Training in progress, step 2554, checkpoint
Browse files- checkpoint-2554/1_Pooling/config.json +10 -0
- checkpoint-2554/README.md +0 -0
- checkpoint-2554/added_tokens.json +3 -0
- checkpoint-2554/config.json +35 -0
- checkpoint-2554/config_sentence_transformers.json +10 -0
- checkpoint-2554/modules.json +14 -0
- checkpoint-2554/optimizer.pt +3 -0
- checkpoint-2554/pytorch_model.bin +3 -0
- checkpoint-2554/rng_state.pth +3 -0
- checkpoint-2554/scheduler.pt +3 -0
- checkpoint-2554/sentence_bert_config.json +4 -0
- checkpoint-2554/special_tokens_map.json +51 -0
- checkpoint-2554/spm.model +3 -0
- checkpoint-2554/tokenizer.json +0 -0
- checkpoint-2554/tokenizer_config.json +65 -0
- checkpoint-2554/trainer_state.json +1608 -0
- checkpoint-2554/training_args.bin +3 -0
checkpoint-2554/1_Pooling/config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"word_embedding_dimension": 768,
|
| 3 |
+
"pooling_mode_cls_token": false,
|
| 4 |
+
"pooling_mode_mean_tokens": true,
|
| 5 |
+
"pooling_mode_max_tokens": false,
|
| 6 |
+
"pooling_mode_mean_sqrt_len_tokens": false,
|
| 7 |
+
"pooling_mode_weightedmean_tokens": false,
|
| 8 |
+
"pooling_mode_lasttoken": false,
|
| 9 |
+
"include_prompt": true
|
| 10 |
+
}
|
checkpoint-2554/README.md
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-2554/added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"[MASK]": 128000
|
| 3 |
+
}
|
checkpoint-2554/config.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "bobox/DeBERTa-ST-AllLayers-v3-checkpoints-tmp",
|
| 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": "4.42.3",
|
| 33 |
+
"type_vocab_size": 0,
|
| 34 |
+
"vocab_size": 128100
|
| 35 |
+
}
|
checkpoint-2554/config_sentence_transformers.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"__version__": {
|
| 3 |
+
"sentence_transformers": "3.0.1",
|
| 4 |
+
"transformers": "4.42.3",
|
| 5 |
+
"pytorch": "2.1.2"
|
| 6 |
+
},
|
| 7 |
+
"prompts": {},
|
| 8 |
+
"default_prompt_name": null,
|
| 9 |
+
"similarity_fn_name": null
|
| 10 |
+
}
|
checkpoint-2554/modules.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"idx": 0,
|
| 4 |
+
"name": "0",
|
| 5 |
+
"path": "",
|
| 6 |
+
"type": "sentence_transformers.models.Transformer"
|
| 7 |
+
},
|
| 8 |
+
{
|
| 9 |
+
"idx": 1,
|
| 10 |
+
"name": "1",
|
| 11 |
+
"path": "1_Pooling",
|
| 12 |
+
"type": "sentence_transformers.models.Pooling"
|
| 13 |
+
}
|
| 14 |
+
]
|
checkpoint-2554/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:66bb1e36276c1c098c52926289b76425a27f3db889d5a7048c1cb263e28e245b
|
| 3 |
+
size 1130520122
|
checkpoint-2554/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8631cd0e79a1f08a0912c11f565aed711b237e95a58d80d9d21672834f172075
|
| 3 |
+
size 565251810
|
checkpoint-2554/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f559b53b673cbaeb2c79890b3798a53f05f92893f5f6881490ea3ce66a7599d9
|
| 3 |
+
size 14244
|
checkpoint-2554/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9dfadb5a50c60e4767cb9734be0c74fcd91fb48041ecf1db5b7da5ffd8146e00
|
| 3 |
+
size 1064
|
checkpoint-2554/sentence_bert_config.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"max_seq_length": 512,
|
| 3 |
+
"do_lower_case": false
|
| 4 |
+
}
|
checkpoint-2554/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-2554/spm.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c679fbf93643d19aab7ee10c0b99e460bdbc02fedf34b92b05af343b4af586fd
|
| 3 |
+
size 2464616
|
checkpoint-2554/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-2554/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": 1000000000000000019884624838656,
|
| 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-2554/trainer_state.json
ADDED
|
@@ -0,0 +1,1608 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.40025074439742986,
|
| 5 |
+
"eval_steps": 320,
|
| 6 |
+
"global_step": 2554,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0050148879485973985,
|
| 13 |
+
"grad_norm": 14.771158218383789,
|
| 14 |
+
"learning_rate": 9.707724425887265e-07,
|
| 15 |
+
"loss": 0.6329,
|
| 16 |
+
"step": 32
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.010029775897194797,
|
| 20 |
+
"grad_norm": 11.052021980285645,
|
| 21 |
+
"learning_rate": 1.9728601252609606e-06,
|
| 22 |
+
"loss": 0.9693,
|
| 23 |
+
"step": 64
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.015044663845792195,
|
| 27 |
+
"grad_norm": 20.26296615600586,
|
| 28 |
+
"learning_rate": 2.9749478079331944e-06,
|
| 29 |
+
"loss": 0.6548,
|
| 30 |
+
"step": 96
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.020059551794389594,
|
| 34 |
+
"grad_norm": 12.62913703918457,
|
| 35 |
+
"learning_rate": 3.945720250521921e-06,
|
| 36 |
+
"loss": 1.1279,
|
| 37 |
+
"step": 128
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.025074439742986992,
|
| 41 |
+
"grad_norm": 12.316486358642578,
|
| 42 |
+
"learning_rate": 4.916492693110647e-06,
|
| 43 |
+
"loss": 1.0017,
|
| 44 |
+
"step": 160
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.03008932769158439,
|
| 48 |
+
"grad_norm": 64.25923919677734,
|
| 49 |
+
"learning_rate": 5.918580375782881e-06,
|
| 50 |
+
"loss": 0.7571,
|
| 51 |
+
"step": 192
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.03510421564018179,
|
| 55 |
+
"grad_norm": 0.8205029368400574,
|
| 56 |
+
"learning_rate": 6.920668058455115e-06,
|
| 57 |
+
"loss": 0.7304,
|
| 58 |
+
"step": 224
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.04011910358877919,
|
| 62 |
+
"grad_norm": 6.598870754241943,
|
| 63 |
+
"learning_rate": 7.922755741127349e-06,
|
| 64 |
+
"loss": 0.7636,
|
| 65 |
+
"step": 256
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.045133991537376586,
|
| 69 |
+
"grad_norm": 8.728073120117188,
|
| 70 |
+
"learning_rate": 8.924843423799583e-06,
|
| 71 |
+
"loss": 0.482,
|
| 72 |
+
"step": 288
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.050148879485973984,
|
| 76 |
+
"grad_norm": 7.645521640777588,
|
| 77 |
+
"learning_rate": 9.926931106471817e-06,
|
| 78 |
+
"loss": 0.6312,
|
| 79 |
+
"step": 320
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.050148879485973984,
|
| 83 |
+
"eval_nli-pairs_loss": 1.0158467292785645,
|
| 84 |
+
"eval_nli-pairs_runtime": 3.7267,
|
| 85 |
+
"eval_nli-pairs_samples_per_second": 26.833,
|
| 86 |
+
"eval_nli-pairs_steps_per_second": 1.073,
|
| 87 |
+
"eval_sts-test_pearson_cosine": 0.7848265412179125,
|
| 88 |
+
"eval_sts-test_pearson_dot": 0.5437080705284749,
|
| 89 |
+
"eval_sts-test_pearson_euclidean": 0.7445845076364892,
|
| 90 |
+
"eval_sts-test_pearson_manhattan": 0.7429239204432232,
|
| 91 |
+
"eval_sts-test_pearson_max": 0.7848265412179125,
|
| 92 |
+
"eval_sts-test_spearman_cosine": 0.7989504707258924,
|
| 93 |
+
"eval_sts-test_spearman_dot": 0.5206855421174118,
|
| 94 |
+
"eval_sts-test_spearman_euclidean": 0.733568982260844,
|
| 95 |
+
"eval_sts-test_spearman_manhattan": 0.7349407257944446,
|
| 96 |
+
"eval_sts-test_spearman_max": 0.7989504707258924,
|
| 97 |
+
"step": 320
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.050148879485973984,
|
| 101 |
+
"eval_vitaminc-pairs_loss": 4.692601680755615,
|
| 102 |
+
"eval_vitaminc-pairs_runtime": 1.1397,
|
| 103 |
+
"eval_vitaminc-pairs_samples_per_second": 74.578,
|
| 104 |
+
"eval_vitaminc-pairs_steps_per_second": 2.632,
|
| 105 |
+
"step": 320
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.050148879485973984,
|
| 109 |
+
"eval_sts-label_loss": 3.5502490997314453,
|
| 110 |
+
"eval_sts-label_runtime": 0.28,
|
| 111 |
+
"eval_sts-label_samples_per_second": 357.117,
|
| 112 |
+
"eval_sts-label_steps_per_second": 14.285,
|
| 113 |
+
"step": 320
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.050148879485973984,
|
| 117 |
+
"eval_qnli-contrastive_loss": 0.16079513728618622,
|
| 118 |
+
"eval_qnli-contrastive_runtime": 0.3646,
|
| 119 |
+
"eval_qnli-contrastive_samples_per_second": 274.299,
|
| 120 |
+
"eval_qnli-contrastive_steps_per_second": 10.972,
|
| 121 |
+
"step": 320
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.050148879485973984,
|
| 125 |
+
"eval_scitail-pairs-qa_loss": 0.07610582560300827,
|
| 126 |
+
"eval_scitail-pairs-qa_runtime": 0.8885,
|
| 127 |
+
"eval_scitail-pairs-qa_samples_per_second": 112.548,
|
| 128 |
+
"eval_scitail-pairs-qa_steps_per_second": 4.502,
