sn14_task1
Browse files- config.json +43 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +51 -0
- tokenizer.json +0 -0
- tokenizer_config.json +58 -0
- trainer_state.json +1153 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "protectai/deberta-v3-base-prompt-injection",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"DebertaV2ForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"hidden_act": "gelu",
|
| 8 |
+
"hidden_dropout_prob": 0.1,
|
| 9 |
+
"hidden_size": 768,
|
| 10 |
+
"id2label": {
|
| 11 |
+
"0": "SAFE",
|
| 12 |
+
"1": "INJECTION"
|
| 13 |
+
},
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 16 |
+
"label2id": {
|
| 17 |
+
"INJECTION": 1,
|
| 18 |
+
"SAFE": 0
|
| 19 |
+
},
|
| 20 |
+
"layer_norm_eps": 1e-07,
|
| 21 |
+
"max_position_embeddings": 512,
|
| 22 |
+
"max_relative_positions": -1,
|
| 23 |
+
"model_type": "deberta-v2",
|
| 24 |
+
"norm_rel_ebd": "layer_norm",
|
| 25 |
+
"num_attention_heads": 12,
|
| 26 |
+
"num_hidden_layers": 12,
|
| 27 |
+
"pad_token_id": 0,
|
| 28 |
+
"pooler_dropout": 0,
|
| 29 |
+
"pooler_hidden_act": "gelu",
|
| 30 |
+
"pooler_hidden_size": 768,
|
| 31 |
+
"pos_att_type": [
|
| 32 |
+
"p2c",
|
| 33 |
+
"c2p"
|
| 34 |
+
],
|
| 35 |
+
"position_biased_input": false,
|
| 36 |
+
"position_buckets": 256,
|
| 37 |
+
"relative_attention": true,
|
| 38 |
+
"share_att_key": true,
|
| 39 |
+
"torch_dtype": "float32",
|
| 40 |
+
"transformers_version": "4.44.0",
|
| 41 |
+
"type_vocab_size": 0,
|
| 42 |
+
"vocab_size": 128100
|
| 43 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8c73e0415b6ca299276e7779bafbdd5bc8be95ee6828305dcbb0aa08a777efc8
|
| 3 |
+
size 737719272
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3da562737df82adf0aab4d3dc3efc8b0f854841225923106f98f22722d6ea6bf
|
| 3 |
+
size 1475558394
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:68d4614bbf939a32ad50791f3b203eba00870c83c9d6e581d47f093a9824950f
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7e0da825d895379b3dbe151f5ffb44bc9609493cccd8b8e47d2c10042ade49ce
|
| 3 |
+
size 1064
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "[CLS]",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"cls_token": {
|
| 10 |
+
"content": "[CLS]",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"eos_token": {
|
| 17 |
+
"content": "[SEP]",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"mask_token": {
|
| 24 |
+
"content": "[MASK]",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 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": "[SEP]",
|
| 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": true,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false
|
| 50 |
+
}
|
| 51 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "[CLS]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "[SEP]",
|
| 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": true,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"128000": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"bos_token": "[CLS]",
|
| 45 |
+
"clean_up_tokenization_spaces": true,
|
| 46 |
+
"cls_token": "[CLS]",
|
| 47 |
+
"do_lower_case": false,
|
| 48 |
+
"eos_token": "[SEP]",
|
| 49 |
+
"mask_token": "[MASK]",
|
| 50 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 51 |
+
"pad_token": "[PAD]",
|
| 52 |
+
"sep_token": "[SEP]",
|
| 53 |
+
"sp_model_kwargs": {},
|
| 54 |
+
"split_by_punct": false,
|
| 55 |
+
"tokenizer_class": "DebertaV2Tokenizer",
|
| 56 |
+
"unk_token": "[UNK]",
|
| 57 |
+
"vocab_type": "spm"
|
| 58 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1153 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9716312056737588,
|
| 3 |
+
"best_model_checkpoint": "./results/checkpoint-3807",
|
| 4 |
+
"epoch": 70.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 5670,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"grad_norm": 6.570446014404297,
|
| 14 |
+
"learning_rate": 1.9728395061728395e-05,
|
| 15 |
+
"loss": 2.6389,
|
| 16 |
+
"step": 81
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 1.0,
|
| 20 |
+
"eval_accuracy": 0.475177304964539,
|
| 21 |
+
"eval_loss": 0.7098350524902344,
|
| 22 |
+
"eval_runtime": 0.3441,
|
| 23 |
+
"eval_samples_per_second": 819.478,
|
| 24 |
+
"eval_steps_per_second": 52.307,
|
| 25 |
+
"step": 81
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 2.0,
|
| 29 |
+
"grad_norm": 4.623419284820557,
|
| 30 |
+
"learning_rate": 1.944268077601411e-05,
|
| 31 |
+
"loss": 0.6477,
|
| 32 |
+
"step": 162
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 2.0,
|
| 36 |
+
"eval_accuracy": 0.48226950354609927,
|
| 37 |
+
"eval_loss": 0.7516428828239441,
|
| 38 |
+
"eval_runtime": 0.3434,
|
| 39 |
+
"eval_samples_per_second": 821.083,
|
| 40 |
+
"eval_steps_per_second": 52.41,
|
| 41 |
+
"step": 162
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 3.0,
|
| 45 |
+
"grad_norm": 10.926794052124023,
|
| 46 |
+
"learning_rate": 1.9156966490299824e-05,
|
| 47 |
+
"loss": 0.6227,
|
| 48 |
+
"step": 243
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 3.0,
|
| 52 |
+
"eval_accuracy": 0.4929078014184397,
|
| 53 |
+
"eval_loss": 0.8317187428474426,
