945619e-05,
|
| 1114 |
+
"loss": 0.0002,
|
| 1115 |
+
"step": 1580
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 1.6,
|
| 1119 |
+
"grad_norm": 1.2702444109891076e-05,
|
| 1120 |
+
"learning_rate": 4.083383685800605e-05,
|
| 1121 |
+
"loss": 0.0,
|
| 1122 |
+
"step": 1590
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 1.61,
|
| 1126 |
+
"grad_norm": 1.6921067924613453e-07,
|
| 1127 |
+
"learning_rate": 4.0712990936555894e-05,
|
| 1128 |
+
"loss": 0.0001,
|
| 1129 |
+
"step": 1600
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 1.62,
|
| 1133 |
+
"grad_norm": 2.8627702874928218e-08,
|
| 1134 |
+
"learning_rate": 4.0592145015105745e-05,
|
| 1135 |
+
"loss": 0.0,
|
| 1136 |
+
"step": 1610
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 1.63,
|
| 1140 |
+
"grad_norm": 6.734685484843794e-06,
|
| 1141 |
+
"learning_rate": 4.047129909365559e-05,
|
| 1142 |
+
"loss": 0.0,
|
| 1143 |
+
"step": 1620
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 1.64,
|
| 1147 |
+
"grad_norm": 1.370299145264653e-07,
|
| 1148 |
+
"learning_rate": 4.035045317220544e-05,
|
| 1149 |
+
"loss": 0.0001,
|
| 1150 |
+
"step": 1630
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 1.65,
|
| 1154 |
+
"grad_norm": 0.00048380409134551883,
|
| 1155 |
+
"learning_rate": 4.0229607250755286e-05,
|
| 1156 |
+
"loss": 0.0059,
|
| 1157 |
+
"step": 1640
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 1.66,
|
| 1161 |
+
"grad_norm": 0.00042098091216757894,
|
| 1162 |
+
"learning_rate": 4.010876132930514e-05,
|
| 1163 |
+
"loss": 0.0001,
|
| 1164 |
+
"step": 1650
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 1.67,
|
| 1168 |
+
"grad_norm": 1.4294737411546521e-05,
|
| 1169 |
+
"learning_rate": 3.998791540785498e-05,
|
| 1170 |
+
"loss": 0.0004,
|
| 1171 |
+
"step": 1660
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 1.68,
|
| 1175 |
+
"grad_norm": 1.3993314285087877e-09,
|
| 1176 |
+
"learning_rate": 3.986706948640484e-05,
|
| 1177 |
+
"loss": 0.0062,
|
| 1178 |
+
"step": 1670
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 1.69,
|
| 1182 |
+
"grad_norm": 0.013060159981250763,
|
| 1183 |
+
"learning_rate": 3.9746223564954684e-05,
|
| 1184 |
+
"loss": 0.0,
|
| 1185 |
+
"step": 1680
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 1.7,
|
| 1189 |
+
"grad_norm": 6.259356450755149e-05,
|
| 1190 |
+
"learning_rate": 3.9625377643504536e-05,
|
| 1191 |
+
"loss": 0.0,
|
| 1192 |
+
"step": 1690
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 1.71,
|
| 1196 |
+
"grad_norm": 9.836193203227594e-05,
|
| 1197 |
+
"learning_rate": 3.950453172205438e-05,
|
| 1198 |
+
"loss": 0.0,
|
| 1199 |
+
"step": 1700
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 1.72,
|
| 1203 |
+
"grad_norm": 1.420707462784776e-08,
|
| 1204 |
+
"learning_rate": 3.938368580060423e-05,
|
| 1205 |
+
"loss": 0.0,
|
| 1206 |
+
"step": 1710
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 1.73,
|
| 1210 |
+
"grad_norm": 1.8018381453543952e-08,
|
| 1211 |
+
"learning_rate": 3.9262839879154076e-05,
|
| 1212 |
+
"loss": 0.0,
|
| 1213 |
+
"step": 1720
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 1.74,
|
| 1217 |
+
"grad_norm": 5.407735148565962e-09,
|
| 1218 |
+
"learning_rate": 3.914199395770393e-05,
|
| 1219 |
+
"loss": 0.0,
|
| 1220 |
+
"step": 1730
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 1.75,
|
| 1224 |
+
"grad_norm": 8.004823826013308e-08,
|
| 1225 |
+
"learning_rate": 3.902114803625378e-05,
|
| 1226 |
+
"loss": 0.0,
|
| 1227 |
+
"step": 1740
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 1.76,
|
| 1231 |
+
"grad_norm": 1.1862041304766535e-07,
|
| 1232 |
+
"learning_rate": 3.890030211480363e-05,
|
| 1233 |
+
"loss": 0.0,
|
| 1234 |
+
"step": 1750
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 1.77,
|
| 1238 |
+
"grad_norm": 3.1350757012660324e-07,
|
| 1239 |
+
"learning_rate": 3.8779456193353475e-05,
|
| 1240 |
+
"loss": 0.0,
|
| 1241 |
+
"step": 1760
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 1.78,
|
| 1245 |
+
"grad_norm": 2.7172923466878274e-08,
|
| 1246 |
+
"learning_rate": 3.8658610271903326e-05,
|
| 1247 |
+
"loss": 0.0,
|
| 1248 |
+
"step": 1770
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 1.79,
|
| 1252 |
+
"grad_norm": 2.770736706736443e-08,
|
| 1253 |
+
"learning_rate": 3.853776435045317e-05,
|
| 1254 |
+
"loss": 0.0,
|
| 1255 |
+
"step": 1780
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 1.8,
|
| 1259 |
+
"grad_norm": 1.3860269598708896e-09,
|
| 1260 |
+
"learning_rate": 3.841691842900302e-05,
|
| 1261 |
+
"loss": 0.0,
|
| 1262 |
+
"step": 1790
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 1.81,
|
| 1266 |
+
"grad_norm": 0.01048044953495264,
|
| 1267 |
+
"learning_rate": 3.8296072507552866e-05,
|
| 1268 |
+
"loss": 0.2012,
|
| 1269 |
+
"step": 1800
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 1.82,
|
| 1273 |
+
"grad_norm": 1.413319807852531e-07,
|
| 1274 |
+
"learning_rate": 3.817522658610272e-05,
|
| 1275 |
+
"loss": 0.0,
|
| 1276 |
+
"step": 1810
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 1.83,
|
| 1280 |
+
"grad_norm": 0.0012239572824910283,
|
| 1281 |
+
"learning_rate": 3.805438066465257e-05,