|
| 129 |
+
"step": 320
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.050148879485973984,
|
| 133 |
+
"eval_scitail-pairs-pos_loss": 0.5141278505325317,
|
| 134 |
+
"eval_scitail-pairs-pos_runtime": 1.3498,
|
| 135 |
+
"eval_scitail-pairs-pos_samples_per_second": 74.085,
|
| 136 |
+
"eval_scitail-pairs-pos_steps_per_second": 2.963,
|
| 137 |
+
"step": 320
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.050148879485973984,
|
| 141 |
+
"eval_xsum-pairs_loss": 0.25581496953964233,
|
| 142 |
+
"eval_xsum-pairs_runtime": 0.9407,
|
| 143 |
+
"eval_xsum-pairs_samples_per_second": 106.304,
|
| 144 |
+
"eval_xsum-pairs_steps_per_second": 4.252,
|
| 145 |
+
"step": 320
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.050148879485973984,
|
| 149 |
+
"eval_compression-pairs_loss": 0.09814296662807465,
|
| 150 |
+
"eval_compression-pairs_runtime": 0.2758,
|
| 151 |
+
"eval_compression-pairs_samples_per_second": 362.517,
|
| 152 |
+
"eval_compression-pairs_steps_per_second": 14.501,
|
| 153 |
+
"step": 320
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.050148879485973984,
|
| 157 |
+
"eval_sciq_pairs_loss": 0.25620242953300476,
|
| 158 |
+
"eval_sciq_pairs_runtime": 4.1155,
|
| 159 |
+
"eval_sciq_pairs_samples_per_second": 24.298,
|
| 160 |
+
"eval_sciq_pairs_steps_per_second": 0.972,
|
| 161 |
+
"step": 320
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 0.050148879485973984,
|
| 165 |
+
"eval_qasc_pairs_loss": 0.2044612169265747,
|
| 166 |
+
"eval_qasc_pairs_runtime": 1.1029,
|
| 167 |
+
"eval_qasc_pairs_samples_per_second": 90.672,
|
| 168 |
+
"eval_qasc_pairs_steps_per_second": 3.627,
|
| 169 |
+
"step": 320
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.050148879485973984,
|
| 173 |
+
"eval_openbookqa_pairs_loss": 1.7537646293640137,
|
| 174 |
+
"eval_openbookqa_pairs_runtime": 0.9037,
|
| 175 |
+
"eval_openbookqa_pairs_samples_per_second": 110.653,
|
| 176 |
+
"eval_openbookqa_pairs_steps_per_second": 4.426,
|
| 177 |
+
"step": 320
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.050148879485973984,
|
| 181 |
+
"eval_msmarco_pairs_loss": 0.5138561725616455,
|
| 182 |
+
"eval_msmarco_pairs_runtime": 2.0511,
|
| 183 |
+
"eval_msmarco_pairs_samples_per_second": 48.754,
|
| 184 |
+
"eval_msmarco_pairs_steps_per_second": 1.95,
|
| 185 |
+
"step": 320
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.050148879485973984,
|
| 189 |
+
"eval_nq_pairs_loss": 0.23510317504405975,
|
| 190 |
+
"eval_nq_pairs_runtime": 4.5293,
|
| 191 |
+
"eval_nq_pairs_samples_per_second": 22.078,
|
| 192 |
+
"eval_nq_pairs_steps_per_second": 0.883,
|
| 193 |
+
"step": 320
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.050148879485973984,
|
| 197 |
+
"eval_trivia_pairs_loss": 0.7808571457862854,
|
| 198 |
+
"eval_trivia_pairs_runtime": 6.5065,
|
| 199 |
+
"eval_trivia_pairs_samples_per_second": 15.369,
|
| 200 |
+
"eval_trivia_pairs_steps_per_second": 0.615,
|
| 201 |
+
"step": 320
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.050148879485973984,
|
| 205 |
+
"eval_quora_pairs_loss": 0.0392119362950325,
|
| 206 |
+
"eval_quora_pairs_runtime": 0.675,
|
| 207 |
+
"eval_quora_pairs_samples_per_second": 148.153,
|
| 208 |
+
"eval_quora_pairs_steps_per_second": 5.926,
|
| 209 |
+
"step": 320
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.050148879485973984,
|
| 213 |
+
"eval_gooaq_pairs_loss": 0.4712902009487152,
|
| 214 |
+
"eval_gooaq_pairs_runtime": 1.4079,
|
| 215 |
+
"eval_gooaq_pairs_samples_per_second": 71.028,
|
| 216 |
+
"eval_gooaq_pairs_steps_per_second": 2.841,
|
| 217 |
+
"step": 320
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.050148879485973984,
|
| 221 |
+
"eval_mrpc_pairs_loss": 0.05498996376991272,
|
| 222 |
+
"eval_mrpc_pairs_runtime": 0.2623,
|
| 223 |
+
"eval_mrpc_pairs_samples_per_second": 381.172,
|
| 224 |
+
"eval_mrpc_pairs_steps_per_second": 15.247,
|
| 225 |
+
"step": 320
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.05516376743457138,
|
| 229 |
+
"grad_norm": 0.34924012422561646,
|
| 230 |
+
"learning_rate": 1.092901878914405e-05,
|
| 231 |
+
"loss": 0.5791,
|
| 232 |
+
"step": 352
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.06017865538316878,
|
| 236 |
+
"grad_norm": 0.36700841784477234,
|
| 237 |
+
"learning_rate": 1.1931106471816284e-05,
|
| 238 |
+
"loss": 0.6413,
|
| 239 |
+
"step": 384
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 0.06519354333176618,
|
| 243 |
+
"grad_norm": 7.559622764587402,
|
| 244 |
+
"learning_rate": 1.2933194154488518e-05,
|
| 245 |
+
"loss": 0.4319,
|
| 246 |
+
"step": 416
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 0.07020843128036358,
|
| 250 |
+
"grad_norm": 7.982416152954102,
|
| 251 |
+
"learning_rate": 1.3935281837160753e-05,
|
| 252 |
+
"loss": 0.6672,
|
| 253 |
+
"step": 448
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 0.07522331922896097,
|
| 257 |
+
"grad_norm": 0.6726166009902954,
|
| 258 |
+
"learning_rate": 1.4937369519832987e-05,
|
| 259 |
+
"loss": 0.459,
|
| 260 |
+
"step": 480
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.08023820717755838,
|
| 264 |
+
"grad_norm": 14.846123695373535,
|
| 265 |
+
"learning_rate": 1.593945720250522e-05,
|
| 266 |
+
"loss": 0.7621,
|
| 267 |
+
"step": 512
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.08525309512615578,
|
| 271 |
+
"grad_norm": 0.7846627831459045,
|
| 272 |
+
"learning_rate": 1.6941544885177454e-05,
|
| 273 |
+
"loss": 0.864,
|
| 274 |
+
"step": 544
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 0.09026798307475317,
|
| 278 |
+
"grad_norm": 0.8993583917617798,
|
| 279 |
+
"learning_rate": 1.7943632567849688e-05,
|
| 280 |
+
"loss": 0.5081,
|
| 281 |
+
"step": 576
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.09528287102335058,
|
| 285 |
+
"grad_norm": 1.4990565776824951,
|
| 286 |
+
"learning_rate": 1.894572025052192e-05,
|
| 287 |
+
"loss": 0.654,
|
| 288 |
+
"step": 608
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.10029775897194797,
|
| 292 |
+
"grad_norm": 15.647976875305176,
|
| 293 |
+
"learning_rate": 1.9947807933194157e-05,
|
| 294 |
+
"loss": 0.6372,
|
| 295 |
+
"step": 640
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.10029775897194797,
|
| 299 |
+
"eval_nli-pairs_loss": 1.0652996301651,
|
| 300 |
+
"eval_nli-pairs_runtime": 3.6326,
|
| 301 |
+
"eval_nli-pairs_samples_per_second": 27.528,
|
| 302 |
+
"eval_nli-pairs_steps_per_second": 1.101,
|
| 303 |
+
"eval_sts-test_pearson_cosine": 0.785263018402905,
|
| 304 |
+
"eval_sts-test_pearson_dot": 0.5290450141477089,
|
| 305 |
+
"eval_sts-test_pearson_euclidean": 0.7433756286425983,
|
| 306 |
+
"eval_sts-test_pearson_manhattan": 0.7411097274300102,
|
| 307 |
+
"eval_sts-test_pearson_max": 0.785263018402905,
|
| 308 |
+
"eval_sts-test_spearman_cosine": 0.7996928912411947,
|
| 309 |
+
"eval_sts-test_spearman_dot": 0.5102571497667188,
|
| 310 |
+
"eval_sts-test_spearman_euclidean": 0.7338969723324641,
|
| 311 |
+
"eval_sts-test_spearman_manhattan": 0.7343494860194358,
|
| 312 |
+
"eval_sts-test_spearman_max": 0.7996928912411947,
|
| 313 |
+
"step": 640
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.10029775897194797,
|
| 317 |
+
"eval_vitaminc-pairs_loss": 4.719416618347168,
|
| 318 |
+
"eval_vitaminc-pairs_runtime": 1.1268,
|
| 319 |
+
"eval_vitaminc-pairs_samples_per_second": 75.437,
|
| 320 |
+
"eval_vitaminc-pairs_steps_per_second": 2.662,
|
| 321 |
+
"step": 640
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.10029775897194797,
|
| 325 |
+
"eval_sts-label_loss": 3.612347364425659,
|
| 326 |
+
"eval_sts-label_runtime": 0.2683,
|
| 327 |
+
"eval_sts-label_samples_per_second": 372.651,
|
| 328 |
+
"eval_sts-label_steps_per_second": 14.906,
|
| 329 |
+
"step": 640
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 0.10029775897194797,
|
| 333 |
+
"eval_qnli-contrastive_loss": 0.15202775597572327,
|
| 334 |
+
"eval_qnli-contrastive_runtime": 0.3528,
|
| 335 |
+
"eval_qnli-contrastive_samples_per_second": 283.457,
|
| 336 |
+
"eval_qnli-contrastive_steps_per_second": 11.338,
|
| 337 |
+
"step": 640
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 0.10029775897194797,
|
| 341 |
+
"eval_scitail-pairs-qa_loss": 0.07544919103384018,