|
| 54 |
+
"eval_runtime": 0.3439,
|
| 55 |
+
"eval_samples_per_second": 820.006,
|
| 56 |
+
"eval_steps_per_second": 52.341,
|
| 57 |
+
"step": 243
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 4.0,
|
| 61 |
+
"grad_norm": 12.648384094238281,
|
| 62 |
+
"learning_rate": 1.887125220458554e-05,
|
| 63 |
+
"loss": 0.5403,
|
| 64 |
+
"step": 324
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 4.0,
|
| 68 |
+
"eval_accuracy": 0.4929078014184397,
|
| 69 |
+
"eval_loss": 1.9380121231079102,
|
| 70 |
+
"eval_runtime": 0.3424,
|
| 71 |
+
"eval_samples_per_second": 823.561,
|
| 72 |
+
"eval_steps_per_second": 52.568,
|
| 73 |
+
"step": 324
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 5.0,
|
| 77 |
+
"grad_norm": 23.567258834838867,
|
| 78 |
+
"learning_rate": 1.8585537918871256e-05,
|
| 79 |
+
"loss": 0.5108,
|
| 80 |
+
"step": 405
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 5.0,
|
| 84 |
+
"eval_accuracy": 0.49645390070921985,
|
| 85 |
+
"eval_loss": 2.270359754562378,
|
| 86 |
+
"eval_runtime": 0.3437,
|
| 87 |
+
"eval_samples_per_second": 820.538,
|
| 88 |
+
"eval_steps_per_second": 52.375,
|
| 89 |
+
"step": 405
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 6.0,
|
| 93 |
+
"grad_norm": 3.5719075202941895,
|
| 94 |
+
"learning_rate": 1.830335097001764e-05,
|
| 95 |
+
"loss": 0.4677,
|
| 96 |
+
"step": 486
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 6.0,
|
| 100 |
+
"eval_accuracy": 0.48936170212765956,
|
| 101 |
+
"eval_loss": 1.6858181953430176,
|
| 102 |
+
"eval_runtime": 0.3432,
|
| 103 |
+
"eval_samples_per_second": 821.693,
|
| 104 |
+
"eval_steps_per_second": 52.449,
|
| 105 |
+
"step": 486
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 7.0,
|
| 109 |
+
"grad_norm": 7.08165168762207,
|
| 110 |
+
"learning_rate": 1.8017636684303353e-05,
|
| 111 |
+
"loss": 0.4798,
|
| 112 |
+
"step": 567
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 7.0,
|
| 116 |
+
"eval_accuracy": 0.49645390070921985,
|
| 117 |
+
"eval_loss": 1.623734712600708,
|
| 118 |
+
"eval_runtime": 0.3436,
|
| 119 |
+
"eval_samples_per_second": 820.682,
|
| 120 |
+
"eval_steps_per_second": 52.384,
|
| 121 |
+
"step": 567
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 8.0,
|
| 125 |
+
"grad_norm": 10.894269943237305,
|
| 126 |
+
"learning_rate": 1.773192239858907e-05,
|
| 127 |
+
"loss": 0.4817,
|
| 128 |
+
"step": 648
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 8.0,
|
| 132 |
+
"eval_accuracy": 0.5141843971631206,
|
| 133 |
+
"eval_loss": 1.3935478925704956,
|
| 134 |
+
"eval_runtime": 0.3435,
|
| 135 |
+
"eval_samples_per_second": 821.029,
|
| 136 |
+
"eval_steps_per_second": 52.406,
|
| 137 |
+
"step": 648
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 9.0,
|
| 141 |
+
"grad_norm": 7.739453315734863,
|
| 142 |
+
"learning_rate": 1.744620811287478e-05,
|
| 143 |
+
"loss": 0.4668,
|
| 144 |
+
"step": 729
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 9.0,
|
| 148 |
+
"eval_accuracy": 0.5177304964539007,
|
| 149 |
+
"eval_loss": 1.259345531463623,
|
| 150 |
+
"eval_runtime": 0.343,
|
| 151 |
+
"eval_samples_per_second": 822.053,
|
| 152 |
+
"eval_steps_per_second": 52.471,
|
| 153 |
+
"step": 729
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 10.0,
|
| 157 |
+
"grad_norm": 17.012800216674805,
|
| 158 |
+
"learning_rate": 1.7160493827160498e-05,
|
| 159 |
+
"loss": 0.4359,
|
| 160 |
+
"step": 810
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 10.0,
|
| 164 |
+
"eval_accuracy": 0.5354609929078015,
|
| 165 |
+
"eval_loss": 1.310729742050171,
|
| 166 |
+
"eval_runtime": 0.3436,
|
| 167 |
+
"eval_samples_per_second": 820.694,
|
| 168 |
+
"eval_steps_per_second": 52.385,
|
| 169 |
+
"step": 810
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 11.0,
|
| 173 |
+
"grad_norm": 1.6642764806747437,
|
| 174 |
+
"learning_rate": 1.687477954144621e-05,
|
| 175 |
+
"loss": 0.3956,
|
| 176 |
+
"step": 891
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 11.0,
|
| 180 |
+
"eval_accuracy": 0.8226950354609929,
|
| 181 |
+
"eval_loss": 0.43421775102615356,
|
| 182 |
+
"eval_runtime": 0.3435,
|
| 183 |
+
"eval_samples_per_second": 820.982,
|
| 184 |
+
"eval_steps_per_second": 52.403,
|
| 185 |
+
"step": 891
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 12.0,
|
| 189 |
+
"grad_norm": 0.3688388168811798,
|
| 190 |
+
"learning_rate": 1.6589065255731923e-05,
|
| 191 |
+
"loss": 0.2906,
|
| 192 |
+
"step": 972
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 12.0,
|
| 196 |
+
"eval_accuracy": 0.9290780141843972,
|
| 197 |
+
"eval_loss": 0.23947754502296448,
|
| 198 |
+
"eval_runtime": 0.3442,
|
| 199 |
+
"eval_samples_per_second": 819.25,
|
| 200 |
+
"eval_steps_per_second": 52.293,
|
| 201 |
+
"step": 972
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 13.0,
|
| 205 |
+
"grad_norm": 37.02349853515625,
|
| 206 |
+
"learning_rate": 1.630335097001764e-05,
|
| 207 |
+
"loss": 0.2146,
|
| 208 |
+
"step": 1053
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 13.0,
|
| 212 |
+
"eval_accuracy": 0.9397163120567376,
|
| 213 |
+
"eval_loss": 0.33284759521484375,
