Training in progress, step 178, checkpoint
Browse files- checkpoint-178/1_Pooling/config.json +10 -0
- checkpoint-178/README.md +0 -0
- checkpoint-178/added_tokens.json +3 -0
- checkpoint-178/config.json +35 -0
- checkpoint-178/config_sentence_transformers.json +10 -0
- checkpoint-178/modules.json +14 -0
- checkpoint-178/optimizer.pt +3 -0
- checkpoint-178/pytorch_model.bin +3 -0
- checkpoint-178/rng_state.pth +3 -0
- checkpoint-178/scheduler.pt +3 -0
- checkpoint-178/sentence_bert_config.json +4 -0
- checkpoint-178/special_tokens_map.json +51 -0
- checkpoint-178/spm.model +3 -0
- checkpoint-178/tokenizer.json +0 -0
- checkpoint-178/tokenizer_config.json +65 -0
- checkpoint-178/trainer_state.json +1729 -0
- checkpoint-178/training_args.bin +3 -0
checkpoint-178/1_Pooling/config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"word_embedding_dimension": 768,
|
| 3 |
+
"pooling_mode_cls_token": false,
|
| 4 |
+
"pooling_mode_mean_tokens": true,
|
| 5 |
+
"pooling_mode_max_tokens": false,
|
| 6 |
+
"pooling_mode_mean_sqrt_len_tokens": false,
|
| 7 |
+
"pooling_mode_weightedmean_tokens": false,
|
| 8 |
+
"pooling_mode_lasttoken": false,
|
| 9 |
+
"include_prompt": true
|
| 10 |
+
}
|
checkpoint-178/README.md
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-178/added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"[MASK]": 128000
|
| 3 |
+
}
|
checkpoint-178/config.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "bobox/DeBERTa-small-ST-v1-test-step2",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"DebertaV2Model"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"hidden_act": "gelu",
|
| 8 |
+
"hidden_dropout_prob": 0.1,
|
| 9 |
+
"hidden_size": 768,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"intermediate_size": 3072,
|
| 12 |
+
"layer_norm_eps": 1e-07,
|
| 13 |
+
"max_position_embeddings": 512,
|
| 14 |
+
"max_relative_positions": -1,
|
| 15 |
+
"model_type": "deberta-v2",
|
| 16 |
+
"norm_rel_ebd": "layer_norm",
|
| 17 |
+
"num_attention_heads": 12,
|
| 18 |
+
"num_hidden_layers": 6,
|
| 19 |
+
"pad_token_id": 0,
|
| 20 |
+
"pooler_dropout": 0,
|
| 21 |
+
"pooler_hidden_act": "gelu",
|
| 22 |
+
"pooler_hidden_size": 768,
|
| 23 |
+
"pos_att_type": [
|
| 24 |
+
"p2c",
|
| 25 |
+
"c2p"
|
| 26 |
+
],
|
| 27 |
+
"position_biased_input": false,
|
| 28 |
+
"position_buckets": 256,
|
| 29 |
+
"relative_attention": true,
|
| 30 |
+
"share_att_key": true,
|
| 31 |
+
"torch_dtype": "float32",
|
| 32 |
+
"transformers_version": "4.44.0",
|
| 33 |
+
"type_vocab_size": 0,
|
| 34 |
+
"vocab_size": 128100
|
| 35 |
+
}
|
checkpoint-178/config_sentence_transformers.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"__version__": {
|
| 3 |
+
"sentence_transformers": "3.0.1",
|
| 4 |
+
"transformers": "4.44.0",
|
| 5 |
+
"pytorch": "2.4.0"
|
| 6 |
+
},
|
| 7 |
+
"prompts": {},
|
| 8 |
+
"default_prompt_name": null,
|
| 9 |
+
"similarity_fn_name": null
|
| 10 |
+
}
|
checkpoint-178/modules.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"idx": 0,
|
| 4 |
+
"name": "0",
|
| 5 |
+
"path": "",
|
| 6 |
+
"type": "sentence_transformers.models.Transformer"
|
| 7 |
+
},
|
| 8 |
+
{
|
| 9 |
+
"idx": 1,
|
| 10 |
+
"name": "1",
|
| 11 |
+
"path": "1_Pooling",
|
| 12 |
+
"type": "sentence_transformers.models.Pooling"
|
| 13 |
+
}
|
| 14 |
+
]
|
checkpoint-178/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6b022738bddc1b2af63d75b1e4d63e5c863ca67d1a70ae786b10917cf4931b37
|
| 3 |
+
size 1130520122
|
checkpoint-178/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:05cf53e50959ea4e585e3fdbb6e6809f56f1462e961349ac4d4d54d6b482b030
|
| 3 |
+
size 565251810
|
checkpoint-178/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9948fc2bcfd705e4d7bb7483fbb1647f9ba68be2e81523048d9fcc4d4c70280d
|
| 3 |
+
size 14244
|
checkpoint-178/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:91b0595725ac19c56c56ccca1ef28743f765072da48e4604839e5cc1c3ddb334
|
| 3 |
+
size 1064
|
checkpoint-178/sentence_bert_config.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"max_seq_length": 512,
|
| 3 |
+
"do_lower_case": false
|
| 4 |
+
}
|
checkpoint-178/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 |
+
}
|
checkpoint-178/spm.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c679fbf93643d19aab7ee10c0b99e460bdbc02fedf34b92b05af343b4af586fd
|
| 3 |
+
size 2464616
|
checkpoint-178/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-178/tokenizer_config.json
ADDED
|
@@ -0,0 +1,65 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"max_length": 512,
|
| 51 |
+
"model_max_length": 512,
|
| 52 |
+
"pad_to_multiple_of": null,
|
| 53 |
+
"pad_token": "[PAD]",
|
| 54 |
+
"pad_token_type_id": 0,
|
| 55 |
+
"padding_side": "right",
|
| 56 |
+
"sep_token": "[SEP]",
|
| 57 |
+
"sp_model_kwargs": {},
|
| 58 |
+
"split_by_punct": false,
|
| 59 |
+
"stride": 0,
|
| 60 |
+
"tokenizer_class": "DebertaV2Tokenizer",
|
| 61 |
+
"truncation_side": "right",
|
| 62 |
+
"truncation_strategy": "longest_first",
|
| 63 |
+
"unk_token": "[UNK]",
|
| 64 |
+
"vocab_type": "spm"
|
| 65 |
+
}
|
checkpoint-178/trainer_state.json
ADDED
|
@@ -0,0 +1,1729 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.4,
|
| 5 |
+
"eval_steps": 45,
|
| 6 |
+
"global_step": 178,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0022471910112359553,
|
| 13 |
+
"grad_norm": 3.9492883682250977,
|
| 14 |
+
"learning_rate": 1.9662921348314604e-07,
|
| 15 |
+
"loss": 0.8103,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0044943820224719105,
|
| 20 |
+
"grad_norm": 4.117438793182373,
|
| 21 |
+
"learning_rate": 3.932584269662921e-07,
|
| 22 |
+
"loss": 0.8803,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.006741573033707865,
|
| 27 |
+
"grad_norm": 3.809002161026001,
|
| 28 |
+
"learning_rate": 5.898876404494381e-07,
|
| 29 |
+
"loss": 0.8219,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.008988764044943821,
|
| 34 |
+
"grad_norm": 0.7417504787445068,
|
| 35 |
+
"learning_rate": 7.865168539325842e-07,
|
| 36 |
+
"loss": 0.0574,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.011235955056179775,
|
| 41 |
+
"grad_norm": 2.707460403442383,
|
| 42 |
+
"learning_rate": 9.831460674157302e-07,
|
| 43 |
+
"loss": 0.3044,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.01348314606741573,
|
| 48 |
+
"grad_norm": 3.082705497741699,
|
| 49 |
+
"learning_rate": 1.1797752808988763e-06,
|
| 50 |
+
"loss": 0.3306,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.015730337078651686,
|
| 55 |
+
"grad_norm": 3.102416753768921,
|
| 56 |
+
"learning_rate": 1.3764044943820223e-06,
|
| 57 |
+
"loss": 0.759,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.017977528089887642,
|
| 62 |
+
"grad_norm": 0.6271047592163086,
|
| 63 |
+
"learning_rate": 1.5730337078651683e-06,
|
| 64 |
+
"loss": 0.0472,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.020224719101123594,
|
| 69 |
+
"grad_norm": 3.1362593173980713,
|
| 70 |
+
"learning_rate": 1.7696629213483144e-06,
|
| 71 |
+
"loss": 0.7782,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.02247191011235955,
|
| 76 |
+
"grad_norm": 1.124997615814209,
|
| 77 |
+
"learning_rate": 1.9662921348314604e-06,
|
| 78 |
+
"loss": 0.0757,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.024719101123595506,
|
| 83 |
+
"grad_norm": 3.194413185119629,
|
| 84 |
+
"learning_rate": 2.1629213483146067e-06,
|
| 85 |
+
"loss": 0.7778,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.02696629213483146,
|
| 90 |
+
"grad_norm": 3.966202974319458,
|
| 91 |
+
"learning_rate": 2.3595505617977525e-06,
|
| 92 |
+
"loss": 0.7111,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.029213483146067417,
|
| 97 |
+
"grad_norm": 3.63393235206604,
|
| 98 |
+
"learning_rate": 2.5561797752808988e-06,
|
| 99 |
+
"loss": 0.6598,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.03146067415730337,
|
| 104 |
+
"grad_norm": 4.087065696716309,
|
| 105 |
+
"learning_rate": 2.7528089887640446e-06,
|
| 106 |
+
"loss": 0.8901,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.033707865168539325,
|
| 111 |
+
"grad_norm": 2.769573211669922,
|
| 112 |
+
"learning_rate": 2.949438202247191e-06,
|
| 113 |
+
"loss": 0.3206,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.035955056179775284,
|
| 118 |
+
"grad_norm": 2.630620002746582,
|
