Training in progress, step 1284, checkpoint
Browse files- checkpoint-1284/1_WeightedPooling/config.json +10 -0
- checkpoint-1284/README.md +0 -0
- checkpoint-1284/added_tokens.json +3 -0
- checkpoint-1284/config.json +35 -0
- checkpoint-1284/config_sentence_transformers.json +10 -0
- checkpoint-1284/modules.json +14 -0
- checkpoint-1284/optimizer.pt +3 -0
- checkpoint-1284/pytorch_model.bin +3 -0
- checkpoint-1284/rng_state.pth +3 -0
- checkpoint-1284/scheduler.pt +3 -0
- checkpoint-1284/sentence_bert_config.json +4 -0
- checkpoint-1284/special_tokens_map.json +51 -0
- checkpoint-1284/spm.model +3 -0
- checkpoint-1284/tokenizer.json +0 -0
- checkpoint-1284/tokenizer_config.json +65 -0
- checkpoint-1284/trainer_state.json +2003 -0
- checkpoint-1284/training_args.bin +3 -0
checkpoint-1284/1_WeightedPooling/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-1284/README.md
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1284/added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"[MASK]": 128000
|
| 3 |
+
}
|
checkpoint-1284/config.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "bobox/DeBERTa-small-ST-v1-test-step3",
|
| 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-1284/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-1284/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_WeightedPooling",
|
| 12 |
+
"type": "__main__"
|
| 13 |
+
}
|
| 14 |
+
]
|
checkpoint-1284/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b846afaae59d9346b9d671bd05aac80d2e2cb2be02eaee79e7784407665d188e
|
| 3 |
+
size 9475334
|
checkpoint-1284/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1886a286545822d70cd17732b6025a52902e7429a8ab7eec28bf9c1a4f143b6c
|
| 3 |
+
size 565251810
|
checkpoint-1284/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a7084ce941a4aa61d84bc31b7d9d487caa9b0c68b38dc61e8fbcd42a5e71fdca
|
| 3 |
+
size 14244
|
checkpoint-1284/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2692faa639a6932dc2740bf42a43cdd479542372a111e651faa5a9247035e038
|
| 3 |
+
size 1064
|
checkpoint-1284/sentence_bert_config.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"max_seq_length": 512,
|
| 3 |
+
"do_lower_case": false
|
| 4 |
+
}
|
checkpoint-1284/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-1284/spm.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c679fbf93643d19aab7ee10c0b99e460bdbc02fedf34b92b05af343b4af586fd
|
| 3 |
+
size 2464616
|
checkpoint-1284/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1284/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-1284/trainer_state.json
ADDED
|
@@ -0,0 +1,2003 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.5017543859649123,
|
| 5 |
+
"eval_steps": 214,
|
| 6 |
+
"global_step": 1284,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.012865497076023392,
|
| 13 |
+
"grad_norm": 0.07311347126960754,
|
| 14 |
+
"learning_rate": 3.5081502480510277e-07,
|
| 15 |
+
"loss": 0.1802,
|
| 16 |
+
"step": 11
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.025730994152046785,
|
| 20 |
+
"grad_norm": 0.3663940727710724,
|
| 21 |
+
"learning_rate": 7.016300496102055e-07,
|
| 22 |
+
"loss": 0.1573,
|
| 23 |
+
"step": 22
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.03859649122807018,
|
| 27 |
+
"grad_norm": 0.22339867055416107,
|
| 28 |
+
"learning_rate": 1.0524450744153084e-06,
|
| 29 |
+
"loss": 0.1184,
|
| 30 |
+
"step": 33
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.05146198830409357,
|
| 34 |
+
"grad_norm": 0.06488268822431564,
|
| 35 |
+
"learning_rate": 1.403260099220411e-06,
|
| 36 |
+
"loss": 0.1456,
|
| 37 |
+
"step": 44
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.06432748538011696,
|
| 41 |
+
"grad_norm": 0.7134692072868347,
|
| 42 |
+
"learning_rate": 1.754075124025514e-06,
|
| 43 |
+
"loss": 0.2102,
|
| 44 |
+
"step": 55
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.07719298245614035,
|
| 48 |
+
"grad_norm": 0.048255760222673416,
|
| 49 |
+
"learning_rate": 2.104890148830617e-06,
|
| 50 |
+
"loss": 0.151,
|
| 51 |
+
"step": 66
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.09005847953216374,
|
| 55 |
+
"grad_norm": 0.43056783080101013,
|
| 56 |
+
"learning_rate": 2.4557051736357195e-06,
|
| 57 |
+
"loss": 0.1356,
|
| 58 |
+
"step": 77
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.10292397660818714,
|
| 62 |
+
"grad_norm": 0.7663404941558838,
|
| 63 |
+
"learning_rate": 2.806520198440822e-06,
|
| 64 |
+
"loss": 0.2292,
|
| 65 |
+
"step": 88
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.11578947368421053,
|
| 69 |
+
"grad_norm": 0.48333388566970825,
|
| 70 |
+
"learning_rate": 3.1573352232459252e-06,
|
| 71 |
+
"loss": 0.126,
|
| 72 |
+
"step": 99
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.1286549707602339,
|
| 76 |
+
"grad_norm": 0.5180703997612,
|
| 77 |
+
"learning_rate": 3.508150248051028e-06,
|
| 78 |
+
"loss": 0.1942,
|
| 79 |
+
"step": 110
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.1415204678362573,
|
| 83 |
+
"grad_norm": 0.6447436809539795,
|
| 84 |
+
"learning_rate": 3.85896527285613e-06,
|
| 85 |
+
"loss": 0.2089,
|
| 86 |
+
"step": 121
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.1543859649122807,
|
| 90 |
+
"grad_norm": 0.5512965321540833,
|
| 91 |
+
"learning_rate": 4.209780297661234e-06,
|
| 92 |
+
"loss": 0.1225,
|
| 93 |
+
"step": 132
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.1672514619883041,
|
| 97 |
+
"grad_norm": 0.5181564688682556,
|
| 98 |
+
"learning_rate": 4.560595322466336e-06,
|
| 99 |
+
"loss": 0.2504,
|
| 100 |
+
"step": 143
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.18011695906432748,
|
| 104 |
+
"grad_norm": 1.549516201019287,
|
| 105 |
+
"learning_rate": 4.911410347271439e-06,
|
| 106 |
+
"loss": 0.1454,
|
| 107 |
+
"step": 154
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.19298245614035087,
|
| 111 |
+
"grad_norm": 0.06744886934757233,
|
| 112 |
+
"learning_rate": 5.262225372076542e-06,
|
| 113 |
+
"loss": 0.2052,
|
| 114 |
+
"step": 165
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.20584795321637428,
|
| 118 |
+
"grad_norm": 0.3412599265575409,
|
| 119 |
+
"learning_rate": 5.613040396881644e-06,
|
| 120 |
+
"loss": 0.1321,
|
| 121 |
+
"step": 176
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.21871345029239767,
|
| 125 |
+
"grad_norm": 0.4878319203853607,
|
| 126 |
+
"learning_rate": 5.963855421686748e-06,
|
| 127 |
+
"loss": 0.1975,
|
| 128 |
+
"step": 187
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.23157894736842105,
|
| 132 |
+
"grad_norm": 0.6517124176025391,
|
| 133 |
+
"learning_rate": 6.3146704464918505e-06,
|
| 134 |
+
"loss": 0.1615,
|
| 135 |
+
"step": 198
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.24444444444444444,
|
| 139 |
+
"grad_norm": 0.1128321960568428,
|
| 140 |
+
"learning_rate": 6.665485471296953e-06,
|
| 141 |
+
"loss": 0.2174,
|
| 142 |
+
"step": 209
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.25029239766081873,
|
| 146 |
+
"eval_Qnli-dev_cosine_accuracy": 0.697265625,
|
| 147 |
+
"eval_Qnli-dev_cosine_accuracy_threshold": 0.6495540738105774,
|
| 148 |
+
"eval_Qnli-dev_cosine_ap": 0.7205205074888317,
|
| 149 |
+
"eval_Qnli-dev_cosine_f1": 0.6815286624203821,
|
| 150 |
+
"eval_Qnli-dev_cosine_f1_threshold": 0.4835514724254608,
|
| 151 |
+
"eval_Qnli-dev_cosine_precision": 0.5459183673469388,
|
| 152 |
+
"eval_Qnli-dev_cosine_recall": 0.9067796610169492,
|
| 153 |
+
"eval_Qnli-dev_dot_accuracy": 0.666015625,
|
| 154 |
+
"eval_Qnli-dev_dot_accuracy_threshold": 307.405517578125,
|
| 155 |
+
"eval_Qnli-dev_dot_ap": 0.676904834512821,
|
| 156 |
+
"eval_Qnli-dev_dot_f1": 0.6718027734976887,
|
| 157 |
+
"eval_Qnli-dev_dot_f1_threshold": 182.19287109375,
|
| 158 |
+
"eval_Qnli-dev_dot_precision": 0.5278450363196125,
|
| 159 |
+
"eval_Qnli-dev_dot_recall": 0.923728813559322,
|
| 160 |
+
"eval_Qnli-dev_euclidean_accuracy": 0.70703125,
|
| 161 |
+
"eval_Qnli-dev_euclidean_accuracy_threshold": 17.179746627807617,
|
| 162 |
+
"eval_Qnli-dev_euclidean_ap": 0.7311696464051982,
|
| 163 |
+
"eval_Qnli-dev_euclidean_f1": 0.6860706860706861,
|
| 164 |
+
"eval_Qnli-dev_euclidean_f1_threshold": 17.50692367553711,
|
| 165 |
+
"eval_Qnli-dev_euclidean_precision": 0.673469387755102,
|
| 166 |
+
"eval_Qnli-dev_euclidean_recall": 0.6991525423728814,
|
| 167 |
+
"eval_Qnli-dev_manhattan_accuracy": 0.703125,
|
| 168 |
+
"eval_Qnli-dev_manhattan_accuracy_threshold": 357.37890625,
|
| 169 |
+
"eval_Qnli-dev_manhattan_ap": 0.7327744127614749,
|
| 170 |
+
"eval_Qnli-dev_manhattan_f1": 0.680161943319838,
|
| 171 |
+
"eval_Qnli-dev_manhattan_f1_threshold": 373.75103759765625,
|
| 172 |
+
"eval_Qnli-dev_manhattan_precision": 0.6511627906976745,
|
| 173 |
+
"eval_Qnli-dev_manhattan_recall": 0.711864406779661,
|
| 174 |
+
"eval_Qnli-dev_max_accuracy": 0.70703125,
|
| 175 |
+
"eval_Qnli-dev_max_accuracy_threshold": 357.37890625,
|
| 176 |
+
"eval_Qnli-dev_max_ap": 0.7327744127614749,
|
| 177 |
+
"eval_Qnli-dev_max_f1": 0.6860706860706861,
|
| 178 |
+
"eval_Qnli-dev_max_f1_threshold": 373.75103759765625,
|
| 179 |
+
"eval_Qnli-dev_max_precision": 0.673469387755102,
|
| 180 |
+
"eval_Qnli-dev_max_recall": 0.923728813559322,
|
| 181 |
+
"eval_allNLI-dev_cosine_accuracy": 0.740234375,
|
| 182 |
+
"eval_allNLI-dev_cosine_accuracy_threshold": 0.7851241827011108,
|
| 183 |
+
"eval_allNLI-dev_cosine_ap": 0.6340322426598262,
|
| 184 |
+
"eval_allNLI-dev_cosine_f1": 0.6371681415929203,
|
| 185 |
+
"eval_allNLI-dev_cosine_f1_threshold": 0.5923386812210083,
|
| 186 |
+
"eval_allNLI-dev_cosine_precision": 0.5161290322580645,
|
| 187 |
+
"eval_allNLI-dev_cosine_recall": 0.8323699421965318,
|
| 188 |
+
"eval_allNLI-dev_dot_accuracy": 0.724609375,
|
| 189 |
+
"eval_allNLI-dev_dot_accuracy_threshold": 294.0272521972656,
|
| 190 |
+
"eval_allNLI-dev_dot_ap": 0.5910094855503549,
|
| 191 |
+
"eval_allNLI-dev_dot_f1": 0.5960264900662251,
|
| 192 |
+
"eval_allNLI-dev_dot_f1_threshold": 227.6922149658203,
|
| 193 |
+
"eval_allNLI-dev_dot_precision": 0.48214285714285715,
|
| 194 |
+
"eval_allNLI-dev_dot_recall": 0.7803468208092486,
|
| 195 |
+
"eval_allNLI-dev_euclidean_accuracy": 0.748046875,
|
| 196 |
+
"eval_allNLI-dev_euclidean_accuracy_threshold": 13.517601013183594,
|
| 197 |
+
"eval_allNLI-dev_euclidean_ap": 0.6422252682731053,
|
| 198 |
+
"eval_allNLI-dev_euclidean_f1": 0.65,
|
| 199 |
+
"eval_allNLI-dev_euclidean_f1_threshold": 17.501617431640625,
|
| 200 |
+
"eval_allNLI-dev_euclidean_precision": 0.5355805243445693,
|
| 201 |
+
"eval_allNLI-dev_euclidean_recall": 0.8265895953757225,
|
| 202 |
+
"eval_allNLI-dev_manhattan_accuracy": 0.74609375,
|
| 203 |
+
"eval_allNLI-dev_manhattan_accuracy_threshold": 298.3919677734375,
|
| 204 |
+
"eval_allNLI-dev_manhattan_ap": 0.6432220315530937,
|
| 205 |
+
"eval_allNLI-dev_manhattan_f1": 0.6618357487922705,
|
| 206 |
+
"eval_allNLI-dev_manhattan_f1_threshold": 358.1668701171875,
|
| 207 |
+
"eval_allNLI-dev_manhattan_precision": 0.5684647302904564,
|
| 208 |
+
"eval_allNLI-dev_manhattan_recall": 0.791907514450867,
|
| 209 |
+
"eval_allNLI-dev_max_accuracy": 0.748046875,
|
| 210 |
+