|
| 214 |
+
"eval_runtime": 0.3437,
|
| 215 |
+
"eval_samples_per_second": 820.462,
|
| 216 |
+
"eval_steps_per_second": 52.37,
|
| 217 |
+
"step": 1053
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 14.0,
|
| 221 |
+
"grad_norm": 0.11939908564090729,
|
| 222 |
+
"learning_rate": 1.601763668430335e-05,
|
| 223 |
+
"loss": 0.1462,
|
| 224 |
+
"step": 1134
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 14.0,
|
| 228 |
+
"eval_accuracy": 0.950354609929078,
|
| 229 |
+
"eval_loss": 0.3009294867515564,
|
| 230 |
+
"eval_runtime": 0.3439,
|
| 231 |
+
"eval_samples_per_second": 819.984,
|
| 232 |
+
"eval_steps_per_second": 52.339,
|
| 233 |
+
"step": 1134
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 15.0,
|
| 237 |
+
"grad_norm": 0.08733003586530685,
|
| 238 |
+
"learning_rate": 1.5731922398589064e-05,
|
| 239 |
+
"loss": 0.1062,
|
| 240 |
+
"step": 1215
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 15.0,
|
| 244 |
+
"eval_accuracy": 0.9290780141843972,
|
| 245 |
+
"eval_loss": 0.21407951414585114,
|
| 246 |
+
"eval_runtime": 0.3436,
|
| 247 |
+
"eval_samples_per_second": 820.805,
|
| 248 |
+
"eval_steps_per_second": 52.392,
|
| 249 |
+
"step": 1215
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 16.0,
|
| 253 |
+
"grad_norm": 0.21886540949344635,
|
| 254 |
+
"learning_rate": 1.544620811287478e-05,
|
| 255 |
+
"loss": 0.0813,
|
| 256 |
+
"step": 1296
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 16.0,
|
| 260 |
+
"eval_accuracy": 0.9432624113475178,
|
| 261 |
+
"eval_loss": 0.34917283058166504,
|
| 262 |
+
"eval_runtime": 0.344,
|
| 263 |
+
"eval_samples_per_second": 819.711,
|
| 264 |
+
"eval_steps_per_second": 52.322,
|
| 265 |
+
"step": 1296
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 17.0,
|
| 269 |
+
"grad_norm": 0.5847246646881104,
|
| 270 |
+
"learning_rate": 1.5160493827160495e-05,
|
| 271 |
+
"loss": 0.1027,
|
| 272 |
+
"step": 1377
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 17.0,
|
| 276 |
+
"eval_accuracy": 0.9219858156028369,
|
| 277 |
+
"eval_loss": 0.3432806432247162,
|
| 278 |
+
"eval_runtime": 0.3446,
|
| 279 |
+
"eval_samples_per_second": 818.425,
|
| 280 |
+
"eval_steps_per_second": 52.24,
|
| 281 |
+
"step": 1377
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 18.0,
|
| 285 |
+
"grad_norm": 0.6198065280914307,
|
| 286 |
+
"learning_rate": 1.4874779541446209e-05,
|
| 287 |
+
"loss": 0.0736,
|
| 288 |
+
"step": 1458
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 18.0,
|
| 292 |
+
"eval_accuracy": 0.9539007092198581,
|
| 293 |
+
"eval_loss": 0.27183273434638977,
|
| 294 |
+
"eval_runtime": 0.3437,
|
| 295 |
+
"eval_samples_per_second": 820.405,
|
| 296 |
+
"eval_steps_per_second": 52.366,
|
| 297 |
+
"step": 1458
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 19.0,
|
| 301 |
+
"grad_norm": 0.5257266163825989,
|
| 302 |
+
"learning_rate": 1.4589065255731925e-05,
|
| 303 |
+
"loss": 0.0684,
|
| 304 |
+
"step": 1539
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 19.0,
|
| 308 |
+
"eval_accuracy": 0.9645390070921985,
|
| 309 |
+
"eval_loss": 0.25684282183647156,
|
| 310 |
+
"eval_runtime": 0.3434,
|
| 311 |
+
"eval_samples_per_second": 821.157,
|
| 312 |
+
"eval_steps_per_second": 52.414,
|
| 313 |
+
"step": 1539
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 20.0,
|
| 317 |
+
"grad_norm": 0.0009818405378609896,
|
| 318 |
+
"learning_rate": 1.4303350970017638e-05,
|
| 319 |
+
"loss": 0.0779,
|
| 320 |
+
"step": 1620
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 20.0,
|
| 324 |
+
"eval_accuracy": 0.9609929078014184,
|
| 325 |
+
"eval_loss": 0.2152564525604248,
|
| 326 |
+
"eval_runtime": 0.3431,
|
| 327 |
+
"eval_samples_per_second": 821.93,
|
| 328 |
+
"eval_steps_per_second": 52.464,
|
| 329 |
+
"step": 1620
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 21.0,
|
| 333 |
+
"grad_norm": 0.4532203674316406,
|
| 334 |
+
"learning_rate": 1.4021164021164022e-05,
|
| 335 |
+
"loss": 0.0745,
|
| 336 |
+
"step": 1701
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 21.0,
|
| 340 |
+
"eval_accuracy": 0.9645390070921985,
|
| 341 |
+
"eval_loss": 0.1914406418800354,
|
| 342 |
+
"eval_runtime": 0.344,
|
| 343 |
+
"eval_samples_per_second": 819.813,
|
| 344 |
+
"eval_steps_per_second": 52.329,
|
| 345 |
+
"step": 1701
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 22.0,
|
| 349 |
+
"grad_norm": 17.428327560424805,
|
| 350 |
+
"learning_rate": 1.3735449735449738e-05,
|
| 351 |
+
"loss": 0.1106,
|
| 352 |
+
"step": 1782
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 22.0,
|
| 356 |
+
"eval_accuracy": 0.9574468085106383,
|
| 357 |
+
"eval_loss": 0.2807099223136902,
|
| 358 |
+
"eval_runtime": 0.3441,
|
| 359 |
+
"eval_samples_per_second": 819.457,
|
| 360 |
+
"eval_steps_per_second": 52.306,
|
| 361 |
+
"step": 1782
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 23.0,
|
| 365 |
+
"grad_norm": 0.00047796443686820567,
|
| 366 |
+
"learning_rate": 1.344973544973545e-05,
|
| 367 |
+
"loss": 0.0755,
|
| 368 |
+
"step": 1863
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 23.0,
|
| 372 |
+