|
| 1282 |
+
"loss": 0.0,
|
| 1283 |
+
"step": 1820
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 1.84,
|
| 1287 |
+
"grad_norm": 0.00016678984684403986,
|
| 1288 |
+
"learning_rate": 3.793353474320242e-05,
|
| 1289 |
+
"loss": 0.0,
|
| 1290 |
+
"step": 1830
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 1.85,
|
| 1294 |
+
"grad_norm": 4.03699864648388e-08,
|
| 1295 |
+
"learning_rate": 3.7812688821752265e-05,
|
| 1296 |
+
"loss": 0.0014,
|
| 1297 |
+
"step": 1840
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 1.86,
|
| 1301 |
+
"grad_norm": 2.3393537048832513e-05,
|
| 1302 |
+
"learning_rate": 3.7691842900302116e-05,
|
| 1303 |
+
"loss": 0.0,
|
| 1304 |
+
"step": 1850
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 1.87,
|
| 1308 |
+
"grad_norm": 0.0013141903327777982,
|
| 1309 |
+
"learning_rate": 3.757099697885196e-05,
|
| 1310 |
+
"loss": 0.0001,
|
| 1311 |
+
"step": 1860
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 1.88,
|
| 1315 |
+
"grad_norm": 0.001294136163778603,
|
| 1316 |
+
"learning_rate": 3.745015105740181e-05,
|
| 1317 |
+
"loss": 0.0,
|
| 1318 |
+
"step": 1870
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 1.89,
|
| 1322 |
+
"grad_norm": 3.9412746843936475e-08,
|
| 1323 |
+
"learning_rate": 3.7329305135951663e-05,
|
| 1324 |
+
"loss": 0.0002,
|
| 1325 |
+
"step": 1880
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 1.9,
|
| 1329 |
+
"grad_norm": 9.437405878998106e-07,
|
| 1330 |
+
"learning_rate": 3.720845921450151e-05,
|
| 1331 |
+
"loss": 0.0008,
|
| 1332 |
+
"step": 1890
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 1.91,
|
| 1336 |
+
"grad_norm": 1.1508973329910077e-06,
|
| 1337 |
+
"learning_rate": 3.7087613293051366e-05,
|
| 1338 |
+
"loss": 0.0002,
|
| 1339 |
+
"step": 1900
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 1.92,
|
| 1343 |
+
"grad_norm": 5.708042749574815e-07,
|
| 1344 |
+
"learning_rate": 3.696676737160121e-05,
|
| 1345 |
+
"loss": 0.0,
|
| 1346 |
+
"step": 1910
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 1.93,
|
| 1350 |
+
"grad_norm": 4.998336407879833e-06,
|
| 1351 |
+
"learning_rate": 3.684592145015106e-05,
|
| 1352 |
+
"loss": 0.0,
|
| 1353 |
+
"step": 1920
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 1.94,
|
| 1357 |
+
"grad_norm": 0.0002045446599368006,
|
| 1358 |
+
"learning_rate": 3.6725075528700906e-05,
|
| 1359 |
+
"loss": 0.0004,
|
| 1360 |
+
"step": 1930
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 1.95,
|
| 1364 |
+
"grad_norm": 1.011339804790623e-06,
|
| 1365 |
+
"learning_rate": 3.660422960725076e-05,
|
| 1366 |
+
"loss": 0.0002,
|
| 1367 |
+
"step": 1940
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 1.96,
|
| 1371 |
+
"grad_norm": 1.1179584191722824e-07,
|
| 1372 |
+
"learning_rate": 3.64833836858006e-05,
|
| 1373 |
+
"loss": 0.0,
|
| 1374 |
+
"step": 1950
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 1.97,
|
| 1378 |
+
"grad_norm": 4.01311623932088e-08,
|
| 1379 |
+
"learning_rate": 3.6362537764350454e-05,
|
| 1380 |
+
"loss": 0.0001,
|
| 1381 |
+
"step": 1960
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 1.98,
|
| 1385 |
+
"grad_norm": 0.00014017950161360204,
|
| 1386 |
+
"learning_rate": 3.6241691842900305e-05,
|
| 1387 |
+
"loss": 0.0,
|
| 1388 |
+
"step": 1970
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 1.99,
|
| 1392 |
+
"grad_norm": 3.3373577025486156e-05,
|
| 1393 |
+
"learning_rate": 3.6120845921450156e-05,
|
| 1394 |
+
"loss": 0.0,
|
| 1395 |
+
"step": 1980
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 2.0,
|
| 1399 |
+
"grad_norm": 6.252674211282283e-05,
|
| 1400 |
+
"learning_rate": 3.6e-05,
|
| 1401 |
+
"loss": 0.0,
|
| 1402 |
+
"step": 1990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 2.01,
|
| 1406 |
+
"grad_norm": 0.0002250933466712013,
|
| 1407 |
+
"learning_rate": 3.587915407854985e-05,
|
| 1408 |
+
"loss": 0.0,
|
| 1409 |
+
"step": 2000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 2.02,
|
| 1413 |
+
"grad_norm": 6.714632849735835e-09,
|
| 1414 |
+
"learning_rate": 3.57583081570997e-05,
|
| 1415 |
+
"loss": 0.0,
|
| 1416 |
+
"step": 2010
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 2.03,
|
| 1420 |
+
"grad_norm": 2.656873050455033e-07,
|
| 1421 |
+
"learning_rate": 3.563746223564955e-05,
|
| 1422 |
+
"loss": 0.0,
|
| 1423 |
+
"step": 2020
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 2.04,
|
| 1427 |
+
"grad_norm": 1.6010590115911327e-05,
|
| 1428 |
+
"learning_rate": 3.551661631419939e-05,
|
| 1429 |
+
"loss": 0.0,
|
| 1430 |
+
"step": 2030
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 2.05,
|
| 1434 |
+
"grad_norm": 2.7113070245832205e-05,
|
| 1435 |
+
"learning_rate": 3.5395770392749244e-05,
|
| 1436 |
+
"loss": 0.0,
|
| 1437 |
+
"step": 2040
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 2.06,
|
| 1441 |
+
"grad_norm": 1.7675635399783118e-09,
|
| 1442 |
+
"learning_rate": 3.5274924471299095e-05,
|
| 1443 |
+
"loss": 0.0,
|
| 1444 |
+
"step": 2050
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 2.07,
|
| 1448 |
+
"grad_norm": 5.97303042582098e-08,
|
| 1449 |
+
"learning_rate": 3.515407854984895e-05,
|