|
| 342 |
+
"eval_scitail-pairs-qa_runtime": 0.8732,
|
| 343 |
+
"eval_scitail-pairs-qa_samples_per_second": 114.517,
|
| 344 |
+
"eval_scitail-pairs-qa_steps_per_second": 4.581,
|
| 345 |
+
"step": 640
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.10029775897194797,
|
| 349 |
+
"eval_scitail-pairs-pos_loss": 0.5404170751571655,
|
| 350 |
+
"eval_scitail-pairs-pos_runtime": 1.3146,
|
| 351 |
+
"eval_scitail-pairs-pos_samples_per_second": 76.067,
|
| 352 |
+
"eval_scitail-pairs-pos_steps_per_second": 3.043,
|
| 353 |
+
"step": 640
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.10029775897194797,
|
| 357 |
+
"eval_xsum-pairs_loss": 0.25958582758903503,
|
| 358 |
+
"eval_xsum-pairs_runtime": 0.9287,
|
| 359 |
+
"eval_xsum-pairs_samples_per_second": 107.679,
|
| 360 |
+
"eval_xsum-pairs_steps_per_second": 4.307,
|
| 361 |
+
"step": 640
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.10029775897194797,
|
| 365 |
+
"eval_compression-pairs_loss": 0.10066353529691696,
|
| 366 |
+
"eval_compression-pairs_runtime": 0.2732,
|
| 367 |
+
"eval_compression-pairs_samples_per_second": 366.076,
|
| 368 |
+
"eval_compression-pairs_steps_per_second": 14.643,
|
| 369 |
+
"step": 640
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.10029775897194797,
|
| 373 |
+
"eval_sciq_pairs_loss": 0.2645374834537506,
|
| 374 |
+
"eval_sciq_pairs_runtime": 4.0725,
|
| 375 |
+
"eval_sciq_pairs_samples_per_second": 24.555,
|
| 376 |
+
"eval_sciq_pairs_steps_per_second": 0.982,
|
| 377 |
+
"step": 640
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.10029775897194797,
|
| 381 |
+
"eval_qasc_pairs_loss": 0.21021947264671326,
|
| 382 |
+
"eval_qasc_pairs_runtime": 1.0743,
|
| 383 |
+
"eval_qasc_pairs_samples_per_second": 93.084,
|
| 384 |
+
"eval_qasc_pairs_steps_per_second": 3.723,
|
| 385 |
+
"step": 640
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 0.10029775897194797,
|
| 389 |
+
"eval_openbookqa_pairs_loss": 1.7905032634735107,
|
| 390 |
+
"eval_openbookqa_pairs_runtime": 0.8886,
|
| 391 |
+
"eval_openbookqa_pairs_samples_per_second": 112.532,
|
| 392 |
+
"eval_openbookqa_pairs_steps_per_second": 4.501,
|
| 393 |
+
"step": 640
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 0.10029775897194797,
|
| 397 |
+
"eval_msmarco_pairs_loss": 0.5102832913398743,
|
| 398 |
+
"eval_msmarco_pairs_runtime": 2.0529,
|
| 399 |
+
"eval_msmarco_pairs_samples_per_second": 48.712,
|
| 400 |
+
"eval_msmarco_pairs_steps_per_second": 1.948,
|
| 401 |
+
"step": 640
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.10029775897194797,
|
| 405 |
+
"eval_nq_pairs_loss": 0.24466972053050995,
|
| 406 |
+
"eval_nq_pairs_runtime": 4.4973,
|
| 407 |
+
"eval_nq_pairs_samples_per_second": 22.235,
|
| 408 |
+
"eval_nq_pairs_steps_per_second": 0.889,
|
| 409 |
+
"step": 640
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.10029775897194797,
|
| 413 |
+
"eval_trivia_pairs_loss": 0.8748095631599426,
|
| 414 |
+
"eval_trivia_pairs_runtime": 6.4825,
|
| 415 |
+
"eval_trivia_pairs_samples_per_second": 15.426,
|
| 416 |
+
"eval_trivia_pairs_steps_per_second": 0.617,
|
| 417 |
+
"step": 640
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.10029775897194797,
|
| 421 |
+
"eval_quora_pairs_loss": 0.07820220291614532,
|
| 422 |
+
"eval_quora_pairs_runtime": 0.6944,
|
| 423 |
+
"eval_quora_pairs_samples_per_second": 144.008,
|
| 424 |
+
"eval_quora_pairs_steps_per_second": 5.76,
|
| 425 |
+
"step": 640
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 0.10029775897194797,
|
| 429 |
+
"eval_gooaq_pairs_loss": 0.5236212611198425,
|
| 430 |
+
"eval_gooaq_pairs_runtime": 1.3899,
|
| 431 |
+
"eval_gooaq_pairs_samples_per_second": 71.949,
|
| 432 |
+
"eval_gooaq_pairs_steps_per_second": 2.878,
|
| 433 |
+
"step": 640
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 0.10029775897194797,
|
| 437 |
+
"eval_mrpc_pairs_loss": 0.05494727939367294,
|
| 438 |
+
"eval_mrpc_pairs_runtime": 0.2598,
|
| 439 |
+
"eval_mrpc_pairs_samples_per_second": 384.941,
|
| 440 |
+
"eval_mrpc_pairs_steps_per_second": 15.398,
|
| 441 |
+
"step": 640
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 0.10531264692054537,
|
| 445 |
+
"grad_norm": 11.01974105834961,
|
| 446 |
+
"learning_rate": 2.0949895615866387e-05,
|
| 447 |
+
"loss": 0.9292,
|
| 448 |
+
"step": 672
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 0.11032753486914276,
|
| 452 |
+
"grad_norm": 0.5542309284210205,
|
| 453 |
+
"learning_rate": 2.1951983298538625e-05,
|
| 454 |
+
"loss": 1.3108,
|
| 455 |
+
"step": 704
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 0.11534242281774017,
|
| 459 |
+
"grad_norm": 15.458569526672363,
|
| 460 |
+
"learning_rate": 2.2954070981210856e-05,
|
| 461 |
+
"loss": 0.9674,
|
| 462 |
+
"step": 736
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 0.12035731076633756,
|
| 466 |
+
"grad_norm": 2.7814478874206543,
|
| 467 |
+
"learning_rate": 2.395615866388309e-05,
|
| 468 |
+
"loss": 0.9226,
|
| 469 |
+
"step": 768
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 0.12537219871493496,
|
| 473 |
+
"grad_norm": 11.393244743347168,
|
| 474 |
+
"learning_rate": 2.4958246346555324e-05,
|
| 475 |
+
"loss": 0.789,
|
| 476 |
+
"step": 800
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.13038708666353235,
|
| 480 |
+
"grad_norm": 9.288290977478027,
|
| 481 |
+
"learning_rate": 2.596033402922756e-05,
|
| 482 |
+
"loss": 0.5186,
|
| 483 |
+
"step": 832
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 0.13540197461212977,
|
| 487 |
+
"grad_norm": 47.65571212768555,
|
| 488 |
+
"learning_rate": 2.6962421711899793e-05,
|
| 489 |
+
"loss": 0.6726,
|
| 490 |
+
"step": 864
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 0.14041686256072716,
|
| 494 |
+
"grad_norm": 12.908064842224121,
|
| 495 |
+
"learning_rate": 2.7964509394572024e-05,
|
| 496 |
+
"loss": 0.5381,
|
| 497 |
+
"step": 896
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 0.14543175050932455,
|
| 501 |
+
"grad_norm": 14.951742172241211,
|
| 502 |
+
"learning_rate": 2.896659707724426e-05,
|
| 503 |
+
"loss": 0.581,
|
| 504 |
+
"step": 928
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.15044663845792194,
|
| 508 |
+
"grad_norm": 20.12006187438965,
|
| 509 |
+
"learning_rate": 2.9968684759916492e-05,
|
| 510 |
+
"loss": 0.9038,
|
| 511 |
+
"step": 960
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 0.15044663845792194,
|
| 515 |
+
"eval_nli-pairs_loss": 1.2173175811767578,
|
| 516 |
+
"eval_nli-pairs_runtime": 3.7098,
|
| 517 |
+
"eval_nli-pairs_samples_per_second": 26.955,
|
| 518 |
+
"eval_nli-pairs_steps_per_second": 1.078,
|
| 519 |
+
"eval_sts-test_pearson_cosine": 0.7840992835675669,
|
| 520 |
+
"eval_sts-test_pearson_dot": 0.5220462136106129,
|
| 521 |
+
"eval_sts-test_pearson_euclidean": 0.7457350047351855,
|
| 522 |
+
"eval_sts-test_pearson_manhattan": 0.7425970830541657,
|
| 523 |
+
"eval_sts-test_pearson_max": 0.7840992835675669,
|
| 524 |
+
"eval_sts-test_spearman_cosine": 0.8006376809572144,
|
| 525 |
+
"eval_sts-test_spearman_dot": 0.5020544543992158,
|
| 526 |
+
"eval_sts-test_spearman_euclidean": 0.7369257710408655,
|
| 527 |
+
"eval_sts-test_spearman_manhattan": 0.7362649758012406,
|
| 528 |
+
"eval_sts-test_spearman_max": 0.8006376809572144,
|
| 529 |
+
"step": 960
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.15044663845792194,
|
| 533 |
+
"eval_vitaminc-pairs_loss": 4.774902820587158,
|
| 534 |
+
"eval_vitaminc-pairs_runtime": 1.1212,
|
| 535 |
+
"eval_vitaminc-pairs_samples_per_second": 75.809,
|
| 536 |
+
"eval_vitaminc-pairs_steps_per_second": 2.676,
|
| 537 |
+
"step": 960
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 0.15044663845792194,
|
| 541 |
+
"eval_sts-label_loss": 3.198556900024414,
|
| 542 |
+
"eval_sts-label_runtime": 0.2678,
|
| 543 |
+
"eval_sts-label_samples_per_second": 373.382,
|
| 544 |
+
"eval_sts-label_steps_per_second": 14.935,
|
| 545 |
+
"step": 960
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 0.15044663845792194,
|
| 549 |
+
"eval_qnli-contrastive_loss": 0.1943340301513672,
|
| 550 |
+
"eval_qnli-contrastive_runtime": 0.3511,
|
| 551 |
+
"eval_qnli-contrastive_samples_per_second": 284.789,
|
| 552 |
+
"eval_qnli-contrastive_steps_per_second": 11.392,