"eval_accuracy": 0.9539007092198581,
|
| 373 |
+
"eval_loss": 0.331978976726532,
|
| 374 |
+
"eval_runtime": 0.3453,
|
| 375 |
+
"eval_samples_per_second": 816.672,
|
| 376 |
+
"eval_steps_per_second": 52.128,
|
| 377 |
+
"step": 1863
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 24.0,
|
| 381 |
+
"grad_norm": 1.006925106048584,
|
| 382 |
+
"learning_rate": 1.3164021164021166e-05,
|
| 383 |
+
"loss": 0.0833,
|
| 384 |
+
"step": 1944
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 24.0,
|
| 388 |
+
"eval_accuracy": 0.9539007092198581,
|
| 389 |
+
"eval_loss": 0.34625303745269775,
|
| 390 |
+
"eval_runtime": 0.3436,
|
| 391 |
+
"eval_samples_per_second": 820.661,
|
| 392 |
+
"eval_steps_per_second": 52.383,
|
| 393 |
+
"step": 1944
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 25.0,
|
| 397 |
+
"grad_norm": 0.506279706954956,
|
| 398 |
+
"learning_rate": 1.288183421516755e-05,
|
| 399 |
+
"loss": 0.0754,
|
| 400 |
+
"step": 2025
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 25.0,
|
| 404 |
+
"eval_accuracy": 0.9432624113475178,
|
| 405 |
+
"eval_loss": 0.34365448355674744,
|
| 406 |
+
"eval_runtime": 0.3432,
|
| 407 |
+
"eval_samples_per_second": 821.691,
|
| 408 |
+
"eval_steps_per_second": 52.448,
|
| 409 |
+
"step": 2025
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 26.0,
|
| 413 |
+
"grad_norm": 0.1998976171016693,
|
| 414 |
+
"learning_rate": 1.2596119929453263e-05,
|
| 415 |
+
"loss": 0.0772,
|
| 416 |
+
"step": 2106
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 26.0,
|
| 420 |
+
"eval_accuracy": 0.950354609929078,
|
| 421 |
+
"eval_loss": 0.3350883424282074,
|
| 422 |
+
"eval_runtime": 0.3435,
|
| 423 |
+
"eval_samples_per_second": 820.852,
|
| 424 |
+
"eval_steps_per_second": 52.395,
|
| 425 |
+
"step": 2106
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 27.0,
|
| 429 |
+
"grad_norm": 0.19478876888751984,
|
| 430 |
+
"learning_rate": 1.2310405643738979e-05,
|
| 431 |
+
"loss": 0.076,
|
| 432 |
+
"step": 2187
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 27.0,
|
| 436 |
+
"eval_accuracy": 0.9468085106382979,
|
| 437 |
+
"eval_loss": 0.4145265519618988,
|
| 438 |
+
"eval_runtime": 0.3445,
|
| 439 |
+
"eval_samples_per_second": 818.483,
|
| 440 |
+
"eval_steps_per_second": 52.244,
|
| 441 |
+
"step": 2187
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 28.0,
|
| 445 |
+
"grad_norm": 0.27469512820243835,
|
| 446 |
+
"learning_rate": 1.2024691358024691e-05,
|
| 447 |
+
"loss": 0.0625,
|
| 448 |
+
"step": 2268
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 28.0,
|
| 452 |
+
"eval_accuracy": 0.950354609929078,
|
| 453 |
+
"eval_loss": 0.44451093673706055,
|
| 454 |
+
"eval_runtime": 0.3439,
|
| 455 |
+
"eval_samples_per_second": 819.913,
|
| 456 |
+
"eval_steps_per_second": 52.335,
|
| 457 |
+
"step": 2268
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 29.0,
|
| 461 |
+
"grad_norm": 26.14291000366211,
|
| 462 |
+
"learning_rate": 1.1738977072310408e-05,
|
| 463 |
+
"loss": 0.0741,
|
| 464 |
+
"step": 2349
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 29.0,
|
| 468 |
+
"eval_accuracy": 0.9468085106382979,
|
| 469 |
+
"eval_loss": 0.29801085591316223,
|
| 470 |
+
"eval_runtime": 0.3448,
|
| 471 |
+
"eval_samples_per_second": 817.812,
|
| 472 |
+
"eval_steps_per_second": 52.201,
|
| 473 |
+
"step": 2349
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 30.0,
|
| 477 |
+
"grad_norm": 0.0004499799106270075,
|
| 478 |
+
"learning_rate": 1.145326278659612e-05,
|
| 479 |
+
"loss": 0.0649,
|
| 480 |
+
"step": 2430
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 30.0,
|
| 484 |
+
"eval_accuracy": 0.9574468085106383,
|
| 485 |
+
"eval_loss": 0.28359255194664,
|
| 486 |
+
"eval_runtime": 0.3442,
|
| 487 |
+
"eval_samples_per_second": 819.247,
|
| 488 |
+
"eval_steps_per_second": 52.292,
|
| 489 |
+
"step": 2430
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 31.0,
|
| 493 |
+
"grad_norm": 0.0018564946949481964,
|
| 494 |
+
"learning_rate": 1.1167548500881835e-05,
|
| 495 |
+
"loss": 0.0688,
|
| 496 |
+
"step": 2511
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 31.0,
|
| 500 |
+
"eval_accuracy": 0.9574468085106383,
|
| 501 |
+
"eval_loss": 0.21793903410434723,
|
| 502 |
+
"eval_runtime": 0.3445,
|
| 503 |
+
"eval_samples_per_second": 818.498,
|
| 504 |
+
"eval_steps_per_second": 52.245,
|
| 505 |
+
"step": 2511
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 32.0,
|
| 509 |
+
"grad_norm": 0.0009469461510889232,
|
| 510 |
+
"learning_rate": 1.088183421516755e-05,
|
| 511 |
+
"loss": 0.0735,
|
| 512 |
+
"step": 2592
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 32.0,
|
| 516 |
+
"eval_accuracy": 0.9539007092198581,
|
| 517 |
+
"eval_loss": 0.22946923971176147,
|
| 518 |
+
"eval_runtime": 0.3449,
|
| 519 |
+
"eval_samples_per_second": 817.666,
|
| 520 |
+
"eval_steps_per_second": 52.191,
|
| 521 |
+
"step": 2592
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 33.0,
|
| 525 |
+
"grad_norm": 0.4778638184070587,
|
| 526 |
+
"learning_rate": 1.0596119929453263e-05,
|
| 527 |
+
"loss": 0.0648,
|
| 528 |
+
"step": 2673