| 1450 |
+
"loss": 0.0001,
|
| 1451 |
+
"step": 2060
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 2.08,
|
| 1455 |
+
"grad_norm": 0.0004335389821790159,
|
| 1456 |
+
"learning_rate": 3.503323262839879e-05,
|
| 1457 |
+
"loss": 0.0,
|
| 1458 |
+
"step": 2070
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 2.09,
|
| 1462 |
+
"grad_norm": 6.485903213615529e-06,
|
| 1463 |
+
"learning_rate": 3.491238670694864e-05,
|
| 1464 |
+
"loss": 0.0,
|
| 1465 |
+
"step": 2080
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 2.1,
|
| 1469 |
+
"grad_norm": 0.0035731520038098097,
|
| 1470 |
+
"learning_rate": 3.479154078549849e-05,
|
| 1471 |
+
"loss": 0.0,
|
| 1472 |
+
"step": 2090
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 2.11,
|
| 1476 |
+
"grad_norm": 4.11334362127036e-08,
|
| 1477 |
+
"learning_rate": 3.467069486404834e-05,
|
| 1478 |
+
"loss": 0.0,
|
| 1479 |
+
"step": 2100
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 2.12,
|
| 1483 |
+
"grad_norm": 5.335771129466593e-05,
|
| 1484 |
+
"learning_rate": 3.454984894259818e-05,
|
| 1485 |
+
"loss": 0.0,
|
| 1486 |
+
"step": 2110
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 2.13,
|
| 1490 |
+
"grad_norm": 1.26767045003362e-05,
|
| 1491 |
+
"learning_rate": 3.442900302114804e-05,
|
| 1492 |
+
"loss": 0.0,
|
| 1493 |
+
"step": 2120
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 2.14,
|
| 1497 |
+
"grad_norm": 2.478699911989679e-09,
|
| 1498 |
+
"learning_rate": 3.4308157099697886e-05,
|
| 1499 |
+
"loss": 0.0,
|
| 1500 |
+
"step": 2130
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 2.15,
|
| 1504 |
+
"grad_norm": 2.9354995945141127e-07,
|
| 1505 |
+
"learning_rate": 3.418731117824774e-05,
|
| 1506 |
+
"loss": 0.0,
|
| 1507 |
+
"step": 2140
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 2.16,
|
| 1511 |
+
"grad_norm": 5.661284831148805e-06,
|
| 1512 |
+
"learning_rate": 3.406646525679758e-05,
|
| 1513 |
+
"loss": 0.0001,
|
| 1514 |
+
"step": 2150
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 2.17,
|
| 1518 |
+
"grad_norm": 0.0009775606449693441,
|
| 1519 |
+
"learning_rate": 3.394561933534743e-05,
|
| 1520 |
+
"loss": 0.0,
|
| 1521 |
+
"step": 2160
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 2.18,
|
| 1525 |
+
"grad_norm": 1.0540851524964978e-09,
|
| 1526 |
+
"learning_rate": 3.382477341389728e-05,
|
| 1527 |
+
"loss": 0.0,
|
| 1528 |
+
"step": 2170
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 2.19,
|
| 1532 |
+
"grad_norm": 4.330029241828015e-06,
|
| 1533 |
+
"learning_rate": 3.370392749244713e-05,
|
| 1534 |
+
"loss": 0.0,
|
| 1535 |
+
"step": 2180
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 2.2,
|
| 1539 |
+
"grad_norm": 1.206312163049006e-06,
|
| 1540 |
+
"learning_rate": 3.358308157099698e-05,
|
| 1541 |
+
"loss": 0.0,
|
| 1542 |
+
"step": 2190
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 2.21,
|
| 1546 |
+
"grad_norm": 2.2324436486087507e-06,
|
| 1547 |
+
"learning_rate": 3.346223564954683e-05,
|
| 1548 |
+
"loss": 0.0,
|
| 1549 |
+
"step": 2200
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 2.22,
|
| 1553 |
+
"grad_norm": 6.707778510417484e-08,
|
| 1554 |
+
"learning_rate": 3.334138972809668e-05,
|
| 1555 |
+
"loss": 0.0,
|
| 1556 |
+
"step": 2210
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 2.23,
|
| 1560 |
+
"grad_norm": 1.0213330181585434e-09,
|
| 1561 |
+
"learning_rate": 3.322054380664653e-05,
|
| 1562 |
+
"loss": 0.0,
|
| 1563 |
+
"step": 2220
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 2.24,
|
| 1567 |
+
"grad_norm": 1.522714967450156e-07,
|
| 1568 |
+
"learning_rate": 3.309969788519638e-05,
|
| 1569 |
+
"loss": 0.0,
|
| 1570 |
+
"step": 2230
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 2.25,
|
| 1574 |
+
"grad_norm": 3.1322219911089633e-06,
|
| 1575 |
+
"learning_rate": 3.297885196374622e-05,
|
| 1576 |
+
"loss": 0.0,
|
| 1577 |
+
"step": 2240
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 2.26,
|
| 1581 |
+
"grad_norm": 0.0,
|
| 1582 |
+
"learning_rate": 3.2858006042296074e-05,
|
| 1583 |
+
"loss": 0.0001,
|
| 1584 |
+
"step": 2250
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 2.27,
|
| 1588 |
+
"grad_norm": 0.0,
|
| 1589 |
+
"learning_rate": 3.273716012084592e-05,
|
| 1590 |
+
"loss": 0.0,
|
| 1591 |
+
"step": 2260
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 2.28,
|
| 1595 |
+
"grad_norm": 3.107200274854449e-09,
|
| 1596 |
+
"learning_rate": 3.261631419939577e-05,
|
| 1597 |
+
"loss": 0.0,
|
| 1598 |
+
"step": 2270
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 2.29,
|
| 1602 |
+
"grad_norm": 9.697516123452488e-10,
|
| 1603 |
+
"learning_rate": 3.249546827794562e-05,
|
| 1604 |
+
"loss": 0.0,
|
| 1605 |
+
"step": 2280
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 2.3,
|
| 1609 |
+
"grad_norm": 1.6279294579035763e-09,
|
| 1610 |
+
"learning_rate": 3.237462235649547e-05,
|
| 1611 |
+
"loss": 0.0,
|
| 1612 |
+
"step": 2290
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 2.31,
|
| 1616 |
+
"grad_norm": 6.846393674209139e-09,
|
| 1617 |
+
"learning_rate": 3.225377643504532e-05,
|
| 1618 |
+
"loss": 0.0,
|
| 1619 |
+
"step": 2300