|
| 553 |
+
"step": 960
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 0.15044663845792194,
|
| 557 |
+
"eval_scitail-pairs-qa_loss": 0.08060617744922638,
|
| 558 |
+
"eval_scitail-pairs-qa_runtime": 0.8778,
|
| 559 |
+
"eval_scitail-pairs-qa_samples_per_second": 113.92,
|
| 560 |
+
"eval_scitail-pairs-qa_steps_per_second": 4.557,
|
| 561 |
+
"step": 960
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 0.15044663845792194,
|
| 565 |
+
"eval_scitail-pairs-pos_loss": 0.4759831428527832,
|
| 566 |
+
"eval_scitail-pairs-pos_runtime": 1.3609,
|
| 567 |
+
"eval_scitail-pairs-pos_samples_per_second": 73.48,
|
| 568 |
+
"eval_scitail-pairs-pos_steps_per_second": 2.939,
|
| 569 |
+
"step": 960
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.15044663845792194,
|
| 573 |
+
"eval_xsum-pairs_loss": 0.27583304047584534,
|
| 574 |
+
"eval_xsum-pairs_runtime": 0.9343,
|
| 575 |
+
"eval_xsum-pairs_samples_per_second": 107.035,
|
| 576 |
+
"eval_xsum-pairs_steps_per_second": 4.281,
|
| 577 |
+
"step": 960
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.15044663845792194,
|
| 581 |
+
"eval_compression-pairs_loss": 0.10094660520553589,
|
| 582 |
+
"eval_compression-pairs_runtime": 0.2739,
|
| 583 |
+
"eval_compression-pairs_samples_per_second": 365.047,
|
| 584 |
+
"eval_compression-pairs_steps_per_second": 14.602,
|
| 585 |
+
"step": 960
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.15044663845792194,
|
| 589 |
+
"eval_sciq_pairs_loss": 0.2688131630420685,
|
| 590 |
+
"eval_sciq_pairs_runtime": 4.0582,
|
| 591 |
+
"eval_sciq_pairs_samples_per_second": 24.641,
|
| 592 |
+
"eval_sciq_pairs_steps_per_second": 0.986,
|
| 593 |
+
"step": 960
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 0.15044663845792194,
|
| 597 |
+
"eval_qasc_pairs_loss": 0.23267821967601776,
|
| 598 |
+
"eval_qasc_pairs_runtime": 1.0554,
|
| 599 |
+
"eval_qasc_pairs_samples_per_second": 94.75,
|
| 600 |
+
"eval_qasc_pairs_steps_per_second": 3.79,
|
| 601 |
+
"step": 960
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 0.15044663845792194,
|
| 605 |
+
"eval_openbookqa_pairs_loss": 1.8053069114685059,
|
| 606 |
+
"eval_openbookqa_pairs_runtime": 0.8871,
|
| 607 |
+
"eval_openbookqa_pairs_samples_per_second": 112.727,
|
| 608 |
+
"eval_openbookqa_pairs_steps_per_second": 4.509,
|
| 609 |
+
"step": 960
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.15044663845792194,
|
| 613 |
+
"eval_msmarco_pairs_loss": 0.5809260606765747,
|
| 614 |
+
"eval_msmarco_pairs_runtime": 2.0498,
|
| 615 |
+
"eval_msmarco_pairs_samples_per_second": 48.786,
|
| 616 |
+
"eval_msmarco_pairs_steps_per_second": 1.951,
|
| 617 |
+
"step": 960
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 0.15044663845792194,
|
| 621 |
+
"eval_nq_pairs_loss": 0.2808491885662079,
|
| 622 |
+
"eval_nq_pairs_runtime": 4.4982,
|
| 623 |
+
"eval_nq_pairs_samples_per_second": 22.231,
|
| 624 |
+
"eval_nq_pairs_steps_per_second": 0.889,
|
| 625 |
+
"step": 960
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.15044663845792194,
|
| 629 |
+
"eval_trivia_pairs_loss": 0.9379808902740479,
|
| 630 |
+
"eval_trivia_pairs_runtime": 6.4578,
|
| 631 |
+
"eval_trivia_pairs_samples_per_second": 15.485,
|
| 632 |
+
"eval_trivia_pairs_steps_per_second": 0.619,
|
| 633 |
+
"step": 960
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.15044663845792194,
|
| 637 |
+
"eval_quora_pairs_loss": 0.0913279801607132,
|
| 638 |
+
"eval_quora_pairs_runtime": 0.6721,
|
| 639 |
+
"eval_quora_pairs_samples_per_second": 148.79,
|
| 640 |
+
"eval_quora_pairs_steps_per_second": 5.952,
|
| 641 |
+
"step": 960
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 0.15044663845792194,
|
| 645 |
+
"eval_gooaq_pairs_loss": 0.5807955265045166,
|
| 646 |
+
"eval_gooaq_pairs_runtime": 1.3915,
|
| 647 |
+
"eval_gooaq_pairs_samples_per_second": 71.865,
|
| 648 |
+
"eval_gooaq_pairs_steps_per_second": 2.875,
|
| 649 |
+
"step": 960
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 0.15044663845792194,
|
| 653 |
+
"eval_mrpc_pairs_loss": 0.05799216777086258,
|
| 654 |
+
"eval_mrpc_pairs_runtime": 0.2571,
|
| 655 |
+
"eval_mrpc_pairs_samples_per_second": 388.998,
|
| 656 |
+
"eval_mrpc_pairs_steps_per_second": 15.56,
|
| 657 |
+
"step": 960
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 0.15546152640651936,
|
| 661 |
+
"grad_norm": 9.773286819458008,
|
| 662 |
+
"learning_rate": 2.9997957904107625e-05,
|
| 663 |
+
"loss": 0.7964,
|
| 664 |
+
"step": 992
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 0.16047641435511675,
|
| 668 |
+
"grad_norm": 19.411075592041016,
|
| 669 |
+
"learning_rate": 2.9991566594209126e-05,
|
| 670 |
+
"loss": 0.8213,
|
| 671 |
+
"step": 1024
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 0.16549130230371414,
|
| 675 |
+
"grad_norm": 3.5282175540924072,
|
| 676 |
+
"learning_rate": 2.9980825799589488e-05,
|
| 677 |
+
"loss": 0.5396,
|
| 678 |
+
"step": 1056
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 0.17050619025231156,
|
| 682 |
+
"grad_norm": 62.66339874267578,
|
| 683 |
+
"learning_rate": 2.996573863646219e-05,
|
| 684 |
+
"loss": 0.9297,
|
| 685 |
+
"step": 1088
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 0.17552107820090895,
|
| 689 |
+
"grad_norm": 8.785274505615234,
|
| 690 |
+
"learning_rate": 2.994630948204727e-05,
|
| 691 |
+
"loss": 1.169,
|
| 692 |
+
"step": 1120
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 0.18053596614950634,
|
| 696 |
+
"grad_norm": 24.10859489440918,
|
| 697 |
+
"learning_rate": 2.992254397330132e-05,
|
| 698 |
+
"loss": 0.7486,
|
| 699 |
+
"step": 1152
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 0.18555085409810373,
|
| 703 |
+
"grad_norm": 25.545284271240234,
|
| 704 |
+
"learning_rate": 2.9894449005282077e-05,
|
| 705 |
+
"loss": 0.6821,
|
| 706 |
+
"step": 1184
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 0.19056574204670115,
|
| 710 |
+
"grad_norm": 0.8675521016120911,
|
| 711 |
+
"learning_rate": 2.9862032729147954e-05,
|
| 712 |
+
"loss": 0.6125,
|
| 713 |
+
"step": 1216
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 0.19558062999529854,
|
| 717 |
+
"grad_norm": 16.122114181518555,
|
| 718 |
+
"learning_rate": 2.9825304549793153e-05,
|
| 719 |
+
"loss": 0.8061,
|
| 720 |
+
"step": 1248
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 0.20059551794389593,
|
| 724 |
+
"grad_norm": 1.0314382314682007,
|
| 725 |
+
"learning_rate": 2.978427512311904e-05,
|
| 726 |
+
"loss": 0.6918,
|
| 727 |
+
"step": 1280
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 0.20059551794389593,
|
| 731 |
+
"eval_nli-pairs_loss": 1.1552109718322754,
|
| 732 |
+
"eval_nli-pairs_runtime": 3.8751,
|
| 733 |
+
"eval_nli-pairs_samples_per_second": 25.806,
|
| 734 |
+
"eval_nli-pairs_steps_per_second": 1.032,
|
| 735 |
+
"eval_sts-test_pearson_cosine": 0.786106976104726,
|
| 736 |
+
"eval_sts-test_pearson_dot": 0.5116758767219935,
|
| 737 |
+
"eval_sts-test_pearson_euclidean": 0.7432891018313416,
|
| 738 |
+
"eval_sts-test_pearson_manhattan": 0.7400929158927781,
|
| 739 |
+
"eval_sts-test_pearson_max": 0.786106976104726,
|
| 740 |
+
"eval_sts-test_spearman_cosine": 0.801377272203007,
|
| 741 |
+
"eval_sts-test_spearman_dot": 0.4921454166952506,
|
| 742 |
+
"eval_sts-test_spearman_euclidean": 0.7343686249967402,
|
| 743 |
+
"eval_sts-test_spearman_manhattan": 0.7331946050808561,
|
| 744 |
+
"eval_sts-test_spearman_max": 0.801377272203007,
|
| 745 |
+
"step": 1280
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.20059551794389593,
|
| 749 |
+
"eval_vitaminc-pairs_loss": 4.6789751052856445,
|
| 750 |
+
"eval_vitaminc-pairs_runtime": 1.1504,
|
| 751 |
+
"eval_vitaminc-pairs_samples_per_second": 73.889,
|
| 752 |
+
"eval_vitaminc-pairs_steps_per_second": 2.608,
|
| 753 |
+
"step": 1280
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.20059551794389593,
|
| 757 |
+
"eval_sts-label_loss": 3.5580556392669678,
|
| 758 |
+
"eval_sts-label_runtime": 0.2834,
|
| 759 |
+
"eval_sts-label_samples_per_second": 352.858,
|
| 760 |
+
"eval_sts-label_steps_per_second": 14.114,
|
| 761 |
+
"step": 1280
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 0.20059551794389593,
|
| 765 |
+
"eval_qnli-contrastive_loss": 0.20369713008403778,