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 33.0,
|
| 532 |
+
"eval_accuracy": 0.9468085106382979,
|
| 533 |
+
"eval_loss": 0.42410480976104736,
|
| 534 |
+
"eval_runtime": 0.3433,
|
| 535 |
+
"eval_samples_per_second": 821.406,
|
| 536 |
+
"eval_steps_per_second": 52.43,
|
| 537 |
+
"step": 2673
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 34.0,
|
| 541 |
+
"grad_norm": 0.21737487614154816,
|
| 542 |
+
"learning_rate": 1.031040564373898e-05,
|
| 543 |
+
"loss": 0.0672,
|
| 544 |
+
"step": 2754
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 34.0,
|
| 548 |
+
"eval_accuracy": 0.9539007092198581,
|
| 549 |
+
"eval_loss": 0.2829430401325226,
|
| 550 |
+
"eval_runtime": 0.3447,
|
| 551 |
+
"eval_samples_per_second": 818.124,
|
| 552 |
+
"eval_steps_per_second": 52.221,
|
| 553 |
+
"step": 2754
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 35.0,
|
| 557 |
+
"grad_norm": 0.08269879966974258,
|
| 558 |
+
"learning_rate": 1.0024691358024692e-05,
|
| 559 |
+
"loss": 0.067,
|
| 560 |
+
"step": 2835
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 35.0,
|
| 564 |
+
"eval_accuracy": 0.9468085106382979,
|
| 565 |
+
"eval_loss": 0.3723122179508209,
|
| 566 |
+
"eval_runtime": 0.3448,
|
| 567 |
+
"eval_samples_per_second": 817.778,
|
| 568 |
+
"eval_steps_per_second": 52.199,
|
| 569 |
+
"step": 2835
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 36.0,
|
| 573 |
+
"grad_norm": 0.3665499687194824,
|
| 574 |
+
"learning_rate": 9.738977072310406e-06,
|
| 575 |
+
"loss": 0.0768,
|
| 576 |
+
"step": 2916
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 36.0,
|
| 580 |
+
"eval_accuracy": 0.9574468085106383,
|
| 581 |
+
"eval_loss": 0.25441667437553406,
|
| 582 |
+
"eval_runtime": 0.3447,
|
| 583 |
+
"eval_samples_per_second": 818.182,
|
| 584 |
+
"eval_steps_per_second": 52.224,
|
| 585 |
+
"step": 2916
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 37.0,
|
| 589 |
+
"grad_norm": 0.11919476091861725,
|
| 590 |
+
"learning_rate": 9.45326278659612e-06,
|
| 591 |
+
"loss": 0.0691,
|
| 592 |
+
"step": 2997
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 37.0,
|
| 596 |
+
"eval_accuracy": 0.9609929078014184,
|
| 597 |
+
"eval_loss": 0.20481815934181213,
|
| 598 |
+
"eval_runtime": 0.3445,
|
| 599 |
+
"eval_samples_per_second": 818.558,
|
| 600 |
+
"eval_steps_per_second": 52.248,
|
| 601 |
+
"step": 2997
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 38.0,
|
| 605 |
+
"grad_norm": 0.0036801116075366735,
|
| 606 |
+
"learning_rate": 9.167548500881835e-06,
|
| 607 |
+
"loss": 0.0661,
|
| 608 |
+
"step": 3078
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 38.0,
|
| 612 |
+
"eval_accuracy": 0.9680851063829787,
|
| 613 |
+
"eval_loss": 0.20478524267673492,
|
| 614 |
+
"eval_runtime": 0.3445,
|
| 615 |
+
"eval_samples_per_second": 818.468,
|
| 616 |
+
"eval_steps_per_second": 52.243,
|
| 617 |
+
"step": 3078
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 39.0,
|
| 621 |
+
"grad_norm": 0.12663815915584564,
|
| 622 |
+
"learning_rate": 8.88183421516755e-06,
|
| 623 |
+
"loss": 0.0409,
|
| 624 |
+
"step": 3159
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 39.0,
|
| 628 |
+
"eval_accuracy": 0.9645390070921985,
|
| 629 |
+
"eval_loss": 0.18502239882946014,
|
| 630 |
+
"eval_runtime": 0.3434,
|
| 631 |
+
"eval_samples_per_second": 821.144,
|
| 632 |
+
"eval_steps_per_second": 52.413,
|
| 633 |
+
"step": 3159
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 40.0,
|
| 637 |
+
"grad_norm": 0.06950168311595917,
|
| 638 |
+
"learning_rate": 8.596119929453264e-06,
|
| 639 |
+
"loss": 0.0424,
|
| 640 |
+
"step": 3240
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 40.0,
|
| 644 |
+
"eval_accuracy": 0.9645390070921985,
|
| 645 |
+
"eval_loss": 0.20747074484825134,
|
| 646 |
+
"eval_runtime": 0.3445,
|
| 647 |
+
"eval_samples_per_second": 818.693,
|
| 648 |
+
"eval_steps_per_second": 52.257,
|
| 649 |
+
"step": 3240
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 41.0,
|
| 653 |
+
"grad_norm": 0.09251494705677032,
|
| 654 |
+
"learning_rate": 8.310405643738978e-06,
|
| 655 |
+
"loss": 0.0381,
|
| 656 |
+
"step": 3321
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 41.0,
|
| 660 |
+
"eval_accuracy": 0.9645390070921985,
|
| 661 |
+
"eval_loss": 0.2633875906467438,
|
| 662 |
+
"eval_runtime": 0.3468,
|
| 663 |
+
"eval_samples_per_second": 813.14,
|
| 664 |
+
"eval_steps_per_second": 51.903,
|
| 665 |
+
"step": 3321
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 42.0,
|
| 669 |
+
"grad_norm": 0.06917154043912888,
|
| 670 |
+
"learning_rate": 8.024691358024692e-06,
|
| 671 |
+
"loss": 0.0383,
|
| 672 |
+
"step": 3402
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 42.0,
|
| 676 |
+
"eval_accuracy": 0.9574468085106383,
|
| 677 |
+
"eval_loss": 0.3520617187023163,
|
| 678 |
+
"eval_runtime": 0.3447,
|
| 679 |
+
"eval_samples_per_second": 818.036,
|
| 680 |
+
"eval_steps_per_second": 52.215,
|
| 681 |
+
"step": 3402
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 43.0,
|
| 685 |
+
"grad_norm": 0.0010325413895770907,
|
| 686 |
+
"learning_rate": 7.738977072310407e-06,