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 2.32,
|
| 1623 |
+
"grad_norm": 1.0791376553243026e-05,
|
| 1624 |
+
"learning_rate": 3.213293051359517e-05,
|
| 1625 |
+
"loss": 0.0,
|
| 1626 |
+
"step": 2310
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 2.33,
|
| 1630 |
+
"grad_norm": 1.0202172688877909e-06,
|
| 1631 |
+
"learning_rate": 3.2012084592145013e-05,
|
| 1632 |
+
"loss": 0.0,
|
| 1633 |
+
"step": 2320
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 2.34,
|
| 1637 |
+
"grad_norm": 3.7638578365140063e-10,
|
| 1638 |
+
"learning_rate": 3.1891238670694865e-05,
|
| 1639 |
+
"loss": 0.0,
|
| 1640 |
+
"step": 2330
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 2.35,
|
| 1644 |
+
"grad_norm": 2.3091282219667164e-09,
|
| 1645 |
+
"learning_rate": 3.177039274924471e-05,
|
| 1646 |
+
"loss": 0.0,
|
| 1647 |
+
"step": 2340
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 2.36,
|
| 1651 |
+
"grad_norm": 0.0,
|
| 1652 |
+
"learning_rate": 3.164954682779457e-05,
|
| 1653 |
+
"loss": 0.0,
|
| 1654 |
+
"step": 2350
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 2.37,
|
| 1658 |
+
"grad_norm": 0.0,
|
| 1659 |
+
"learning_rate": 3.152870090634441e-05,
|
| 1660 |
+
"loss": 0.0,
|
| 1661 |
+
"step": 2360
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 2.38,
|
| 1665 |
+
"grad_norm": 0.0,
|
| 1666 |
+
"learning_rate": 3.140785498489426e-05,
|
| 1667 |
+
"loss": 0.0,
|
| 1668 |
+
"step": 2370
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 2.39,
|
| 1672 |
+
"grad_norm": 0.0,
|
| 1673 |
+
"learning_rate": 3.128700906344411e-05,
|
| 1674 |
+
"loss": 0.0,
|
| 1675 |
+
"step": 2380
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 2.41,
|
| 1679 |
+
"grad_norm": 0.0,
|
| 1680 |
+
"learning_rate": 3.116616314199396e-05,
|
| 1681 |
+
"loss": 0.0,
|
| 1682 |
+
"step": 2390
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 2.42,
|
| 1686 |
+
"grad_norm": 2.4167900846805423e-05,
|
| 1687 |
+
"learning_rate": 3.1045317220543804e-05,
|
| 1688 |
+
"loss": 0.0,
|
| 1689 |
+
"step": 2400
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 2.43,
|
| 1693 |
+
"grad_norm": 2.4693702016520547e-07,
|
| 1694 |
+
"learning_rate": 3.0924471299093655e-05,
|
| 1695 |
+
"loss": 0.0,
|
| 1696 |
+
"step": 2410
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 2.44,
|
| 1700 |
+
"grad_norm": 1.0320581056433298e-09,
|
| 1701 |
+
"learning_rate": 3.08036253776435e-05,
|
| 1702 |
+
"loss": 0.0,
|
| 1703 |
+
"step": 2420
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 2.45,
|
| 1707 |
+
"grad_norm": 8.062679768272574e-08,
|
| 1708 |
+
"learning_rate": 3.068277945619336e-05,
|
| 1709 |
+
"loss": 0.0,
|
| 1710 |
+
"step": 2430
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 2.46,
|
| 1714 |
+
"grad_norm": 2.012851219035383e-09,
|
| 1715 |
+
"learning_rate": 3.05619335347432e-05,
|
| 1716 |
+
"loss": 0.0,
|
| 1717 |
+
"step": 2440
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 2.47,
|
| 1721 |
+
"grad_norm": 0.0,
|
| 1722 |
+
"learning_rate": 3.0441087613293054e-05,
|
| 1723 |
+
"loss": 0.0,
|
| 1724 |
+
"step": 2450
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 2.48,
|
| 1728 |
+
"grad_norm": 3.1564242330972547e-09,
|
| 1729 |
+
"learning_rate": 3.0320241691842898e-05,
|
| 1730 |
+
"loss": 0.0,
|
| 1731 |
+
"step": 2460
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 2.49,
|
| 1735 |
+
"grad_norm": 0.0,
|
| 1736 |
+
"learning_rate": 3.019939577039275e-05,
|
| 1737 |
+
"loss": 0.0,
|
| 1738 |
+
"step": 2470
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 2.5,
|
| 1742 |
+
"grad_norm": 3.624436430982314e-05,
|
| 1743 |
+
"learning_rate": 3.0078549848942597e-05,
|
| 1744 |
+
"loss": 0.0,
|
| 1745 |
+
"step": 2480
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 2.51,
|
| 1749 |
+
"grad_norm": 1.0910243730677394e-07,
|
| 1750 |
+
"learning_rate": 2.995770392749245e-05,
|
| 1751 |
+
"loss": 0.0,
|
| 1752 |
+
"step": 2490
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 2.52,
|
| 1756 |
+
"grad_norm": 0.0,
|
| 1757 |
+
"learning_rate": 2.9836858006042297e-05,
|
| 1758 |
+
"loss": 0.0,
|
| 1759 |
+
"step": 2500
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 2.53,
|
| 1763 |
+
"grad_norm": 2.473066729180573e-08,
|
| 1764 |
+
"learning_rate": 2.9716012084592145e-05,
|
| 1765 |
+
"loss": 0.0,
|
| 1766 |
+
"step": 2510
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 2.54,
|
| 1770 |
+
"grad_norm": 3.971327089402621e-08,
|
| 1771 |
+
"learning_rate": 2.9595166163141996e-05,
|
| 1772 |
+
"loss": 0.0,
|
| 1773 |
+
"step": 2520
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 2.55,
|
| 1777 |
+
"grad_norm": 0.0,
|
| 1778 |
+
"learning_rate": 2.9474320241691844e-05,
|
| 1779 |
+
"loss": 0.0,
|
| 1780 |
+
"step": 2530
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 2.56,
|
| 1784 |
+
"grad_norm": 2.1878437109990045e-05,
|
| 1785 |
+
"learning_rate": 2.9353474320241692e-05,
|
| 1786 |
+
"loss": 0.0,
|
| 1787 |
+
"step": 2540
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 2.57,
|
| 1791 |
+
"grad_norm": 0.00010030780686065555,
|
| 1792 |
+
"learning_rate": 2.923262839879154e-05,
|
| 1793 |
+
"loss": 0.0,
|
| 1794 |