"eval_allNLI-dev_max_accuracy_threshold": 298.3919677734375,
|
| 211 |
+
"eval_allNLI-dev_max_ap": 0.6432220315530937,
|
| 212 |
+
"eval_allNLI-dev_max_f1": 0.6618357487922705,
|
| 213 |
+
"eval_allNLI-dev_max_f1_threshold": 358.1668701171875,
|
| 214 |
+
"eval_allNLI-dev_max_precision": 0.5684647302904564,
|
| 215 |
+
"eval_allNLI-dev_max_recall": 0.8323699421965318,
|
| 216 |
+
"eval_sequential_score": 0.7327744127614749,
|
| 217 |
+
"eval_sts-test_pearson_cosine": 0.8852985023833193,
|
| 218 |
+
"eval_sts-test_pearson_dot": 0.8698187375798812,
|
| 219 |
+
"eval_sts-test_pearson_euclidean": 0.9048697523934177,
|
| 220 |
+
"eval_sts-test_pearson_manhattan": 0.9054784772409876,
|
| 221 |
+
"eval_sts-test_pearson_max": 0.9054784772409876,
|
| 222 |
+
"eval_sts-test_spearman_cosine": 0.9079225861122531,
|
| 223 |
+
"eval_sts-test_spearman_dot": 0.8661416896957076,
|
| 224 |
+
"eval_sts-test_spearman_euclidean": 0.9007419777144345,
|
| 225 |
+
"eval_sts-test_spearman_manhattan": 0.9013345291665179,
|
| 226 |
+
"eval_sts-test_spearman_max": 0.9079225861122531,
|
| 227 |
+
"eval_vitaminc-pairs_loss": 2.7610974311828613,
|
| 228 |
+
"eval_vitaminc-pairs_runtime": 3.8706,
|
| 229 |
+
"eval_vitaminc-pairs_samples_per_second": 33.07,
|
| 230 |
+
"eval_vitaminc-pairs_steps_per_second": 0.258,
|
| 231 |
+
"step": 214
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 0.25029239766081873,
|
| 235 |
+
"eval_negation-triplets_loss": 1.0606263875961304,
|
| 236 |
+
"eval_negation-triplets_runtime": 0.6073,
|
| 237 |
+
"eval_negation-triplets_samples_per_second": 210.757,
|
| 238 |
+
"eval_negation-triplets_steps_per_second": 1.647,
|
| 239 |
+
"step": 214
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 0.25029239766081873,
|
| 243 |
+
"eval_scitail-pairs-pos_loss": 0.059617433696985245,
|
| 244 |
+
"eval_scitail-pairs-pos_runtime": 0.9151,
|
| 245 |
+
"eval_scitail-pairs-pos_samples_per_second": 139.868,
|
| 246 |
+
"eval_scitail-pairs-pos_steps_per_second": 1.093,
|
| 247 |
+
"step": 214
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.25029239766081873,
|
| 251 |
+
"eval_scitail-pairs-qa_loss": 2.513320396246854e-07,
|
| 252 |
+
"eval_scitail-pairs-qa_runtime": 0.5906,
|
| 253 |
+
"eval_scitail-pairs-qa_samples_per_second": 216.739,
|
| 254 |
+
"eval_scitail-pairs-qa_steps_per_second": 1.693,
|
| 255 |
+
"step": 214
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.25029239766081873,
|
| 259 |
+
"eval_xsum-pairs_loss": 0.004431054927408695,
|
| 260 |
+
"eval_xsum-pairs_runtime": 2.7873,
|
| 261 |
+
"eval_xsum-pairs_samples_per_second": 45.922,
|
| 262 |
+
"eval_xsum-pairs_steps_per_second": 0.359,
|
| 263 |
+
"step": 214
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 0.25029239766081873,
|
| 267 |
+
"eval_sciq_pairs_loss": 0.013232991099357605,
|
| 268 |
+
"eval_sciq_pairs_runtime": 4.1862,
|
| 269 |
+
"eval_sciq_pairs_samples_per_second": 30.576,
|
| 270 |
+
"eval_sciq_pairs_steps_per_second": 0.239,
|
| 271 |
+
"step": 214
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.25029239766081873,
|
| 275 |
+
"eval_qasc_pairs_loss": 0.049007147550582886,
|
| 276 |
+
"eval_qasc_pairs_runtime": 0.6288,
|
| 277 |
+
"eval_qasc_pairs_samples_per_second": 203.554,
|
| 278 |
+
"eval_qasc_pairs_steps_per_second": 1.59,
|
| 279 |
+
"step": 214
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.25029239766081873,
|
| 283 |
+
"eval_openbookqa_pairs_loss": 1.3196656703948975,
|
| 284 |
+
"eval_openbookqa_pairs_runtime": 0.5848,
|
| 285 |
+
"eval_openbookqa_pairs_samples_per_second": 218.897,
|
| 286 |
+
"eval_openbookqa_pairs_steps_per_second": 1.71,
|
| 287 |
+
"step": 214
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.25029239766081873,
|
| 291 |
+
"eval_msmarco_pairs_loss": 0.13587385416030884,
|
| 292 |
+
"eval_msmarco_pairs_runtime": 1.341,
|
| 293 |
+
"eval_msmarco_pairs_samples_per_second": 95.453,
|
| 294 |
+
"eval_msmarco_pairs_steps_per_second": 0.746,
|
| 295 |
+
"step": 214
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.25029239766081873,
|
| 299 |
+
"eval_nq_pairs_loss": 0.04134061187505722,
|
| 300 |
+
"eval_nq_pairs_runtime": 2.9296,
|
| 301 |
+
"eval_nq_pairs_samples_per_second": 43.693,
|
| 302 |
+
"eval_nq_pairs_steps_per_second": 0.341,
|
| 303 |
+
"step": 214
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.25029239766081873,
|
| 307 |
+
"eval_trivia_pairs_loss": 0.211629718542099,
|
| 308 |
+
"eval_trivia_pairs_runtime": 3.2033,
|
| 309 |
+
"eval_trivia_pairs_samples_per_second": 39.958,
|
| 310 |
+
"eval_trivia_pairs_steps_per_second": 0.312,
|
| 311 |
+
"step": 214
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.25029239766081873,
|
| 315 |
+
"eval_gooaq_pairs_loss": 0.11869887262582779,
|
| 316 |
+
"eval_gooaq_pairs_runtime": 0.869,
|
| 317 |
+
"eval_gooaq_pairs_samples_per_second": 147.291,
|
| 318 |
+
"eval_gooaq_pairs_steps_per_second": 1.151,
|
| 319 |
+
"step": 214
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 0.25029239766081873,
|
| 323 |
+
"eval_paws-pos_loss": 0.04590649530291557,
|
| 324 |
+
"eval_paws-pos_runtime": 0.6652,
|
| 325 |
+
"eval_paws-pos_samples_per_second": 192.433,
|
| 326 |
+
"eval_paws-pos_steps_per_second": 1.503,
|
| 327 |
+
"step": 214
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 0.25029239766081873,
|
| 331 |
+
"eval_global_dataset_loss": 0.21873071789741516,
|
| 332 |
+
"eval_global_dataset_runtime": 24.2472,
|
| 333 |
+
"eval_global_dataset_samples_per_second": 27.343,
|
| 334 |
+
"eval_global_dataset_steps_per_second": 0.247,
|
| 335 |
+
"step": 214
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 0.2573099415204678,
|
| 339 |
+
"grad_norm": 0.2186487466096878,
|
| 340 |
+
"learning_rate": 7.016300496102056e-06,
|
| 341 |
+
"loss": 0.1545,
|
| 342 |
+
"step": 220
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 0.27017543859649124,
|
| 346 |
+
"grad_norm": 0.16798946261405945,
|
| 347 |
+
"learning_rate": 7.3671155209071584e-06,
|
| 348 |
+
"loss": 0.1201,
|
| 349 |
+
"step": 231
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 0.2830409356725146,
|
| 353 |
+
"grad_norm": 0.5161852836608887,
|
| 354 |
+
"learning_rate": 7.71793054571226e-06,
|
| 355 |
+
"loss": 0.1389,
|
| 356 |
+
"step": 242
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.295906432748538,
|
| 360 |
+
"grad_norm": 0.6402292251586914,
|
| 361 |
+
"learning_rate": 8.068745570517365e-06,
|
| 362 |
+
"loss": 0.1657,
|
| 363 |
+
"step": 253
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 0.3087719298245614,
|
| 367 |
+
"grad_norm": 0.8064345717430115,
|
| 368 |
+
"learning_rate": 8.419560595322467e-06,
|
| 369 |
+
"loss": 0.237,
|
| 370 |
+
"step": 264
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 0.3216374269005848,
|
| 374 |
+
"grad_norm": 0.8077338337898254,
|
| 375 |
+
"learning_rate": 8.770375620127569e-06,
|
| 376 |
+
"loss": 0.1094,
|
| 377 |
+
"step": 275
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.3345029239766082,
|
| 381 |
+
"grad_norm": 0.23748837411403656,
|
| 382 |
+
"learning_rate": 9.121190644932673e-06,
|
| 383 |
+
"loss": 0.196,
|
| 384 |
+
"step": 286
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.3473684210526316,
|
| 388 |
+
"grad_norm": 0.26720044016838074,
|
| 389 |
+
"learning_rate": 9.472005669737774e-06,
|
| 390 |
+
"loss": 0.2164,
|
| 391 |
+
"step": 297
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 0.36023391812865496,
|
| 395 |
+
"grad_norm": 0.8082326650619507,
|
| 396 |
+
"learning_rate": 9.822820694542878e-06,
|
| 397 |
+
"loss": 0.1793,
|
| 398 |
+
"step": 308
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.3730994152046784,
|
| 402 |
+
"grad_norm": 0.8150445818901062,
|
| 403 |
+
"learning_rate": 1.017363571934798e-05,
|
| 404 |
+
"loss": 0.2878,
|
| 405 |
+
"step": 319
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 0.38596491228070173,
|
| 409 |
+
"grad_norm": 0.2097979038953781,
|
| 410 |
+
"learning_rate": 1.0524450744153083e-05,
|
| 411 |
+
"loss": 0.1189,
|
| 412 |
+
"step": 330
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 0.39883040935672515,
|
| 416 |
+
"grad_norm": 1.0323094129562378,
|
| 417 |
+
"learning_rate": 1.0875265768958187e-05,
|
| 418 |
+
"loss": 0.1475,
|
| 419 |
+
"step": 341
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 0.41169590643274856,
|
| 423 |
+
"grad_norm": 0.41917282342910767,
|
| 424 |
+
"learning_rate": 1.1226080793763289e-05,
|
| 425 |
+
"loss": 0.1019,
|
| 426 |
+
"step": 352
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 0.4245614035087719,
|
| 430 |
+
"grad_norm": 0.17783042788505554,
|
| 431 |
+
"learning_rate": 1.1576895818568394e-05,
|
| 432 |
+
"loss": 0.1587,
|
| 433 |
+
"step": 363
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 0.43742690058479533,
|
| 437 |
+
"grad_norm": 0.8943735957145691,
|
| 438 |
+
"learning_rate": 1.1927710843373496e-05,
|
| 439 |
+
"loss": 0.2483,
|
| 440 |
+
"step": 374
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.4502923976608187,
|
| 444 |
+
"grad_norm": 0.7853964567184448,
|
| 445 |
+
"learning_rate": 1.2278525868178599e-05,
|
| 446 |
+
"loss": 0.1427,
|
| 447 |
+
"step": 385
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 0.4631578947368421,
|
| 451 |
+
"grad_norm": 0.4753592908382416,
|
| 452 |
+
"learning_rate": 1.2629340892983701e-05,
|
| 453 |
+
"loss": 0.1199,
|
| 454 |
+
"step": 396
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 0.4760233918128655,
|
| 458 |
+
"grad_norm": 0.47377559542655945,
|
| 459 |
+
"learning_rate": 1.2980155917788803e-05,
|
| 460 |
+
"loss": 0.2037,
|
| 461 |
+
"step": 407
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 0.4888888888888889,
|
| 465 |
+
"grad_norm": 0.6249018311500549,
|
| 466 |
+
"learning_rate": 1.3330970942593906e-05,
|
| 467 |
+
"loss": 0.1317,
|
| 468 |
+
"step": 418
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 0.5005847953216375,
|
| 472 |
+
"eval_Qnli-dev_cosine_accuracy": 0.6953125,
|
| 473 |
+
"eval_Qnli-dev_cosine_accuracy_threshold": 0.6539117097854614,
|
| 474 |
+
"eval_Qnli-dev_cosine_ap": 0.7174538904795291,
|
| 475 |
+
"eval_Qnli-dev_cosine_f1": 0.6773675762439807,
|
| 476 |
+
"eval_Qnli-dev_cosine_f1_threshold": 0.4876832365989685,
|
| 477 |
+
"eval_Qnli-dev_cosine_precision": 0.5452196382428941,
|
| 478 |
+
"eval_Qnli-dev_cosine_recall": 0.8940677966101694,
|
| 479 |
+
"eval_Qnli-dev_dot_accuracy": 0.673828125,
|
| 480 |
+
"eval_Qnli-dev_dot_accuracy_threshold": 248.8419647216797,
|
| 481 |
+
"eval_Qnli-dev_dot_ap": 0.6692436401997881,
|
| 482 |
+
"eval_Qnli-dev_dot_f1": 0.6758832565284177,
|
| 483 |
+
"eval_Qnli-dev_dot_f1_threshold": 174.55992126464844,
|
| 484 |
+
"eval_Qnli-dev_dot_precision": 0.5301204819277109,
|
| 485 |
+
"eval_Qnli-dev_dot_recall": 0.9322033898305084,
|
| 486 |
+
"eval_Qnli-dev_euclidean_accuracy": 0.70703125,
|
| 487 |
+
"eval_Qnli-dev_euclidean_accuracy_threshold": 16.61339569091797,
|
| 488 |
+
"eval_Qnli-dev_euclidean_ap": 0.7278403442362325,
|
| 489 |
+
"eval_Qnli-dev_euclidean_f1": 0.6821052631578948,
|
| 490 |
+
"eval_Qnli-dev_euclidean_f1_threshold": 17.059036254882812,
|
| 491 |
+