|
| 687 |
+
"loss": 0.0288,
|
| 688 |
+
"step": 3483
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 43.0,
|
| 692 |
+
"eval_accuracy": 0.9680851063829787,
|
| 693 |
+
"eval_loss": 0.2726523280143738,
|
| 694 |
+
"eval_runtime": 0.3428,
|
| 695 |
+
"eval_samples_per_second": 822.588,
|
| 696 |
+
"eval_steps_per_second": 52.506,
|
| 697 |
+
"step": 3483
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 44.0,
|
| 701 |
+
"grad_norm": 0.04726780578494072,
|
| 702 |
+
"learning_rate": 7.45326278659612e-06,
|
| 703 |
+
"loss": 0.035,
|
| 704 |
+
"step": 3564
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 44.0,
|
| 708 |
+
"eval_accuracy": 0.9645390070921985,
|
| 709 |
+
"eval_loss": 0.2995310127735138,
|
| 710 |
+
"eval_runtime": 0.3442,
|
| 711 |
+
"eval_samples_per_second": 819.308,
|
| 712 |
+
"eval_steps_per_second": 52.296,
|
| 713 |
+
"step": 3564
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 45.0,
|
| 717 |
+
"grad_norm": 0.09283600747585297,
|
| 718 |
+
"learning_rate": 7.167548500881835e-06,
|
| 719 |
+
"loss": 0.0265,
|
| 720 |
+
"step": 3645
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 45.0,
|
| 724 |
+
"eval_accuracy": 0.9609929078014184,
|
| 725 |
+
"eval_loss": 0.33694958686828613,
|
| 726 |
+
"eval_runtime": 0.3443,
|
| 727 |
+
"eval_samples_per_second": 818.994,
|
| 728 |
+
"eval_steps_per_second": 52.276,
|
| 729 |
+
"step": 3645
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 46.0,
|
| 733 |
+
"grad_norm": 0.03685113787651062,
|
| 734 |
+
"learning_rate": 6.881834215167549e-06,
|
| 735 |
+
"loss": 0.0217,
|
| 736 |
+
"step": 3726
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 46.0,
|
| 740 |
+
"eval_accuracy": 0.9609929078014184,
|
| 741 |
+
"eval_loss": 0.35722091794013977,
|
| 742 |
+
"eval_runtime": 0.3438,
|
| 743 |
+
"eval_samples_per_second": 820.281,
|
| 744 |
+
"eval_steps_per_second": 52.358,
|
| 745 |
+
"step": 3726
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 47.0,
|
| 749 |
+
"grad_norm": 0.04708189144730568,
|
| 750 |
+
"learning_rate": 6.596119929453263e-06,
|
| 751 |
+
"loss": 0.0259,
|
| 752 |
+
"step": 3807
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 47.0,
|
| 756 |
+
"eval_accuracy": 0.9716312056737588,
|
| 757 |
+
"eval_loss": 0.21833930909633636,
|
| 758 |
+
"eval_runtime": 0.3427,
|
| 759 |
+
"eval_samples_per_second": 822.913,
|
| 760 |
+
"eval_steps_per_second": 52.526,
|
| 761 |
+
"step": 3807
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 48.0,
|
| 765 |
+
"grad_norm": 0.06329997628927231,
|
| 766 |
+
"learning_rate": 6.310405643738977e-06,
|
| 767 |
+
"loss": 0.0264,
|
| 768 |
+
"step": 3888
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 48.0,
|
| 772 |
+
"eval_accuracy": 0.9609929078014184,
|
| 773 |
+
"eval_loss": 0.2745024561882019,
|
| 774 |
+
"eval_runtime": 0.3436,
|
| 775 |
+
"eval_samples_per_second": 820.777,
|
| 776 |
+
"eval_steps_per_second": 52.39,
|
| 777 |
+
"step": 3888
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 49.0,
|
| 781 |
+
"grad_norm": 0.13020673394203186,
|
| 782 |
+
"learning_rate": 6.024691358024692e-06,
|
| 783 |
+
"loss": 0.027,
|
| 784 |
+
"step": 3969
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 49.0,
|
| 788 |
+
"eval_accuracy": 0.9539007092198581,
|
| 789 |
+
"eval_loss": 0.3425739109516144,
|
| 790 |
+
"eval_runtime": 0.3449,
|
| 791 |
+
"eval_samples_per_second": 817.548,
|
| 792 |
+
"eval_steps_per_second": 52.184,
|
| 793 |
+
"step": 3969
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 50.0,
|
| 797 |
+
"grad_norm": 0.04181819409132004,
|
| 798 |
+
"learning_rate": 5.7389770723104065e-06,
|
| 799 |
+
"loss": 0.023,
|
| 800 |
+
"step": 4050
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 50.0,
|
| 804 |
+
"eval_accuracy": 0.950354609929078,
|
| 805 |
+
"eval_loss": 0.37068530917167664,
|
| 806 |
+
"eval_runtime": 0.3441,
|
| 807 |
+
"eval_samples_per_second": 819.471,
|
| 808 |
+
"eval_steps_per_second": 52.307,
|
| 809 |
+
"step": 4050
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 51.0,
|
| 813 |
+
"grad_norm": 0.03754027560353279,
|
| 814 |
+
"learning_rate": 5.453262786596121e-06,
|
| 815 |
+
"loss": 0.0241,
|
| 816 |
+
"step": 4131
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 51.0,
|
| 820 |
+
"eval_accuracy": 0.9645390070921985,
|
| 821 |
+
"eval_loss": 0.3041815459728241,
|
| 822 |
+
"eval_runtime": 0.3443,
|
| 823 |
+
"eval_samples_per_second": 819.127,
|
| 824 |
+
"eval_steps_per_second": 52.285,
|
| 825 |
+
"step": 4131
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 52.0,
|
| 829 |
+
"grad_norm": 0.06724414229393005,
|
| 830 |
+
"learning_rate": 5.167548500881835e-06,
|
| 831 |
+
"loss": 0.0248,
|
| 832 |
+
"step": 4212
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 52.0,
|
| 836 |
+
"eval_accuracy": 0.9609929078014184,
|
| 837 |
+
"eval_loss": 0.3282240927219391,
|
| 838 |
+
"eval_runtime": 0.3433,
|
| 839 |
+
"eval_samples_per_second": 821.512,
|
| 840 |
+
"eval_steps_per_second": 52.437,
|
| 841 |
+
"step": 4212
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 53.0,
|
| 845 |
+