| 119 |
+
"learning_rate": 3.1460674157303367e-06,
|
| 120 |
+
"loss": 0.3408,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.038202247191011236,
|
| 125 |
+
"grad_norm": 2.9570937156677246,
|
| 126 |
+
"learning_rate": 3.342696629213483e-06,
|
| 127 |
+
"loss": 0.5623,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.04044943820224719,
|
| 132 |
+
"grad_norm": 1.0999970436096191,
|
| 133 |
+
"learning_rate": 3.5393258426966288e-06,
|
| 134 |
+
"loss": 0.0758,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.04269662921348315,
|
| 139 |
+
"grad_norm": 5.516472816467285,
|
| 140 |
+
"learning_rate": 3.735955056179775e-06,
|
| 141 |
+
"loss": 0.994,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.0449438202247191,
|
| 146 |
+
"grad_norm": 6.245299816131592,
|
| 147 |
+
"learning_rate": 3.932584269662921e-06,
|
| 148 |
+
"loss": 2.4196,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.04719101123595506,
|
| 153 |
+
"grad_norm": 0.546605110168457,
|
| 154 |
+
"learning_rate": 4.129213483146067e-06,
|
| 155 |
+
"loss": 0.0561,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.04943820224719101,
|
| 160 |
+
"grad_norm": 0.7049635648727417,
|
| 161 |
+
"learning_rate": 4.325842696629213e-06,
|
| 162 |
+
"loss": 0.0827,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.051685393258426963,
|
| 167 |
+
"grad_norm": 3.1022439002990723,
|
| 168 |
+
"learning_rate": 4.522471910112359e-06,
|
| 169 |
+
"loss": 0.7405,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.05393258426966292,
|
| 174 |
+
"grad_norm": 4.534759044647217,
|
| 175 |
+
"learning_rate": 4.719101123595505e-06,
|
| 176 |
+
"loss": 0.9656,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.056179775280898875,
|
| 181 |
+
"grad_norm": 3.0486032962799072,
|
| 182 |
+
"learning_rate": 4.915730337078652e-06,
|
| 183 |
+
"loss": 0.7855,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.058426966292134834,
|
| 188 |
+
"grad_norm": 3.7457478046417236,
|
| 189 |
+
"learning_rate": 5.1123595505617975e-06,
|
| 190 |
+
"loss": 0.6349,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.060674157303370786,
|
| 195 |
+
"grad_norm": 3.2051479816436768,
|
| 196 |
+
"learning_rate": 5.308988764044943e-06,
|
| 197 |
+
"loss": 0.8087,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.06292134831460675,
|
| 202 |
+
"grad_norm": 4.389094829559326,
|
| 203 |
+
"learning_rate": 5.505617977528089e-06,
|
| 204 |
+
"loss": 0.9282,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.0651685393258427,
|
| 209 |
+
"grad_norm": 2.920410394668579,
|
| 210 |
+
"learning_rate": 5.702247191011236e-06,
|
| 211 |
+
"loss": 0.3377,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.06741573033707865,
|
| 216 |
+
"grad_norm": 2.7193148136138916,
|
| 217 |
+
"learning_rate": 5.898876404494382e-06,
|
| 218 |
+
"loss": 0.3289,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.0696629213483146,
|
| 223 |
+
"grad_norm": 4.0008225440979,
|
| 224 |
+
"learning_rate": 6.0955056179775275e-06,
|
| 225 |
+
"loss": 0.6314,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.07191011235955057,
|
| 230 |
+
"grad_norm": 0.5842159390449524,
|
| 231 |
+
"learning_rate": 6.292134831460673e-06,
|
| 232 |
+
"loss": 0.0611,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.07415730337078652,
|
| 237 |
+
"grad_norm": 3.1256043910980225,
|
| 238 |
+
"learning_rate": 6.48876404494382e-06,
|
| 239 |
+
"loss": 0.8942,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.07640449438202247,
|
| 244 |
+
"grad_norm": 0.9526051878929138,
|
| 245 |
+
"learning_rate": 6.685393258426966e-06,
|
| 246 |
+
"loss": 0.0701,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.07865168539325842,
|
| 251 |
+
"grad_norm": 4.061926364898682,
|
| 252 |
+
"learning_rate": 6.882022471910112e-06,
|
| 253 |
+
"loss": 0.8506,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.08089887640449438,
|
| 258 |
+
"grad_norm": 2.8898491859436035,
|
| 259 |
+
"learning_rate": 7.0786516853932575e-06,
|
| 260 |
+
"loss": 0.3386,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.08314606741573034,
|
| 265 |
+
"grad_norm": 0.9806709289550781,
|
| 266 |
+
"learning_rate": 7.275280898876404e-06,
|
| 267 |
+
"loss": 0.0701,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.0853932584269663,
|
| 272 |
+
"grad_norm": 3.8004391193389893,
|
| 273 |
+
"learning_rate": 7.47191011235955e-06,
|
| 274 |
+
"loss": 0.8042,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.08764044943820225,
|
| 279 |
+
"grad_norm": 4.089083194732666,
|
| 280 |
+
"learning_rate": 7.668539325842697e-06,
|
| 281 |
+
"loss": 0.8744,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.0898876404494382,
|
| 286 |
+
"grad_norm": 3.419440984725952,
|
| 287 |
+
"learning_rate": 7.865168539325842e-06,
|
| 288 |
+
"loss": 0.8644,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.09213483146067415,
|
| 293 |
+
"grad_norm": 4.094921588897705,
|
| 294 |
+
"learning_rate": 8.061797752808988e-06,
|
| 295 |
+
"loss": 0.8647,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.09438202247191012,
|
| 300 |
+
"grad_norm": 3.9199764728546143,
|
| 301 |
+
"learning_rate": 8.258426966292133e-06,
|
| 302 |
+
"loss": 0.7916,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.09662921348314607,
|
| 307 |
+
"grad_norm": 4.082360744476318,
|
| 308 |
+
"learning_rate": 8.45505617977528e-06,
|
| 309 |
+
"loss": 0.8599,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.09887640449438202,
|
| 314 |
+
"grad_norm": 0.6443855166435242,
|
| 315 |
+
"learning_rate": 8.651685393258427e-06,
|
| 316 |
+
"loss": 0.0523,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.10112359550561797,
|
| 321 |
+
"grad_norm": 4.051048278808594,
|
| 322 |
+
"learning_rate": 8.848314606741572e-06,
|
| 323 |
+
"loss": 0.6968,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.10112359550561797,
|
| 328 |
+
"eval_VitaminC_cosine_accuracy": 0.55859375,
|
| 329 |
+
"eval_VitaminC_cosine_accuracy_threshold": 0.8300318121910095,
|
| 330 |
+
"eval_VitaminC_cosine_ap": 0.5514483751609435,
|
| 331 |
+
"eval_VitaminC_cosine_f1": 0.6657718120805369,
|
| 332 |
+
"eval_VitaminC_cosine_f1_threshold": 0.37456807494163513,
|
| 333 |
+
"eval_VitaminC_cosine_precision": 0.5020242914979757,
|
| 334 |
+
"eval_VitaminC_cosine_recall": 0.9880478087649402,
|
| 335 |
+
"eval_VitaminC_dot_accuracy": 0.5546875,
|
| 336 |
+
"eval_VitaminC_dot_accuracy_threshold": 311.380615234375,
|
| 337 |
+
"eval_VitaminC_dot_ap": 0.5333497363350208,
|
| 338 |
+
"eval_VitaminC_dot_f1": 0.6684709066305818,
|
| 339 |
+
"eval_VitaminC_dot_f1_threshold": 144.8927001953125,
|
| 340 |
+
"eval_VitaminC_dot_precision": 0.5061475409836066,
|
| 341 |
+
"eval_VitaminC_dot_recall": 0.9840637450199203,
|
| 342 |
+
"eval_VitaminC_euclidean_accuracy": 0.556640625,
|
| 343 |
+
"eval_VitaminC_euclidean_accuracy_threshold": 11.028482437133789,
|
| 344 |
+
"eval_VitaminC_euclidean_ap": 0.5544340410314673,
|
| 345 |
+
"eval_VitaminC_euclidean_f1": 0.6649006622516557,
|
| 346 |
+
"eval_VitaminC_euclidean_f1_threshold": 23.38451385498047,
|
| 347 |
+
"eval_VitaminC_euclidean_precision": 0.498015873015873,
|
| 348 |
+
"eval_VitaminC_euclidean_recall": 1.0,
|
| 349 |
+
"eval_VitaminC_manhattan_accuracy": 0.556640625,
|
| 350 |
+
"eval_VitaminC_manhattan_accuracy_threshold": 232.38790893554688,
|
| 351 |
+
"eval_VitaminC_manhattan_ap": 0.5515569514532939,
|
| 352 |
+
"eval_VitaminC_manhattan_f1": 0.6649006622516557,
|
| 353 |
+
"eval_VitaminC_manhattan_f1_threshold": 498.126220703125,
|
| 354 |
+
"eval_VitaminC_manhattan_precision": 0.498015873015873,
|
| 355 |
+
"eval_VitaminC_manhattan_recall": 1.0,
|
| 356 |
+
"eval_VitaminC_max_accuracy": 0.55859375,
|
| 357 |
+
"eval_VitaminC_max_accuracy_threshold": 311.380615234375,
|
| 358 |
+
"eval_VitaminC_max_ap": 0.5544340410314673,
|
| 359 |
+
"eval_VitaminC_max_f1": 0.6684709066305818,
|
| 360 |
+
"eval_VitaminC_max_f1_threshold": 498.126220703125,
|
| 361 |
+
"eval_VitaminC_max_precision": 0.5061475409836066,
|
| 362 |
+
"eval_VitaminC_max_recall": 1.0,
|
| 363 |
+
"eval_sequential_score": 0.5544340410314673,
|
| 364 |
+
"eval_sts-test_pearson_cosine": 0.8803067271464453,
|
| 365 |
+
"eval_sts-test_pearson_dot": 0.8698285291814508,
|
| 366 |
+
"eval_sts-test_pearson_euclidean": 0.9023937835918766,