|
| 766 |
+
"eval_qnli-contrastive_runtime": 0.358,
|
| 767 |
+
"eval_qnli-contrastive_samples_per_second": 279.331,
|
| 768 |
+
"eval_qnli-contrastive_steps_per_second": 11.173,
|
| 769 |
+
"step": 1280
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 0.20059551794389593,
|
| 773 |
+
"eval_scitail-pairs-qa_loss": 0.07465875148773193,
|
| 774 |
+
"eval_scitail-pairs-qa_runtime": 0.9504,
|
| 775 |
+
"eval_scitail-pairs-qa_samples_per_second": 105.214,
|
| 776 |
+
"eval_scitail-pairs-qa_steps_per_second": 4.209,
|
| 777 |
+
"step": 1280
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 0.20059551794389593,
|
| 781 |
+
"eval_scitail-pairs-pos_loss": 0.49434563517570496,
|
| 782 |
+
"eval_scitail-pairs-pos_runtime": 1.6041,
|
| 783 |
+
"eval_scitail-pairs-pos_samples_per_second": 62.339,
|
| 784 |
+
"eval_scitail-pairs-pos_steps_per_second": 2.494,
|
| 785 |
+
"step": 1280
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 0.20059551794389593,
|
| 789 |
+
"eval_xsum-pairs_loss": 0.28282061219215393,
|
| 790 |
+
"eval_xsum-pairs_runtime": 0.9316,
|
| 791 |
+
"eval_xsum-pairs_samples_per_second": 107.346,
|
| 792 |
+
"eval_xsum-pairs_steps_per_second": 4.294,
|
| 793 |
+
"step": 1280
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.20059551794389593,
|
| 797 |
+
"eval_compression-pairs_loss": 0.097385473549366,
|
| 798 |
+
"eval_compression-pairs_runtime": 0.2754,
|
| 799 |
+
"eval_compression-pairs_samples_per_second": 363.1,
|
| 800 |
+
"eval_compression-pairs_steps_per_second": 14.524,
|
| 801 |
+
"step": 1280
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.20059551794389593,
|
| 805 |
+
"eval_sciq_pairs_loss": 0.2762215733528137,
|
| 806 |
+
"eval_sciq_pairs_runtime": 4.2307,
|
| 807 |
+
"eval_sciq_pairs_samples_per_second": 23.637,
|
| 808 |
+
"eval_sciq_pairs_steps_per_second": 0.945,
|
| 809 |
+
"step": 1280
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 0.20059551794389593,
|
| 813 |
+
"eval_qasc_pairs_loss": 0.19347424805164337,
|
| 814 |
+
"eval_qasc_pairs_runtime": 1.2282,
|
| 815 |
+
"eval_qasc_pairs_samples_per_second": 81.421,
|
| 816 |
+
"eval_qasc_pairs_steps_per_second": 3.257,
|
| 817 |
+
"step": 1280
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 0.20059551794389593,
|
| 821 |
+
"eval_openbookqa_pairs_loss": 1.6875064373016357,
|
| 822 |
+
"eval_openbookqa_pairs_runtime": 1.1661,
|
| 823 |
+
"eval_openbookqa_pairs_samples_per_second": 85.754,
|
| 824 |
+
"eval_openbookqa_pairs_steps_per_second": 3.43,
|
| 825 |
+
"step": 1280
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 0.20059551794389593,
|
| 829 |
+
"eval_msmarco_pairs_loss": 0.5743877291679382,
|
| 830 |
+
"eval_msmarco_pairs_runtime": 2.1428,
|
| 831 |
+
"eval_msmarco_pairs_samples_per_second": 46.669,
|
| 832 |
+
"eval_msmarco_pairs_steps_per_second": 1.867,
|
| 833 |
+
"step": 1280
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 0.20059551794389593,
|
| 837 |
+
"eval_nq_pairs_loss": 0.30348217487335205,
|
| 838 |
+
"eval_nq_pairs_runtime": 4.5543,
|
| 839 |
+
"eval_nq_pairs_samples_per_second": 21.957,
|
| 840 |
+
"eval_nq_pairs_steps_per_second": 0.878,
|
| 841 |
+
"step": 1280
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.20059551794389593,
|
| 845 |
+
"eval_trivia_pairs_loss": 0.9221765995025635,
|
| 846 |
+
"eval_trivia_pairs_runtime": 6.6513,
|
| 847 |
+
"eval_trivia_pairs_samples_per_second": 15.035,
|
| 848 |
+
"eval_trivia_pairs_steps_per_second": 0.601,
|
| 849 |
+
"step": 1280
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.20059551794389593,
|
| 853 |
+
"eval_quora_pairs_loss": 0.03854631260037422,
|
| 854 |
+
"eval_quora_pairs_runtime": 0.7822,
|
| 855 |
+
"eval_quora_pairs_samples_per_second": 127.852,
|
| 856 |
+
"eval_quora_pairs_steps_per_second": 5.114,
|
| 857 |
+
"step": 1280
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.20059551794389593,
|
| 861 |
+
"eval_gooaq_pairs_loss": 0.528398334980011,
|
| 862 |
+
"eval_gooaq_pairs_runtime": 1.4882,
|
| 863 |
+
"eval_gooaq_pairs_samples_per_second": 67.194,
|
| 864 |
+
"eval_gooaq_pairs_steps_per_second": 2.688,
|
| 865 |
+
"step": 1280
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 0.20059551794389593,
|
| 869 |
+
"eval_mrpc_pairs_loss": 0.05623970925807953,
|
| 870 |
+
"eval_mrpc_pairs_runtime": 0.2698,
|
| 871 |
+
"eval_mrpc_pairs_samples_per_second": 370.713,
|
| 872 |
+
"eval_mrpc_pairs_steps_per_second": 14.829,
|
| 873 |
+
"step": 1280
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 0.20561040589249335,
|
| 877 |
+
"grad_norm": 0.6042119860649109,
|
| 878 |
+
"learning_rate": 2.9738956352942557e-05,
|
| 879 |
+
"loss": 0.9421,
|
| 880 |
+
"step": 1312
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 0.21062529384109074,
|
| 884 |
+
"grad_norm": 13.87867546081543,
|
| 885 |
+
"learning_rate": 2.968936138754259e-05,
|
| 886 |
+
"loss": 0.8641,
|
| 887 |
+
"step": 1344
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 0.21564018178968813,
|
| 891 |
+
"grad_norm": 44.48640441894531,
|
| 892 |
+
"learning_rate": 2.9635504615845257e-05,
|
| 893 |
+
"loss": 1.157,
|
| 894 |
+
"step": 1376
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 0.22065506973828553,
|
| 898 |
+
"grad_norm": 15.554729461669922,
|
| 899 |
+
"learning_rate": 2.957928148945977e-05,
|
| 900 |
+
"loss": 0.8772,
|
| 901 |
+
"step": 1408
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 0.22566995768688294,
|
| 905 |
+
"grad_norm": 16.644670486450195,
|
| 906 |
+
"learning_rate": 2.9517081112297707e-05,
|
| 907 |
+
"loss": 1.0496,
|
| 908 |
+
"step": 1440
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 0.23068484563548033,
|
| 912 |
+
"grad_norm": 13.053145408630371,
|
| 913 |
+
"learning_rate": 2.9450668912302004e-05,
|
| 914 |
+
"loss": 0.589,
|
| 915 |
+
"step": 1472
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 0.23569973358407773,
|
| 919 |
+
"grad_norm": 7.827791213989258,
|
| 920 |
+
"learning_rate": 2.9380064157562306e-05,
|
| 921 |
+
"loss": 0.8234,
|
| 922 |
+
"step": 1504
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 0.24071462153267512,
|
| 926 |
+
"grad_norm": 15.598438262939453,
|
| 927 |
+
"learning_rate": 2.930528733254901e-05,
|
| 928 |
+
"loss": 0.7365,
|
| 929 |
+
"step": 1536
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 0.24572950948127253,
|
| 933 |
+
"grad_norm": 13.723180770874023,
|
| 934 |
+
"learning_rate": 2.9226360132170112e-05,
|
| 935 |
+
"loss": 0.5076,
|
| 936 |
+
"step": 1568
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 0.2507443974298699,
|
| 940 |
+
"grad_norm": 10.20022964477539,
|
| 941 |
+
"learning_rate": 2.9143305455476866e-05,
|
| 942 |
+
"loss": 1.0329,
|
| 943 |
+
"step": 1600
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 0.2507443974298699,
|
| 947 |
+
"eval_nli-pairs_loss": 1.0577216148376465,
|
| 948 |
+
"eval_nli-pairs_runtime": 3.6476,
|
| 949 |
+
"eval_nli-pairs_samples_per_second": 27.415,
|
| 950 |
+
"eval_nli-pairs_steps_per_second": 1.097,
|
| 951 |
+
"eval_sts-test_pearson_cosine": 0.7876359552191669,
|
| 952 |
+
"eval_sts-test_pearson_dot": 0.5220803655074544,
|
| 953 |
+
"eval_sts-test_pearson_euclidean": 0.7444632413869628,
|
| 954 |
+
"eval_sts-test_pearson_manhattan": 0.7418744760088763,
|
| 955 |
+
"eval_sts-test_pearson_max": 0.7876359552191669,
|
| 956 |
+
"eval_sts-test_spearman_cosine": 0.8018874000525117,
|
| 957 |
+
"eval_sts-test_spearman_dot": 0.5034518981121652,
|
| 958 |
+
"eval_sts-test_spearman_euclidean": 0.7344750702387959,
|
| 959 |
+
"eval_sts-test_spearman_manhattan": 0.7332804063416474,
|
| 960 |
+
"eval_sts-test_spearman_max": 0.8018874000525117,
|
| 961 |
+
"step": 1600
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.2507443974298699,
|
| 965 |
+
"eval_vitaminc-pairs_loss": 4.784573554992676,
|
| 966 |
+
"eval_vitaminc-pairs_runtime": 1.145,
|
| 967 |
+
"eval_vitaminc-pairs_samples_per_second": 74.235,
|
| 968 |
+
"eval_vitaminc-pairs_steps_per_second": 2.62,
|
| 969 |
+
"step": 1600
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.2507443974298699,
|
| 973 |
+
"eval_sts-label_loss": 3.6113080978393555,
|
| 974 |
+
"eval_sts-label_runtime": 0.2746,
|
| 975 |
+
"eval_sts-label_samples_per_second": 364.172,
|