"eval_Qnli-dev_euclidean_precision": 0.6778242677824268,
|
| 492 |
+
"eval_Qnli-dev_euclidean_recall": 0.6864406779661016,
|
| 493 |
+
"eval_Qnli-dev_manhattan_accuracy": 0.697265625,
|
| 494 |
+
"eval_Qnli-dev_manhattan_accuracy_threshold": 348.561279296875,
|
| 495 |
+
"eval_Qnli-dev_manhattan_ap": 0.7293997108333973,
|
| 496 |
+
"eval_Qnli-dev_manhattan_f1": 0.6797642436149313,
|
| 497 |
+
"eval_Qnli-dev_manhattan_f1_threshold": 371.4664306640625,
|
| 498 |
+
"eval_Qnli-dev_manhattan_precision": 0.6336996336996337,
|
| 499 |
+
"eval_Qnli-dev_manhattan_recall": 0.7330508474576272,
|
| 500 |
+
"eval_Qnli-dev_max_accuracy": 0.70703125,
|
| 501 |
+
"eval_Qnli-dev_max_accuracy_threshold": 348.561279296875,
|
| 502 |
+
"eval_Qnli-dev_max_ap": 0.7293997108333973,
|
| 503 |
+
"eval_Qnli-dev_max_f1": 0.6821052631578948,
|
| 504 |
+
"eval_Qnli-dev_max_f1_threshold": 371.4664306640625,
|
| 505 |
+
"eval_Qnli-dev_max_precision": 0.6778242677824268,
|
| 506 |
+
"eval_Qnli-dev_max_recall": 0.9322033898305084,
|
| 507 |
+
"eval_allNLI-dev_cosine_accuracy": 0.740234375,
|
| 508 |
+
"eval_allNLI-dev_cosine_accuracy_threshold": 0.789513111114502,
|
| 509 |
+
"eval_allNLI-dev_cosine_ap": 0.6326730829325957,
|
| 510 |
+
"eval_allNLI-dev_cosine_f1": 0.6369710467706012,
|
| 511 |
+
"eval_allNLI-dev_cosine_f1_threshold": 0.6001436710357666,
|
| 512 |
+
"eval_allNLI-dev_cosine_precision": 0.5181159420289855,
|
| 513 |
+
"eval_allNLI-dev_cosine_recall": 0.8265895953757225,
|
| 514 |
+
"eval_allNLI-dev_dot_accuracy": 0.728515625,
|
| 515 |
+
"eval_allNLI-dev_dot_accuracy_threshold": 284.5158996582031,
|
| 516 |
+
"eval_allNLI-dev_dot_ap": 0.587991265887336,
|
| 517 |
+
"eval_allNLI-dev_dot_f1": 0.5996204933586338,
|
| 518 |
+
"eval_allNLI-dev_dot_f1_threshold": 187.17926025390625,
|
| 519 |
+
"eval_allNLI-dev_dot_precision": 0.4463276836158192,
|
| 520 |
+
"eval_allNLI-dev_dot_recall": 0.9132947976878613,
|
| 521 |
+
"eval_allNLI-dev_euclidean_accuracy": 0.740234375,
|
| 522 |
+
"eval_allNLI-dev_euclidean_accuracy_threshold": 13.375718116760254,
|
| 523 |
+
"eval_allNLI-dev_euclidean_ap": 0.6391067622382595,
|
| 524 |
+
"eval_allNLI-dev_euclidean_f1": 0.6478260869565217,
|
| 525 |
+
"eval_allNLI-dev_euclidean_f1_threshold": 17.581111907958984,
|
| 526 |
+
"eval_allNLI-dev_euclidean_precision": 0.519163763066202,
|
| 527 |
+
"eval_allNLI-dev_euclidean_recall": 0.861271676300578,
|
| 528 |
+
"eval_allNLI-dev_manhattan_accuracy": 0.740234375,
|
| 529 |
+
"eval_allNLI-dev_manhattan_accuracy_threshold": 271.5396728515625,
|
| 530 |
+
"eval_allNLI-dev_manhattan_ap": 0.6397141445378819,
|
| 531 |
+
"eval_allNLI-dev_manhattan_f1": 0.6558139534883721,
|
| 532 |
+
"eval_allNLI-dev_manhattan_f1_threshold": 356.73675537109375,
|
| 533 |
+
"eval_allNLI-dev_manhattan_precision": 0.5486381322957199,
|
| 534 |
+
"eval_allNLI-dev_manhattan_recall": 0.815028901734104,
|
| 535 |
+
"eval_allNLI-dev_max_accuracy": 0.740234375,
|
| 536 |
+
"eval_allNLI-dev_max_accuracy_threshold": 284.5158996582031,
|
| 537 |
+
"eval_allNLI-dev_max_ap": 0.6397141445378819,
|
| 538 |
+
"eval_allNLI-dev_max_f1": 0.6558139534883721,
|
| 539 |
+
"eval_allNLI-dev_max_f1_threshold": 356.73675537109375,
|
| 540 |
+
"eval_allNLI-dev_max_precision": 0.5486381322957199,
|
| 541 |
+
"eval_allNLI-dev_max_recall": 0.9132947976878613,
|
| 542 |
+
"eval_sequential_score": 0.7293997108333973,
|
| 543 |
+
"eval_sts-test_pearson_cosine": 0.8826342923852873,
|
| 544 |
+
"eval_sts-test_pearson_dot": 0.8663977738237862,
|
| 545 |
+
"eval_sts-test_pearson_euclidean": 0.9007138184358896,
|
| 546 |
+
"eval_sts-test_pearson_manhattan": 0.9014088314200603,
|
| 547 |
+
"eval_sts-test_pearson_max": 0.9014088314200603,
|
| 548 |
+
"eval_sts-test_spearman_cosine": 0.9062742181501622,
|
| 549 |
+
"eval_sts-test_spearman_dot": 0.8628408814067393,
|
| 550 |
+
"eval_sts-test_spearman_euclidean": 0.897460680975433,
|
| 551 |
+
"eval_sts-test_spearman_manhattan": 0.8979340598184311,
|
| 552 |
+
"eval_sts-test_spearman_max": 0.9062742181501622,
|
| 553 |
+
"eval_vitaminc-pairs_loss": 2.7317824363708496,
|
| 554 |
+
"eval_vitaminc-pairs_runtime": 3.8325,
|
| 555 |
+
"eval_vitaminc-pairs_samples_per_second": 33.398,
|
| 556 |
+
"eval_vitaminc-pairs_steps_per_second": 0.261,
|
| 557 |
+
"step": 428
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 0.5005847953216375,
|
| 561 |
+
"eval_negation-triplets_loss": 1.0631705522537231,
|
| 562 |
+
"eval_negation-triplets_runtime": 0.6008,
|
| 563 |
+
"eval_negation-triplets_samples_per_second": 213.032,
|
| 564 |
+
"eval_negation-triplets_steps_per_second": 1.664,
|
| 565 |
+
"step": 428
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 0.5005847953216375,
|
| 569 |
+
"eval_scitail-pairs-pos_loss": 0.06051468849182129,
|
| 570 |
+
"eval_scitail-pairs-pos_runtime": 0.8927,
|
| 571 |
+
"eval_scitail-pairs-pos_samples_per_second": 143.381,
|
| 572 |
+
"eval_scitail-pairs-pos_steps_per_second": 1.12,
|
| 573 |
+
"step": 428
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 0.5005847953216375,
|
| 577 |
+
"eval_scitail-pairs-qa_loss": 3.731894366865163e-07,
|
| 578 |
+
"eval_scitail-pairs-qa_runtime": 0.5829,
|
| 579 |
+
"eval_scitail-pairs-qa_samples_per_second": 219.584,
|
| 580 |
+
"eval_scitail-pairs-qa_steps_per_second": 1.716,
|
| 581 |
+
"step": 428
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 0.5005847953216375,
|
| 585 |
+
"eval_xsum-pairs_loss": 0.004530374892055988,
|
| 586 |
+
"eval_xsum-pairs_runtime": 2.7882,
|
| 587 |
+
"eval_xsum-pairs_samples_per_second": 45.908,
|
| 588 |
+
"eval_xsum-pairs_steps_per_second": 0.359,
|
| 589 |
+
"step": 428
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 0.5005847953216375,
|
| 593 |
+
"eval_sciq_pairs_loss": 0.0028694826178252697,
|
| 594 |
+
"eval_sciq_pairs_runtime": 4.1341,
|
| 595 |
+
"eval_sciq_pairs_samples_per_second": 30.962,
|
| 596 |
+
"eval_sciq_pairs_steps_per_second": 0.242,
|
| 597 |
+
"step": 428
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.5005847953216375,
|
| 601 |
+
"eval_qasc_pairs_loss": 0.0601225420832634,
|
| 602 |
+
"eval_qasc_pairs_runtime": 0.6017,
|
| 603 |
+
"eval_qasc_pairs_samples_per_second": 212.728,
|
| 604 |
+
"eval_qasc_pairs_steps_per_second": 1.662,
|
| 605 |
+
"step": 428
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.5005847953216375,
|
| 609 |
+
"eval_openbookqa_pairs_loss": 1.2716314792633057,
|
| 610 |
+
"eval_openbookqa_pairs_runtime": 0.5746,
|
| 611 |
+
"eval_openbookqa_pairs_samples_per_second": 222.757,
|
| 612 |
+
"eval_openbookqa_pairs_steps_per_second": 1.74,
|
| 613 |
+
"step": 428
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 0.5005847953216375,
|
| 617 |
+
"eval_msmarco_pairs_loss": 0.16218337416648865,
|
| 618 |
+
"eval_msmarco_pairs_runtime": 1.3355,
|
| 619 |
+
"eval_msmarco_pairs_samples_per_second": 95.842,
|
| 620 |
+
"eval_msmarco_pairs_steps_per_second": 0.749,
|
| 621 |
+
"step": 428
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 0.5005847953216375,
|
| 625 |
+
"eval_nq_pairs_loss": 0.039985254406929016,
|
| 626 |
+
"eval_nq_pairs_runtime": 2.9238,
|
| 627 |
+
"eval_nq_pairs_samples_per_second": 43.779,
|
| 628 |
+
"eval_nq_pairs_steps_per_second": 0.342,
|
| 629 |
+
"step": 428
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 0.5005847953216375,
|
| 633 |
+
"eval_trivia_pairs_loss": 0.20936143398284912,
|
| 634 |
+
"eval_trivia_pairs_runtime": 3.1822,
|
| 635 |
+
"eval_trivia_pairs_samples_per_second": 40.223,
|
| 636 |
+
"eval_trivia_pairs_steps_per_second": 0.314,
|
| 637 |
+
"step": 428
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 0.5005847953216375,
|
| 641 |
+
"eval_gooaq_pairs_loss": 0.12705038487911224,
|
| 642 |
+
"eval_gooaq_pairs_runtime": 0.8602,
|
| 643 |
+
"eval_gooaq_pairs_samples_per_second": 148.795,
|
| 644 |
+
"eval_gooaq_pairs_steps_per_second": 1.162,
|
| 645 |
+
"step": 428
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 0.5005847953216375,
|
| 649 |
+
"eval_paws-pos_loss": 0.046078916639089584,
|
| 650 |
+
"eval_paws-pos_runtime": 0.6578,
|
| 651 |
+
"eval_paws-pos_samples_per_second": 194.575,
|
| 652 |
+
"eval_paws-pos_steps_per_second": 1.52,
|
| 653 |
+
"step": 428
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.5005847953216375,
|
| 657 |
+
"eval_global_dataset_loss": 0.21221815049648285,
|
| 658 |
+
"eval_global_dataset_runtime": 24.2782,
|
| 659 |
+
"eval_global_dataset_samples_per_second": 27.308,
|
| 660 |
+
"eval_global_dataset_steps_per_second": 0.247,
|
| 661 |
+
"step": 428
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.5017543859649123,
|
| 665 |
+
"grad_norm": 0.6901716589927673,
|
| 666 |
+
"learning_rate": 1.3681785967399008e-05,
|
| 667 |
+
"loss": 0.1293,
|
| 668 |
+
"step": 429
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.5146198830409356,
|
| 672 |
+
"grad_norm": 1.1465657949447632,
|
| 673 |
+
"learning_rate": 1.4032600992204112e-05,
|
| 674 |
+
"loss": 0.1902,
|
| 675 |
+
"step": 440
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.5274853801169591,
|
| 679 |
+
"grad_norm": 1.0795351266860962,
|
| 680 |
+
"learning_rate": 1.4383416017009213e-05,
|
| 681 |
+
"loss": 0.1429,
|
| 682 |
+
"step": 451
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.5403508771929825,
|
| 686 |
+
"grad_norm": 0.9797111749649048,
|
| 687 |
+
"learning_rate": 1.4734231041814317e-05,
|
| 688 |
+
"loss": 0.2446,
|
| 689 |
+
"step": 462
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.5532163742690058,
|
| 693 |
+
"grad_norm": 0.21088336408138275,
|
| 694 |
+
"learning_rate": 1.5085046066619419e-05,
|
| 695 |
+
"loss": 0.1623,
|
| 696 |
+
"step": 473
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.5660818713450292,
|
| 700 |
+
"grad_norm": 0.4258130192756653,
|
| 701 |
+
"learning_rate": 1.543586109142452e-05,
|
| 702 |
+
"loss": 0.0707,
|
| 703 |
+
"step": 484
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.5789473684210527,
|
| 707 |
+
"grad_norm": 0.8029544949531555,
|
| 708 |
+
"learning_rate": 1.5786676116229624e-05,
|
| 709 |
+
"loss": 0.1557,
|
| 710 |
+
"step": 495
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.591812865497076,
|
| 714 |
+
"grad_norm": 2.734902858734131,
|
| 715 |
+
"learning_rate": 1.613749114103473e-05,
|
| 716 |
+
"loss": 0.2016,
|
| 717 |
+
"step": 506
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.6046783625730994,
|
| 721 |
+
"grad_norm": 0.4968857169151306,
|
| 722 |
+
"learning_rate": 1.648830616583983e-05,
|
| 723 |
+
"loss": 0.1018,
|
| 724 |
+
"step": 517
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.6175438596491228,
|
| 728 |
+
"grad_norm": 1.5250153541564941,
|
| 729 |
+
"learning_rate": 1.6839121190644935e-05,
|
| 730 |
+
"loss": 0.1821,
|
| 731 |
+
"step": 528
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.6304093567251462,
|
| 735 |
+
"grad_norm": 0.7500655055046082,
|
| 736 |
+
"learning_rate": 1.7189936215450038e-05,
|
| 737 |
+
"loss": 0.1437,
|
| 738 |
+
"step": 539
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.6432748538011696,
|
| 742 |
+
"grad_norm": 0.19580113887786865,
|
| 743 |
+
"learning_rate": 1.7540751240255138e-05,
|
| 744 |