"grad_norm": 0.044111430644989014,
|
| 846 |
+
"learning_rate": 4.881834215167549e-06,
|
| 847 |
+
"loss": 0.0267,
|
| 848 |
+
"step": 4293
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 53.0,
|
| 852 |
+
"eval_accuracy": 0.9680851063829787,
|
| 853 |
+
"eval_loss": 0.2480100840330124,
|
| 854 |
+
"eval_runtime": 0.3438,
|
| 855 |
+
"eval_samples_per_second": 820.176,
|
| 856 |
+
"eval_steps_per_second": 52.352,
|
| 857 |
+
"step": 4293
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 54.0,
|
| 861 |
+
"grad_norm": 0.09385800361633301,
|
| 862 |
+
"learning_rate": 4.596119929453263e-06,
|
| 863 |
+
"loss": 0.019,
|
| 864 |
+
"step": 4374
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 54.0,
|
| 868 |
+
"eval_accuracy": 0.9680851063829787,
|
| 869 |
+
"eval_loss": 0.2954387366771698,
|
| 870 |
+
"eval_runtime": 0.3444,
|
| 871 |
+
"eval_samples_per_second": 818.748,
|
| 872 |
+
"eval_steps_per_second": 52.261,
|
| 873 |
+
"step": 4374
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 55.0,
|
| 877 |
+
"grad_norm": 0.00036285247188061476,
|
| 878 |
+
"learning_rate": 4.3104056437389775e-06,
|
| 879 |
+
"loss": 0.0233,
|
| 880 |
+
"step": 4455
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 55.0,
|
| 884 |
+
"eval_accuracy": 0.9645390070921985,
|
| 885 |
+
"eval_loss": 0.26300373673439026,
|
| 886 |
+
"eval_runtime": 0.3483,
|
| 887 |
+
"eval_samples_per_second": 809.563,
|
| 888 |
+
"eval_steps_per_second": 51.674,
|
| 889 |
+
"step": 4455
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 56.0,
|
| 893 |
+
"grad_norm": 0.03549063578248024,
|
| 894 |
+
"learning_rate": 4.024691358024692e-06,
|
| 895 |
+
"loss": 0.0231,
|
| 896 |
+
"step": 4536
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 56.0,
|
| 900 |
+
"eval_accuracy": 0.9645390070921985,
|
| 901 |
+
"eval_loss": 0.26614007353782654,
|
| 902 |
+
"eval_runtime": 0.3434,
|
| 903 |
+
"eval_samples_per_second": 821.294,
|
| 904 |
+
"eval_steps_per_second": 52.423,
|
| 905 |
+
"step": 4536
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 57.0,
|
| 909 |
+
"grad_norm": 0.0008688592351973057,
|
| 910 |
+
"learning_rate": 3.7389770723104058e-06,
|
| 911 |
+
"loss": 0.0188,
|
| 912 |
+
"step": 4617
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 57.0,
|
| 916 |
+
"eval_accuracy": 0.9574468085106383,
|
| 917 |
+
"eval_loss": 0.3676702678203583,
|
| 918 |
+
"eval_runtime": 0.3441,
|
| 919 |
+
"eval_samples_per_second": 819.514,
|
| 920 |
+
"eval_steps_per_second": 52.309,
|
| 921 |
+
"step": 4617
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 58.0,
|
| 925 |
+
"grad_norm": 0.00031407736241817474,
|
| 926 |
+
"learning_rate": 3.4532627865961205e-06,
|
| 927 |
+
"loss": 0.0263,
|
| 928 |
+
"step": 4698
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 58.0,
|
| 932 |
+
"eval_accuracy": 0.9539007092198581,
|
| 933 |
+
"eval_loss": 0.36925771832466125,
|
| 934 |
+
"eval_runtime": 0.348,
|
| 935 |
+
"eval_samples_per_second": 810.368,
|
| 936 |
+
"eval_steps_per_second": 51.726,
|
| 937 |
+
"step": 4698
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 59.0,
|
| 941 |
+
"grad_norm": 0.040128860622644424,
|
| 942 |
+
"learning_rate": 3.1675485008818345e-06,
|
| 943 |
+
"loss": 0.019,
|
| 944 |
+
"step": 4779
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 59.0,
|
| 948 |
+
"eval_accuracy": 0.9574468085106383,
|
| 949 |
+
"eval_loss": 0.35094693303108215,
|
| 950 |
+
"eval_runtime": 0.3436,
|
| 951 |
+
"eval_samples_per_second": 820.815,
|
| 952 |
+
"eval_steps_per_second": 52.392,
|
| 953 |
+
"step": 4779
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 60.0,
|
| 957 |
+
"grad_norm": 0.0004439246258698404,
|
| 958 |
+
"learning_rate": 2.881834215167549e-06,
|
| 959 |
+
"loss": 0.0202,
|
| 960 |
+
"step": 4860
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 60.0,
|
| 964 |
+
"eval_accuracy": 0.9609929078014184,
|
| 965 |
+
"eval_loss": 0.3040333092212677,
|
| 966 |
+
"eval_runtime": 0.3445,
|
| 967 |
+
"eval_samples_per_second": 818.559,
|
| 968 |
+
"eval_steps_per_second": 52.248,
|
| 969 |
+
"step": 4860
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 61.0,
|
| 973 |
+
"grad_norm": 0.07529360055923462,
|
| 974 |
+
"learning_rate": 2.5961199294532628e-06,
|
| 975 |
+
"loss": 0.0208,
|
| 976 |
+
"step": 4941
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 61.0,
|
| 980 |
+
"eval_accuracy": 0.9468085106382979,
|
| 981 |
+
"eval_loss": 0.5039365887641907,
|
| 982 |
+
"eval_runtime": 0.3439,
|
| 983 |
+
"eval_samples_per_second": 819.902,
|
| 984 |
+
"eval_steps_per_second": 52.334,
|
| 985 |
+
"step": 4941
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 62.0,
|
| 989 |
+
"grad_norm": 0.00026053638430312276,
|
| 990 |
+
"learning_rate": 2.310405643738977e-06,
|
| 991 |
+
"loss": 0.0242,
|
| 992 |
+
"step": 5022
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 62.0,
|
| 996 |
+
"eval_accuracy": 0.950354609929078,
|
| 997 |
+
"eval_loss": 0.4803861677646637,
|
| 998 |
+
"eval_runtime": 0.3445,
|
| 999 |
+
"eval_samples_per_second": 818.64,
|
| 1000 |
+
"eval_steps_per_second": 52.254,
|
| 1001 |