|
| 367 |
+
"eval_sts-test_pearson_manhattan": 0.9020751259156048,
|
| 368 |
+
"eval_sts-test_pearson_max": 0.9023937835918766,
|
| 369 |
+
"eval_sts-test_spearman_cosine": 0.9038005474254912,
|
| 370 |
+
"eval_sts-test_spearman_dot": 0.8707897794601254,
|
| 371 |
+
"eval_sts-test_spearman_euclidean": 0.8989733631129851,
|
| 372 |
+
"eval_sts-test_spearman_manhattan": 0.8980189529612906,
|
| 373 |
+
"eval_sts-test_spearman_max": 0.9038005474254912,
|
| 374 |
+
"eval_vitaminc-pairs_loss": 1.7273772954940796,
|
| 375 |
+
"eval_vitaminc-pairs_runtime": 1.8924,
|
| 376 |
+
"eval_vitaminc-pairs_samples_per_second": 57.071,
|
| 377 |
+
"eval_vitaminc-pairs_steps_per_second": 1.057,
|
| 378 |
+
"step": 45
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 0.10112359550561797,
|
| 382 |
+
"eval_negation-triplets_loss": 0.9174526929855347,
|
| 383 |
+
"eval_negation-triplets_runtime": 0.2972,
|
| 384 |
+
"eval_negation-triplets_samples_per_second": 215.314,
|
| 385 |
+
"eval_negation-triplets_steps_per_second": 3.364,
|
| 386 |
+
"step": 45
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.10112359550561797,
|
| 390 |
+
"eval_scitail-pairs-pos_loss": 0.07368183881044388,
|
| 391 |
+
"eval_scitail-pairs-pos_runtime": 0.379,
|
| 392 |
+
"eval_scitail-pairs-pos_samples_per_second": 142.492,
|
| 393 |
+
"eval_scitail-pairs-pos_steps_per_second": 2.639,
|
| 394 |
+
"step": 45
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.10112359550561797,
|
| 398 |
+
"eval_scitail-pairs-qa_loss": 0.001584450714290142,
|
| 399 |
+
"eval_scitail-pairs-qa_runtime": 0.5178,
|
| 400 |
+
"eval_scitail-pairs-qa_samples_per_second": 247.198,
|
| 401 |
+
"eval_scitail-pairs-qa_steps_per_second": 3.862,
|
| 402 |
+
"step": 45
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.10112359550561797,
|
| 406 |
+
"eval_xsum-pairs_loss": 0.038235221058130264,
|
| 407 |
+
"eval_xsum-pairs_runtime": 2.7268,
|
| 408 |
+
"eval_xsum-pairs_samples_per_second": 46.941,
|
| 409 |
+
"eval_xsum-pairs_steps_per_second": 0.733,
|
| 410 |
+
"step": 45
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.10112359550561797,
|
| 414 |
+
"eval_sciq_pairs_loss": 0.01538097020238638,
|
| 415 |
+
"eval_sciq_pairs_runtime": 2.7808,
|
| 416 |
+
"eval_sciq_pairs_samples_per_second": 46.029,
|
| 417 |
+
"eval_sciq_pairs_steps_per_second": 0.719,
|
| 418 |
+
"step": 45
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 0.10112359550561797,
|
| 422 |
+
"eval_qasc_pairs_loss": 0.09078988432884216,
|
| 423 |
+
"eval_qasc_pairs_runtime": 0.6473,
|
| 424 |
+
"eval_qasc_pairs_samples_per_second": 197.758,
|
| 425 |
+
"eval_qasc_pairs_steps_per_second": 3.09,
|
| 426 |
+
"step": 45
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 0.10112359550561797,
|
| 430 |
+
"eval_openbookqa_pairs_loss": 0.6754768490791321,
|
| 431 |
+
"eval_openbookqa_pairs_runtime": 0.573,
|
| 432 |
+
"eval_openbookqa_pairs_samples_per_second": 223.397,
|
| 433 |
+
"eval_openbookqa_pairs_steps_per_second": 3.491,
|
| 434 |
+
"step": 45
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.10112359550561797,
|
| 438 |
+
"eval_msmarco_pairs_loss": 0.15991328656673431,
|
| 439 |
+
"eval_msmarco_pairs_runtime": 1.487,
|
| 440 |
+
"eval_msmarco_pairs_samples_per_second": 86.078,
|
| 441 |
+
"eval_msmarco_pairs_steps_per_second": 1.345,
|
| 442 |
+
"step": 45
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 0.10112359550561797,
|
| 446 |
+
"eval_nq_pairs_loss": 0.09591890126466751,
|
| 447 |
+
"eval_nq_pairs_runtime": 2.3943,
|
| 448 |
+
"eval_nq_pairs_samples_per_second": 53.459,
|
| 449 |
+
"eval_nq_pairs_steps_per_second": 0.835,
|
| 450 |
+
"step": 45
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.10112359550561797,
|
| 454 |
+
"eval_trivia_pairs_loss": 0.5305934548377991,
|
| 455 |
+
"eval_trivia_pairs_runtime": 3.5752,
|
| 456 |
+
"eval_trivia_pairs_samples_per_second": 35.802,
|
| 457 |
+
"eval_trivia_pairs_steps_per_second": 0.559,
|
| 458 |
+
"step": 45
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.10112359550561797,
|
| 462 |
+
"eval_gooaq_pairs_loss": 0.29681000113487244,
|
| 463 |
+
"eval_gooaq_pairs_runtime": 0.9087,
|
| 464 |
+
"eval_gooaq_pairs_samples_per_second": 140.861,
|
| 465 |
+
"eval_gooaq_pairs_steps_per_second": 2.201,
|
| 466 |
+
"step": 45
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.10112359550561797,
|
| 470 |
+
"eval_paws-pos_loss": 0.024501051753759384,
|
| 471 |
+
"eval_paws-pos_runtime": 0.6773,
|
| 472 |
+
"eval_paws-pos_samples_per_second": 188.996,
|
| 473 |
+
"eval_paws-pos_steps_per_second": 2.953,
|
| 474 |
+
"step": 45
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.10337078651685393,
|
| 478 |
+
"grad_norm": 2.9021923542022705,
|
| 479 |
+
"learning_rate": 9.044943820224718e-06,
|
| 480 |
+
"loss": 0.3376,
|
| 481 |
+
"step": 46
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 0.10561797752808989,
|
| 485 |
+
"grad_norm": 3.179288625717163,
|
| 486 |
+
"learning_rate": 9.241573033707863e-06,
|
| 487 |
+
"loss": 0.5174,
|
| 488 |
+
"step": 47
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 0.10786516853932585,
|
| 492 |
+
"grad_norm": 3.1919493675231934,
|
| 493 |
+
"learning_rate": 9.43820224719101e-06,
|
| 494 |
+
"loss": 0.8162,
|
| 495 |
+
"step": 48
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 0.1101123595505618,
|
| 499 |
+
"grad_norm": 2.8602521419525146,
|
| 500 |
+
"learning_rate": 9.634831460674157e-06,
|
| 501 |
+
"loss": 0.3545,
|
| 502 |
+
"step": 49
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 0.11235955056179775,
|
| 506 |
+
"grad_norm": 2.7570478916168213,
|
| 507 |
+
"learning_rate": 9.831460674157303e-06,
|
| 508 |
+
"loss": 0.315,
|
| 509 |
+
"step": 50
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 0.1146067415730337,
|
| 513 |
+
"grad_norm": 0.8641514778137207,
|
| 514 |
+
"learning_rate": 1.0028089887640448e-05,
|
| 515 |
+
"loss": 0.0627,
|
| 516 |
+
"step": 51
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 0.11685393258426967,
|
| 520 |
+
"grad_norm": 3.9437484741210938,
|
| 521 |
+
"learning_rate": 1.0224719101123595e-05,
|
| 522 |
+
"loss": 0.8851,
|
| 523 |
+
"step": 52
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 0.11910112359550562,
|
| 527 |
+
"grad_norm": 4.144773006439209,
|
| 528 |
+
"learning_rate": 1.042134831460674e-05,
|
| 529 |
+
"loss": 0.8382,
|
| 530 |
+
"step": 53
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 0.12134831460674157,
|
| 534 |
+
"grad_norm": 4.277736186981201,
|
| 535 |
+
"learning_rate": 1.0617977528089887e-05,
|
| 536 |
+
"loss": 0.733,
|
| 537 |
+
"step": 54
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 0.12359550561797752,
|
| 541 |
+
"grad_norm": 4.025904178619385,
|
| 542 |
+
"learning_rate": 1.0814606741573032e-05,
|
| 543 |
+
"loss": 0.7173,
|
| 544 |
+
"step": 55
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 0.1258426966292135,
|
| 548 |
+
"grad_norm": 3.923046827316284,
|
| 549 |
+
"learning_rate": 1.1011235955056178e-05,
|
| 550 |
+
"loss": 0.7659,
|
| 551 |
+
"step": 56
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 0.12808988764044943,
|
| 555 |
+
"grad_norm": 3.2707138061523438,
|
| 556 |
+
"learning_rate": 1.1207865168539325e-05,
|
| 557 |
+
"loss": 0.793,
|
| 558 |
+
"step": 57
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 0.1303370786516854,
|
| 562 |
+
"grad_norm": 3.1660959720611572,
|
| 563 |
+
"learning_rate": 1.1404494382022472e-05,
|
| 564 |
+
"loss": 0.5426,
|
| 565 |
+
"step": 58
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 0.13258426966292136,
|
| 569 |
+
"grad_norm": 4.5236663818359375,
|
| 570 |
+
"learning_rate": 1.1601123595505617e-05,
|
| 571 |
+
"loss": 0.7641,
|
| 572 |
+
"step": 59
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 0.1348314606741573,
|
| 576 |
+
"grad_norm": 0.5771021246910095,
|
| 577 |
+
"learning_rate": 1.1797752808988763e-05,
|
| 578 |
+
"loss": 0.0657,
|
| 579 |
+
"step": 60
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 0.13707865168539327,
|
| 583 |
+
"grad_norm": 3.8541343212127686,
|
| 584 |
+
"learning_rate": 1.1994382022471908e-05,
|
| 585 |
+
"loss": 0.7836,
|
| 586 |
+
"step": 61
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 0.1393258426966292,
|
| 590 |
+
"grad_norm": 4.284148693084717,
|
| 591 |