+
"loss": 0.1112,
|
| 745 |
+
"step": 550
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.656140350877193,
|
| 749 |
+
"grad_norm": 0.1376626342535019,
|
| 750 |
+
"learning_rate": 1.789156626506024e-05,
|
| 751 |
+
"loss": 0.12,
|
| 752 |
+
"step": 561
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.6690058479532164,
|
| 756 |
+
"grad_norm": 0.49516016244888306,
|
| 757 |
+
"learning_rate": 1.8242381289865345e-05,
|
| 758 |
+
"loss": 0.0933,
|
| 759 |
+
"step": 572
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.6818713450292397,
|
| 763 |
+
"grad_norm": 0.20120573043823242,
|
| 764 |
+
"learning_rate": 1.859319631467045e-05,
|
| 765 |
+
"loss": 0.0939,
|
| 766 |
+
"step": 583
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.6947368421052632,
|
| 770 |
+
"grad_norm": 0.3648889362812042,
|
| 771 |
+
"learning_rate": 1.894401133947555e-05,
|
| 772 |
+
"loss": 0.2064,
|
| 773 |
+
"step": 594
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.7076023391812866,
|
| 777 |
+
"grad_norm": 0.07958167791366577,
|
| 778 |
+
"learning_rate": 1.9294826364280652e-05,
|
| 779 |
+
"loss": 0.131,
|
| 780 |
+
"step": 605
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.7204678362573099,
|
| 784 |
+
"grad_norm": 0.7020072340965271,
|
| 785 |
+
"learning_rate": 1.9645641389085756e-05,
|
| 786 |
+
"loss": 0.161,
|
| 787 |
+
"step": 616
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.7333333333333333,
|
| 791 |
+
"grad_norm": 1.3057806491851807,
|
| 792 |
+
"learning_rate": 1.999645641389086e-05,
|
| 793 |
+
"loss": 0.213,
|
| 794 |
+
"step": 627
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.7461988304093568,
|
| 798 |
+
"grad_norm": 0.7668664455413818,
|
| 799 |
+
"learning_rate": 2.034727143869596e-05,
|
| 800 |
+
"loss": 0.1853,
|
| 801 |
+
"step": 638
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.7508771929824561,
|
| 805 |
+
"eval_Qnli-dev_cosine_accuracy": 0.693359375,
|
| 806 |
+
"eval_Qnli-dev_cosine_accuracy_threshold": 0.6550551652908325,
|
| 807 |
+
"eval_Qnli-dev_cosine_ap": 0.7155098805507838,
|
| 808 |
+
"eval_Qnli-dev_cosine_f1": 0.6794871794871794,
|
| 809 |
+
"eval_Qnli-dev_cosine_f1_threshold": 0.4866919219493866,
|
| 810 |
+
"eval_Qnli-dev_cosine_precision": 0.5463917525773195,
|
| 811 |
+
"eval_Qnli-dev_cosine_recall": 0.8983050847457628,
|
| 812 |
+
"eval_Qnli-dev_dot_accuracy": 0.67578125,
|
| 813 |
+
"eval_Qnli-dev_dot_accuracy_threshold": 258.037841796875,
|
| 814 |
+
"eval_Qnli-dev_dot_ap": 0.672701841309456,
|
| 815 |
+
"eval_Qnli-dev_dot_f1": 0.6731078904991948,
|
| 816 |
+
"eval_Qnli-dev_dot_f1_threshold": 197.6945343017578,
|
| 817 |
+
"eval_Qnli-dev_dot_precision": 0.5428571428571428,
|
| 818 |
+
"eval_Qnli-dev_dot_recall": 0.885593220338983,
|
| 819 |
+
"eval_Qnli-dev_euclidean_accuracy": 0.697265625,
|
| 820 |
+
"eval_Qnli-dev_euclidean_accuracy_threshold": 16.745851516723633,
|
| 821 |
+
"eval_Qnli-dev_euclidean_ap": 0.7236455475550527,
|
| 822 |
+
"eval_Qnli-dev_euclidean_f1": 0.6750524109014674,
|
| 823 |
+
"eval_Qnli-dev_euclidean_f1_threshold": 17.19139862060547,
|
| 824 |
+
"eval_Qnli-dev_euclidean_precision": 0.6680497925311203,
|
| 825 |
+
"eval_Qnli-dev_euclidean_recall": 0.6822033898305084,
|
| 826 |
+
"eval_Qnli-dev_manhattan_accuracy": 0.697265625,
|
| 827 |
+
"eval_Qnli-dev_manhattan_accuracy_threshold": 362.6970520019531,
|
| 828 |
+
"eval_Qnli-dev_manhattan_ap": 0.7257321921740054,
|
| 829 |
+
"eval_Qnli-dev_manhattan_f1": 0.6774847870182557,
|
| 830 |
+
"eval_Qnli-dev_manhattan_f1_threshold": 369.7593994140625,
|
| 831 |
+
"eval_Qnli-dev_manhattan_precision": 0.6498054474708171,
|
| 832 |
+
"eval_Qnli-dev_manhattan_recall": 0.7076271186440678,
|
| 833 |
+
"eval_Qnli-dev_max_accuracy": 0.697265625,
|
| 834 |
+
"eval_Qnli-dev_max_accuracy_threshold": 362.6970520019531,
|
| 835 |
+
"eval_Qnli-dev_max_ap": 0.7257321921740054,
|
| 836 |
+
"eval_Qnli-dev_max_f1": 0.6794871794871794,
|
| 837 |
+
"eval_Qnli-dev_max_f1_threshold": 369.7593994140625,
|
| 838 |
+
"eval_Qnli-dev_max_precision": 0.6680497925311203,
|
| 839 |
+
"eval_Qnli-dev_max_recall": 0.8983050847457628,
|
| 840 |
+
"eval_allNLI-dev_cosine_accuracy": 0.73828125,
|
| 841 |
+
"eval_allNLI-dev_cosine_accuracy_threshold": 0.7900433540344238,
|
| 842 |
+
"eval_allNLI-dev_cosine_ap": 0.6359348988299178,
|
| 843 |
+
"eval_allNLI-dev_cosine_f1": 0.639269406392694,
|
| 844 |
+
"eval_allNLI-dev_cosine_f1_threshold": 0.6079586744308472,
|
| 845 |
+
"eval_allNLI-dev_cosine_precision": 0.5283018867924528,
|
| 846 |
+
"eval_allNLI-dev_cosine_recall": 0.8092485549132948,
|
| 847 |
+
"eval_allNLI-dev_dot_accuracy": 0.7265625,
|
| 848 |
+
"eval_allNLI-dev_dot_accuracy_threshold": 282.57489013671875,
|
| 849 |
+
"eval_allNLI-dev_dot_ap": 0.589207838274076,
|
| 850 |
+
"eval_allNLI-dev_dot_f1": 0.6059405940594059,
|
| 851 |
+
"eval_allNLI-dev_dot_f1_threshold": 203.0603485107422,
|
| 852 |
+
"eval_allNLI-dev_dot_precision": 0.4608433734939759,
|
| 853 |
+
"eval_allNLI-dev_dot_recall": 0.884393063583815,
|
| 854 |
+
"eval_allNLI-dev_euclidean_accuracy": 0.744140625,
|
| 855 |
+
"eval_allNLI-dev_euclidean_accuracy_threshold": 13.129840850830078,
|
| 856 |
+
"eval_allNLI-dev_euclidean_ap": 0.6431587171633473,
|
| 857 |
+
"eval_allNLI-dev_euclidean_f1": 0.6506550218340611,
|
| 858 |
+
"eval_allNLI-dev_euclidean_f1_threshold": 17.757301330566406,
|
| 859 |
+
"eval_allNLI-dev_euclidean_precision": 0.5228070175438596,
|
| 860 |
+
"eval_allNLI-dev_euclidean_recall": 0.861271676300578,
|
| 861 |
+
"eval_allNLI-dev_manhattan_accuracy": 0.74609375,
|
| 862 |
+
"eval_allNLI-dev_manhattan_accuracy_threshold": 276.65826416015625,
|
| 863 |
+
"eval_allNLI-dev_manhattan_ap": 0.6424686977323816,
|
| 864 |
+
"eval_allNLI-dev_manhattan_f1": 0.6589861751152074,
|
| 865 |
+
"eval_allNLI-dev_manhattan_f1_threshold": 362.68414306640625,
|
| 866 |
+
"eval_allNLI-dev_manhattan_precision": 0.5478927203065134,
|
| 867 |
+
"eval_allNLI-dev_manhattan_recall": 0.8265895953757225,
|
| 868 |
+
"eval_allNLI-dev_max_accuracy": 0.74609375,
|
| 869 |
+
"eval_allNLI-dev_max_accuracy_threshold": 282.57489013671875,
|
| 870 |
+
"eval_allNLI-dev_max_ap": 0.6431587171633473,
|
| 871 |
+
"eval_allNLI-dev_max_f1": 0.6589861751152074,
|
| 872 |
+
"eval_allNLI-dev_max_f1_threshold": 362.68414306640625,
|
| 873 |
+
"eval_allNLI-dev_max_precision": 0.5478927203065134,
|
| 874 |
+
"eval_allNLI-dev_max_recall": 0.884393063583815,
|
| 875 |
+
"eval_sequential_score": 0.7257321921740054,
|
| 876 |
+
"eval_sts-test_pearson_cosine": 0.8848372605057793,
|
| 877 |
+
"eval_sts-test_pearson_dot": 0.8709019996065221,
|
| 878 |
+
"eval_sts-test_pearson_euclidean": 0.9038234321498232,
|
| 879 |
+
"eval_sts-test_pearson_manhattan": 0.9044737426180963,
|
| 880 |
+
"eval_sts-test_pearson_max": 0.9044737426180963,
|
| 881 |
+
"eval_sts-test_spearman_cosine": 0.9073185783372533,
|
| 882 |
+
"eval_sts-test_spearman_dot": 0.8668432552205406,
|
| 883 |
+
"eval_sts-test_spearman_euclidean": 0.899716547310912,
|
| 884 |
+
"eval_sts-test_spearman_manhattan": 0.9006361409812439,
|
| 885 |
+
"eval_sts-test_spearman_max": 0.9073185783372533,
|
| 886 |
+
"eval_vitaminc-pairs_loss": 2.7074272632598877,
|
| 887 |
+
"eval_vitaminc-pairs_runtime": 3.7948,
|
| 888 |
+
"eval_vitaminc-pairs_samples_per_second": 33.73,
|
| 889 |
+
"eval_vitaminc-pairs_steps_per_second": 0.264,
|
| 890 |
+
"step": 642
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 0.7508771929824561,
|
| 894 |
+
"eval_negation-triplets_loss": 1.0683540105819702,
|
| 895 |
+
"eval_negation-triplets_runtime": 0.5897,
|
| 896 |
+
"eval_negation-triplets_samples_per_second": 217.072,
|
| 897 |
+
"eval_negation-triplets_steps_per_second": 1.696,
|
| 898 |
+
"step": 642
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.7508771929824561,
|
| 902 |
+
"eval_scitail-pairs-pos_loss": 0.05928462743759155,
|
| 903 |
+
"eval_scitail-pairs-pos_runtime": 0.8353,
|
| 904 |
+
"eval_scitail-pairs-pos_samples_per_second": 153.247,
|
| 905 |
+
"eval_scitail-pairs-pos_steps_per_second": 1.197,
|
| 906 |
+
"step": 642
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.7508771929824561,
|
| 910 |
+
"eval_scitail-pairs-qa_loss": 2.8886071845590777e-07,
|
| 911 |
+
"eval_scitail-pairs-qa_runtime": 0.58,
|
| 912 |
+
"eval_scitail-pairs-qa_samples_per_second": 220.677,
|
| 913 |
+
"eval_scitail-pairs-qa_steps_per_second": 1.724,
|
| 914 |
+
"step": 642
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 0.7508771929824561,
|
| 918 |
+
"eval_xsum-pairs_loss": 0.005155234597623348,
|
| 919 |
+
"eval_xsum-pairs_runtime": 2.8136,
|
| 920 |
+
"eval_xsum-pairs_samples_per_second": 45.494,
|
| 921 |
+
"eval_xsum-pairs_steps_per_second": 0.355,
|
| 922 |
+
"step": 642
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 0.7508771929824561,
|
| 926 |
+
"eval_sciq_pairs_loss": 0.0030471039935946465,
|
| 927 |
+
"eval_sciq_pairs_runtime": 4.0962,
|
| 928 |
+
"eval_sciq_pairs_samples_per_second": 31.248,
|
| 929 |
+
"eval_sciq_pairs_steps_per_second": 0.244,
|
| 930 |
+
"step": 642
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 0.7508771929824561,
|
| 934 |
+
"eval_qasc_pairs_loss": 0.05629788711667061,
|
| 935 |
+
"eval_qasc_pairs_runtime": 0.5944,
|
| 936 |
+
"eval_qasc_pairs_samples_per_second": 215.353,
|
| 937 |
+
"eval_qasc_pairs_steps_per_second": 1.682,
|
| 938 |
+
"step": 642
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 0.7508771929824561,
|
| 942 |
+
"eval_openbookqa_pairs_loss": 1.2897449731826782,
|
| 943 |
+
"eval_openbookqa_pairs_runtime": 0.5597,
|
| 944 |
+
"eval_openbookqa_pairs_samples_per_second": 228.687,
|
| 945 |
+
"eval_openbookqa_pairs_steps_per_second": 1.787,
|
| 946 |
+
"step": 642
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 0.7508771929824561,
|
| 950 |
+
"eval_msmarco_pairs_loss": 0.14785167574882507,
|
| 951 |
+
"eval_msmarco_pairs_runtime": 1.3218,
|
| 952 |
+
"eval_msmarco_pairs_samples_per_second": 96.836,
|
| 953 |
+
"eval_msmarco_pairs_steps_per_second": 0.757,
|
| 954 |
+
"step": 642
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.7508771929824561,
|
| 958 |
+
"eval_nq_pairs_loss": 0.04229257255792618,
|
| 959 |
+
"eval_nq_pairs_runtime": 2.9207,
|
| 960 |
+
"eval_nq_pairs_samples_per_second": 43.825,
|
| 961 |
+
"eval_nq_pairs_steps_per_second": 0.342,
|
| 962 |
+
"step": 642
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.7508771929824561,
|
| 966 |
+
"eval_trivia_pairs_loss": 0.2112799882888794,
|
| 967 |
+
"eval_trivia_pairs_runtime": 3.1802,
|
| 968 |
+
"eval_trivia_pairs_samples_per_second": 40.249,
|
| 969 |
+
"eval_trivia_pairs_steps_per_second": 0.314,
|
| 970 |
+
"step": 642
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 0.7508771929824561,
|
| 974 |
+
"eval_gooaq_pairs_loss": 0.11469888687133789,
|
| 975 |
+
"eval_gooaq_pairs_runtime": 0.8601,
|
| 976 |
+
"eval_gooaq_pairs_samples_per_second": 148.825,
|
| 977 |
+
"eval_gooaq_pairs_steps_per_second": 1.163,
|
| 978 |
+
"step": 642
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 0.7508771929824561,
|
| 982 |
+
"eval_paws-pos_loss": 0.0460214801132679,
|
| 983 |
+