+
"step": 2550
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 2.58,
|
| 1798 |
+
"grad_norm": 4.3228740587153425e-09,
|
| 1799 |
+
"learning_rate": 2.911178247734139e-05,
|
| 1800 |
+
"loss": 0.0,
|
| 1801 |
+
"step": 2560
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 2.59,
|
| 1805 |
+
"grad_norm": 0.0,
|
| 1806 |
+
"learning_rate": 2.899093655589124e-05,
|
| 1807 |
+
"loss": 0.0,
|
| 1808 |
+
"step": 2570
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 2.6,
|
| 1812 |
+
"grad_norm": 3.8086006570736686e-10,
|
| 1813 |
+
"learning_rate": 2.8870090634441087e-05,
|
| 1814 |
+
"loss": 0.0,
|
| 1815 |
+
"step": 2580
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 2.61,
|
| 1819 |
+
"grad_norm": 6.88414825145145e-10,
|
| 1820 |
+
"learning_rate": 2.8749244712990935e-05,
|
| 1821 |
+
"loss": 0.0,
|
| 1822 |
+
"step": 2590
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 2.62,
|
| 1826 |
+
"grad_norm": 2.8818004693675903e-07,
|
| 1827 |
+
"learning_rate": 2.8628398791540786e-05,
|
| 1828 |
+
"loss": 0.002,
|
| 1829 |
+
"step": 2600
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 2.63,
|
| 1833 |
+
"grad_norm": 3.844630569460605e-08,
|
| 1834 |
+
"learning_rate": 2.8507552870090634e-05,
|
| 1835 |
+
"loss": 0.0,
|
| 1836 |
+
"step": 2610
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 2.64,
|
| 1840 |
+
"grad_norm": 0.026187095791101456,
|
| 1841 |
+
"learning_rate": 2.8386706948640482e-05,
|
| 1842 |
+
"loss": 0.0,
|
| 1843 |
+
"step": 2620
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 2.65,
|
| 1847 |
+
"grad_norm": 3.006335624888834e-09,
|
| 1848 |
+
"learning_rate": 2.826586102719033e-05,
|
| 1849 |
+
"loss": 0.0,
|
| 1850 |
+
"step": 2630
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 2.66,
|
| 1854 |
+
"grad_norm": 1.2555399564462277e-07,
|
| 1855 |
+
"learning_rate": 2.814501510574018e-05,
|
| 1856 |
+
"loss": 0.0,
|
| 1857 |
+
"step": 2640
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 2.67,
|
| 1861 |
+
"grad_norm": 0.03935589641332626,
|
| 1862 |
+
"learning_rate": 2.802416918429003e-05,
|
| 1863 |
+
"loss": 0.0,
|
| 1864 |
+
"step": 2650
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 2.68,
|
| 1868 |
+
"grad_norm": 4.552068730845349e-07,
|
| 1869 |
+
"learning_rate": 2.7903323262839877e-05,
|
| 1870 |
+
"loss": 0.0,
|
| 1871 |
+
"step": 2660
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 2.69,
|
| 1875 |
+
"grad_norm": 60.47336959838867,
|
| 1876 |
+
"learning_rate": 2.7782477341389732e-05,
|
| 1877 |
+
"loss": 0.0803,
|
| 1878 |
+
"step": 2670
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 2.7,
|
| 1882 |
+
"grad_norm": 1.4243897616950107e-08,
|
| 1883 |
+
"learning_rate": 2.766163141993958e-05,
|
| 1884 |
+
"loss": 0.0017,
|
| 1885 |
+
"step": 2680
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 2.71,
|
| 1889 |
+
"grad_norm": 8.210067790059838e-06,
|
| 1890 |
+
"learning_rate": 2.7540785498489428e-05,
|
| 1891 |
+
"loss": 0.0,
|
| 1892 |
+
"step": 2690
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 2.72,
|
| 1896 |
+
"grad_norm": 1.0214007417630455e-08,
|
| 1897 |
+
"learning_rate": 2.7419939577039276e-05,
|
| 1898 |
+
"loss": 0.0,
|
| 1899 |
+
"step": 2700
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 2.73,
|
| 1903 |
+
"grad_norm": 1.5932839048105052e-08,
|
| 1904 |
+
"learning_rate": 2.7299093655589127e-05,
|
| 1905 |
+
"loss": 0.0,
|
| 1906 |
+
"step": 2710
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 2.74,
|
| 1910 |
+
"grad_norm": 1.8762943976291524e-10,
|
| 1911 |
+
"learning_rate": 2.7178247734138975e-05,
|
| 1912 |
+
"loss": 0.0,
|
| 1913 |
+
"step": 2720
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 2.75,
|
| 1917 |
+
"grad_norm": 4.0266687051371264e-07,
|
| 1918 |
+
"learning_rate": 2.7057401812688823e-05,
|
| 1919 |
+
"loss": 0.0,
|
| 1920 |
+
"step": 2730
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 2.76,
|
| 1924 |
+
"grad_norm": 4.4115214592466145e-08,
|
| 1925 |
+
"learning_rate": 2.693655589123867e-05,
|
| 1926 |
+
"loss": 0.0,
|
| 1927 |
+
"step": 2740
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 2.77,
|
| 1931 |
+
"grad_norm": 0.0,
|
| 1932 |
+
"learning_rate": 2.6815709969788522e-05,
|
| 1933 |
+
"loss": 0.0,
|
| 1934 |
+
"step": 2750
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 2.78,
|
| 1938 |
+
"grad_norm": 1.6773222810684274e-08,
|
| 1939 |
+
"learning_rate": 2.669486404833837e-05,
|
| 1940 |
+
"loss": 0.0001,
|
| 1941 |
+
"step": 2760
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 2.79,
|
| 1945 |
+
"grad_norm": 0.0,
|
| 1946 |
+
"learning_rate": 2.6574018126888218e-05,
|
| 1947 |
+
"loss": 0.0,
|
| 1948 |
+
"step": 2770
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 2.8,
|
| 1952 |
+
"grad_norm": 0.002830307465046644,
|
| 1953 |
+
"learning_rate": 2.6453172205438066e-05,
|
| 1954 |
+
"loss": 0.0538,
|
| 1955 |
+
"step": 2780
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 2.81,
|
| 1959 |
+
"grad_norm": 1.8111581123747555e-09,
|
| 1960 |
+
"learning_rate": 2.6332326283987917e-05,
|
| 1961 |
+
"loss": 0.0,
|
| 1962 |
+
"step": 2790
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 2.82,