+
"learning_rate": 1.2191011235955055e-05,
|
| 592 |
+
"loss": 0.9306,
|
| 593 |
+
"step": 62
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 0.14157303370786517,
|
| 597 |
+
"grad_norm": 4.175032615661621,
|
| 598 |
+
"learning_rate": 1.23876404494382e-05,
|
| 599 |
+
"loss": 0.8673,
|
| 600 |
+
"step": 63
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 0.14382022471910114,
|
| 604 |
+
"grad_norm": 5.025452136993408,
|
| 605 |
+
"learning_rate": 1.2584269662921347e-05,
|
| 606 |
+
"loss": 0.9296,
|
| 607 |
+
"step": 64
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 0.14606741573033707,
|
| 611 |
+
"grad_norm": 3.970745086669922,
|
| 612 |
+
"learning_rate": 1.2780898876404493e-05,
|
| 613 |
+
"loss": 0.8211,
|
| 614 |
+
"step": 65
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.14831460674157304,
|
| 618 |
+
"grad_norm": 3.150197744369507,
|
| 619 |
+
"learning_rate": 1.297752808988764e-05,
|
| 620 |
+
"loss": 0.7685,
|
| 621 |
+
"step": 66
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 0.15056179775280898,
|
| 625 |
+
"grad_norm": 4.280994415283203,
|
| 626 |
+
"learning_rate": 1.3174157303370785e-05,
|
| 627 |
+
"loss": 0.7139,
|
| 628 |
+
"step": 67
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 0.15280898876404495,
|
| 632 |
+
"grad_norm": 4.288730621337891,
|
| 633 |
+
"learning_rate": 1.3370786516853932e-05,
|
| 634 |
+
"loss": 0.8241,
|
| 635 |
+
"step": 68
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 0.1550561797752809,
|
| 639 |
+
"grad_norm": 3.7402424812316895,
|
| 640 |
+
"learning_rate": 1.3567415730337077e-05,
|
| 641 |
+
"loss": 0.6256,
|
| 642 |
+
"step": 69
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 0.15730337078651685,
|
| 646 |
+
"grad_norm": 4.478890895843506,
|
| 647 |
+
"learning_rate": 1.3764044943820223e-05,
|
| 648 |
+
"loss": 0.8842,
|
| 649 |
+
"step": 70
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 0.15955056179775282,
|
| 653 |
+
"grad_norm": 3.8147876262664795,
|
| 654 |
+
"learning_rate": 1.3960674157303368e-05,
|
| 655 |
+
"loss": 0.804,
|
| 656 |
+
"step": 71
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 0.16179775280898875,
|
| 660 |
+
"grad_norm": 0.7314035296440125,
|
| 661 |
+
"learning_rate": 1.4157303370786515e-05,
|
| 662 |
+
"loss": 0.0989,
|
| 663 |
+
"step": 72
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 0.16404494382022472,
|
| 667 |
+
"grad_norm": 3.074303150177002,
|
| 668 |
+
"learning_rate": 1.4353932584269662e-05,
|
| 669 |
+
"loss": 0.332,
|
| 670 |
+
"step": 73
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 0.1662921348314607,
|
| 674 |
+
"grad_norm": 3.414987325668335,
|
| 675 |
+
"learning_rate": 1.4550561797752808e-05,
|
| 676 |
+
"loss": 0.5736,
|
| 677 |
+
"step": 74
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 0.16853932584269662,
|
| 681 |
+
"grad_norm": 3.7946674823760986,
|
| 682 |
+
"learning_rate": 1.4747191011235953e-05,
|
| 683 |
+
"loss": 0.8285,
|
| 684 |
+
"step": 75
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 0.1707865168539326,
|
| 688 |
+
"grad_norm": 4.310474395751953,
|
| 689 |
+
"learning_rate": 1.49438202247191e-05,
|
| 690 |
+
"loss": 0.9561,
|
| 691 |
+
"step": 76
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 0.17303370786516853,
|
| 695 |
+
"grad_norm": 0.9791378974914551,
|
| 696 |
+
"learning_rate": 1.5140449438202245e-05,
|
| 697 |
+
"loss": 0.0633,
|
| 698 |
+
"step": 77
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 0.1752808988764045,
|
| 702 |
+
"grad_norm": 0.6351795196533203,
|
| 703 |
+
"learning_rate": 1.5337078651685393e-05,
|
| 704 |
+
"loss": 0.0848,
|
| 705 |
+
"step": 78
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 0.17752808988764046,
|
| 709 |
+
"grad_norm": 3.4832303524017334,
|
| 710 |
+
"learning_rate": 1.553370786516854e-05,
|
| 711 |
+
"loss": 0.8325,
|
| 712 |
+
"step": 79
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 0.1797752808988764,
|
| 716 |
+
"grad_norm": 5.115800380706787,
|
| 717 |
+
"learning_rate": 1.5730337078651683e-05,
|
| 718 |
+
"loss": 1.0011,
|
| 719 |
+
"step": 80
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 0.18202247191011237,
|
| 723 |
+
"grad_norm": 3.552396297454834,
|
| 724 |
+
"learning_rate": 1.592696629213483e-05,
|
| 725 |
+
"loss": 0.8697,
|
| 726 |
+
"step": 81
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 0.1842696629213483,
|
| 730 |
+
"grad_norm": 4.491541862487793,
|
| 731 |
+
"learning_rate": 1.6123595505617977e-05,
|
| 732 |
+
"loss": 0.8344,
|
| 733 |
+
"step": 82
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 0.18651685393258427,
|
| 737 |
+
"grad_norm": 4.73278284072876,
|
| 738 |
+
"learning_rate": 1.6320224719101122e-05,
|
| 739 |
+
"loss": 0.9967,
|
| 740 |
+
"step": 83
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 0.18876404494382024,
|
| 744 |
+
"grad_norm": 2.994192123413086,
|
| 745 |
+
"learning_rate": 1.6516853932584267e-05,
|
| 746 |
+
"loss": 0.4638,
|
| 747 |
+
"step": 84
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 0.19101123595505617,
|
| 751 |
+
"grad_norm": 4.142394542694092,
|
| 752 |
+
"learning_rate": 1.6713483146067415e-05,
|
| 753 |
+
"loss": 0.8994,
|
| 754 |
+
"step": 85
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 0.19325842696629214,
|
| 758 |
+
"grad_norm": 4.149839401245117,
|
| 759 |
+
"learning_rate": 1.691011235955056e-05,
|
| 760 |
+
"loss": 0.7789,
|
| 761 |
+
"step": 86
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 0.19550561797752808,
|
| 765 |
+
"grad_norm": 0.45795938372612,
|
| 766 |
+
"learning_rate": 1.7106741573033705e-05,
|
| 767 |
+
"loss": 0.0555,
|
| 768 |
+
"step": 87
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 0.19775280898876405,
|
| 772 |
+
"grad_norm": 3.4293618202209473,
|
| 773 |
+
"learning_rate": 1.7303370786516853e-05,
|
| 774 |
+
"loss": 0.3778,
|
| 775 |
+
"step": 88
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 0.2,
|
| 779 |
+
"grad_norm": 4.041529655456543,
|
| 780 |
+
"learning_rate": 1.75e-05,
|
| 781 |
+
"loss": 0.708,
|
| 782 |
+
"step": 89
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 0.20224719101123595,
|
| 786 |
+
"grad_norm": 0.6160458922386169,
|
| 787 |
+
"learning_rate": 1.7696629213483143e-05,
|
| 788 |
+
"loss": 0.0689,
|
| 789 |
+
"step": 90
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 0.20224719101123595,
|
| 793 |
+
"eval_VitaminC_cosine_accuracy": 0.556640625,
|
| 794 |
+
"eval_VitaminC_cosine_accuracy_threshold": 0.74173903465271,
|
| 795 |
+
"eval_VitaminC_cosine_ap": 0.5513770735348443,
|
| 796 |
+
"eval_VitaminC_cosine_f1": 0.6675531914893617,
|
| 797 |
+
"eval_VitaminC_cosine_f1_threshold": 0.32480987906455994,
|
| 798 |
+
"eval_VitaminC_cosine_precision": 0.500998003992016,
|
| 799 |
+
"eval_VitaminC_cosine_recall": 1.0,
|
| 800 |
+
"eval_VitaminC_dot_accuracy": 0.560546875,
|
| 801 |
+
"eval_VitaminC_dot_accuracy_threshold": 297.664794921875,
|
| 802 |
+
"eval_VitaminC_dot_ap": 0.5340088824099496,
|
| 803 |
+
"eval_VitaminC_dot_f1": 0.6666666666666667,
|
| 804 |
+
"eval_VitaminC_dot_f1_threshold": 126.67618560791016,
|
| 805 |
+
"eval_VitaminC_dot_precision": 0.501002004008016,
|
| 806 |
+
"eval_VitaminC_dot_recall": 0.9960159362549801,
|
| 807 |
+
"eval_VitaminC_euclidean_accuracy": 0.55859375,
|
| 808 |
+
"eval_VitaminC_euclidean_accuracy_threshold": 14.345688819885254,
|
| 809 |
+
"eval_VitaminC_euclidean_ap": 0.5542145004976253,
|
| 810 |
+
"eval_VitaminC_euclidean_f1": 0.6675531914893617,
|
| 811 |
+
"eval_VitaminC_euclidean_f1_threshold": 23.381019592285156,
|
| 812 |
+
"eval_VitaminC_euclidean_precision": 0.500998003992016,
|
| 813 |
+
"eval_VitaminC_euclidean_recall": 1.0,
|
| 814 |
+
"eval_VitaminC_manhattan_accuracy": 0.552734375,
|
| 815 |
+
"eval_VitaminC_manhattan_accuracy_threshold": 232.7296142578125,
|
| 816 |
+
"eval_VitaminC_manhattan_ap": 0.5523953693907266,
|
| 817 |
+
"eval_VitaminC_manhattan_f1": 0.6657824933687002,
|
| 818 |
+
"eval_VitaminC_manhattan_f1_threshold": 496.4290466308594,
|
| 819 |
+
"eval_VitaminC_manhattan_precision": 0.4990059642147117,
|
| 820 |
+
"eval_VitaminC_manhattan_recall": 1.0,
|
| 821 |
+
"eval_VitaminC_max_accuracy": 0.560546875,
|
| 822 |
+
"eval_VitaminC_max_accuracy_threshold": 297.664794921875,
|
| 823 |
+