"eval_paws-pos_runtime": 0.6612,
|
| 984 |
+
"eval_paws-pos_samples_per_second": 193.581,
|
| 985 |
+
"eval_paws-pos_steps_per_second": 1.512,
|
| 986 |
+
"step": 642
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 0.7508771929824561,
|
| 990 |
+
"eval_global_dataset_loss": 0.20304369926452637,
|
| 991 |
+
"eval_global_dataset_runtime": 24.1397,
|
| 992 |
+
"eval_global_dataset_samples_per_second": 27.465,
|
| 993 |
+
"eval_global_dataset_steps_per_second": 0.249,
|
| 994 |
+
"step": 642
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 0.7590643274853801,
|
| 998 |
+
"grad_norm": 0.5546406507492065,
|
| 999 |
+
"learning_rate": 2.0698086463501066e-05,
|
| 1000 |
+
"loss": 0.1919,
|
| 1001 |
+
"step": 649
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 0.7719298245614035,
|
| 1005 |
+
"grad_norm": 0.16179773211479187,
|
| 1006 |
+
"learning_rate": 2.1048901488306166e-05,
|
| 1007 |
+
"loss": 0.1395,
|
| 1008 |
+
"step": 660
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 0.7847953216374269,
|
| 1012 |
+
"grad_norm": 0.4947220981121063,
|
| 1013 |
+
"learning_rate": 2.139971651311127e-05,
|
| 1014 |
+
"loss": 0.2047,
|
| 1015 |
+
"step": 671
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 0.7976608187134503,
|
| 1019 |
+
"grad_norm": 0.4697045683860779,
|
| 1020 |
+
"learning_rate": 2.1750531537916374e-05,
|
| 1021 |
+
"loss": 0.1421,
|
| 1022 |
+
"step": 682
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 0.8105263157894737,
|
| 1026 |
+
"grad_norm": 0.7013148665428162,
|
| 1027 |
+
"learning_rate": 2.2101346562721477e-05,
|
| 1028 |
+
"loss": 0.1227,
|
| 1029 |
+
"step": 693
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 0.8233918128654971,
|
| 1033 |
+
"grad_norm": 1.1045671701431274,
|
| 1034 |
+
"learning_rate": 2.2452161587526577e-05,
|
| 1035 |
+
"loss": 0.1235,
|
| 1036 |
+
"step": 704
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 0.8362573099415205,
|
| 1040 |
+
"grad_norm": 1.5613625049591064,
|
| 1041 |
+
"learning_rate": 2.280297661233168e-05,
|
| 1042 |
+
"loss": 0.2004,
|
| 1043 |
+
"step": 715
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 0.8491228070175438,
|
| 1047 |
+
"grad_norm": 0.23709431290626526,
|
| 1048 |
+
"learning_rate": 2.3153791637136788e-05,
|
| 1049 |
+
"loss": 0.1568,
|
| 1050 |
+
"step": 726
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 0.8619883040935673,
|
| 1054 |
+
"grad_norm": 1.1035298109054565,
|
| 1055 |
+
"learning_rate": 2.3504606661941888e-05,
|
| 1056 |
+
"loss": 0.1598,
|
| 1057 |
+
"step": 737
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 0.8748538011695907,
|
| 1061 |
+
"grad_norm": 0.7472051978111267,
|
| 1062 |
+
"learning_rate": 2.385542168674699e-05,
|
| 1063 |
+
"loss": 0.1328,
|
| 1064 |
+
"step": 748
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 0.887719298245614,
|
| 1068 |
+
"grad_norm": 0.57217937707901,
|
| 1069 |
+
"learning_rate": 2.420623671155209e-05,
|
| 1070 |
+
"loss": 0.0999,
|
| 1071 |
+
"step": 759
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 0.9005847953216374,
|
| 1075 |
+
"grad_norm": 0.3447597920894623,
|
| 1076 |
+
"learning_rate": 2.4557051736357198e-05,
|
| 1077 |
+
"loss": 0.1058,
|
| 1078 |
+
"step": 770
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 0.9134502923976608,
|
| 1082 |
+
"grad_norm": 1.0243078470230103,
|
| 1083 |
+
"learning_rate": 2.4907866761162295e-05,
|
| 1084 |
+
"loss": 0.1673,
|
| 1085 |
+
"step": 781
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 0.9263157894736842,
|
| 1089 |
+
"grad_norm": 1.0685806274414062,
|
| 1090 |
+
"learning_rate": 2.5258681785967402e-05,
|
| 1091 |
+
"loss": 0.1905,
|
| 1092 |
+
"step": 792
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 0.9391812865497076,
|
| 1096 |
+
"grad_norm": 1.1175950765609741,
|
| 1097 |
+
"learning_rate": 2.5609496810772502e-05,
|
| 1098 |
+
"loss": 0.1463,
|
| 1099 |
+
"step": 803
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"epoch": 0.952046783625731,
|
| 1103 |
+
"grad_norm": 0.13670021295547485,
|
| 1104 |
+
"learning_rate": 2.5960311835577605e-05,
|
| 1105 |
+
"loss": 0.1294,
|
| 1106 |
+
"step": 814
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 0.9649122807017544,
|
| 1110 |
+
"grad_norm": 0.6462442278862,
|
| 1111 |
+
"learning_rate": 2.6311126860382706e-05,
|
| 1112 |
+
"loss": 0.1312,
|
| 1113 |
+
"step": 825
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 0.9777777777777777,
|
| 1117 |
+
"grad_norm": 0.1836312711238861,
|
| 1118 |
+
"learning_rate": 2.6661941885187812e-05,
|
| 1119 |
+
"loss": 0.1308,
|
| 1120 |
+
"step": 836
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 0.9906432748538012,
|
| 1124 |
+
"grad_norm": 4.2791407395270653e-07,
|
| 1125 |
+
"learning_rate": 2.7012756909992913e-05,
|
| 1126 |
+
"loss": 0.1076,
|
| 1127 |
+
"step": 847
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 1.001169590643275,
|
| 1131 |
+
"eval_Qnli-dev_cosine_accuracy": 0.69140625,
|
| 1132 |
+
"eval_Qnli-dev_cosine_accuracy_threshold": 0.657210111618042,
|
| 1133 |
+
"eval_Qnli-dev_cosine_ap": 0.714041144664227,
|
| 1134 |
+
"eval_Qnli-dev_cosine_f1": 0.6783439490445861,
|
| 1135 |
+
"eval_Qnli-dev_cosine_f1_threshold": 0.48659271001815796,
|
| 1136 |
+
"eval_Qnli-dev_cosine_precision": 0.5433673469387755,
|
| 1137 |
+
"eval_Qnli-dev_cosine_recall": 0.902542372881356,
|
| 1138 |
+
"eval_Qnli-dev_dot_accuracy": 0.669921875,
|
| 1139 |
+
"eval_Qnli-dev_dot_accuracy_threshold": 256.53155517578125,
|
| 1140 |
+
"eval_Qnli-dev_dot_ap": 0.6648303273464243,
|
| 1141 |
+
"eval_Qnli-dev_dot_f1": 0.678125,
|
| 1142 |
+
"eval_Qnli-dev_dot_f1_threshold": 178.63389587402344,
|
| 1143 |
+
"eval_Qnli-dev_dot_precision": 0.5371287128712872,
|
| 1144 |
+
"eval_Qnli-dev_dot_recall": 0.9194915254237288,
|
| 1145 |
+
"eval_Qnli-dev_euclidean_accuracy": 0.6953125,
|
| 1146 |
+
"eval_Qnli-dev_euclidean_accuracy_threshold": 16.298263549804688,
|
| 1147 |
+
"eval_Qnli-dev_euclidean_ap": 0.7198419743004867,
|
| 1148 |
+
"eval_Qnli-dev_euclidean_f1": 0.6771929824561403,
|
| 1149 |
+
"eval_Qnli-dev_euclidean_f1_threshold": 18.734228134155273,
|
| 1150 |
+
"eval_Qnli-dev_euclidean_precision": 0.5778443113772455,
|
| 1151 |
+
"eval_Qnli-dev_euclidean_recall": 0.8177966101694916,
|
| 1152 |
+
"eval_Qnli-dev_manhattan_accuracy": 0.697265625,
|
| 1153 |
+
"eval_Qnli-dev_manhattan_accuracy_threshold": 350.3825988769531,
|
| 1154 |
+
"eval_Qnli-dev_manhattan_ap": 0.7225583351616828,
|
| 1155 |
+
"eval_Qnli-dev_manhattan_f1": 0.6733466933867736,
|
| 1156 |
+
"eval_Qnli-dev_manhattan_f1_threshold": 364.5922546386719,
|
| 1157 |
+
"eval_Qnli-dev_manhattan_precision": 0.6387832699619772,
|
| 1158 |
+
"eval_Qnli-dev_manhattan_recall": 0.711864406779661,
|
| 1159 |
+
"eval_Qnli-dev_max_accuracy": 0.697265625,
|
| 1160 |
+
"eval_Qnli-dev_max_accuracy_threshold": 350.3825988769531,
|
| 1161 |
+
"eval_Qnli-dev_max_ap": 0.7225583351616828,
|
| 1162 |
+
"eval_Qnli-dev_max_f1": 0.6783439490445861,
|
| 1163 |
+
"eval_Qnli-dev_max_f1_threshold": 364.5922546386719,
|
| 1164 |
+
"eval_Qnli-dev_max_precision": 0.6387832699619772,
|
| 1165 |
+
"eval_Qnli-dev_max_recall": 0.9194915254237288,
|
| 1166 |
+
"eval_allNLI-dev_cosine_accuracy": 0.73828125,
|
| 1167 |
+
"eval_allNLI-dev_cosine_accuracy_threshold": 0.7839766144752502,
|
| 1168 |
+
"eval_allNLI-dev_cosine_ap": 0.6377040389504978,
|
| 1169 |
+
"eval_allNLI-dev_cosine_f1": 0.6371681415929203,
|
| 1170 |
+
"eval_allNLI-dev_cosine_f1_threshold": 0.5921227931976318,
|
| 1171 |
+
"eval_allNLI-dev_cosine_precision": 0.5161290322580645,
|
| 1172 |
+
"eval_allNLI-dev_cosine_recall": 0.8323699421965318,
|
| 1173 |
+
"eval_allNLI-dev_dot_accuracy": 0.72265625,
|
| 1174 |
+
"eval_allNLI-dev_dot_accuracy_threshold": 287.0504150390625,
|
| 1175 |
+
"eval_allNLI-dev_dot_ap": 0.5976338107788919,
|
| 1176 |
+
"eval_allNLI-dev_dot_f1": 0.615,
|
| 1177 |
+
"eval_allNLI-dev_dot_f1_threshold": 242.96902465820312,
|
| 1178 |
+
"eval_allNLI-dev_dot_precision": 0.5418502202643172,
|
| 1179 |
+
"eval_allNLI-dev_dot_recall": 0.7109826589595376,
|
| 1180 |
+
"eval_allNLI-dev_euclidean_accuracy": 0.740234375,
|
| 1181 |
+
"eval_allNLI-dev_euclidean_accuracy_threshold": 13.528583526611328,
|
| 1182 |
+
"eval_allNLI-dev_euclidean_ap": 0.6364974679625237,
|
| 1183 |
+
"eval_allNLI-dev_euclidean_f1": 0.6419753086419753,
|
| 1184 |
+
"eval_allNLI-dev_euclidean_f1_threshold": 16.527751922607422,
|
| 1185 |
+
"eval_allNLI-dev_euclidean_precision": 0.5603448275862069,
|
| 1186 |
+
"eval_allNLI-dev_euclidean_recall": 0.7514450867052023,
|
| 1187 |
+
"eval_allNLI-dev_manhattan_accuracy": 0.73828125,
|
| 1188 |
+
"eval_allNLI-dev_manhattan_accuracy_threshold": 287.6244812011719,
|
| 1189 |
+
"eval_allNLI-dev_manhattan_ap": 0.6359733229236403,
|
| 1190 |
+
"eval_allNLI-dev_manhattan_f1": 0.6461538461538462,
|
| 1191 |
+
"eval_allNLI-dev_manhattan_f1_threshold": 341.6584167480469,
|
| 1192 |
+
"eval_allNLI-dev_manhattan_precision": 0.5806451612903226,
|
| 1193 |
+
"eval_allNLI-dev_manhattan_recall": 0.7283236994219653,
|
| 1194 |
+
"eval_allNLI-dev_max_accuracy": 0.740234375,
|
| 1195 |
+
"eval_allNLI-dev_max_accuracy_threshold": 287.6244812011719,
|
| 1196 |
+
"eval_allNLI-dev_max_ap": 0.6377040389504978,
|
| 1197 |
+
"eval_allNLI-dev_max_f1": 0.6461538461538462,
|
| 1198 |
+
"eval_allNLI-dev_max_f1_threshold": 341.6584167480469,
|
| 1199 |
+
"eval_allNLI-dev_max_precision": 0.5806451612903226,
|
| 1200 |
+
"eval_allNLI-dev_max_recall": 0.8323699421965318,
|
| 1201 |
+
"eval_sequential_score": 0.7225583351616828,
|
| 1202 |
+
"eval_sts-test_pearson_cosine": 0.8846526004146165,
|
| 1203 |
+
"eval_sts-test_pearson_dot": 0.8694917588757636,
|
| 1204 |
+
"eval_sts-test_pearson_euclidean": 0.903648547377558,
|
| 1205 |
+
"eval_sts-test_pearson_manhattan": 0.9046900352481024,
|
| 1206 |
+
"eval_sts-test_pearson_max": 0.9046900352481024,
|
| 1207 |
+
"eval_sts-test_spearman_cosine": 0.9069678627017289,
|
| 1208 |
+
"eval_sts-test_spearman_dot": 0.8661496554202353,
|
| 1209 |
+
"eval_sts-test_spearman_euclidean": 0.8991392112877015,
|
| 1210 |
+
"eval_sts-test_spearman_manhattan": 0.9004067012640901,
|
| 1211 |
+
"eval_sts-test_spearman_max": 0.9069678627017289,
|
| 1212 |
+
"eval_vitaminc-pairs_loss": 2.658703327178955,
|
| 1213 |
+
"eval_vitaminc-pairs_runtime": 3.7894,
|
| 1214 |
+
"eval_vitaminc-pairs_samples_per_second": 33.778,
|
| 1215 |
+
"eval_vitaminc-pairs_steps_per_second": 0.264,
|
| 1216 |
+
"step": 856
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 1.001169590643275,
|
| 1220 |
+
"eval_negation-triplets_loss": 1.048051357269287,
|
| 1221 |
+
"eval_negation-triplets_runtime": 0.5948,
|
| 1222 |
+
"eval_negation-triplets_samples_per_second": 215.195,
|
| 1223 |
+
"eval_negation-triplets_steps_per_second": 1.681,
|
| 1224 |
+
"step": 856
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 1.001169590643275,
|
| 1228 |
+
"eval_scitail-pairs-pos_loss": 0.05958803743124008,
|
| 1229 |
+
"eval_scitail-pairs-pos_runtime": 0.8408,
|
| 1230 |
+
"eval_scitail-pairs-pos_samples_per_second": 152.231,
|
| 1231 |
+