| 976 |
+
"eval_sts-label_steps_per_second": 14.567,
|
| 977 |
+
"step": 1600
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 0.2507443974298699,
|
| 981 |
+
"eval_qnli-contrastive_loss": 0.18593625724315643,
|
| 982 |
+
"eval_qnli-contrastive_runtime": 0.3541,
|
| 983 |
+
"eval_qnli-contrastive_samples_per_second": 282.413,
|
| 984 |
+
"eval_qnli-contrastive_steps_per_second": 11.297,
|
| 985 |
+
"step": 1600
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 0.2507443974298699,
|
| 989 |
+
"eval_scitail-pairs-qa_loss": 0.07545661181211472,
|
| 990 |
+
"eval_scitail-pairs-qa_runtime": 0.8854,
|
| 991 |
+
"eval_scitail-pairs-qa_samples_per_second": 112.941,
|
| 992 |
+
"eval_scitail-pairs-qa_steps_per_second": 4.518,
|
| 993 |
+
"step": 1600
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 0.2507443974298699,
|
| 997 |
+
"eval_scitail-pairs-pos_loss": 0.5018333792686462,
|
| 998 |
+
"eval_scitail-pairs-pos_runtime": 1.3443,
|
| 999 |
+
"eval_scitail-pairs-pos_samples_per_second": 74.386,
|
| 1000 |
+
"eval_scitail-pairs-pos_steps_per_second": 2.975,
|
| 1001 |
+
"step": 1600
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 0.2507443974298699,
|
| 1005 |
+
"eval_xsum-pairs_loss": 0.2749001085758209,
|
| 1006 |
+
"eval_xsum-pairs_runtime": 0.9439,
|
| 1007 |
+
"eval_xsum-pairs_samples_per_second": 105.939,
|
| 1008 |
+
"eval_xsum-pairs_steps_per_second": 4.238,
|
| 1009 |
+
"step": 1600
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 0.2507443974298699,
|
| 1013 |
+
"eval_compression-pairs_loss": 0.09735233336687088,
|
| 1014 |
+
"eval_compression-pairs_runtime": 0.2764,
|
| 1015 |
+
"eval_compression-pairs_samples_per_second": 361.753,
|
| 1016 |
+
"eval_compression-pairs_steps_per_second": 14.47,
|
| 1017 |
+
"step": 1600
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.2507443974298699,
|
| 1021 |
+
"eval_sciq_pairs_loss": 0.2648228108882904,
|
| 1022 |
+
"eval_sciq_pairs_runtime": 4.1207,
|
| 1023 |
+
"eval_sciq_pairs_samples_per_second": 24.268,
|
| 1024 |
+
"eval_sciq_pairs_steps_per_second": 0.971,
|
| 1025 |
+
"step": 1600
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.2507443974298699,
|
| 1029 |
+
"eval_qasc_pairs_loss": 0.21318012475967407,
|
| 1030 |
+
"eval_qasc_pairs_runtime": 1.0917,
|
| 1031 |
+
"eval_qasc_pairs_samples_per_second": 91.604,
|
| 1032 |
+
"eval_qasc_pairs_steps_per_second": 3.664,
|
| 1033 |
+
"step": 1600
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 0.2507443974298699,
|
| 1037 |
+
"eval_openbookqa_pairs_loss": 1.790009617805481,
|
| 1038 |
+
"eval_openbookqa_pairs_runtime": 0.8969,
|
| 1039 |
+
"eval_openbookqa_pairs_samples_per_second": 111.496,
|
| 1040 |
+
"eval_openbookqa_pairs_steps_per_second": 4.46,
|
| 1041 |
+
"step": 1600
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 0.2507443974298699,
|
| 1045 |
+
"eval_msmarco_pairs_loss": 0.57186359167099,
|
| 1046 |
+
"eval_msmarco_pairs_runtime": 2.0592,
|
| 1047 |
+
"eval_msmarco_pairs_samples_per_second": 48.563,
|
| 1048 |
+
"eval_msmarco_pairs_steps_per_second": 1.943,
|
| 1049 |
+
"step": 1600
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 0.2507443974298699,
|
| 1053 |
+
"eval_nq_pairs_loss": 0.2738310396671295,
|
| 1054 |
+
"eval_nq_pairs_runtime": 4.5092,
|
| 1055 |
+
"eval_nq_pairs_samples_per_second": 22.177,
|
| 1056 |
+
"eval_nq_pairs_steps_per_second": 0.887,
|
| 1057 |
+
"step": 1600
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 0.2507443974298699,
|
| 1061 |
+
"eval_trivia_pairs_loss": 0.8291679620742798,
|
| 1062 |
+
"eval_trivia_pairs_runtime": 6.526,
|
| 1063 |
+
"eval_trivia_pairs_samples_per_second": 15.323,
|
| 1064 |
+
"eval_trivia_pairs_steps_per_second": 0.613,
|
| 1065 |
+
"step": 1600
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 0.2507443974298699,
|
| 1069 |
+
"eval_quora_pairs_loss": 0.08000540733337402,
|
| 1070 |
+
"eval_quora_pairs_runtime": 0.6761,
|
| 1071 |
+
"eval_quora_pairs_samples_per_second": 147.909,
|
| 1072 |
+
"eval_quora_pairs_steps_per_second": 5.916,
|
| 1073 |
+
"step": 1600
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.2507443974298699,
|
| 1077 |
+
"eval_gooaq_pairs_loss": 0.5998037457466125,
|
| 1078 |
+
"eval_gooaq_pairs_runtime": 1.3978,
|
| 1079 |
+
"eval_gooaq_pairs_samples_per_second": 71.541,
|
| 1080 |
+
"eval_gooaq_pairs_steps_per_second": 2.862,
|
| 1081 |
+
"step": 1600
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.2507443974298699,
|
| 1085 |
+
"eval_mrpc_pairs_loss": 0.05507182702422142,
|
| 1086 |
+
"eval_mrpc_pairs_runtime": 0.2617,
|
| 1087 |
+
"eval_mrpc_pairs_samples_per_second": 382.156,
|
| 1088 |
+
"eval_mrpc_pairs_steps_per_second": 15.286,
|
| 1089 |
+
"step": 1600
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 0.2557592853784673,
|
| 1093 |
+
"grad_norm": 8.05022144317627,
|
| 1094 |
+
"learning_rate": 2.9056147399020182e-05,
|
| 1095 |
+
"loss": 1.4006,
|
| 1096 |
+
"step": 1632
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 0.2607741733270647,
|
| 1100 |
+
"grad_norm": 0.38224154710769653,
|
| 1101 |
+
"learning_rate": 2.8964911249859437e-05,
|
| 1102 |
+
"loss": 0.5963,
|
| 1103 |
+
"step": 1664
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 0.2657890612756621,
|
| 1107 |
+
"grad_norm": 0.46655791997909546,
|
| 1108 |
+
"learning_rate": 2.886962347822604e-05,
|
| 1109 |
+
"loss": 0.7488,
|
| 1110 |
+
"step": 1696
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 0.27080394922425954,
|
| 1114 |
+
"grad_norm": 8.102537155151367,
|
| 1115 |
+
"learning_rate": 2.8770311729843616e-05,
|
| 1116 |
+
"loss": 0.8548,
|
| 1117 |
+
"step": 1728
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 0.27581883717285693,
|
| 1121 |
+
"grad_norm": 11.803775787353516,
|
| 1122 |
+
"learning_rate": 2.86670048179072e-05,
|
| 1123 |
+
"loss": 1.3324,
|
| 1124 |
+
"step": 1760
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 0.2808337251214543,
|
| 1128 |
+
"grad_norm": 16.266756057739258,
|
| 1129 |
+
"learning_rate": 2.8559732714723715e-05,
|
| 1130 |
+
"loss": 0.5804,
|
| 1131 |
+
"step": 1792
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 0.2858486130700517,
|
| 1135 |
+
"grad_norm": 2.8448822498321533,
|
| 1136 |
+
"learning_rate": 2.8448526543016114e-05,
|
| 1137 |
+
"loss": 0.7827,
|
| 1138 |
+
"step": 1824
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 0.2908635010186491,
|
| 1142 |
+
"grad_norm": 21.346328735351562,
|
| 1143 |
+
"learning_rate": 2.8333418566893796e-05,
|
| 1144 |
+
"loss": 0.5448,
|
| 1145 |
+
"step": 1856
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 0.2958783889672465,
|
| 1149 |
+
"grad_norm": 3.4379029273986816,
|
| 1150 |
+
"learning_rate": 2.8214442182491866e-05,
|
| 1151 |
+
"loss": 0.7368,
|
| 1152 |
+
"step": 1888
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 0.3008932769158439,
|
| 1156 |
+
"grad_norm": 17.05881690979004,
|
| 1157 |
+
"learning_rate": 2.8091631908281963e-05,
|
| 1158 |
+
"loss": 0.5657,
|
| 1159 |
+
"step": 1920
|
| 1160 |
+
},
|
| 1161 |
+
{
|
| 1162 |
+
"epoch": 0.3008932769158439,
|
| 1163 |
+
"eval_nli-pairs_loss": 1.0244356393814087,
|
| 1164 |
+
"eval_nli-pairs_runtime": 3.6217,
|
| 1165 |
+
"eval_nli-pairs_samples_per_second": 27.612,
|
| 1166 |
+
"eval_nli-pairs_steps_per_second": 1.104,
|
| 1167 |
+
"eval_sts-test_pearson_cosine": 0.781915957368962,
|
| 1168 |
+
"eval_sts-test_pearson_dot": 0.49821032356844613,
|
| 1169 |
+
"eval_sts-test_pearson_euclidean": 0.7329308897504494,
|
| 1170 |
+
"eval_sts-test_pearson_manhattan": 0.7292186092506918,
|
| 1171 |
+
"eval_sts-test_pearson_max": 0.781915957368962,
|
| 1172 |
+
"eval_sts-test_spearman_cosine": 0.7983596570250642,
|
| 1173 |
+
"eval_sts-test_spearman_dot": 0.4812350313638781,
|
| 1174 |
+
"eval_sts-test_spearman_euclidean": 0.7265758267352669,
|
| 1175 |
+
"eval_sts-test_spearman_manhattan": 0.7259264140902829,
|
| 1176 |
+
"eval_sts-test_spearman_max": 0.7983596570250642,
|
| 1177 |
+
"step": 1920
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 0.3008932769158439,
|
| 1181 |
+
"eval_vitaminc-pairs_loss": 4.698296070098877,
|
| 1182 |
+
"eval_vitaminc-pairs_runtime": 1.1338,
|
| 1183 |
+
"eval_vitaminc-pairs_samples_per_second": 74.97,
|
| 1184 |
+