"eval_VitaminC_max_ap": 0.5542145004976253,
|
| 824 |
+
"eval_VitaminC_max_f1": 0.6675531914893617,
|
| 825 |
+
"eval_VitaminC_max_f1_threshold": 496.4290466308594,
|
| 826 |
+
"eval_VitaminC_max_precision": 0.501002004008016,
|
| 827 |
+
"eval_VitaminC_max_recall": 1.0,
|
| 828 |
+
"eval_sequential_score": 0.5542145004976253,
|
| 829 |
+
"eval_sts-test_pearson_cosine": 0.8800782580988616,
|
| 830 |
+
"eval_sts-test_pearson_dot": 0.8687642290872662,
|
| 831 |
+
"eval_sts-test_pearson_euclidean": 0.9034088230546415,
|
| 832 |
+
"eval_sts-test_pearson_manhattan": 0.9030146212284895,
|
| 833 |
+
"eval_sts-test_pearson_max": 0.9034088230546415,
|
| 834 |
+
"eval_sts-test_spearman_cosine": 0.904560289590133,
|
| 835 |
+
"eval_sts-test_spearman_dot": 0.8705944849554133,
|
| 836 |
+
"eval_sts-test_spearman_euclidean": 0.8998959103665689,
|
| 837 |
+
"eval_sts-test_spearman_manhattan": 0.8995891404697307,
|
| 838 |
+
"eval_sts-test_spearman_max": 0.904560289590133,
|
| 839 |
+
"eval_vitaminc-pairs_loss": 1.6141985654830933,
|
| 840 |
+
"eval_vitaminc-pairs_runtime": 1.864,
|
| 841 |
+
"eval_vitaminc-pairs_samples_per_second": 57.94,
|
| 842 |
+
"eval_vitaminc-pairs_steps_per_second": 1.073,
|
| 843 |
+
"step": 90
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.20224719101123595,
|
| 847 |
+
"eval_negation-triplets_loss": 0.9220322370529175,
|
| 848 |
+
"eval_negation-triplets_runtime": 0.3199,
|
| 849 |
+
"eval_negation-triplets_samples_per_second": 200.043,
|
| 850 |
+
"eval_negation-triplets_steps_per_second": 3.126,
|
| 851 |
+
"step": 90
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 0.20224719101123595,
|
| 855 |
+
"eval_scitail-pairs-pos_loss": 0.0654294565320015,
|
| 856 |
+
"eval_scitail-pairs-pos_runtime": 0.4625,
|
| 857 |
+
"eval_scitail-pairs-pos_samples_per_second": 116.76,
|
| 858 |
+
"eval_scitail-pairs-pos_steps_per_second": 2.162,
|
| 859 |
+
"step": 90
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 0.20224719101123595,
|
| 863 |
+
"eval_scitail-pairs-qa_loss": 0.0015887805493548512,
|
| 864 |
+
"eval_scitail-pairs-qa_runtime": 0.5768,
|
| 865 |
+
"eval_scitail-pairs-qa_samples_per_second": 221.899,
|
| 866 |
+
"eval_scitail-pairs-qa_steps_per_second": 3.467,
|
| 867 |
+
"step": 90
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 0.20224719101123595,
|
| 871 |
+
"eval_xsum-pairs_loss": 0.03991687670350075,
|
| 872 |
+
"eval_xsum-pairs_runtime": 2.7403,
|
| 873 |
+
"eval_xsum-pairs_samples_per_second": 46.71,
|
| 874 |
+
"eval_xsum-pairs_steps_per_second": 0.73,
|
| 875 |
+
"step": 90
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 0.20224719101123595,
|
| 879 |
+
"eval_sciq_pairs_loss": 0.01584962010383606,
|
| 880 |
+
"eval_sciq_pairs_runtime": 2.8429,
|
| 881 |
+
"eval_sciq_pairs_samples_per_second": 45.024,
|
| 882 |
+
"eval_sciq_pairs_steps_per_second": 0.703,
|
| 883 |
+
"step": 90
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 0.20224719101123595,
|
| 887 |
+
"eval_qasc_pairs_loss": 0.09112343192100525,
|
| 888 |
+
"eval_qasc_pairs_runtime": 0.6492,
|
| 889 |
+
"eval_qasc_pairs_samples_per_second": 197.154,
|
| 890 |
+
"eval_qasc_pairs_steps_per_second": 3.081,
|
| 891 |
+
"step": 90
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.20224719101123595,
|
| 895 |
+
"eval_openbookqa_pairs_loss": 0.7132729887962341,
|
| 896 |
+
"eval_openbookqa_pairs_runtime": 0.5847,
|
| 897 |
+
"eval_openbookqa_pairs_samples_per_second": 218.922,
|
| 898 |
+
"eval_openbookqa_pairs_steps_per_second": 3.421,
|
| 899 |
+
"step": 90
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.20224719101123595,
|
| 903 |
+
"eval_msmarco_pairs_loss": 0.15173853933811188,
|
| 904 |
+
"eval_msmarco_pairs_runtime": 1.4966,
|
| 905 |
+
"eval_msmarco_pairs_samples_per_second": 85.527,
|
| 906 |
+
"eval_msmarco_pairs_steps_per_second": 1.336,
|
| 907 |
+
"step": 90
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 0.20224719101123595,
|
| 911 |
+
"eval_nq_pairs_loss": 0.09653442353010178,
|
| 912 |
+
"eval_nq_pairs_runtime": 2.3749,
|
| 913 |
+
"eval_nq_pairs_samples_per_second": 53.897,
|
| 914 |
+
"eval_nq_pairs_steps_per_second": 0.842,
|
| 915 |
+
"step": 90
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 0.20224719101123595,
|
| 919 |
+
"eval_trivia_pairs_loss": 0.5191965699195862,
|
| 920 |
+
"eval_trivia_pairs_runtime": 3.6006,
|
| 921 |
+
"eval_trivia_pairs_samples_per_second": 35.55,
|
| 922 |
+
"eval_trivia_pairs_steps_per_second": 0.555,
|
| 923 |
+
"step": 90
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 0.20224719101123595,
|
| 927 |
+
"eval_gooaq_pairs_loss": 0.30713126063346863,
|
| 928 |
+
"eval_gooaq_pairs_runtime": 0.9131,
|
| 929 |
+
"eval_gooaq_pairs_samples_per_second": 140.178,
|
| 930 |
+
"eval_gooaq_pairs_steps_per_second": 2.19,
|
| 931 |
+
"step": 90
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 0.20224719101123595,
|
| 935 |
+
"eval_paws-pos_loss": 0.024471310898661613,
|
| 936 |
+
"eval_paws-pos_runtime": 0.6872,
|
| 937 |
+
"eval_paws-pos_samples_per_second": 186.254,
|
| 938 |
+
"eval_paws-pos_steps_per_second": 2.91,
|
| 939 |
+
"step": 90
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 0.20449438202247192,
|
| 943 |
+
"grad_norm": 6.209661483764648,
|
| 944 |
+
"learning_rate": 1.7893258426966292e-05,
|
| 945 |
+
"loss": 2.3489,
|
| 946 |
+
"step": 91
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 0.20674157303370785,
|
| 950 |
+
"grad_norm": 3.1821141242980957,
|
| 951 |
+
"learning_rate": 1.8089887640449437e-05,
|
| 952 |
+
"loss": 0.741,
|
| 953 |
+
"step": 92
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 0.20898876404494382,
|
| 957 |
+
"grad_norm": 3.871994972229004,
|
| 958 |
+
"learning_rate": 1.8286516853932585e-05,
|
| 959 |
+
"loss": 0.7729,
|
| 960 |
+
"step": 93
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 0.21123595505617979,
|
| 964 |
+
"grad_norm": 0.5280765891075134,
|
| 965 |
+
"learning_rate": 1.8483146067415727e-05,
|
| 966 |
+
"loss": 0.0631,
|
| 967 |
+
"step": 94
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 0.21348314606741572,
|
| 971 |
+
"grad_norm": 4.475915431976318,
|
| 972 |
+
"learning_rate": 1.8679775280898875e-05,
|
| 973 |
+
"loss": 0.9342,
|
| 974 |
+
"step": 95
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 0.2157303370786517,
|
| 978 |
+
"grad_norm": 3.949381113052368,
|
| 979 |
+
"learning_rate": 1.887640449438202e-05,
|
| 980 |
+
"loss": 0.8581,
|
| 981 |
+
"step": 96
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 0.21797752808988763,
|
| 985 |
+
"grad_norm": 2.910426616668701,
|
| 986 |
+
"learning_rate": 1.907303370786517e-05,
|
| 987 |
+
"loss": 0.5198,
|
| 988 |
+
"step": 97
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 0.2202247191011236,
|
| 992 |
+
"grad_norm": 4.028941631317139,
|
| 993 |
+
"learning_rate": 1.9269662921348313e-05,
|
| 994 |
+
"loss": 0.846,
|
| 995 |
+
"step": 98
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 0.22247191011235956,
|
| 999 |
+
"grad_norm": 4.183433532714844,
|
| 1000 |
+
"learning_rate": 1.946629213483146e-05,
|
| 1001 |
+
"loss": 0.6581,
|
| 1002 |
+
"step": 99
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 0.2247191011235955,
|
| 1006 |
+
"grad_norm": 3.348114252090454,
|
| 1007 |
+
"learning_rate": 1.9662921348314607e-05,
|
| 1008 |
+
"loss": 0.3579,
|
| 1009 |
+
"step": 100
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 0.22696629213483147,
|
| 1013 |
+
"grad_norm": 4.055211544036865,
|
| 1014 |
+
"learning_rate": 1.9859550561797752e-05,
|
| 1015 |
+
"loss": 0.908,
|
| 1016 |
+
"step": 101
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 0.2292134831460674,
|
| 1020 |
+
"grad_norm": 1.0024710893630981,
|
| 1021 |
+
"learning_rate": 2.0056179775280897e-05,
|
| 1022 |
+
"loss": 0.0664,
|
| 1023 |
+
"step": 102
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 0.23146067415730337,
|
| 1027 |
+
"grad_norm": 3.582249641418457,
|
| 1028 |
+
"learning_rate": 2.0252808988764042e-05,
|
| 1029 |
+
"loss": 0.5411,
|
| 1030 |
+
"step": 103
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 0.23370786516853934,
|
| 1034 |
+
"grad_norm": 4.226349830627441,
|
| 1035 |
+
"learning_rate": 2.044943820224719e-05,
|
| 1036 |
+
"loss": 0.9163,
|
| 1037 |
+
"step": 104