|
| 1966 |
+
"grad_norm": 2.044587290583877e-07,
|
| 1967 |
+
"learning_rate": 2.6211480362537765e-05,
|
| 1968 |
+
"loss": 0.0,
|
| 1969 |
+
"step": 2800
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 2.83,
|
| 1973 |
+
"grad_norm": 0.0,
|
| 1974 |
+
"learning_rate": 2.6090634441087613e-05,
|
| 1975 |
+
"loss": 0.0,
|
| 1976 |
+
"step": 2810
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 2.84,
|
| 1980 |
+
"grad_norm": 1.001442484493964e-09,
|
| 1981 |
+
"learning_rate": 2.596978851963746e-05,
|
| 1982 |
+
"loss": 0.0,
|
| 1983 |
+
"step": 2820
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 2.85,
|
| 1987 |
+
"grad_norm": 0.0,
|
| 1988 |
+
"learning_rate": 2.5848942598187313e-05,
|
| 1989 |
+
"loss": 0.0,
|
| 1990 |
+
"step": 2830
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 2.86,
|
| 1994 |
+
"grad_norm": 1.0242295900297904e-08,
|
| 1995 |
+
"learning_rate": 2.572809667673716e-05,
|
| 1996 |
+
"loss": 0.0,
|
| 1997 |
+
"step": 2840
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 2.87,
|
| 2001 |
+
"grad_norm": 3.850387564341418e-09,
|
| 2002 |
+
"learning_rate": 2.560725075528701e-05,
|
| 2003 |
+
"loss": 0.0,
|
| 2004 |
+
"step": 2850
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 2.88,
|
| 2008 |
+
"grad_norm": 1.869227411743779e-10,
|
| 2009 |
+
"learning_rate": 2.548640483383686e-05,
|
| 2010 |
+
"loss": 0.0,
|
| 2011 |
+
"step": 2860
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 2.89,
|
| 2015 |
+
"grad_norm": 1.6829256210826316e-08,
|
| 2016 |
+
"learning_rate": 2.5365558912386708e-05,
|
| 2017 |
+
"loss": 0.0,
|
| 2018 |
+
"step": 2870
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 2.9,
|
| 2022 |
+
"grad_norm": 3.8250621514634986e-07,
|
| 2023 |
+
"learning_rate": 2.5244712990936556e-05,
|
| 2024 |
+
"loss": 0.0,
|
| 2025 |
+
"step": 2880
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 2.91,
|
| 2029 |
+
"grad_norm": 2.593100623116129e-09,
|
| 2030 |
+
"learning_rate": 2.5123867069486404e-05,
|
| 2031 |
+
"loss": 0.0,
|
| 2032 |
+
"step": 2890
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 2.92,
|
| 2036 |
+
"grad_norm": 3.2663322713233356e-08,
|
| 2037 |
+
"learning_rate": 2.5003021148036255e-05,
|
| 2038 |
+
"loss": 0.0,
|
| 2039 |
+
"step": 2900
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 2.93,
|
| 2043 |
+
"grad_norm": 3.1544191703147817e-09,
|
| 2044 |
+
"learning_rate": 2.4882175226586103e-05,
|
| 2045 |
+
"loss": 0.0,
|
| 2046 |
+
"step": 2910
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 2.94,
|
| 2050 |
+
"grad_norm": 4.537993092412762e-09,
|
| 2051 |
+
"learning_rate": 2.476132930513595e-05,
|
| 2052 |
+
"loss": 0.0,
|
| 2053 |
+
"step": 2920
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 2.95,
|
| 2057 |
+
"grad_norm": 2.1482501324499026e-05,
|
| 2058 |
+
"learning_rate": 2.46404833836858e-05,
|
| 2059 |
+
"loss": 0.0,
|
| 2060 |
+
"step": 2930
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 2.96,
|
| 2064 |
+
"grad_norm": 3.4461805853425176e-08,
|
| 2065 |
+
"learning_rate": 2.451963746223565e-05,
|
| 2066 |
+
"loss": 0.0,
|
| 2067 |
+
"step": 2940
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 2.97,
|
| 2071 |
+
"grad_norm": 1.910395930337927e-08,
|
| 2072 |
+
"learning_rate": 2.4398791540785498e-05,
|
| 2073 |
+
"loss": 0.0,
|
| 2074 |
+
"step": 2950
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 2.98,
|
| 2078 |
+
"grad_norm": 1.0935653271815227e-08,
|
| 2079 |
+
"learning_rate": 2.4277945619335346e-05,
|
| 2080 |
+
"loss": 0.0,
|
| 2081 |
+
"step": 2960
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 2.99,
|
| 2085 |
+
"grad_norm": 1.6578489692165022e-09,
|
| 2086 |
+
"learning_rate": 2.4157099697885194e-05,
|
| 2087 |
+
"loss": 0.0,
|
| 2088 |
+
"step": 2970
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 3.0,
|
| 2092 |
+
"grad_norm": 4.757405349664623e-07,
|
| 2093 |
+
"learning_rate": 2.4036253776435045e-05,
|
| 2094 |
+
"loss": 0.0,
|
| 2095 |
+
"step": 2980
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 3.01,
|
| 2099 |
+
"grad_norm": 0.0,
|
| 2100 |
+
"learning_rate": 2.3915407854984897e-05,
|
| 2101 |
+
"loss": 0.0,
|
| 2102 |
+
"step": 2990
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 3.02,
|
| 2106 |
+
"grad_norm": 1.9106341397900906e-08,
|
| 2107 |
+
"learning_rate": 2.3794561933534744e-05,
|
| 2108 |
+
"loss": 0.0,
|
| 2109 |
+
"step": 3000
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 3.03,
|
| 2113 |
+
"grad_norm": 2.297590118160997e-09,
|
| 2114 |
+
"learning_rate": 2.3673716012084596e-05,
|
| 2115 |
+
"loss": 0.0,
|
| 2116 |
+
"step": 3010
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 3.04,
|
| 2120 |
+
"grad_norm": 0.0,
|
| 2121 |
+
"learning_rate": 2.3552870090634444e-05,
|
| 2122 |
+
"loss": 0.0,
|
| 2123 |
+
"step": 3020
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 3.05,
|
| 2127 |
+
"grad_norm": 5.0260878197150305e-05,
|
| 2128 |
+
"learning_rate": 2.343202416918429e-05,
|
| 2129 |
+
"loss": 0.0,
|
| 2130 |
+
"step": 3030
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 3.06,
|
| 2134 |
+
"grad_norm": 1.84212534115602e-10,
|
| 2135 |
+