"eval_scitail-pairs-pos_steps_per_second": 1.189,
|
| 1232 |
+
"step": 856
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 1.001169590643275,
|
| 1236 |
+
"eval_scitail-pairs-qa_loss": 4.770551527144562e-07,
|
| 1237 |
+
"eval_scitail-pairs-qa_runtime": 0.5629,
|
| 1238 |
+
"eval_scitail-pairs-qa_samples_per_second": 227.394,
|
| 1239 |
+
"eval_scitail-pairs-qa_steps_per_second": 1.777,
|
| 1240 |
+
"step": 856
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 1.001169590643275,
|
| 1244 |
+
"eval_xsum-pairs_loss": 0.005623215343803167,
|
| 1245 |
+
"eval_xsum-pairs_runtime": 2.7656,
|
| 1246 |
+
"eval_xsum-pairs_samples_per_second": 46.283,
|
| 1247 |
+
"eval_xsum-pairs_steps_per_second": 0.362,
|
| 1248 |
+
"step": 856
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 1.001169590643275,
|
| 1252 |
+
"eval_sciq_pairs_loss": 0.0031388087663799524,
|
| 1253 |
+
"eval_sciq_pairs_runtime": 4.099,
|
| 1254 |
+
"eval_sciq_pairs_samples_per_second": 31.227,
|
| 1255 |
+
"eval_sciq_pairs_steps_per_second": 0.244,
|
| 1256 |
+
"step": 856
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.001169590643275,
|
| 1260 |
+
"eval_qasc_pairs_loss": 0.05423235148191452,
|
| 1261 |
+
"eval_qasc_pairs_runtime": 0.596,
|
| 1262 |
+
"eval_qasc_pairs_samples_per_second": 214.767,
|
| 1263 |
+
"eval_qasc_pairs_steps_per_second": 1.678,
|
| 1264 |
+
"step": 856
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 1.001169590643275,
|
| 1268 |
+
"eval_openbookqa_pairs_loss": 1.2992339134216309,
|
| 1269 |
+
"eval_openbookqa_pairs_runtime": 0.5529,
|
| 1270 |
+
"eval_openbookqa_pairs_samples_per_second": 231.507,
|
| 1271 |
+
"eval_openbookqa_pairs_steps_per_second": 1.809,
|
| 1272 |
+
"step": 856
|
| 1273 |
+
},
|
| 1274 |
+
{
|
| 1275 |
+
"epoch": 1.001169590643275,
|
| 1276 |
+
"eval_msmarco_pairs_loss": 0.15928690135478973,
|
| 1277 |
+
"eval_msmarco_pairs_runtime": 1.3201,
|
| 1278 |
+
"eval_msmarco_pairs_samples_per_second": 96.963,
|
| 1279 |
+
"eval_msmarco_pairs_steps_per_second": 0.758,
|
| 1280 |
+
"step": 856
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 1.001169590643275,
|
| 1284 |
+
"eval_nq_pairs_loss": 0.04052837938070297,
|
| 1285 |
+
"eval_nq_pairs_runtime": 2.9252,
|
| 1286 |
+
"eval_nq_pairs_samples_per_second": 43.758,
|
| 1287 |
+
"eval_nq_pairs_steps_per_second": 0.342,
|
| 1288 |
+
"step": 856
|
| 1289 |
+
},
|
| 1290 |
+
{
|
| 1291 |
+
"epoch": 1.001169590643275,
|
| 1292 |
+
"eval_trivia_pairs_loss": 0.21009953320026398,
|
| 1293 |
+
"eval_trivia_pairs_runtime": 3.1824,
|
| 1294 |
+
"eval_trivia_pairs_samples_per_second": 40.221,
|
| 1295 |
+
"eval_trivia_pairs_steps_per_second": 0.314,
|
| 1296 |
+
"step": 856
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 1.001169590643275,
|
| 1300 |
+
"eval_gooaq_pairs_loss": 0.11401581764221191,
|
| 1301 |
+
"eval_gooaq_pairs_runtime": 0.8553,
|
| 1302 |
+
"eval_gooaq_pairs_samples_per_second": 149.66,
|
| 1303 |
+
"eval_gooaq_pairs_steps_per_second": 1.169,
|
| 1304 |
+
"step": 856
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 1.001169590643275,
|
| 1308 |
+
"eval_paws-pos_loss": 0.04620524123311043,
|
| 1309 |
+
"eval_paws-pos_runtime": 0.6561,
|
| 1310 |
+
"eval_paws-pos_samples_per_second": 195.099,
|
| 1311 |
+
"eval_paws-pos_steps_per_second": 1.524,
|
| 1312 |
+
"step": 856
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.001169590643275,
|
| 1316 |
+
"eval_global_dataset_loss": 0.20265640318393707,
|
| 1317 |
+
"eval_global_dataset_runtime": 24.1741,
|
| 1318 |
+
"eval_global_dataset_samples_per_second": 27.426,
|
| 1319 |
+
"eval_global_dataset_steps_per_second": 0.248,
|
| 1320 |
+
"step": 856
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 1.0035087719298246,
|
| 1324 |
+
"grad_norm": 1.2066627740859985,
|
| 1325 |
+
"learning_rate": 2.7363571934798016e-05,
|
| 1326 |
+
"loss": 0.1085,
|
| 1327 |
+
"step": 858
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 1.016374269005848,
|
| 1331 |
+
"grad_norm": 0.3020640015602112,
|
| 1332 |
+
"learning_rate": 2.7714386959603123e-05,
|
| 1333 |
+
"loss": 0.2214,
|
| 1334 |
+
"step": 869
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 1.0292397660818713,
|
| 1338 |
+
"grad_norm": 0.9783104062080383,
|
| 1339 |
+
"learning_rate": 2.8065201984408223e-05,
|
| 1340 |
+
"loss": 0.1214,
|
| 1341 |
+
"step": 880
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"epoch": 1.0421052631578946,
|
| 1345 |
+
"grad_norm": 0.24499095976352692,
|
| 1346 |
+
"learning_rate": 2.8416017009213327e-05,
|
| 1347 |
+
"loss": 0.1049,
|
| 1348 |
+
"step": 891
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"epoch": 1.0549707602339182,
|
| 1352 |
+
"grad_norm": 0.5718978643417358,
|
| 1353 |
+
"learning_rate": 2.8766832034018427e-05,
|
| 1354 |
+
"loss": 0.1897,
|
| 1355 |
+
"step": 902
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 1.0678362573099416,
|
| 1359 |
+
"grad_norm": 0.558549165725708,
|
| 1360 |
+
"learning_rate": 2.9117647058823534e-05,
|
| 1361 |
+
"loss": 0.1273,
|
| 1362 |
+
"step": 913
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"epoch": 1.080701754385965,
|
| 1366 |
+
"grad_norm": 0.014839976094663143,
|
| 1367 |
+
"learning_rate": 2.9468462083628634e-05,
|
| 1368 |
+
"loss": 0.1474,
|
| 1369 |
+
"step": 924
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 1.0935672514619883,
|
| 1373 |
+
"grad_norm": 0.28388065099716187,
|
| 1374 |
+
"learning_rate": 2.9819277108433737e-05,
|
| 1375 |
+
"loss": 0.1313,
|
| 1376 |
+
"step": 935
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 1.1064327485380117,
|
| 1380 |
+
"grad_norm": 0.28269803524017334,
|
| 1381 |
+
"learning_rate": 3.0170092133238837e-05,
|
| 1382 |
+
"loss": 0.1769,
|
| 1383 |
+
"step": 946
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 1.119298245614035,
|
| 1387 |
+
"grad_norm": 0.910582959651947,
|
| 1388 |
+
"learning_rate": 3.052090715804394e-05,
|
| 1389 |
+
"loss": 0.143,
|
| 1390 |
+
"step": 957
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 1.1321637426900586,
|
| 1394 |
+
"grad_norm": 0.5537557601928711,
|
| 1395 |
+
"learning_rate": 3.087172218284904e-05,
|
| 1396 |
+
"loss": 0.1968,
|
| 1397 |
+
"step": 968
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 1.145029239766082,
|
| 1401 |
+
"grad_norm": 0.9547837376594543,
|
| 1402 |
+
"learning_rate": 3.122253720765415e-05,
|
| 1403 |
+
"loss": 0.1771,
|
| 1404 |
+
"step": 979
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 1.1578947368421053,
|
| 1408 |
+
"grad_norm": 1.1063166856765747,
|
| 1409 |
+
"learning_rate": 3.157335223245925e-05,
|
| 1410 |
+
"loss": 0.1822,
|
| 1411 |
+
"step": 990
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 1.1707602339181287,
|
| 1415 |
+
"grad_norm": 0.3083917796611786,
|
| 1416 |
+
"learning_rate": 3.1924167257264355e-05,
|
| 1417 |
+
"loss": 0.2467,
|
| 1418 |
+
"step": 1001
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 1.183625730994152,
|
| 1422 |
+
"grad_norm": 0.8678072690963745,
|
| 1423 |
+
"learning_rate": 3.227498228206946e-05,
|
| 1424 |
+
"loss": 0.1419,
|
| 1425 |
+
"step": 1012
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 1.1964912280701754,
|
| 1429 |
+
"grad_norm": 1.0385596752166748,
|
| 1430 |
+
"learning_rate": 3.262579730687456e-05,
|
| 1431 |
+
"loss": 0.1782,
|
| 1432 |
+
"step": 1023
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 1.2093567251461987,
|
| 1436 |
+
"grad_norm": 1.0005829334259033,
|
| 1437 |
+
"learning_rate": 3.297661233167966e-05,
|
| 1438 |
+
"loss": 0.1297,
|
| 1439 |
+
"step": 1034
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 1.2222222222222223,
|
| 1443 |
+
"grad_norm": 0.7221047282218933,
|
| 1444 |
+
"learning_rate": 3.332742735648476e-05,
|
| 1445 |
+
"loss": 0.1972,
|
| 1446 |
+
"step": 1045
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 1.2350877192982457,
|
| 1450 |
+
"grad_norm": 0.20262663066387177,
|
| 1451 |
+
"learning_rate": 3.367824238128987e-05,
|
| 1452 |
+
"loss": 0.1491,
|
| 1453 |
+
"step": 1056
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 1.247953216374269,
|
| 1457 |
+
"grad_norm": 0.7044623494148254,
|
| 1458 |
+
"learning_rate": 3.402905740609497e-05,
|
| 1459 |
+
"loss": 0.1721,
|
| 1460 |
+
"step": 1067
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 1.2514619883040936,
|
| 1464 |
+
"eval_Qnli-dev_cosine_accuracy": 0.69140625,
|
| 1465 |
+
"eval_Qnli-dev_cosine_accuracy_threshold": 0.6546923518180847,
|
| 1466 |
+
"eval_Qnli-dev_cosine_ap": 0.7153621655342896,
|
| 1467 |
+
"eval_Qnli-dev_cosine_f1": 0.6783439490445861,
|
| 1468 |
+
"eval_Qnli-dev_cosine_f1_threshold": 0.4858614206314087,
|
| 1469 |
+
"eval_Qnli-dev_cosine_precision": 0.5433673469387755,
|
| 1470 |
+
"eval_Qnli-dev_cosine_recall": 0.902542372881356,
|
| 1471 |
+
"eval_Qnli-dev_dot_accuracy": 0.671875,
|
| 1472 |
+
"eval_Qnli-dev_dot_accuracy_threshold": 254.48748779296875,
|
| 1473 |
+
"eval_Qnli-dev_dot_ap": 0.6630076505293763,
|
| 1474 |
+
"eval_Qnli-dev_dot_f1": 0.6740157480314961,
|
| 1475 |
+
"eval_Qnli-dev_dot_f1_threshold": 183.4345703125,
|
| 1476 |
+
"eval_Qnli-dev_dot_precision": 0.5363408521303258,
|
| 1477 |
+
"eval_Qnli-dev_dot_recall": 0.9067796610169492,
|
| 1478 |
+
"eval_Qnli-dev_euclidean_accuracy": 0.701171875,
|
| 1479 |
+
"eval_Qnli-dev_euclidean_accuracy_threshold": 16.410430908203125,
|
| 1480 |
+
"eval_Qnli-dev_euclidean_ap": 0.7221448194401613,
|
| 1481 |
+
"eval_Qnli-dev_euclidean_f1": 0.6811881188118811,
|
| 1482 |
+
"eval_Qnli-dev_euclidean_f1_threshold": 17.367481231689453,
|
| 1483 |
+
"eval_Qnli-dev_euclidean_precision": 0.6394052044609665,
|
| 1484 |
+
"eval_Qnli-dev_euclidean_recall": 0.7288135593220338,
|
| 1485 |
+
"eval_Qnli-dev_manhattan_accuracy": 0.6953125,
|
| 1486 |
+
"eval_Qnli-dev_manhattan_accuracy_threshold": 353.0638122558594,
|
| 1487 |
+
"eval_Qnli-dev_manhattan_ap": 0.7247492174165924,
|
| 1488 |
+
"eval_Qnli-dev_manhattan_f1": 0.6756238003838773,
|
| 1489 |
+
"eval_Qnli-dev_manhattan_f1_threshold": 373.6475524902344,
|
| 1490 |
+
"eval_Qnli-dev_manhattan_precision": 0.6175438596491228,
|
| 1491 |
+
"eval_Qnli-dev_manhattan_recall": 0.7457627118644068,
|
| 1492 |
+
"eval_Qnli-dev_max_accuracy": 0.701171875,
|
| 1493 |
+
"eval_Qnli-dev_max_accuracy_threshold": 353.0638122558594,
|
| 1494 |
+
"eval_Qnli-dev_max_ap": 0.7247492174165924,
|
| 1495 |
+
"eval_Qnli-dev_max_f1": 0.6811881188118811,
|
| 1496 |
+
"eval_Qnli-dev_max_f1_threshold": 373.6475524902344,
|
| 1497 |
+
"eval_Qnli-dev_max_precision": 0.6394052044609665,
|
| 1498 |
+
"eval_Qnli-dev_max_recall": 0.9067796610169492,
|
| 1499 |
+
"eval_allNLI-dev_cosine_accuracy": 0.740234375,
|
| 1500 |
+
"eval_allNLI-dev_cosine_accuracy_threshold": 0.7620664834976196,
|
| 1501 |
+
"eval_allNLI-dev_cosine_ap": 0.6352023549250958,
|
| 1502 |
+
"eval_allNLI-dev_cosine_f1": 0.6401766004415012,
|
| 1503 |
+
"eval_allNLI-dev_cosine_f1_threshold": 0.5983242988586426,
|
| 1504 |
+
"eval_allNLI-dev_cosine_precision": 0.5178571428571429,
|
| 1505 |
+
"eval_allNLI-dev_cosine_recall": 0.838150289017341,
|
| 1506 |
+
"eval_allNLI-dev_dot_accuracy": 0.732421875,
|
| 1507 |
+