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 0.23595505617977527,
|
| 1041 |
+
"grad_norm": 3.002727508544922,
|
| 1042 |
+
"learning_rate": 2.0646067415730335e-05,
|
| 1043 |
+
"loss": 0.7975,
|
| 1044 |
+
"step": 105
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 0.23820224719101124,
|
| 1048 |
+
"grad_norm": 3.5497515201568604,
|
| 1049 |
+
"learning_rate": 2.084269662921348e-05,
|
| 1050 |
+
"loss": 0.37,
|
| 1051 |
+
"step": 106
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 0.24044943820224718,
|
| 1055 |
+
"grad_norm": 4.381045341491699,
|
| 1056 |
+
"learning_rate": 2.103932584269663e-05,
|
| 1057 |
+
"loss": 0.8495,
|
| 1058 |
+
"step": 107
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 0.24269662921348314,
|
| 1062 |
+
"grad_norm": 3.926840305328369,
|
| 1063 |
+
"learning_rate": 2.1235955056179773e-05,
|
| 1064 |
+
"loss": 0.8073,
|
| 1065 |
+
"step": 108
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 0.2449438202247191,
|
| 1069 |
+
"grad_norm": 3.0835390090942383,
|
| 1070 |
+
"learning_rate": 2.1432584269662922e-05,
|
| 1071 |
+
"loss": 0.7563,
|
| 1072 |
+
"step": 109
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 0.24719101123595505,
|
| 1076 |
+
"grad_norm": 4.230669975280762,
|
| 1077 |
+
"learning_rate": 2.1629213483146063e-05,
|
| 1078 |
+
"loss": 0.6585,
|
| 1079 |
+
"step": 110
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 0.24943820224719102,
|
| 1083 |
+
"grad_norm": 2.8849070072174072,
|
| 1084 |
+
"learning_rate": 2.1825842696629212e-05,
|
| 1085 |
+
"loss": 0.3246,
|
| 1086 |
+
"step": 111
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 0.251685393258427,
|
| 1090 |
+
"grad_norm": 4.796951770782471,
|
| 1091 |
+
"learning_rate": 2.2022471910112357e-05,
|
| 1092 |
+
"loss": 0.9718,
|
| 1093 |
+
"step": 112
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 0.2539325842696629,
|
| 1097 |
+
"grad_norm": 4.60318660736084,
|
| 1098 |
+
"learning_rate": 2.2219101123595505e-05,
|
| 1099 |
+
"loss": 0.8584,
|
| 1100 |
+
"step": 113
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 0.25617977528089886,
|
| 1104 |
+
"grad_norm": 3.098703384399414,
|
| 1105 |
+
"learning_rate": 2.241573033707865e-05,
|
| 1106 |
+
"loss": 0.3385,
|
| 1107 |
+
"step": 114
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 0.25842696629213485,
|
| 1111 |
+
"grad_norm": 2.9519224166870117,
|
| 1112 |
+
"learning_rate": 2.2612359550561795e-05,
|
| 1113 |
+
"loss": 0.323,
|
| 1114 |
+
"step": 115
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 0.2606741573033708,
|
| 1118 |
+
"grad_norm": 2.913742780685425,
|
| 1119 |
+
"learning_rate": 2.2808988764044944e-05,
|
| 1120 |
+
"loss": 0.3359,
|
| 1121 |
+
"step": 116
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 0.26292134831460673,
|
| 1125 |
+
"grad_norm": 4.148440837860107,
|
| 1126 |
+
"learning_rate": 2.300561797752809e-05,
|
| 1127 |
+
"loss": 0.6955,
|
| 1128 |
+
"step": 117
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 0.2651685393258427,
|
| 1132 |
+
"grad_norm": 0.8463248610496521,
|
| 1133 |
+
"learning_rate": 2.3202247191011234e-05,
|
| 1134 |
+
"loss": 0.0539,
|
| 1135 |
+
"step": 118
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 0.26741573033707866,
|
| 1139 |
+
"grad_norm": 0.7284589409828186,
|
| 1140 |
+
"learning_rate": 2.339887640449438e-05,
|
| 1141 |
+
"loss": 0.0507,
|
| 1142 |
+
"step": 119
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 0.2696629213483146,
|
| 1146 |
+
"grad_norm": 3.615086317062378,
|
| 1147 |
+
"learning_rate": 2.3595505617977527e-05,
|
| 1148 |
+
"loss": 0.314,
|
| 1149 |
+
"step": 120
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 0.27191011235955054,
|
| 1153 |
+
"grad_norm": 5.229820728302002,
|
| 1154 |
+
"learning_rate": 2.3792134831460672e-05,
|
| 1155 |
+
"loss": 1.0339,
|
| 1156 |
+
"step": 121
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 0.27415730337078653,
|
| 1160 |
+
"grad_norm": 3.6847782135009766,
|
| 1161 |
+
"learning_rate": 2.3988764044943817e-05,
|
| 1162 |
+
"loss": 0.3158,
|
| 1163 |
+
"step": 122
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 0.27640449438202247,
|
| 1167 |
+
"grad_norm": 4.280517578125,
|
| 1168 |
+
"learning_rate": 2.4185393258426965e-05,
|
| 1169 |
+
"loss": 0.7809,
|
| 1170 |
+
"step": 123
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 0.2786516853932584,
|
| 1174 |
+
"grad_norm": 4.476150035858154,
|
| 1175 |
+
"learning_rate": 2.438202247191011e-05,
|
| 1176 |
+
"loss": 0.9516,
|
| 1177 |
+
"step": 124
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 0.2808988764044944,
|
| 1181 |
+
"grad_norm": 2.7380239963531494,
|
| 1182 |
+
"learning_rate": 2.457865168539326e-05,
|
| 1183 |
+
"loss": 0.3117,
|
| 1184 |
+
"step": 125
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 0.28314606741573034,
|
| 1188 |
+
"grad_norm": 3.9667162895202637,
|
| 1189 |
+
"learning_rate": 2.47752808988764e-05,
|
| 1190 |
+
"loss": 0.8366,
|
| 1191 |
+
"step": 126
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 0.2853932584269663,
|
| 1195 |
+
"grad_norm": 4.552999019622803,
|
| 1196 |
+
"learning_rate": 2.497191011235955e-05,
|
| 1197 |
+
"loss": 0.8033,
|
| 1198 |
+
"step": 127
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 0.2876404494382023,
|
| 1202 |
+
"grad_norm": 3.4238576889038086,
|
| 1203 |
+
"learning_rate": 2.5168539325842694e-05,
|
| 1204 |
+
"loss": 0.7253,
|
| 1205 |
+
"step": 128
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 0.2898876404494382,
|
| 1209 |
+
"grad_norm": 4.677807331085205,
|
| 1210 |
+
"learning_rate": 2.5365168539325842e-05,
|
| 1211 |
+
"loss": 0.8345,
|
| 1212 |
+
"step": 129
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"epoch": 0.29213483146067415,
|
| 1216 |
+
"grad_norm": 4.282113075256348,
|
| 1217 |
+
"learning_rate": 2.5561797752808987e-05,
|
| 1218 |
+
"loss": 0.7532,
|
| 1219 |
+
"step": 130
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 0.2943820224719101,
|
| 1223 |
+
"grad_norm": 4.375221252441406,
|
| 1224 |
+
"learning_rate": 2.5758426966292132e-05,
|
| 1225 |
+
"loss": 0.8247,
|
| 1226 |
+
"step": 131
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 0.2966292134831461,
|
| 1230 |
+
"grad_norm": 3.2591633796691895,
|
| 1231 |
+
"learning_rate": 2.595505617977528e-05,
|
| 1232 |
+
"loss": 0.5175,
|
| 1233 |
+
"step": 132
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 0.298876404494382,
|
| 1237 |
+
"grad_norm": 4.146636962890625,
|
| 1238 |
+
"learning_rate": 2.6151685393258425e-05,
|
| 1239 |
+
"loss": 0.7813,
|
| 1240 |
+
"step": 133
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 0.30112359550561796,
|
| 1244 |
+
"grad_norm": 4.2413249015808105,
|
| 1245 |
+
"learning_rate": 2.634831460674157e-05,
|
| 1246 |
+
"loss": 0.6582,
|
| 1247 |
+
"step": 134
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 0.30337078651685395,
|
| 1251 |
+
"grad_norm": 4.541455268859863,
|
| 1252 |
+
"learning_rate": 2.6544943820224715e-05,
|
| 1253 |
+
"loss": 0.3484,
|
| 1254 |
+
"step": 135
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 0.30337078651685395,
|
| 1258 |
+
"eval_VitaminC_cosine_accuracy": 0.560546875,
|
| 1259 |
+
"eval_VitaminC_cosine_accuracy_threshold": 0.7956135272979736,
|
| 1260 |
+
"eval_VitaminC_cosine_ap": 0.5505565383154402,
|
| 1261 |
+
"eval_VitaminC_cosine_f1": 0.6684709066305818,
|
| 1262 |
+
"eval_VitaminC_cosine_f1_threshold": 0.40466147661209106,
|
| 1263 |
+
"eval_VitaminC_cosine_precision": 0.5061475409836066,
|
| 1264 |
+
"eval_VitaminC_cosine_recall": 0.9840637450199203,
|
| 1265 |
+
"eval_VitaminC_dot_accuracy": 0.5546875,
|
| 1266 |
+
"eval_VitaminC_dot_accuracy_threshold": 312.2774658203125,
|
| 1267 |
+
"eval_VitaminC_dot_ap": 0.5365135091766033,
|
| 1268 |
+
"eval_VitaminC_dot_f1": 0.6684856753069577,
|
| 1269 |
+
"eval_VitaminC_dot_f1_threshold": 157.33203125,
|
| 1270 |
+
"eval_VitaminC_dot_precision": 0.508298755186722,
|
| 1271 |
+
"eval_VitaminC_dot_recall": 0.9760956175298805,
|
| 1272 |
+
"eval_VitaminC_euclidean_accuracy": 0.5546875,
|
| 1273 |
+
"eval_VitaminC_euclidean_accuracy_threshold": 12.184114456176758,
|
| 1274 |
+
"eval_VitaminC_euclidean_ap": 0.5517706579195627,
|
| 1275 |
+
"eval_VitaminC_euclidean_f1": 0.6649006622516557,
|
| 1276 |
+