"eval_vitaminc-pairs_steps_per_second": 2.646,
|
| 1185 |
+
"step": 1920
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.3008932769158439,
|
| 1189 |
+
"eval_sts-label_loss": 3.1822261810302734,
|
| 1190 |
+
"eval_sts-label_runtime": 0.2702,
|
| 1191 |
+
"eval_sts-label_samples_per_second": 370.09,
|
| 1192 |
+
"eval_sts-label_steps_per_second": 14.804,
|
| 1193 |
+
"step": 1920
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.3008932769158439,
|
| 1197 |
+
"eval_qnli-contrastive_loss": 0.11326340585947037,
|
| 1198 |
+
"eval_qnli-contrastive_runtime": 0.3581,
|
| 1199 |
+
"eval_qnli-contrastive_samples_per_second": 279.28,
|
| 1200 |
+
"eval_qnli-contrastive_steps_per_second": 11.171,
|
| 1201 |
+
"step": 1920
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 0.3008932769158439,
|
| 1205 |
+
"eval_scitail-pairs-qa_loss": 0.07009608298540115,
|
| 1206 |
+
"eval_scitail-pairs-qa_runtime": 0.8816,
|
| 1207 |
+
"eval_scitail-pairs-qa_samples_per_second": 113.424,
|
| 1208 |
+
"eval_scitail-pairs-qa_steps_per_second": 4.537,
|
| 1209 |
+
"step": 1920
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 0.3008932769158439,
|
| 1213 |
+
"eval_scitail-pairs-pos_loss": 0.49156129360198975,
|
| 1214 |
+
"eval_scitail-pairs-pos_runtime": 1.3759,
|
| 1215 |
+
"eval_scitail-pairs-pos_samples_per_second": 72.678,
|
| 1216 |
+
"eval_scitail-pairs-pos_steps_per_second": 2.907,
|
| 1217 |
+
"step": 1920
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 0.3008932769158439,
|
| 1221 |
+
"eval_xsum-pairs_loss": 0.25940877199172974,
|
| 1222 |
+
"eval_xsum-pairs_runtime": 0.9373,
|
| 1223 |
+
"eval_xsum-pairs_samples_per_second": 106.695,
|
| 1224 |
+
"eval_xsum-pairs_steps_per_second": 4.268,
|
| 1225 |
+
"step": 1920
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 0.3008932769158439,
|
| 1229 |
+
"eval_compression-pairs_loss": 0.0919649675488472,
|
| 1230 |
+
"eval_compression-pairs_runtime": 0.2738,
|
| 1231 |
+
"eval_compression-pairs_samples_per_second": 365.291,
|
| 1232 |
+
"eval_compression-pairs_steps_per_second": 14.612,
|
| 1233 |
+
"step": 1920
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 0.3008932769158439,
|
| 1237 |
+
"eval_sciq_pairs_loss": 0.29138606786727905,
|
| 1238 |
+
"eval_sciq_pairs_runtime": 4.1059,
|
| 1239 |
+
"eval_sciq_pairs_samples_per_second": 24.355,
|
| 1240 |
+
"eval_sciq_pairs_steps_per_second": 0.974,
|
| 1241 |
+
"step": 1920
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.3008932769158439,
|
| 1245 |
+
"eval_qasc_pairs_loss": 0.19625085592269897,
|
| 1246 |
+
"eval_qasc_pairs_runtime": 1.0611,
|
| 1247 |
+
"eval_qasc_pairs_samples_per_second": 94.24,
|
| 1248 |
+
"eval_qasc_pairs_steps_per_second": 3.77,
|
| 1249 |
+
"step": 1920
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.3008932769158439,
|
| 1253 |
+
"eval_openbookqa_pairs_loss": 1.7960456609725952,
|
| 1254 |
+
"eval_openbookqa_pairs_runtime": 0.9042,
|
| 1255 |
+
"eval_openbookqa_pairs_samples_per_second": 110.601,
|
| 1256 |
+
"eval_openbookqa_pairs_steps_per_second": 4.424,
|
| 1257 |
+
"step": 1920
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 0.3008932769158439,
|
| 1261 |
+
"eval_msmarco_pairs_loss": 0.5171416997909546,
|
| 1262 |
+
"eval_msmarco_pairs_runtime": 2.0637,
|
| 1263 |
+
"eval_msmarco_pairs_samples_per_second": 48.457,
|
| 1264 |
+
"eval_msmarco_pairs_steps_per_second": 1.938,
|
| 1265 |
+
"step": 1920
|
| 1266 |
+
},
|
| 1267 |
+
{
|
| 1268 |
+
"epoch": 0.3008932769158439,
|
| 1269 |
+
"eval_nq_pairs_loss": 0.24809740483760834,
|
| 1270 |
+
"eval_nq_pairs_runtime": 4.529,
|
| 1271 |
+
"eval_nq_pairs_samples_per_second": 22.08,
|
| 1272 |
+
"eval_nq_pairs_steps_per_second": 0.883,
|
| 1273 |
+
"step": 1920
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 0.3008932769158439,
|
| 1277 |
+
"eval_trivia_pairs_loss": 0.9041999578475952,
|
| 1278 |
+
"eval_trivia_pairs_runtime": 6.5257,
|
| 1279 |
+
"eval_trivia_pairs_samples_per_second": 15.324,
|
| 1280 |
+
"eval_trivia_pairs_steps_per_second": 0.613,
|
| 1281 |
+
"step": 1920
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 0.3008932769158439,
|
| 1285 |
+
"eval_quora_pairs_loss": 0.03601976856589317,
|
| 1286 |
+
"eval_quora_pairs_runtime": 0.6811,
|
| 1287 |
+
"eval_quora_pairs_samples_per_second": 146.827,
|
| 1288 |
+
"eval_quora_pairs_steps_per_second": 5.873,
|
| 1289 |
+
"step": 1920
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 0.3008932769158439,
|
| 1293 |
+
"eval_gooaq_pairs_loss": 0.5626399517059326,
|
| 1294 |
+
"eval_gooaq_pairs_runtime": 1.3943,
|
| 1295 |
+
"eval_gooaq_pairs_samples_per_second": 71.72,
|
| 1296 |
+
"eval_gooaq_pairs_steps_per_second": 2.869,
|
| 1297 |
+
"step": 1920
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.3008932769158439,
|
| 1301 |
+
"eval_mrpc_pairs_loss": 0.04984402656555176,
|
| 1302 |
+
"eval_mrpc_pairs_runtime": 0.2579,
|
| 1303 |
+
"eval_mrpc_pairs_samples_per_second": 387.725,
|
| 1304 |
+
"eval_mrpc_pairs_steps_per_second": 15.509,
|
| 1305 |
+
"step": 1920
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.30590816486444133,
|
| 1309 |
+
"grad_norm": 22.65591812133789,
|
| 1310 |
+
"learning_rate": 2.796502337505742e-05,
|
| 1311 |
+
"loss": 0.7425,
|
| 1312 |
+
"step": 1952
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.3109230528130387,
|
| 1316 |
+
"grad_norm": 10.119640350341797,
|
| 1317 |
+
"learning_rate": 2.78346533155958e-05,
|
| 1318 |
+
"loss": 0.7819,
|
| 1319 |
+
"step": 1984
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.3159379407616361,
|
| 1323 |
+
"grad_norm": 8.690531730651855,
|
| 1324 |
+
"learning_rate": 2.770055955400161e-05,
|
| 1325 |
+
"loss": 0.5937,
|
| 1326 |
+
"step": 2016
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.3209528287102335,
|
| 1330 |
+
"grad_norm": 0.8992699384689331,
|
| 1331 |
+
"learning_rate": 2.7562780994732476e-05,
|
| 1332 |
+
"loss": 0.8133,
|
| 1333 |
+
"step": 2048
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.3259677166588309,
|
| 1337 |
+
"grad_norm": 10.619684219360352,
|
| 1338 |
+
"learning_rate": 2.7421357611311824e-05,
|
| 1339 |
+
"loss": 1.0674,
|
| 1340 |
+
"step": 2080
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.3309826046074283,
|
| 1344 |
+
"grad_norm": 7.222084045410156,
|
| 1345 |
+
"learning_rate": 2.727633043473141e-05,
|
| 1346 |
+
"loss": 0.6288,
|
| 1347 |
+
"step": 2112
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.3359974925560257,
|
| 1351 |
+
"grad_norm": 10.166888236999512,
|
| 1352 |
+
"learning_rate": 2.712774154154707e-05,
|
| 1353 |
+
"loss": 0.5866,
|
| 1354 |
+
"step": 2144
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.3410123805046231,
|
| 1358 |
+
"grad_norm": 0.36360761523246765,
|
| 1359 |
+
"learning_rate": 2.6975634041671052e-05,
|
| 1360 |
+
"loss": 0.6962,
|
| 1361 |
+
"step": 2176
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.3460272684532205,
|
| 1365 |
+
"grad_norm": 9.586665153503418,
|
| 1366 |
+
"learning_rate": 2.6820052065864665e-05,
|
| 1367 |
+
"loss": 0.5562,
|
| 1368 |
+
"step": 2208
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.3510421564018179,
|
| 1372 |
+
"grad_norm": 1.1307642459869385,
|
| 1373 |
+
"learning_rate": 2.6661040752934594e-05,
|
| 1374 |
+
"loss": 0.8871,
|
| 1375 |
+
"step": 2240
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.3510421564018179,
|
| 1379 |
+
"eval_nli-pairs_loss": 1.0147591829299927,
|
| 1380 |
+
"eval_nli-pairs_runtime": 3.7201,
|
| 1381 |
+
"eval_nli-pairs_samples_per_second": 26.881,
|
| 1382 |
+
"eval_nli-pairs_steps_per_second": 1.075,
|
| 1383 |
+
"eval_sts-test_pearson_cosine": 0.7872126529181761,
|
| 1384 |
+
"eval_sts-test_pearson_dot": 0.5062045289861089,
|
| 1385 |
+
"eval_sts-test_pearson_euclidean": 0.7351473988633473,
|
| 1386 |
+
"eval_sts-test_pearson_manhattan": 0.7310226402088944,
|
| 1387 |
+
"eval_sts-test_pearson_max": 0.7872126529181761,
|
| 1388 |
+
"eval_sts-test_spearman_cosine": 0.801487068999052,
|
| 1389 |
+
"eval_sts-test_spearman_dot": 0.4912205722904683,
|
| 1390 |
+
"eval_sts-test_spearman_euclidean": 0.7267262355024484,
|
| 1391 |
+
"eval_sts-test_spearman_manhattan": 0.72510169253649,
|
| 1392 |
+
"eval_sts-test_spearman_max": 0.801487068999052,
|