+
"step": 5022
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 63.0,
|
| 1005 |
+
"grad_norm": 0.06742388755083084,
|
| 1006 |
+
"learning_rate": 2.0246913580246915e-06,
|
| 1007 |
+
"loss": 0.023,
|
| 1008 |
+
"step": 5103
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 63.0,
|
| 1012 |
+
"eval_accuracy": 0.9609929078014184,
|
| 1013 |
+
"eval_loss": 0.3538144826889038,
|
| 1014 |
+
"eval_runtime": 0.3445,
|
| 1015 |
+
"eval_samples_per_second": 818.51,
|
| 1016 |
+
"eval_steps_per_second": 52.245,
|
| 1017 |
+
"step": 5103
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 64.0,
|
| 1021 |
+
"grad_norm": 0.00042550539365038276,
|
| 1022 |
+
"learning_rate": 1.7389770723104056e-06,
|
| 1023 |
+
"loss": 0.0189,
|
| 1024 |
+
"step": 5184
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 64.0,
|
| 1028 |
+
"eval_accuracy": 0.9574468085106383,
|
| 1029 |
+
"eval_loss": 0.37617096304893494,
|
| 1030 |
+
"eval_runtime": 0.3442,
|
| 1031 |
+
"eval_samples_per_second": 819.198,
|
| 1032 |
+
"eval_steps_per_second": 52.289,
|
| 1033 |
+
"step": 5184
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 65.0,
|
| 1037 |
+
"grad_norm": 0.02407378889620304,
|
| 1038 |
+
"learning_rate": 1.45326278659612e-06,
|
| 1039 |
+
"loss": 0.0209,
|
| 1040 |
+
"step": 5265
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 65.0,
|
| 1044 |
+
"eval_accuracy": 0.950354609929078,
|
| 1045 |
+
"eval_loss": 0.43608424067497253,
|
| 1046 |
+
"eval_runtime": 0.3438,
|
| 1047 |
+
"eval_samples_per_second": 820.243,
|
| 1048 |
+
"eval_steps_per_second": 52.356,
|
| 1049 |
+
"step": 5265
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 66.0,
|
| 1053 |
+
"grad_norm": 0.054311446845531464,
|
| 1054 |
+
"learning_rate": 1.1675485008818344e-06,
|
| 1055 |
+
"loss": 0.0209,
|
| 1056 |
+
"step": 5346
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 66.0,
|
| 1060 |
+
"eval_accuracy": 0.950354609929078,
|
| 1061 |
+
"eval_loss": 0.41794532537460327,
|
| 1062 |
+
"eval_runtime": 0.3436,
|
| 1063 |
+
"eval_samples_per_second": 820.791,
|
| 1064 |
+
"eval_steps_per_second": 52.391,
|
| 1065 |
+
"step": 5346
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 67.0,
|
| 1069 |
+
"grad_norm": 0.04109662398695946,
|
| 1070 |
+
"learning_rate": 8.818342151675485e-07,
|
| 1071 |
+
"loss": 0.0198,
|
| 1072 |
+
"step": 5427
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 67.0,
|
| 1076 |
+
"eval_accuracy": 0.9539007092198581,
|
| 1077 |
+
"eval_loss": 0.3815895617008209,
|
| 1078 |
+
"eval_runtime": 0.3443,
|
| 1079 |
+
"eval_samples_per_second": 819.013,
|
| 1080 |
+
"eval_steps_per_second": 52.277,
|
| 1081 |
+
"step": 5427
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 68.0,
|
| 1085 |
+
"grad_norm": 0.13629287481307983,
|
| 1086 |
+
"learning_rate": 5.961199294532629e-07,
|
| 1087 |
+
"loss": 0.0197,
|
| 1088 |
+
"step": 5508
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 68.0,
|
| 1092 |
+
"eval_accuracy": 0.950354609929078,
|
| 1093 |
+
"eval_loss": 0.39786896109580994,
|
| 1094 |
+
"eval_runtime": 0.3445,
|
| 1095 |
+
"eval_samples_per_second": 818.46,
|
| 1096 |
+
"eval_steps_per_second": 52.242,
|
| 1097 |
+
"step": 5508
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 69.0,
|
| 1101 |
+
"grad_norm": 0.039983708411455154,
|
| 1102 |
+
"learning_rate": 3.104056437389771e-07,
|
| 1103 |
+
"loss": 0.0192,
|
| 1104 |
+
"step": 5589
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 69.0,
|
| 1108 |
+
"eval_accuracy": 0.950354609929078,
|
| 1109 |
+
"eval_loss": 0.411296546459198,
|
| 1110 |
+
"eval_runtime": 0.3435,
|
| 1111 |
+
"eval_samples_per_second": 820.901,
|
| 1112 |
+
"eval_steps_per_second": 52.398,
|
| 1113 |
+
"step": 5589
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 70.0,
|
| 1117 |
+
"grad_norm": 0.00027353325276635587,
|
| 1118 |
+
"learning_rate": 2.469135802469136e-08,
|
| 1119 |
+
"loss": 0.0177,
|
| 1120 |
+
"step": 5670
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 70.0,
|
| 1124 |
+
"eval_accuracy": 0.9539007092198581,
|
| 1125 |
+
"eval_loss": 0.40772485733032227,
|
| 1126 |
+
"eval_runtime": 0.3437,
|
| 1127 |
+
"eval_samples_per_second": 820.466,
|
| 1128 |
+
"eval_steps_per_second": 52.37,
|
| 1129 |
+
"step": 5670
|
| 1130 |
+
}
|
| 1131 |
+
],
|
| 1132 |
+
"logging_steps": 100,
|
| 1133 |
+
"max_steps": 5670,
|
| 1134 |
+
"num_input_tokens_seen": 0,
|
| 1135 |
+
"num_train_epochs": 70,
|
| 1136 |
+
"save_steps": 500,
|
| 1137 |
+
"stateful_callbacks": {
|
| 1138 |
+
"TrainerControl": {
|
| 1139 |
+
"args": {
|
| 1140 |
+
"should_epoch_stop": false,
|
| 1141 |
+
"should_evaluate": false,
|
| 1142 |
+
"should_log": false,
|
| 1143 |
+
"should_save": true,
|
| 1144 |
+
"should_training_stop": true
|
| 1145 |
+
},
|
| 1146 |
+
"attributes": {}
|
| 1147 |
+
}
|
| 1148 |
+
},
|
| 1149 |
+
"total_flos": 9735501528974304.0,
|
| 1150 |
+
"train_batch_size": 16,
|
| 1151 |
+
"trial_name": null,
|
| 1152 |
+
"trial_params": null
|
| 1153 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:122d1be97ec5b07fbecea5e4a86ddec4f045fe64d647dec340002d311d768f01
|
| 3 |
+
size 5112
|