"eval_VitaminC_euclidean_f1_threshold": 23.68879508972168,
|
| 1277 |
+
"eval_VitaminC_euclidean_precision": 0.498015873015873,
|
| 1278 |
+
"eval_VitaminC_euclidean_recall": 1.0,
|
| 1279 |
+
"eval_VitaminC_manhattan_accuracy": 0.5546875,
|
| 1280 |
+
"eval_VitaminC_manhattan_accuracy_threshold": 241.22061157226562,
|
| 1281 |
+
"eval_VitaminC_manhattan_ap": 0.5494156168773414,
|
| 1282 |
+
"eval_VitaminC_manhattan_f1": 0.6649006622516557,
|
| 1283 |
+
"eval_VitaminC_manhattan_f1_threshold": 510.2530212402344,
|
| 1284 |
+
"eval_VitaminC_manhattan_precision": 0.498015873015873,
|
| 1285 |
+
"eval_VitaminC_manhattan_recall": 1.0,
|
| 1286 |
+
"eval_VitaminC_max_accuracy": 0.560546875,
|
| 1287 |
+
"eval_VitaminC_max_accuracy_threshold": 312.2774658203125,
|
| 1288 |
+
"eval_VitaminC_max_ap": 0.5517706579195627,
|
| 1289 |
+
"eval_VitaminC_max_f1": 0.6684856753069577,
|
| 1290 |
+
"eval_VitaminC_max_f1_threshold": 510.2530212402344,
|
| 1291 |
+
"eval_VitaminC_max_precision": 0.508298755186722,
|
| 1292 |
+
"eval_VitaminC_max_recall": 1.0,
|
| 1293 |
+
"eval_sequential_score": 0.5517706579195627,
|
| 1294 |
+
"eval_sts-test_pearson_cosine": 0.8812438499723412,
|
| 1295 |
+
"eval_sts-test_pearson_dot": 0.8695651753004092,
|
| 1296 |
+
"eval_sts-test_pearson_euclidean": 0.9036940037118162,
|
| 1297 |
+
"eval_sts-test_pearson_manhattan": 0.9035516699922166,
|
| 1298 |
+
"eval_sts-test_pearson_max": 0.9036940037118162,
|
| 1299 |
+
"eval_sts-test_spearman_cosine": 0.9049742835092648,
|
| 1300 |
+
"eval_sts-test_spearman_dot": 0.8707925987895928,
|
| 1301 |
+
"eval_sts-test_spearman_euclidean": 0.9003956924537878,
|
| 1302 |
+
"eval_sts-test_spearman_manhattan": 0.9002747745455083,
|
| 1303 |
+
"eval_sts-test_spearman_max": 0.9049742835092648,
|
| 1304 |
+
"eval_vitaminc-pairs_loss": 1.5520410537719727,
|
| 1305 |
+
"eval_vitaminc-pairs_runtime": 1.8323,
|
| 1306 |
+
"eval_vitaminc-pairs_samples_per_second": 58.943,
|
| 1307 |
+
"eval_vitaminc-pairs_steps_per_second": 1.092,
|
| 1308 |
+
"step": 135
|
| 1309 |
+
},
|
| 1310 |
+
{
|
| 1311 |
+
"epoch": 0.30337078651685395,
|
| 1312 |
+
"eval_negation-triplets_loss": 0.9211694002151489,
|
| 1313 |
+
"eval_negation-triplets_runtime": 0.2923,
|
| 1314 |
+
"eval_negation-triplets_samples_per_second": 218.93,
|
| 1315 |
+
"eval_negation-triplets_steps_per_second": 3.421,
|
| 1316 |
+
"step": 135
|
| 1317 |
+
},
|
| 1318 |
+
{
|
| 1319 |
+
"epoch": 0.30337078651685395,
|
| 1320 |
+
"eval_scitail-pairs-pos_loss": 0.07377135753631592,
|
| 1321 |
+
"eval_scitail-pairs-pos_runtime": 0.3681,
|
| 1322 |
+
"eval_scitail-pairs-pos_samples_per_second": 146.691,
|
| 1323 |
+
"eval_scitail-pairs-pos_steps_per_second": 2.716,
|
| 1324 |
+
"step": 135
|
| 1325 |
+
},
|
| 1326 |
+
{
|
| 1327 |
+
"epoch": 0.30337078651685395,
|
| 1328 |
+
"eval_scitail-pairs-qa_loss": 0.00150959100574255,
|
| 1329 |
+
"eval_scitail-pairs-qa_runtime": 0.5123,
|
| 1330 |
+
"eval_scitail-pairs-qa_samples_per_second": 249.842,
|
| 1331 |
+
"eval_scitail-pairs-qa_steps_per_second": 3.904,
|
| 1332 |
+
"step": 135
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.30337078651685395,
|
| 1336 |
+
"eval_xsum-pairs_loss": 0.036599572747945786,
|
| 1337 |
+
"eval_xsum-pairs_runtime": 2.7238,
|
| 1338 |
+
"eval_xsum-pairs_samples_per_second": 46.994,
|
| 1339 |
+
"eval_xsum-pairs_steps_per_second": 0.734,
|
| 1340 |
+
"step": 135
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.30337078651685395,
|
| 1344 |
+
"eval_sciq_pairs_loss": 0.01615014858543873,
|
| 1345 |
+
"eval_sciq_pairs_runtime": 2.8064,
|
| 1346 |
+
"eval_sciq_pairs_samples_per_second": 45.61,
|
| 1347 |
+
"eval_sciq_pairs_steps_per_second": 0.713,
|
| 1348 |
+
"step": 135
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"epoch": 0.30337078651685395,
|
| 1352 |
+
"eval_qasc_pairs_loss": 0.09235507994890213,
|
| 1353 |
+
"eval_qasc_pairs_runtime": 0.6488,
|
| 1354 |
+
"eval_qasc_pairs_samples_per_second": 197.276,
|
| 1355 |
+
"eval_qasc_pairs_steps_per_second": 3.082,
|
| 1356 |
+
"step": 135
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 0.30337078651685395,
|
| 1360 |
+
"eval_openbookqa_pairs_loss": 0.6891775727272034,
|
| 1361 |
+
"eval_openbookqa_pairs_runtime": 0.5698,
|
| 1362 |
+
"eval_openbookqa_pairs_samples_per_second": 224.641,
|
| 1363 |
+
"eval_openbookqa_pairs_steps_per_second": 3.51,
|
| 1364 |
+
"step": 135
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"epoch": 0.30337078651685395,
|
| 1368 |
+
"eval_msmarco_pairs_loss": 0.16766037046909332,
|
| 1369 |
+
"eval_msmarco_pairs_runtime": 1.4798,
|
| 1370 |
+
"eval_msmarco_pairs_samples_per_second": 86.499,
|
| 1371 |
+
"eval_msmarco_pairs_steps_per_second": 1.352,
|
| 1372 |
+
"step": 135
|
| 1373 |
+
},
|
| 1374 |
+
{
|
| 1375 |
+
"epoch": 0.30337078651685395,
|
| 1376 |
+
"eval_nq_pairs_loss": 0.09737721085548401,
|
| 1377 |
+
"eval_nq_pairs_runtime": 2.3409,
|
| 1378 |
+
"eval_nq_pairs_samples_per_second": 54.68,
|
| 1379 |
+
"eval_nq_pairs_steps_per_second": 0.854,
|
| 1380 |
+
"step": 135
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 0.30337078651685395,
|
| 1384 |
+
"eval_trivia_pairs_loss": 0.5458433032035828,
|
| 1385 |
+
"eval_trivia_pairs_runtime": 3.5771,
|
| 1386 |
+
"eval_trivia_pairs_samples_per_second": 35.783,
|
| 1387 |
+
"eval_trivia_pairs_steps_per_second": 0.559,
|
| 1388 |
+
"step": 135
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.30337078651685395,
|
| 1392 |
+
"eval_gooaq_pairs_loss": 0.3082329332828522,
|
| 1393 |
+
"eval_gooaq_pairs_runtime": 0.9181,
|
| 1394 |
+
"eval_gooaq_pairs_samples_per_second": 139.413,
|
| 1395 |
+
"eval_gooaq_pairs_steps_per_second": 2.178,
|
| 1396 |
+
"step": 135
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.30337078651685395,
|
| 1400 |
+
"eval_paws-pos_loss": 0.02423396334052086,
|
| 1401 |
+
"eval_paws-pos_runtime": 0.6827,
|
| 1402 |
+
"eval_paws-pos_samples_per_second": 187.501,
|
| 1403 |
+
"eval_paws-pos_steps_per_second": 2.93,
|
| 1404 |
+
"step": 135
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 0.3056179775280899,
|
| 1408 |
+
"grad_norm": 4.549901485443115,
|
| 1409 |
+
"learning_rate": 2.6741573033707864e-05,
|
| 1410 |
+
"loss": 0.7648,
|
| 1411 |
+
"step": 136
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 0.30786516853932583,
|
| 1415 |
+
"grad_norm": 3.225851535797119,
|
| 1416 |
+
"learning_rate": 2.693820224719101e-05,
|
| 1417 |
+
"loss": 0.7554,
|
| 1418 |
+
"step": 137
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 0.3101123595505618,
|
| 1422 |
+
"grad_norm": 0.6228423118591309,
|
| 1423 |
+
"learning_rate": 2.7134831460674154e-05,
|
| 1424 |
+
"loss": 0.0753,
|
| 1425 |
+
"step": 138
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 0.31235955056179776,
|
| 1429 |
+
"grad_norm": 3.12802791595459,
|
| 1430 |
+
"learning_rate": 2.7331460674157302e-05,
|
| 1431 |
+
"loss": 0.4987,
|
| 1432 |
+
"step": 139
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 0.3146067415730337,
|
| 1436 |
+
"grad_norm": 4.1997880935668945,
|
| 1437 |
+
"learning_rate": 2.7528089887640447e-05,
|
| 1438 |
+
"loss": 0.8543,
|
| 1439 |
+
"step": 140
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 0.31685393258426964,
|
| 1443 |
+
"grad_norm": 4.3362860679626465,
|
| 1444 |
+
"learning_rate": 2.7724719101123595e-05,
|
| 1445 |
+
"loss": 0.9425,
|
| 1446 |
+
"step": 141
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 0.31910112359550563,
|
| 1450 |
+
"grad_norm": 0.5599316954612732,
|
| 1451 |
+
"learning_rate": 2.7921348314606737e-05,
|
| 1452 |
+
"loss": 0.0472,
|
| 1453 |
+
"step": 142
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 0.32134831460674157,
|
| 1457 |
+
"grad_norm": 3.503603458404541,
|
| 1458 |
+
"learning_rate": 2.8117977528089885e-05,
|
| 1459 |
+
"loss": 0.848,
|
| 1460 |
+
"step": 143
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 0.3235955056179775,
|
| 1464 |
+
"grad_norm": 4.712310314178467,
|
| 1465 |
+
"learning_rate": 2.831460674157303e-05,
|
| 1466 |
+
"loss": 0.8946,
|
| 1467 |
+
"step": 144
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 0.3258426966292135,
|
| 1471 |
+
"grad_norm": 3.1823527812957764,
|
| 1472 |
+
"learning_rate": 2.851123595505618e-05,
|
| 1473 |
+
"loss": 0.7841,
|
| 1474 |
+
"step": 145