"eval_allNLI-dev_dot_accuracy_threshold": 282.25836181640625,
|
| 1508 |
+
"eval_allNLI-dev_dot_ap": 0.5938709922161639,
|
| 1509 |
+
"eval_allNLI-dev_dot_f1": 0.612987012987013,
|
| 1510 |
+
"eval_allNLI-dev_dot_f1_threshold": 245.4638671875,
|
| 1511 |
+
"eval_allNLI-dev_dot_precision": 0.5566037735849056,
|
| 1512 |
+
"eval_allNLI-dev_dot_recall": 0.6820809248554913,
|
| 1513 |
+
"eval_allNLI-dev_euclidean_accuracy": 0.740234375,
|
| 1514 |
+
"eval_allNLI-dev_euclidean_accuracy_threshold": 12.869304656982422,
|
| 1515 |
+
"eval_allNLI-dev_euclidean_ap": 0.6367777728048029,
|
| 1516 |
+
"eval_allNLI-dev_euclidean_f1": 0.6458797327394209,
|
| 1517 |
+
"eval_allNLI-dev_euclidean_f1_threshold": 17.27260971069336,
|
| 1518 |
+
"eval_allNLI-dev_euclidean_precision": 0.5253623188405797,
|
| 1519 |
+
"eval_allNLI-dev_euclidean_recall": 0.838150289017341,
|
| 1520 |
+
"eval_allNLI-dev_manhattan_accuracy": 0.7421875,
|
| 1521 |
+
"eval_allNLI-dev_manhattan_accuracy_threshold": 297.8420715332031,
|
| 1522 |
+
"eval_allNLI-dev_manhattan_ap": 0.6368910837780656,
|
| 1523 |
+
"eval_allNLI-dev_manhattan_f1": 0.6477024070021882,
|
| 1524 |
+
"eval_allNLI-dev_manhattan_f1_threshold": 369.8564453125,
|
| 1525 |
+
"eval_allNLI-dev_manhattan_precision": 0.5211267605633803,
|
| 1526 |
+
"eval_allNLI-dev_manhattan_recall": 0.8554913294797688,
|
| 1527 |
+
"eval_allNLI-dev_max_accuracy": 0.7421875,
|
| 1528 |
+
"eval_allNLI-dev_max_accuracy_threshold": 297.8420715332031,
|
| 1529 |
+
"eval_allNLI-dev_max_ap": 0.6368910837780656,
|
| 1530 |
+
"eval_allNLI-dev_max_f1": 0.6477024070021882,
|
| 1531 |
+
"eval_allNLI-dev_max_f1_threshold": 369.8564453125,
|
| 1532 |
+
"eval_allNLI-dev_max_precision": 0.5566037735849056,
|
| 1533 |
+
"eval_allNLI-dev_max_recall": 0.8554913294797688,
|
| 1534 |
+
"eval_sequential_score": 0.7247492174165924,
|
| 1535 |
+
"eval_sts-test_pearson_cosine": 0.884621677997079,
|
| 1536 |
+
"eval_sts-test_pearson_dot": 0.8662022984817861,
|
| 1537 |
+
"eval_sts-test_pearson_euclidean": 0.9042360469838505,
|
| 1538 |
+
"eval_sts-test_pearson_manhattan": 0.9051133756387458,
|
| 1539 |
+
"eval_sts-test_pearson_max": 0.9051133756387458,
|
| 1540 |
+
"eval_sts-test_spearman_cosine": 0.9071986449567242,
|
| 1541 |
+
"eval_sts-test_spearman_dot": 0.8602733672869306,
|
| 1542 |
+
"eval_sts-test_spearman_euclidean": 0.8999521179508763,
|
| 1543 |
+
"eval_sts-test_spearman_manhattan": 0.901083564092635,
|
| 1544 |
+
"eval_sts-test_spearman_max": 0.9071986449567242,
|
| 1545 |
+
"eval_vitaminc-pairs_loss": 2.655998468399048,
|
| 1546 |
+
"eval_vitaminc-pairs_runtime": 3.8396,
|
| 1547 |
+
"eval_vitaminc-pairs_samples_per_second": 33.337,
|
| 1548 |
+
"eval_vitaminc-pairs_steps_per_second": 0.26,
|
| 1549 |
+
"step": 1070
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 1.2514619883040936,
|
| 1553 |
+
"eval_negation-triplets_loss": 1.0599219799041748,
|
| 1554 |
+
"eval_negation-triplets_runtime": 0.6284,
|
| 1555 |
+
"eval_negation-triplets_samples_per_second": 203.693,
|
| 1556 |
+
"eval_negation-triplets_steps_per_second": 1.591,
|
| 1557 |
+
"step": 1070
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 1.2514619883040936,
|
| 1561 |
+
"eval_scitail-pairs-pos_loss": 0.061028748750686646,
|
| 1562 |
+
"eval_scitail-pairs-pos_runtime": 0.9206,
|
| 1563 |
+
"eval_scitail-pairs-pos_samples_per_second": 139.044,
|
| 1564 |
+
"eval_scitail-pairs-pos_steps_per_second": 1.086,
|
| 1565 |
+
"step": 1070
|
| 1566 |
+
},
|
| 1567 |
+
{
|
| 1568 |
+
"epoch": 1.2514619883040936,
|
| 1569 |
+
"eval_scitail-pairs-qa_loss": 4.0994524397319765e-07,
|
| 1570 |
+
"eval_scitail-pairs-qa_runtime": 0.5941,
|
| 1571 |
+
"eval_scitail-pairs-qa_samples_per_second": 215.464,
|
| 1572 |
+
"eval_scitail-pairs-qa_steps_per_second": 1.683,
|
| 1573 |
+
"step": 1070
|
| 1574 |
+
},
|
| 1575 |
+
{
|
| 1576 |
+
"epoch": 1.2514619883040936,
|
| 1577 |
+
"eval_xsum-pairs_loss": 0.005265752784907818,
|
| 1578 |
+
"eval_xsum-pairs_runtime": 2.8061,
|
| 1579 |
+
"eval_xsum-pairs_samples_per_second": 45.615,
|
| 1580 |
+
"eval_xsum-pairs_steps_per_second": 0.356,
|
| 1581 |
+
"step": 1070
|
| 1582 |
+
},
|
| 1583 |
+
{
|
| 1584 |
+
"epoch": 1.2514619883040936,
|
| 1585 |
+
"eval_sciq_pairs_loss": 0.0029235396068543196,
|
| 1586 |
+
"eval_sciq_pairs_runtime": 4.2052,
|
| 1587 |
+
"eval_sciq_pairs_samples_per_second": 30.438,
|
| 1588 |
+
"eval_sciq_pairs_steps_per_second": 0.238,
|
| 1589 |
+
"step": 1070
|
| 1590 |
+
},
|
| 1591 |
+
{
|
| 1592 |
+
"epoch": 1.2514619883040936,
|
| 1593 |
+
"eval_qasc_pairs_loss": 0.05088697001338005,
|
| 1594 |
+
"eval_qasc_pairs_runtime": 0.6097,
|
| 1595 |
+
"eval_qasc_pairs_samples_per_second": 209.926,
|
| 1596 |
+
"eval_qasc_pairs_steps_per_second": 1.64,
|
| 1597 |
+
"step": 1070
|
| 1598 |
+
},
|
| 1599 |
+
{
|
| 1600 |
+
"epoch": 1.2514619883040936,
|
| 1601 |
+
"eval_openbookqa_pairs_loss": 1.281742811203003,
|
| 1602 |
+
"eval_openbookqa_pairs_runtime": 0.5692,
|
| 1603 |
+
"eval_openbookqa_pairs_samples_per_second": 224.86,
|
| 1604 |
+
"eval_openbookqa_pairs_steps_per_second": 1.757,
|
| 1605 |
+
"step": 1070
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 1.2514619883040936,
|
| 1609 |
+
"eval_msmarco_pairs_loss": 0.1445852667093277,
|
| 1610 |
+
"eval_msmarco_pairs_runtime": 1.3418,
|
| 1611 |
+
"eval_msmarco_pairs_samples_per_second": 95.392,
|
| 1612 |
+
"eval_msmarco_pairs_steps_per_second": 0.745,
|
| 1613 |
+
"step": 1070
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 1.2514619883040936,
|
| 1617 |
+
"eval_nq_pairs_loss": 0.041951049119234085,
|
| 1618 |
+
"eval_nq_pairs_runtime": 2.9501,
|
| 1619 |
+
"eval_nq_pairs_samples_per_second": 43.389,
|
| 1620 |
+
"eval_nq_pairs_steps_per_second": 0.339,
|
| 1621 |
+
"step": 1070
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 1.2514619883040936,
|
| 1625 |
+
"eval_trivia_pairs_loss": 0.2087937891483307,
|
| 1626 |
+
"eval_trivia_pairs_runtime": 3.2077,
|
| 1627 |
+
"eval_trivia_pairs_samples_per_second": 39.904,
|
| 1628 |
+
"eval_trivia_pairs_steps_per_second": 0.312,
|
| 1629 |
+
"step": 1070
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 1.2514619883040936,
|
| 1633 |
+
"eval_gooaq_pairs_loss": 0.11122985929250717,
|
| 1634 |
+
"eval_gooaq_pairs_runtime": 0.8735,
|
| 1635 |
+
"eval_gooaq_pairs_samples_per_second": 146.542,
|
| 1636 |
+
"eval_gooaq_pairs_steps_per_second": 1.145,
|
| 1637 |
+
"step": 1070
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"epoch": 1.2514619883040936,
|
| 1641 |
+
"eval_paws-pos_loss": 0.04601467400789261,
|
| 1642 |
+
"eval_paws-pos_runtime": 0.6727,
|
| 1643 |
+
"eval_paws-pos_samples_per_second": 190.278,
|
| 1644 |
+
"eval_paws-pos_steps_per_second": 1.487,
|
| 1645 |
+
"step": 1070
|
| 1646 |
+
},
|
| 1647 |
+
{
|
| 1648 |
+
"epoch": 1.2514619883040936,
|
| 1649 |
+
"eval_global_dataset_loss": 0.21354959905147552,
|
| 1650 |
+
"eval_global_dataset_runtime": 24.3059,
|
| 1651 |
+
"eval_global_dataset_samples_per_second": 27.277,
|
| 1652 |
+
"eval_global_dataset_steps_per_second": 0.247,
|
| 1653 |
+
"step": 1070
|
| 1654 |
+
},
|
| 1655 |
+
{
|
| 1656 |
+
"epoch": 1.2608187134502924,
|
| 1657 |
+
"grad_norm": 0.5193822383880615,
|
| 1658 |
+
"learning_rate": 3.4379872430900076e-05,
|
| 1659 |
+
"loss": 0.1279,
|
| 1660 |
+
"step": 1078
|
| 1661 |
+
},
|
| 1662 |
+
{
|
| 1663 |
+
"epoch": 1.2736842105263158,
|
| 1664 |
+
"grad_norm": 1.694514513015747,
|
| 1665 |
+
"learning_rate": 3.4730687455705176e-05,
|
| 1666 |
+
"loss": 0.106,
|
| 1667 |
+
"step": 1089
|
| 1668 |
+
},
|
| 1669 |
+
{
|
| 1670 |
+
"epoch": 1.286549707602339,
|
| 1671 |
+
"grad_norm": 0.8599225282669067,
|
| 1672 |
+
"learning_rate": 3.5081502480510276e-05,
|
| 1673 |
+
"loss": 0.1597,
|
| 1674 |
+
"step": 1100
|
| 1675 |
+
},
|
| 1676 |
+
{
|
| 1677 |
+
"epoch": 1.2994152046783625,
|
| 1678 |
+
"grad_norm": 0.7980732917785645,
|
| 1679 |
+
"learning_rate": 3.5432317505315376e-05,
|
| 1680 |
+
"loss": 0.192,
|
| 1681 |
+
"step": 1111
|
| 1682 |
+
},
|
| 1683 |
+
{
|
| 1684 |
+
"epoch": 1.312280701754386,
|
| 1685 |
+
"grad_norm": 0.33019399642944336,
|
| 1686 |
+
"learning_rate": 3.578313253012048e-05,
|
| 1687 |
+
"loss": 0.165,
|
| 1688 |
+
"step": 1122
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 1.3251461988304094,
|
| 1692 |
+
"grad_norm": 0.006413115654140711,
|
| 1693 |
+
"learning_rate": 3.6133947554925583e-05,
|
| 1694 |
+
"loss": 0.1472,
|
| 1695 |
+
"step": 1133
|
| 1696 |
+
},
|
| 1697 |
+
{
|
| 1698 |
+
"epoch": 1.3380116959064328,
|
| 1699 |
+
"grad_norm": 0.38985970616340637,
|
| 1700 |
+
"learning_rate": 3.648476257973069e-05,
|
| 1701 |
+
"loss": 0.1528,
|
| 1702 |
+
"step": 1144
|
| 1703 |
+
},
|
| 1704 |
+
{
|
| 1705 |
+
"epoch": 1.3508771929824561,
|
| 1706 |
+
"grad_norm": 1.1716166734695435,
|
| 1707 |
+
"learning_rate": 3.68355776045358e-05,
|
| 1708 |
+
"loss": 0.202,
|
| 1709 |
+
"step": 1155
|
| 1710 |
+
},
|
| 1711 |
+
{
|
| 1712 |
+
"epoch": 1.3637426900584795,
|
| 1713 |
+
"grad_norm": 0.6012507081031799,
|
| 1714 |
+
"learning_rate": 3.71863926293409e-05,
|
| 1715 |
+
"loss": 0.1974,
|
| 1716 |
+
"step": 1166
|
| 1717 |
+
},
|
| 1718 |
+
{
|
| 1719 |
+
"epoch": 1.376608187134503,
|
| 1720 |
+
"grad_norm": 0.15670593082904816,
|
| 1721 |
+
"learning_rate": 3.7537207654146e-05,
|
| 1722 |
+
"loss": 0.2229,
|
| 1723 |
+
"step": 1177
|
| 1724 |
+
},
|
| 1725 |
+
{
|
| 1726 |
+
"epoch": 1.3894736842105262,
|
| 1727 |
+
"grad_norm": 0.5632084608078003,
|
| 1728 |
+
"learning_rate": 3.78880226789511e-05,
|
| 1729 |
+
"loss": 0.1104,
|
| 1730 |
+
"step": 1188
|
| 1731 |
+
},
|
| 1732 |
+
{
|
| 1733 |
+
"epoch": 1.4023391812865498,
|
| 1734 |
+
"grad_norm": 0.7236007452011108,
|
| 1735 |
+
"learning_rate": 3.8238837703756205e-05,
|
| 1736 |
+
"loss": 0.1544,
|
| 1737 |
+
"step": 1199
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"epoch": 1.4152046783625731,
|
| 1741 |
+
"grad_norm": 0.22364391386508942,
|
| 1742 |
+
"learning_rate": 3.8589652728561305e-05,
|
| 1743 |
+
"loss": 0.0875,
|
| 1744 |
+
"step": 1210
|
| 1745 |
+
},
|
| 1746 |
+
{
|
| 1747 |
+
"epoch": 1.4280701754385965,
|
| 1748 |
+
"grad_norm": 0.23043423891067505,
|
| 1749 |
+
"learning_rate": 3.894046775336641e-05,
|
| 1750 |
+
"loss": 0.1607,
|
| 1751 |
+
"step": 1221
|
| 1752 |
+
},
|
| 1753 |
+
{
|
| 1754 |
+
"epoch": 1.4409356725146198,
|
| 1755 |
+
"grad_norm": 0.7280920147895813,
|
| 1756 |
+
"learning_rate": 3.929128277817151e-05,
|
| 1757 |
+
"loss": 0.2026,
|
| 1758 |
+
"step": 1232
|
| 1759 |
+
},
|
| 1760 |
+
{
|
| 1761 |
+
"epoch": 1.4538011695906432,
|
| 1762 |
+
"grad_norm": 0.1936722993850708,
|
| 1763 |
+
"learning_rate": 3.964209780297662e-05,
|
| 1764 |
+
"loss": 0.185,
|
| 1765 |
+
"step": 1243
|
| 1766 |
+
},
|
| 1767 |
+
{
|
| 1768 |
+
"epoch": 1.4666666666666668,
|
| 1769 |
+
"grad_norm": 0.4526028633117676,
|
| 1770 |
+
"learning_rate": 3.999291282778172e-05,
|
| 1771 |
+
"loss": 0.1114,
|
| 1772 |
+
"step": 1254
|
| 1773 |
+
},
|
| 1774 |
+
{
|
| 1775 |
+
"epoch": 1.47953216374269,
|
| 1776 |
+
"grad_norm": 0.2208670973777771,
|
| 1777 |
+
"learning_rate": 4.034372785258682e-05,