"learning_rate": 2.331117824773414e-05,
|
| 2136 |
+
"loss": 0.0,
|
| 2137 |
+
"step": 3040
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 3.07,
|
| 2141 |
+
"grad_norm": 1.035463603749065e-09,
|
| 2142 |
+
"learning_rate": 2.319033232628399e-05,
|
| 2143 |
+
"loss": 0.0,
|
| 2144 |
+
"step": 3050
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 3.08,
|
| 2148 |
+
"grad_norm": 8.718736843604802e-09,
|
| 2149 |
+
"learning_rate": 2.306948640483384e-05,
|
| 2150 |
+
"loss": 0.0,
|
| 2151 |
+
"step": 3060
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 3.09,
|
| 2155 |
+
"grad_norm": 3.523397129612249e-10,
|
| 2156 |
+
"learning_rate": 2.2948640483383687e-05,
|
| 2157 |
+
"loss": 0.0,
|
| 2158 |
+
"step": 3070
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 3.1,
|
| 2162 |
+
"grad_norm": 3.516629210054134e-10,
|
| 2163 |
+
"learning_rate": 2.2827794561933535e-05,
|
| 2164 |
+
"loss": 0.0,
|
| 2165 |
+
"step": 3080
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 3.11,
|
| 2169 |
+
"grad_norm": 1.1817105072253753e-09,
|
| 2170 |
+
"learning_rate": 2.2706948640483386e-05,
|
| 2171 |
+
"loss": 0.0,
|
| 2172 |
+
"step": 3090
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 3.12,
|
| 2176 |
+
"grad_norm": 6.393580065378046e-08,
|
| 2177 |
+
"learning_rate": 2.2586102719033234e-05,
|
| 2178 |
+
"loss": 0.0,
|
| 2179 |
+
"step": 3100
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 3.13,
|
| 2183 |
+
"grad_norm": 0.0,
|
| 2184 |
+
"learning_rate": 2.2465256797583082e-05,
|
| 2185 |
+
"loss": 0.0,
|
| 2186 |
+
"step": 3110
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 3.14,
|
| 2190 |
+
"grad_norm": 2.3444219010571032e-08,
|
| 2191 |
+
"learning_rate": 2.234441087613293e-05,
|
| 2192 |
+
"loss": 0.0,
|
| 2193 |
+
"step": 3120
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 3.15,
|
| 2197 |
+
"grad_norm": 5.386530688156199e-08,
|
| 2198 |
+
"learning_rate": 2.222356495468278e-05,
|
| 2199 |
+
"loss": 0.0,
|
| 2200 |
+
"step": 3130
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 3.16,
|
| 2204 |
+
"grad_norm": 7.222023867825555e-08,
|
| 2205 |
+
"learning_rate": 2.210271903323263e-05,
|
| 2206 |
+
"loss": 0.0,
|
| 2207 |
+
"step": 3140
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 3.17,
|
| 2211 |
+
"grad_norm": 2.3319643993602313e-09,
|
| 2212 |
+
"learning_rate": 2.1981873111782477e-05,
|
| 2213 |
+
"loss": 0.0,
|
| 2214 |
+
"step": 3150
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 3.18,
|
| 2218 |
+
"grad_norm": 1.8931477219208404e-10,
|
| 2219 |
+
"learning_rate": 2.1861027190332325e-05,
|
| 2220 |
+
"loss": 0.0,
|
| 2221 |
+
"step": 3160
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 3.19,
|
| 2225 |
+
"grad_norm": 0.014633001759648323,
|
| 2226 |
+
"learning_rate": 2.1740181268882176e-05,
|
| 2227 |
+
"loss": 0.0,
|
| 2228 |
+
"step": 3170
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 3.2,
|
| 2232 |
+
"grad_norm": 2.1926523913862184e-05,
|
| 2233 |
+
"learning_rate": 2.1619335347432024e-05,
|
| 2234 |
+
"loss": 0.0,
|
| 2235 |
+
"step": 3180
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 3.21,
|
| 2239 |
+
"grad_norm": 2.996988213155305e-09,
|
| 2240 |
+
"learning_rate": 2.1498489425981872e-05,
|
| 2241 |
+
"loss": 0.0,
|
| 2242 |
+
"step": 3190
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 3.22,
|
| 2246 |
+
"grad_norm": 7.307365649467101e-06,
|
| 2247 |
+
"learning_rate": 2.1377643504531724e-05,
|
| 2248 |
+
"loss": 0.0,
|
| 2249 |
+
"step": 3200
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 3.23,
|
| 2253 |
+
"grad_norm": 1.264578168047592e-05,
|
| 2254 |
+
"learning_rate": 2.125679758308157e-05,
|
| 2255 |
+
"loss": 0.0,
|
| 2256 |
+
"step": 3210
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 3.24,
|
| 2260 |
+
"grad_norm": 1.8541245481173974e-06,
|
| 2261 |
+
"learning_rate": 2.113595166163142e-05,
|
| 2262 |
+
"loss": 0.0,
|
| 2263 |
+
"step": 3220
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 3.25,
|
| 2267 |
+
"grad_norm": 7.540985080822793e-08,
|
| 2268 |
+
"learning_rate": 2.1015105740181267e-05,
|
| 2269 |
+
"loss": 0.0,
|
| 2270 |
+
"step": 3230
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 3.26,
|
| 2274 |
+
"grad_norm": 5.4537132143650524e-08,
|
| 2275 |
+
"learning_rate": 2.089425981873112e-05,
|
| 2276 |
+
"loss": 0.0,
|
| 2277 |
+
"step": 3240
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 3.27,
|
| 2281 |
+
"grad_norm": 7.338199338846607e-06,
|
| 2282 |
+
"learning_rate": 2.0773413897280967e-05,
|
| 2283 |
+
"loss": 0.0,
|
| 2284 |
+
"step": 3250
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 3.28,
|
| 2288 |
+
"grad_norm": 0.0,
|
| 2289 |
+
"learning_rate": 2.0652567975830815e-05,
|
| 2290 |
+
"loss": 0.0,
|
| 2291 |
+
"step": 3260
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 3.29,
|
| 2295 |
+
"grad_norm": 0.0,
|
| 2296 |
+
"learning_rate": 2.0531722054380663e-05,
|
| 2297 |
+
"loss": 0.0,
|
| 2298 |
+
"step": 3270
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 3.3,
|
| 2302 |
+
"grad_norm": 2.116165687127136e-09,
|
| 2303 |
+
"learning_rate": 2.0410876132930514e-05,
|
| 2304 |
+
"loss": 0.0,
|
| 2305 |
+