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 0.32808988764044944,
|
| 1478 |
+
"grad_norm": 4.423196315765381,
|
| 1479 |
+
"learning_rate": 2.8707865168539324e-05,
|
| 1480 |
+
"loss": 0.6653,
|
| 1481 |
+
"step": 146
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 0.3303370786516854,
|
| 1485 |
+
"grad_norm": 4.137822151184082,
|
| 1486 |
+
"learning_rate": 2.890449438202247e-05,
|
| 1487 |
+
"loss": 0.3522,
|
| 1488 |
+
"step": 147
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 0.3325842696629214,
|
| 1492 |
+
"grad_norm": 2.997777223587036,
|
| 1493 |
+
"learning_rate": 2.9101123595505617e-05,
|
| 1494 |
+
"loss": 0.4853,
|
| 1495 |
+
"step": 148
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 0.3348314606741573,
|
| 1499 |
+
"grad_norm": 2.89650559425354,
|
| 1500 |
+
"learning_rate": 2.9297752808988762e-05,
|
| 1501 |
+
"loss": 0.4726,
|
| 1502 |
+
"step": 149
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 0.33707865168539325,
|
| 1506 |
+
"grad_norm": 5.486624717712402,
|
| 1507 |
+
"learning_rate": 2.9494382022471907e-05,
|
| 1508 |
+
"loss": 0.8693,
|
| 1509 |
+
"step": 150
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 0.3393258426966292,
|
| 1513 |
+
"grad_norm": 4.800889015197754,
|
| 1514 |
+
"learning_rate": 2.9691011235955052e-05,
|
| 1515 |
+
"loss": 0.8124,
|
| 1516 |
+
"step": 151
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 0.3415730337078652,
|
| 1520 |
+
"grad_norm": 4.188066005706787,
|
| 1521 |
+
"learning_rate": 2.98876404494382e-05,
|
| 1522 |
+
"loss": 0.8206,
|
| 1523 |
+
"step": 152
|
| 1524 |
+
},
|
| 1525 |
+
{
|
| 1526 |
+
"epoch": 0.3438202247191011,
|
| 1527 |
+
"grad_norm": 4.340461254119873,
|
| 1528 |
+
"learning_rate": 3.0084269662921345e-05,
|
| 1529 |
+
"loss": 0.9406,
|
| 1530 |
+
"step": 153
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"epoch": 0.34606741573033706,
|
| 1534 |
+
"grad_norm": 4.658304214477539,
|
| 1535 |
+
"learning_rate": 3.028089887640449e-05,
|
| 1536 |
+
"loss": 0.7944,
|
| 1537 |
+
"step": 154
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 0.34831460674157305,
|
| 1541 |
+
"grad_norm": 0.6266987919807434,
|
| 1542 |
+
"learning_rate": 3.047752808988764e-05,
|
| 1543 |
+
"loss": 0.0766,
|
| 1544 |
+
"step": 155
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 0.350561797752809,
|
| 1548 |
+
"grad_norm": 4.252346515655518,
|
| 1549 |
+
"learning_rate": 3.067415730337079e-05,
|
| 1550 |
+
"loss": 0.8609,
|
| 1551 |
+
"step": 156
|
| 1552 |
+
},
|
| 1553 |
+
{
|
| 1554 |
+
"epoch": 0.35280898876404493,
|
| 1555 |
+
"grad_norm": 4.9649658203125,
|
| 1556 |
+
"learning_rate": 3.087078651685393e-05,
|
| 1557 |
+
"loss": 1.0533,
|
| 1558 |
+
"step": 157
|
| 1559 |
+
},
|
| 1560 |
+
{
|
| 1561 |
+
"epoch": 0.3550561797752809,
|
| 1562 |
+
"grad_norm": 4.485607624053955,
|
| 1563 |
+
"learning_rate": 3.106741573033708e-05,
|
| 1564 |
+
"loss": 0.8396,
|
| 1565 |
+
"step": 158
|
| 1566 |
+
},
|
| 1567 |
+
{
|
| 1568 |
+
"epoch": 0.35730337078651686,
|
| 1569 |
+
"grad_norm": 3.241231918334961,
|
| 1570 |
+
"learning_rate": 3.126404494382022e-05,
|
| 1571 |
+
"loss": 0.7865,
|
| 1572 |
+
"step": 159
|
| 1573 |
+
},
|
| 1574 |
+
{
|
| 1575 |
+
"epoch": 0.3595505617977528,
|
| 1576 |
+
"grad_norm": 6.846582889556885,
|
| 1577 |
+
"learning_rate": 3.146067415730337e-05,
|
| 1578 |
+
"loss": 2.4616,
|
| 1579 |
+
"step": 160
|
| 1580 |
+
},
|
| 1581 |
+
{
|
| 1582 |
+
"epoch": 0.36179775280898874,
|
| 1583 |
+
"grad_norm": 0.5514687895774841,
|
| 1584 |
+
"learning_rate": 3.165730337078651e-05,
|
| 1585 |
+
"loss": 0.0556,
|
| 1586 |
+
"step": 161
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 0.36404494382022473,
|
| 1590 |
+
"grad_norm": 3.7877562046051025,
|
| 1591 |
+
"learning_rate": 3.185393258426966e-05,
|
| 1592 |
+
"loss": 0.3758,
|
| 1593 |
+
"step": 162
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 0.36629213483146067,
|
| 1597 |
+
"grad_norm": 5.397939682006836,
|
| 1598 |
+
"learning_rate": 3.205056179775281e-05,
|
| 1599 |
+
"loss": 0.9312,
|
| 1600 |
+
"step": 163
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 0.3685393258426966,
|
| 1604 |
+
"grad_norm": 4.301459312438965,
|
| 1605 |
+
"learning_rate": 3.2247191011235954e-05,
|
| 1606 |
+
"loss": 0.7993,
|
| 1607 |
+
"step": 164
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 0.3707865168539326,
|
| 1611 |
+
"grad_norm": 4.49428129196167,
|
| 1612 |
+
"learning_rate": 3.24438202247191e-05,
|
| 1613 |
+
"loss": 0.8104,
|
| 1614 |
+
"step": 165
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 0.37303370786516854,
|
| 1618 |
+
"grad_norm": 3.2210912704467773,
|
| 1619 |
+
"learning_rate": 3.2640449438202244e-05,
|
| 1620 |
+
"loss": 0.8199,
|
| 1621 |
+
"step": 166
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 0.3752808988764045,
|
| 1625 |
+
"grad_norm": 5.359859466552734,
|
| 1626 |
+
"learning_rate": 3.283707865168539e-05,
|
| 1627 |
+
"loss": 1.0724,
|
| 1628 |
+
"step": 167
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 0.3775280898876405,
|
| 1632 |
+
"grad_norm": 4.00059700012207,
|
| 1633 |
+
"learning_rate": 3.3033707865168534e-05,
|
| 1634 |
+
"loss": 0.3521,
|
| 1635 |
+
"step": 168
|
| 1636 |
+
},
|
| 1637 |
+
{
|
| 1638 |
+
"epoch": 0.3797752808988764,
|
| 1639 |
+
"grad_norm": 4.418768882751465,
|
| 1640 |
+
"learning_rate": 3.3230337078651685e-05,
|
| 1641 |
+
"loss": 0.8536,
|
| 1642 |
+
"step": 169
|
| 1643 |
+
},
|
| 1644 |
+
{
|
| 1645 |
+
"epoch": 0.38202247191011235,
|
| 1646 |
+
"grad_norm": 4.15454626083374,
|
| 1647 |
+
"learning_rate": 3.342696629213483e-05,
|
| 1648 |
+
"loss": 0.872,
|
| 1649 |
+
"step": 170
|
| 1650 |
+
},
|
| 1651 |
+
{
|
| 1652 |
+
"epoch": 0.3842696629213483,
|
| 1653 |
+
"grad_norm": 3.8060054779052734,
|
| 1654 |
+
"learning_rate": 3.3623595505617975e-05,
|
| 1655 |
+
"loss": 0.8009,
|
| 1656 |
+
"step": 171
|
| 1657 |
+
},
|
| 1658 |
+
{
|
| 1659 |
+
"epoch": 0.3865168539325843,
|
| 1660 |
+
"grad_norm": 3.584745407104492,
|
| 1661 |
+
"learning_rate": 3.382022471910112e-05,
|
| 1662 |
+
"loss": 0.7798,
|
| 1663 |
+
"step": 172
|
| 1664 |
+
},
|
| 1665 |
+
{
|
| 1666 |
+
"epoch": 0.3887640449438202,
|
| 1667 |
+
"grad_norm": 4.861410140991211,
|
| 1668 |
+
"learning_rate": 3.4016853932584265e-05,
|
| 1669 |
+
"loss": 0.5953,
|
| 1670 |
+
"step": 173
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 0.39101123595505616,
|
| 1674 |
+
"grad_norm": 3.983793020248413,
|
| 1675 |
+
"learning_rate": 3.421348314606741e-05,
|
| 1676 |
+
"loss": 0.7562,
|
| 1677 |
+
"step": 174
|
| 1678 |
+
},
|
| 1679 |
+
{
|
| 1680 |
+
"epoch": 0.39325842696629215,
|
| 1681 |
+
"grad_norm": 4.841738224029541,
|
| 1682 |
+
"learning_rate": 3.4410112359550555e-05,
|
| 1683 |
+
"loss": 0.7227,
|
| 1684 |
+
"step": 175
|
| 1685 |
+
},
|
| 1686 |
+
{
|
| 1687 |
+
"epoch": 0.3955056179775281,
|
| 1688 |
+
"grad_norm": 4.787370204925537,
|
| 1689 |
+
"learning_rate": 3.460674157303371e-05,
|
| 1690 |
+
"loss": 0.8953,
|
| 1691 |
+
"step": 176
|
| 1692 |
+
},
|
| 1693 |
+
{
|
| 1694 |
+
"epoch": 0.39775280898876403,
|
| 1695 |
+
"grad_norm": 4.337812900543213,
|
| 1696 |
+
"learning_rate": 3.480337078651685e-05,
|
| 1697 |
+
"loss": 0.7102,
|
| 1698 |
+
"step": 177
|
| 1699 |
+
},
|
| 1700 |
+
{
|
| 1701 |
+
"epoch": 0.4,
|
| 1702 |
+
"grad_norm": 0.9599294662475586,
|
| 1703 |
+
"learning_rate": 3.5e-05,
|
| 1704 |
+
"loss": 0.0667,
|
| 1705 |
+
"step": 178
|
| 1706 |
+
}
|
| 1707 |
+
],
|
| 1708 |
+
"logging_steps": 1,
|
| 1709 |
+
"max_steps": 890,
|
| 1710 |
+
"num_input_tokens_seen": 0,
|
| 1711 |
+
"num_train_epochs": 2,
|
| 1712 |
+
"save_steps": 89,
|
| 1713 |
+
"stateful_callbacks": {
|
| 1714 |
+
"TrainerControl": {
|
| 1715 |
+
"args": {
|
| 1716 |
+
"should_epoch_stop": false,
|
| 1717 |
+
"should_evaluate": false,
|
| 1718 |
+
"should_log": false,
|
| 1719 |
+
"should_save": true,
|
| 1720 |
+
"should_training_stop": false
|
| 1721 |
+
},
|
| 1722 |
+
"attributes": {}
|
| 1723 |
+
}
|
| 1724 |
+
},
|
| 1725 |
+
"total_flos": 0.0,
|
| 1726 |
+
"train_batch_size": 640,
|
| 1727 |
+
"trial_name": null,
|
| 1728 |
+
"trial_params": null
|
| 1729 |
+
}
|
checkpoint-178/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3db14a166daba939b5d6d6e272be7e74379c2defcff24dc5638ec137086aaf24
|
| 3 |
+
size 5752
|