| 1393 |
+
"step": 2240
|
| 1394 |
+
},
|
| 1395 |
+
{
|
| 1396 |
+
"epoch": 0.3510421564018179,
|
| 1397 |
+
"eval_vitaminc-pairs_loss": 4.644638538360596,
|
| 1398 |
+
"eval_vitaminc-pairs_runtime": 1.1453,
|
| 1399 |
+
"eval_vitaminc-pairs_samples_per_second": 74.215,
|
| 1400 |
+
"eval_vitaminc-pairs_steps_per_second": 2.619,
|
| 1401 |
+
"step": 2240
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 0.3510421564018179,
|
| 1405 |
+
"eval_sts-label_loss": 3.915343999862671,
|
| 1406 |
+
"eval_sts-label_runtime": 0.2807,
|
| 1407 |
+
"eval_sts-label_samples_per_second": 356.217,
|
| 1408 |
+
"eval_sts-label_steps_per_second": 14.249,
|
| 1409 |
+
"step": 2240
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.3510421564018179,
|
| 1413 |
+
"eval_qnli-contrastive_loss": 0.11220741271972656,
|
| 1414 |
+
"eval_qnli-contrastive_runtime": 0.3614,
|
| 1415 |
+
"eval_qnli-contrastive_samples_per_second": 276.705,
|
| 1416 |
+
"eval_qnli-contrastive_steps_per_second": 11.068,
|
| 1417 |
+
"step": 2240
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.3510421564018179,
|
| 1421 |
+
"eval_scitail-pairs-qa_loss": 0.06635177880525589,
|
| 1422 |
+
"eval_scitail-pairs-qa_runtime": 0.8881,
|
| 1423 |
+
"eval_scitail-pairs-qa_samples_per_second": 112.594,
|
| 1424 |
+
"eval_scitail-pairs-qa_steps_per_second": 4.504,
|
| 1425 |
+
"step": 2240
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 0.3510421564018179,
|
| 1429 |
+
"eval_scitail-pairs-pos_loss": 0.5765587687492371,
|
| 1430 |
+
"eval_scitail-pairs-pos_runtime": 1.3496,
|
| 1431 |
+
"eval_scitail-pairs-pos_samples_per_second": 74.097,
|
| 1432 |
+
"eval_scitail-pairs-pos_steps_per_second": 2.964,
|
| 1433 |
+
"step": 2240
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 0.3510421564018179,
|
| 1437 |
+
"eval_xsum-pairs_loss": 0.2595808804035187,
|
| 1438 |
+
"eval_xsum-pairs_runtime": 0.9377,
|
| 1439 |
+
"eval_xsum-pairs_samples_per_second": 106.641,
|
| 1440 |
+
"eval_xsum-pairs_steps_per_second": 4.266,
|
| 1441 |
+
"step": 2240
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 0.3510421564018179,
|
| 1445 |
+
"eval_compression-pairs_loss": 0.0918564721941948,
|
| 1446 |
+
"eval_compression-pairs_runtime": 0.2755,
|
| 1447 |
+
"eval_compression-pairs_samples_per_second": 363.032,
|
| 1448 |
+
"eval_compression-pairs_steps_per_second": 14.521,
|
| 1449 |
+
"step": 2240
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 0.3510421564018179,
|
| 1453 |
+
"eval_sciq_pairs_loss": 0.284303218126297,
|
| 1454 |
+
"eval_sciq_pairs_runtime": 4.1289,
|
| 1455 |
+
"eval_sciq_pairs_samples_per_second": 24.22,
|
| 1456 |
+
"eval_sciq_pairs_steps_per_second": 0.969,
|
| 1457 |
+
"step": 2240
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 0.3510421564018179,
|
| 1461 |
+
"eval_qasc_pairs_loss": 0.19232892990112305,
|
| 1462 |
+
"eval_qasc_pairs_runtime": 1.0709,
|
| 1463 |
+
"eval_qasc_pairs_samples_per_second": 93.384,
|
| 1464 |
+
"eval_qasc_pairs_steps_per_second": 3.735,
|
| 1465 |
+
"step": 2240
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.3510421564018179,
|
| 1469 |
+
"eval_openbookqa_pairs_loss": 1.6234371662139893,
|
| 1470 |
+
"eval_openbookqa_pairs_runtime": 0.9558,
|
| 1471 |
+
"eval_openbookqa_pairs_samples_per_second": 104.62,
|
| 1472 |
+
"eval_openbookqa_pairs_steps_per_second": 4.185,
|
| 1473 |
+
"step": 2240
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.3510421564018179,
|
| 1477 |
+
"eval_msmarco_pairs_loss": 0.5325217247009277,
|
| 1478 |
+
"eval_msmarco_pairs_runtime": 2.0971,
|
| 1479 |
+
"eval_msmarco_pairs_samples_per_second": 47.685,
|
| 1480 |
+
"eval_msmarco_pairs_steps_per_second": 1.907,
|
| 1481 |
+
"step": 2240
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 0.3510421564018179,
|
| 1485 |
+
"eval_nq_pairs_loss": 0.2721095681190491,
|
| 1486 |
+
"eval_nq_pairs_runtime": 4.5393,
|
| 1487 |
+
"eval_nq_pairs_samples_per_second": 22.03,
|
| 1488 |
+
"eval_nq_pairs_steps_per_second": 0.881,
|
| 1489 |
+
"step": 2240
|
| 1490 |
+
},
|
| 1491 |
+
{
|
| 1492 |
+
"epoch": 0.3510421564018179,
|
| 1493 |
+
"eval_trivia_pairs_loss": 0.8544899821281433,
|
| 1494 |
+
"eval_trivia_pairs_runtime": 6.4668,
|
| 1495 |
+
"eval_trivia_pairs_samples_per_second": 15.464,
|
| 1496 |
+
"eval_trivia_pairs_steps_per_second": 0.619,
|
| 1497 |
+
"step": 2240
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 0.3510421564018179,
|
| 1501 |
+
"eval_quora_pairs_loss": 0.08441996574401855,
|
| 1502 |
+
"eval_quora_pairs_runtime": 0.6933,
|
| 1503 |
+
"eval_quora_pairs_samples_per_second": 144.233,
|
| 1504 |
+
"eval_quora_pairs_steps_per_second": 5.769,
|
| 1505 |
+
"step": 2240
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 0.3510421564018179,
|
| 1509 |
+
"eval_gooaq_pairs_loss": 0.5711588859558105,
|
| 1510 |
+
"eval_gooaq_pairs_runtime": 1.3941,
|
| 1511 |
+
"eval_gooaq_pairs_samples_per_second": 71.733,
|
| 1512 |
+
"eval_gooaq_pairs_steps_per_second": 2.869,
|
| 1513 |
+
"step": 2240
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 0.3510421564018179,
|
| 1517 |
+
"eval_mrpc_pairs_loss": 0.05093960464000702,
|
| 1518 |
+
"eval_mrpc_pairs_runtime": 0.2633,
|
| 1519 |
+
"eval_mrpc_pairs_samples_per_second": 379.777,
|
| 1520 |
+
"eval_mrpc_pairs_steps_per_second": 15.191,
|
| 1521 |
+
"step": 2240
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.3560570443504153,
|
| 1525 |
+
"grad_norm": 0.39178094267845154,
|
| 1526 |
+
"learning_rate": 2.6498646236636892e-05,
|
| 1527 |
+
"loss": 0.6805,
|
| 1528 |
+
"step": 2272
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.3610719322990127,
|
| 1532 |
+
"grad_norm": 7.91475248336792,
|
| 1533 |
+
"learning_rate": 2.6332915632292237e-05,
|
| 1534 |
+
"loss": 1.0451,
|
| 1535 |
+
"step": 2304
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.3660868202476101,
|
| 1539 |
+
"grad_norm": 31.54157066345215,
|
| 1540 |
+
"learning_rate": 2.616389702311641e-05,
|
| 1541 |
+
"loss": 1.0603,
|
| 1542 |
+
"step": 2336
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.37110170819620747,
|
| 1546 |
+
"grad_norm": 8.400779724121094,
|
| 1547 |
+
"learning_rate": 2.5991639446269964e-05,
|
| 1548 |
+
"loss": 0.8142,
|
| 1549 |
+
"step": 2368
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.3761165961448049,
|
| 1553 |
+
"grad_norm": 20.99441146850586,
|
| 1554 |
+
"learning_rate": 2.5816192878631166e-05,
|
| 1555 |
+
"loss": 1.7211,
|
| 1556 |
+
"step": 2400
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.3811314840934023,
|
| 1560 |
+
"grad_norm": 10.574430465698242,
|
| 1561 |
+
"learning_rate": 2.5637608222296237e-05,
|
| 1562 |
+
"loss": 0.7523,
|
| 1563 |
+
"step": 2432
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.3861463720419997,
|
| 1567 |
+
"grad_norm": 0.8941424489021301,
|
| 1568 |
+
"learning_rate": 2.5455937289811207e-05,
|
| 1569 |
+
"loss": 0.8053,
|
| 1570 |
+
"step": 2464
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.3911612599905971,
|
| 1574 |
+
"grad_norm": 1.9402281045913696,
|
| 1575 |
+
"learning_rate": 2.5271232789139587e-05,
|
| 1576 |
+
"loss": 0.8427,
|
| 1577 |
+
"step": 2496
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.3961761479391945,
|
| 1581 |
+
"grad_norm": 23.42873764038086,
|
| 1582 |
+
"learning_rate": 2.5083548308370296e-05,
|
| 1583 |
+
"loss": 0.8204,
|
| 1584 |
+
"step": 2528
|
| 1585 |
+
}
|
| 1586 |
+
],
|
| 1587 |
+
"logging_steps": 32,
|
| 1588 |
+
"max_steps": 12762,
|
| 1589 |
+
"num_input_tokens_seen": 0,
|
| 1590 |
+
"num_train_epochs": 2,
|
| 1591 |
+
"save_steps": 1277,
|
| 1592 |
+
"stateful_callbacks": {
|
| 1593 |
+
"TrainerControl": {
|
| 1594 |
+
"args": {
|
| 1595 |
+
"should_epoch_stop": false,
|
| 1596 |
+
"should_evaluate": false,
|
| 1597 |
+
"should_log": false,
|
| 1598 |
+
"should_save": true,
|
| 1599 |
+
"should_training_stop": false
|
| 1600 |
+
},
|
| 1601 |
+
"attributes": {}
|
| 1602 |
+
}
|
| 1603 |
+
},
|
| 1604 |
+
"total_flos": 0.0,
|
| 1605 |
+
"train_batch_size": 32,
|
| 1606 |
+
"trial_name": null,
|
| 1607 |
+
"trial_params": null
|
| 1608 |
+
}
|
checkpoint-2554/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:383c5bf6513da1dfbcd1294f0c8e85ce43118bc61e2de49d9b5d1e28eb653003
|
| 3 |
+
size 5624
|