"step": 3280
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 3.31,
|
| 2309 |
+
"grad_norm": 2.066717907922566e-09,
|
| 2310 |
+
"learning_rate": 2.0290030211480362e-05,
|
| 2311 |
+
"loss": 0.0,
|
| 2312 |
+
"step": 3290
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 3.32,
|
| 2316 |
+
"grad_norm": 1.8276996582855531e-10,
|
| 2317 |
+
"learning_rate": 2.016918429003021e-05,
|
| 2318 |
+
"loss": 0.0,
|
| 2319 |
+
"step": 3300
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 3.33,
|
| 2323 |
+
"grad_norm": 3.0077771384640073e-09,
|
| 2324 |
+
"learning_rate": 2.004833836858006e-05,
|
| 2325 |
+
"loss": 0.0,
|
| 2326 |
+
"step": 3310
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 3.34,
|
| 2330 |
+
"grad_norm": 0.0,
|
| 2331 |
+
"learning_rate": 1.9927492447129912e-05,
|
| 2332 |
+
"loss": 0.0,
|
| 2333 |
+
"step": 3320
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 3.35,
|
| 2337 |
+
"grad_norm": 0.0,
|
| 2338 |
+
"learning_rate": 1.980664652567976e-05,
|
| 2339 |
+
"loss": 0.0003,
|
| 2340 |
+
"step": 3330
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 3.36,
|
| 2344 |
+
"grad_norm": 0.007729419972747564,
|
| 2345 |
+
"learning_rate": 1.968580060422961e-05,
|
| 2346 |
+
"loss": 0.0,
|
| 2347 |
+
"step": 3340
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 3.37,
|
| 2351 |
+
"grad_norm": 8.718690514797345e-05,
|
| 2352 |
+
"learning_rate": 1.9564954682779456e-05,
|
| 2353 |
+
"loss": 0.0,
|
| 2354 |
+
"step": 3350
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 3.38,
|
| 2358 |
+
"grad_norm": 4.8525894271733705e-06,
|
| 2359 |
+
"learning_rate": 1.9444108761329308e-05,
|
| 2360 |
+
"loss": 0.0,
|
| 2361 |
+
"step": 3360
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 3.39,
|
| 2365 |
+
"grad_norm": 2.319741509992923e-09,
|
| 2366 |
+
"learning_rate": 1.9323262839879156e-05,
|
| 2367 |
+
"loss": 0.0,
|
| 2368 |
+
"step": 3370
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 3.4,
|
| 2372 |
+
"grad_norm": 9.715724758052602e-08,
|
| 2373 |
+
"learning_rate": 1.9202416918429003e-05,
|
| 2374 |
+
"loss": 0.0,
|
| 2375 |
+
"step": 3380
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 3.41,
|
| 2379 |
+
"grad_norm": 5.839451091560477e-07,
|
| 2380 |
+
"learning_rate": 1.9081570996978855e-05,
|
| 2381 |
+
"loss": 0.0,
|
| 2382 |
+
"step": 3390
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 3.42,
|
| 2386 |
+
"grad_norm": 1.4994407138146926e-05,
|
| 2387 |
+
"learning_rate": 1.8960725075528703e-05,
|
| 2388 |
+
"loss": 0.0,
|
| 2389 |
+
"step": 3400
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 3.43,
|
| 2393 |
+
"grad_norm": 3.994861685896467e-08,
|
| 2394 |
+
"learning_rate": 1.883987915407855e-05,
|
| 2395 |
+
"loss": 0.0,
|
| 2396 |
+
"step": 3410
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 3.44,
|
| 2400 |
+
"grad_norm": 0.0011337869800627232,
|
| 2401 |
+
"learning_rate": 1.87190332326284e-05,
|
| 2402 |
+
"loss": 0.0,
|
| 2403 |
+
"step": 3420
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 3.45,
|
| 2407 |
+
"grad_norm": 2.8960034370584253e-08,
|
| 2408 |
+
"learning_rate": 1.859818731117825e-05,
|
| 2409 |
+
"loss": 0.0,
|
| 2410 |
+
"step": 3430
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 3.46,
|
| 2414 |
+
"grad_norm": 9.850285209722642e-08,
|
| 2415 |
+
"learning_rate": 1.8477341389728098e-05,
|
| 2416 |
+
"loss": 0.0,
|
| 2417 |
+
"step": 3440
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 3.47,
|
| 2421 |
+
"grad_norm": 6.7992941694683395e-06,
|
| 2422 |
+
"learning_rate": 1.8356495468277946e-05,
|
| 2423 |
+
"loss": 0.0,
|
| 2424 |
+
"step": 3450
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 3.48,
|
| 2428 |
+
"grad_norm": 0.0,
|
| 2429 |
+
"learning_rate": 1.8235649546827794e-05,
|
| 2430 |
+
"loss": 0.0,
|
| 2431 |
+
"step": 3460
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 3.49,
|
| 2435 |
+
"grad_norm": 5.028818250707445e-09,
|
| 2436 |
+
"learning_rate": 1.8114803625377645e-05,
|
| 2437 |
+
"loss": 0.0,
|
| 2438 |
+
"step": 3470
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 3.5,
|
| 2442 |
+
"grad_norm": 2.1626638613270188e-07,
|
| 2443 |
+
"learning_rate": 1.7993957703927493e-05,
|
| 2444 |
+
"loss": 0.0,
|
| 2445 |
+
"step": 3480
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 3.51,
|
| 2449 |
+
"grad_norm": 1.7985012090715458e-10,
|
| 2450 |
+
"learning_rate": 1.787311178247734e-05,
|
| 2451 |
+
"loss": 0.0,
|
| 2452 |
+
"step": 3490
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 3.52,
|
| 2456 |
+
"grad_norm": 3.205403231731907e-08,
|
| 2457 |
+
"learning_rate": 1.775226586102719e-05,
|
| 2458 |
+
"loss": 0.0,
|
| 2459 |
+
"step": 3500
|
| 2460 |
+
}
|
| 2461 |
+
],
|
| 2462 |
+
"logging_steps": 10,
|
| 2463 |
+
"max_steps": 4965,
|
| 2464 |
+
"num_input_tokens_seen": 0,
|
| 2465 |
+
"num_train_epochs": 5,
|
| 2466 |
+
"save_steps": 500,
|
| 2467 |
+
"total_flos": 0.0,
|
| 2468 |
+
"train_batch_size": 1,
|
| 2469 |
+
"trial_name": null,
|
| 2470 |
+
"trial_params": null
|
| 2471 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2cbf70232036733fb3e6e678055d099760eade07cedf5bdfdb22fcbf30e444d7
|
| 3 |
+
size 4920
|