|
| 1778 |
+
"loss": 0.2033,
|
| 1779 |
+
"step": 1265
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 1.4923976608187135,
|
| 1783 |
+
"grad_norm": 1.5043717622756958,
|
| 1784 |
+
"learning_rate": 4.069454287739192e-05,
|
| 1785 |
+
"loss": 0.1216,
|
| 1786 |
+
"step": 1276
|
| 1787 |
+
},
|
| 1788 |
+
{
|
| 1789 |
+
"epoch": 1.5017543859649123,
|
| 1790 |
+
"eval_Qnli-dev_cosine_accuracy": 0.69140625,
|
| 1791 |
+
"eval_Qnli-dev_cosine_accuracy_threshold": 0.6479891538619995,
|
| 1792 |
+
"eval_Qnli-dev_cosine_ap": 0.7145834476822306,
|
| 1793 |
+
"eval_Qnli-dev_cosine_f1": 0.6772655007949125,
|
| 1794 |
+
"eval_Qnli-dev_cosine_f1_threshold": 0.4831158518791199,
|
| 1795 |
+
"eval_Qnli-dev_cosine_precision": 0.5419847328244275,
|
| 1796 |
+
"eval_Qnli-dev_cosine_recall": 0.902542372881356,
|
| 1797 |
+
"eval_Qnli-dev_dot_accuracy": 0.669921875,
|
| 1798 |
+
"eval_Qnli-dev_dot_accuracy_threshold": 253.99070739746094,
|
| 1799 |
+
"eval_Qnli-dev_dot_ap": 0.6600609959091455,
|
| 1800 |
+
"eval_Qnli-dev_dot_f1": 0.6749226006191951,
|
| 1801 |
+
"eval_Qnli-dev_dot_f1_threshold": 177.03799438476562,
|
| 1802 |
+
"eval_Qnli-dev_dot_precision": 0.5317073170731708,
|
| 1803 |
+
"eval_Qnli-dev_dot_recall": 0.923728813559322,
|
| 1804 |
+
"eval_Qnli-dev_euclidean_accuracy": 0.703125,
|
| 1805 |
+
"eval_Qnli-dev_euclidean_accuracy_threshold": 16.42089080810547,
|
| 1806 |
+
"eval_Qnli-dev_euclidean_ap": 0.7233470144987824,
|
| 1807 |
+
"eval_Qnli-dev_euclidean_f1": 0.6786427145708582,
|
| 1808 |
+
"eval_Qnli-dev_euclidean_f1_threshold": 17.447650909423828,
|
| 1809 |
+
"eval_Qnli-dev_euclidean_precision": 0.6415094339622641,
|
| 1810 |
+
"eval_Qnli-dev_euclidean_recall": 0.7203389830508474,
|
| 1811 |
+
"eval_Qnli-dev_manhattan_accuracy": 0.69921875,
|
| 1812 |
+
"eval_Qnli-dev_manhattan_accuracy_threshold": 348.24969482421875,
|
| 1813 |
+
"eval_Qnli-dev_manhattan_ap": 0.7248042664576803,
|
| 1814 |
+
"eval_Qnli-dev_manhattan_f1": 0.6735966735966736,
|
| 1815 |
+
"eval_Qnli-dev_manhattan_f1_threshold": 363.0526123046875,
|
| 1816 |
+
"eval_Qnli-dev_manhattan_precision": 0.6612244897959184,
|
| 1817 |
+
"eval_Qnli-dev_manhattan_recall": 0.6864406779661016,
|
| 1818 |
+
"eval_Qnli-dev_max_accuracy": 0.703125,
|
| 1819 |
+
"eval_Qnli-dev_max_accuracy_threshold": 348.24969482421875,
|
| 1820 |
+
"eval_Qnli-dev_max_ap": 0.7248042664576803,
|
| 1821 |
+
"eval_Qnli-dev_max_f1": 0.6786427145708582,
|
| 1822 |
+
"eval_Qnli-dev_max_f1_threshold": 363.0526123046875,
|
| 1823 |
+
"eval_Qnli-dev_max_precision": 0.6612244897959184,
|
| 1824 |
+
"eval_Qnli-dev_max_recall": 0.923728813559322,
|
| 1825 |
+
"eval_allNLI-dev_cosine_accuracy": 0.740234375,
|
| 1826 |
+
"eval_allNLI-dev_cosine_accuracy_threshold": 0.7793502807617188,
|
| 1827 |
+
"eval_allNLI-dev_cosine_ap": 0.6356059008560961,
|
| 1828 |
+
"eval_allNLI-dev_cosine_f1": 0.6412556053811659,
|
| 1829 |
+
"eval_allNLI-dev_cosine_f1_threshold": 0.5976245999336243,
|
| 1830 |
+
"eval_allNLI-dev_cosine_precision": 0.5238095238095238,
|
| 1831 |
+
"eval_allNLI-dev_cosine_recall": 0.8265895953757225,
|
| 1832 |
+
"eval_allNLI-dev_dot_accuracy": 0.728515625,
|
| 1833 |
+
"eval_allNLI-dev_dot_accuracy_threshold": 289.59637451171875,
|
| 1834 |
+
"eval_allNLI-dev_dot_ap": 0.6011407360027551,
|
| 1835 |
+
"eval_allNLI-dev_dot_f1": 0.6177215189873418,
|
| 1836 |
+
"eval_allNLI-dev_dot_f1_threshold": 246.49258422851562,
|
| 1837 |
+
"eval_allNLI-dev_dot_precision": 0.5495495495495496,
|
| 1838 |
+
"eval_allNLI-dev_dot_recall": 0.7052023121387283,
|
| 1839 |
+
"eval_allNLI-dev_euclidean_accuracy": 0.744140625,
|
| 1840 |
+
"eval_allNLI-dev_euclidean_accuracy_threshold": 13.19027328491211,
|
| 1841 |
+
"eval_allNLI-dev_euclidean_ap": 0.6358049693529386,
|
| 1842 |
+
"eval_allNLI-dev_euclidean_f1": 0.6430155210643015,
|
| 1843 |
+
"eval_allNLI-dev_euclidean_f1_threshold": 17.634546279907227,
|
| 1844 |
+
"eval_allNLI-dev_euclidean_precision": 0.5215827338129496,
|
| 1845 |
+
"eval_allNLI-dev_euclidean_recall": 0.838150289017341,
|
| 1846 |
+
"eval_allNLI-dev_manhattan_accuracy": 0.73828125,
|
| 1847 |
+
"eval_allNLI-dev_manhattan_accuracy_threshold": 281.0400085449219,
|
| 1848 |
+
"eval_allNLI-dev_manhattan_ap": 0.6342668702497214,
|
| 1849 |
+
"eval_allNLI-dev_manhattan_f1": 0.6512820512820513,
|
| 1850 |
+
"eval_allNLI-dev_manhattan_f1_threshold": 342.90478515625,
|
| 1851 |
+
"eval_allNLI-dev_manhattan_precision": 0.5852534562211982,
|
| 1852 |
+
"eval_allNLI-dev_manhattan_recall": 0.7341040462427746,
|
| 1853 |
+
"eval_allNLI-dev_max_accuracy": 0.744140625,
|
| 1854 |
+
"eval_allNLI-dev_max_accuracy_threshold": 289.59637451171875,
|
| 1855 |
+
"eval_allNLI-dev_max_ap": 0.6358049693529386,
|
| 1856 |
+
"eval_allNLI-dev_max_f1": 0.6512820512820513,
|
| 1857 |
+
"eval_allNLI-dev_max_f1_threshold": 342.90478515625,
|
| 1858 |
+
"eval_allNLI-dev_max_precision": 0.5852534562211982,
|
| 1859 |
+
"eval_allNLI-dev_max_recall": 0.838150289017341,
|
| 1860 |
+
"eval_sequential_score": 0.7248042664576803,
|
| 1861 |
+
"eval_sts-test_pearson_cosine": 0.8852081089755639,
|
| 1862 |
+
"eval_sts-test_pearson_dot": 0.8700618346105049,
|
| 1863 |
+
"eval_sts-test_pearson_euclidean": 0.904503930603612,
|
| 1864 |
+
"eval_sts-test_pearson_manhattan": 0.905215331612789,
|
| 1865 |
+
"eval_sts-test_pearson_max": 0.905215331612789,
|
| 1866 |
+
"eval_sts-test_spearman_cosine": 0.9076046731513284,
|
| 1867 |
+
"eval_sts-test_spearman_dot": 0.8651591757518594,
|
| 1868 |
+
"eval_sts-test_spearman_euclidean": 0.9000229144463976,
|
| 1869 |
+
"eval_sts-test_spearman_manhattan": 0.9012659254828049,
|
| 1870 |
+
"eval_sts-test_spearman_max": 0.9076046731513284,
|
| 1871 |
+
"eval_vitaminc-pairs_loss": 2.694180727005005,
|
| 1872 |
+
"eval_vitaminc-pairs_runtime": 3.8321,
|
| 1873 |
+
"eval_vitaminc-pairs_samples_per_second": 33.402,
|
| 1874 |
+
"eval_vitaminc-pairs_steps_per_second": 0.261,
|
| 1875 |
+
"step": 1284
|
| 1876 |
+
},
|
| 1877 |
+
{
|
| 1878 |
+
"epoch": 1.5017543859649123,
|
| 1879 |
+
"eval_negation-triplets_loss": 1.049867033958435,
|
| 1880 |
+
"eval_negation-triplets_runtime": 0.6019,
|
| 1881 |
+
"eval_negation-triplets_samples_per_second": 212.664,
|
| 1882 |
+
"eval_negation-triplets_steps_per_second": 1.661,
|
| 1883 |
+
"step": 1284
|
| 1884 |
+
},
|
| 1885 |
+
{
|
| 1886 |
+
"epoch": 1.5017543859649123,
|
| 1887 |
+
"eval_scitail-pairs-pos_loss": 0.05977838858962059,
|
| 1888 |
+
"eval_scitail-pairs-pos_runtime": 0.8746,
|
| 1889 |
+
"eval_scitail-pairs-pos_samples_per_second": 146.356,
|
| 1890 |
+
"eval_scitail-pairs-pos_steps_per_second": 1.143,
|
| 1891 |
+
"step": 1284
|
| 1892 |
+
},
|
| 1893 |
+
{
|
| 1894 |
+
"epoch": 1.5017543859649123,
|
| 1895 |
+
"eval_scitail-pairs-qa_loss": 3.577365532692056e-07,
|
| 1896 |
+
"eval_scitail-pairs-qa_runtime": 0.577,
|
| 1897 |
+
"eval_scitail-pairs-qa_samples_per_second": 221.823,
|
| 1898 |
+
"eval_scitail-pairs-qa_steps_per_second": 1.733,
|
| 1899 |
+
"step": 1284
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 1.5017543859649123,
|
| 1903 |
+
"eval_xsum-pairs_loss": 0.005321314092725515,
|
| 1904 |
+
"eval_xsum-pairs_runtime": 2.7947,
|
| 1905 |
+
"eval_xsum-pairs_samples_per_second": 45.801,
|
| 1906 |
+
"eval_xsum-pairs_steps_per_second": 0.358,
|
| 1907 |
+
"step": 1284
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 1.5017543859649123,
|
| 1911 |
+
"eval_sciq_pairs_loss": 0.003324751742184162,
|
| 1912 |
+
"eval_sciq_pairs_runtime": 4.2166,
|
| 1913 |
+
"eval_sciq_pairs_samples_per_second": 30.356,
|
| 1914 |
+
"eval_sciq_pairs_steps_per_second": 0.237,
|
| 1915 |
+
"step": 1284
|
| 1916 |
+
},
|
| 1917 |
+
{
|
| 1918 |
+
"epoch": 1.5017543859649123,
|
| 1919 |
+
"eval_qasc_pairs_loss": 0.0436801016330719,
|
| 1920 |
+
"eval_qasc_pairs_runtime": 0.6277,
|
| 1921 |
+
"eval_qasc_pairs_samples_per_second": 203.908,
|
| 1922 |
+
"eval_qasc_pairs_steps_per_second": 1.593,
|
| 1923 |
+
"step": 1284
|
| 1924 |
+
},
|
| 1925 |
+
{
|
| 1926 |
+
"epoch": 1.5017543859649123,
|
| 1927 |
+
"eval_openbookqa_pairs_loss": 1.3073405027389526,
|
| 1928 |
+
"eval_openbookqa_pairs_runtime": 0.5694,
|
| 1929 |
+
"eval_openbookqa_pairs_samples_per_second": 224.81,
|
| 1930 |
+
"eval_openbookqa_pairs_steps_per_second": 1.756,
|
| 1931 |
+
"step": 1284
|
| 1932 |
+
},
|
| 1933 |
+
{
|
| 1934 |
+
"epoch": 1.5017543859649123,
|
| 1935 |
+
"eval_msmarco_pairs_loss": 0.15479393303394318,
|
| 1936 |
+
"eval_msmarco_pairs_runtime": 1.3314,
|
| 1937 |
+
"eval_msmarco_pairs_samples_per_second": 96.137,
|
| 1938 |
+
"eval_msmarco_pairs_steps_per_second": 0.751,
|
| 1939 |
+
"step": 1284
|
| 1940 |
+
},
|
| 1941 |
+
{
|
| 1942 |
+
"epoch": 1.5017543859649123,
|
| 1943 |
+
"eval_nq_pairs_loss": 0.03904910385608673,
|
| 1944 |
+
"eval_nq_pairs_runtime": 2.9759,
|
| 1945 |
+
"eval_nq_pairs_samples_per_second": 43.012,
|
| 1946 |
+
"eval_nq_pairs_steps_per_second": 0.336,
|
| 1947 |
+
"step": 1284
|
| 1948 |
+
},
|
| 1949 |
+
{
|
| 1950 |
+
"epoch": 1.5017543859649123,
|
| 1951 |
+
"eval_trivia_pairs_loss": 0.2038673609495163,
|
| 1952 |
+
"eval_trivia_pairs_runtime": 3.1963,
|
| 1953 |
+
"eval_trivia_pairs_samples_per_second": 40.047,
|
| 1954 |
+
"eval_trivia_pairs_steps_per_second": 0.313,
|
| 1955 |
+
"step": 1284
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 1.5017543859649123,
|
| 1959 |
+
"eval_gooaq_pairs_loss": 0.10658062994480133,
|
| 1960 |
+
"eval_gooaq_pairs_runtime": 0.8625,
|
| 1961 |
+
"eval_gooaq_pairs_samples_per_second": 148.407,
|
| 1962 |
+
"eval_gooaq_pairs_steps_per_second": 1.159,
|
| 1963 |
+
"step": 1284
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 1.5017543859649123,
|
| 1967 |
+
"eval_paws-pos_loss": 0.04552536457777023,
|
| 1968 |
+
"eval_paws-pos_runtime": 0.6659,
|
| 1969 |
+
"eval_paws-pos_samples_per_second": 192.226,
|
| 1970 |
+
"eval_paws-pos_steps_per_second": 1.502,
|
| 1971 |
+
"step": 1284
|
| 1972 |
+
},
|
| 1973 |
+
{
|
| 1974 |
+
"epoch": 1.5017543859649123,
|
| 1975 |
+
"eval_global_dataset_loss": 0.2039254754781723,
|
| 1976 |
+
"eval_global_dataset_runtime": 24.2218,
|
| 1977 |
+
"eval_global_dataset_samples_per_second": 27.372,
|
| 1978 |
+
"eval_global_dataset_steps_per_second": 0.248,
|
| 1979 |
+
"step": 1284
|
| 1980 |
+
}
|
| 1981 |
+
],
|
| 1982 |
+
"logging_steps": 11,
|
| 1983 |
+
"max_steps": 4275,
|
| 1984 |
+
"num_input_tokens_seen": 0,
|
| 1985 |
+
"num_train_epochs": 5,
|
| 1986 |
+
"save_steps": 214,
|
| 1987 |
+
"stateful_callbacks": {
|
| 1988 |
+
"TrainerControl": {
|
| 1989 |
+
"args": {
|
| 1990 |
+
"should_epoch_stop": false,
|
| 1991 |
+
"should_evaluate": false,
|
| 1992 |
+
"should_log": false,
|
| 1993 |
+
"should_save": true,
|
| 1994 |
+
"should_training_stop": false
|
| 1995 |
+
},
|
| 1996 |
+
"attributes": {}
|
| 1997 |
+
}
|
| 1998 |
+
},
|
| 1999 |
+
"total_flos": 0.0,
|
| 2000 |
+
"train_batch_size": 96,
|
| 2001 |
+
"trial_name": null,
|
| 2002 |
+
"trial_params": null
|
| 2003 |
+
}
|
checkpoint-1284/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aed6626d5e4516eecaff94019f86ab02eb588df865930f776737c69908518829
|
| 3 |
+
size 5624
|