Training in progress, step 3826, checkpoint
Browse files- checkpoint-3826/1_Pooling/config.json +10 -0
- checkpoint-3826/README.md +0 -0
- checkpoint-3826/added_tokens.json +3 -0
- checkpoint-3826/config.json +35 -0
- checkpoint-3826/config_sentence_transformers.json +10 -0
- checkpoint-3826/modules.json +14 -0
- checkpoint-3826/optimizer.pt +3 -0
- checkpoint-3826/pytorch_model.bin +3 -0
- checkpoint-3826/rng_state.pth +3 -0
- checkpoint-3826/scheduler.pt +3 -0
- checkpoint-3826/sentence_bert_config.json +4 -0
- checkpoint-3826/special_tokens_map.json +15 -0
- checkpoint-3826/spm.model +3 -0
- checkpoint-3826/tokenizer.json +0 -0
- checkpoint-3826/tokenizer_config.json +58 -0
- checkpoint-3826/trainer_state.json +1608 -0
- checkpoint-3826/training_args.bin +3 -0
checkpoint-3826/1_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-3826/README.md
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-3826/added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"[MASK]": 128000
|
| 3 |
+
}
|
checkpoint-3826/config.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "microsoft/deberta-v3-small",
|
| 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": "4.41.2",
|
| 33 |
+
"type_vocab_size": 0,
|
| 34 |
+
"vocab_size": 128100
|
| 35 |
+
}
|
checkpoint-3826/config_sentence_transformers.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"__version__": {
|
| 3 |
+
"sentence_transformers": "3.0.1",
|
| 4 |
+
"transformers": "4.41.2",
|
| 5 |
+
"pytorch": "2.3.0+cu121"
|
| 6 |
+
},
|
| 7 |
+
"prompts": {},
|
| 8 |
+
"default_prompt_name": null,
|
| 9 |
+
"similarity_fn_name": null
|
| 10 |
+
}
|
checkpoint-3826/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-3826/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b89fef9092f3a36aca2b966210474265e5aa8960faf07aa7bd9831167eb7f9c7
|
| 3 |
+
size 1130520122
|
checkpoint-3826/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:65f9fa0b83ce5ff8450c20c9194ed05a623b9a9fc236f5c72a815112ed128557
|
| 3 |
+
size 565251810
|
checkpoint-3826/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e3541e9b0e99aaa7497aabeb4b6ec201c30ddbb32d3aeb84028e7be5da2fd3a9
|
| 3 |
+
size 14244
|
checkpoint-3826/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3af235f62fb2726407814a285f8b4de2655388716473ed2b4cd4bf7950425cc0
|
| 3 |
+
size 1064
|
checkpoint-3826/sentence_bert_config.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"max_seq_length": 512,
|
| 3 |
+
"do_lower_case": false
|
| 4 |
+
}
|
checkpoint-3826/special_tokens_map.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "[CLS]",
|
| 3 |
+
"cls_token": "[CLS]",
|
| 4 |
+
"eos_token": "[SEP]",
|
| 5 |
+
"mask_token": "[MASK]",
|
| 6 |
+
"pad_token": "[PAD]",
|
| 7 |
+
"sep_token": "[SEP]",
|
| 8 |
+
"unk_token": {
|
| 9 |
+
"content": "[UNK]",
|
| 10 |
+
"lstrip": false,
|
| 11 |
+
"normalized": true,
|
| 12 |
+
"rstrip": false,
|
| 13 |
+
"single_word": false
|
| 14 |
+
}
|
| 15 |
+
}
|
checkpoint-3826/spm.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c679fbf93643d19aab7ee10c0b99e460bdbc02fedf34b92b05af343b4af586fd
|
| 3 |
+
size 2464616
|
checkpoint-3826/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-3826/tokenizer_config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "[CLS]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "[SEP]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "[UNK]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": true,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"128000": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"bos_token": "[CLS]",
|
| 45 |
+
"clean_up_tokenization_spaces": true,
|
| 46 |
+
"cls_token": "[CLS]",
|
| 47 |
+
"do_lower_case": false,
|
| 48 |
+
"eos_token": "[SEP]",
|
| 49 |
+
"mask_token": "[MASK]",
|
| 50 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 51 |
+
"pad_token": "[PAD]",
|
| 52 |
+
"sep_token": "[SEP]",
|
| 53 |
+
"sp_model_kwargs": {},
|
| 54 |
+
"split_by_punct": false,
|
| 55 |
+
"tokenizer_class": "DebertaV2Tokenizer",
|
| 56 |
+
"unk_token": "[UNK]",
|
| 57 |
+
"vocab_type": "spm"
|
| 58 |
+
}
|
checkpoint-3826/trainer_state.json
ADDED
|
@@ -0,0 +1,1608 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.6000627352572145,
|
| 5 |
+
"eval_steps": 479,
|
| 6 |
+
"global_step": 3826,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0075282308657465494,
|
| 13 |
+
"grad_norm": 87.12931823730469,
|
| 14 |
+
"learning_rate": 3.147218736930155e-07,
|
| 15 |
+
"loss": 12.3074,
|
| 16 |
+
"step": 48
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.015056461731493099,
|
| 20 |
+
"grad_norm": 64.15751647949219,
|
| 21 |
+
"learning_rate": 6.660393140945211e-07,
|
| 22 |
+
"loss": 15.7221,
|
| 23 |
+
"step": 96
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.02258469259723965,
|
| 27 |
+
"grad_norm": 81.39651489257812,
|
| 28 |
+
"learning_rate": 1.0173567544960265e-06,
|
| 29 |
+
"loss": 10.8027,
|
| 30 |
+
"step": 144
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.030112923462986198,
|
| 34 |
+
"grad_norm": 22.048904418945312,
|
| 35 |
+
"learning_rate": 1.3686741948975323e-06,
|
| 36 |
+
"loss": 8.9559,
|
| 37 |
+
"step": 192
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.037641154328732745,
|
| 41 |
+
"grad_norm": 11.542724609375,
|
| 42 |
+
"learning_rate": 1.7126725219573398e-06,
|
| 43 |
+
"loss": 8.8511,
|
| 44 |
+
"step": 240
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.0451693851944793,
|
| 48 |
+
"grad_norm": 10.625059127807617,
|
| 49 |
+
"learning_rate": 2.063989962358846e-06,
|
| 50 |
+
"loss": 9.3478,
|
| 51 |
+
"step": 288
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.05269761606022585,
|
| 55 |
+
"grad_norm": 14.21434211730957,
|
| 56 |
+
"learning_rate": 2.415307402760351e-06,
|
| 57 |
+
"loss": 8.8892,
|
| 58 |
+
"step": 336
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.060225846925972396,
|
| 62 |
+
"grad_norm": 13.216053009033203,
|
| 63 |
+
"learning_rate": 2.7666248431618565e-06,
|
| 64 |
+
"loss": 8.3008,
|
| 65 |
+
"step": 384
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.06775407779171895,
|
| 69 |
+
"grad_norm": 22.503334045410156,
|
| 70 |
+
"learning_rate": 3.117942283563362e-06,
|
| 71 |
+
"loss": 7.3455,
|
| 72 |
+
"step": 432
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.07512547051442911,
|
| 76 |
+
"eval_nli-pairs_loss": 6.591032028198242,
|
| 77 |
+
"eval_nli-pairs_runtime": 4.3469,
|
| 78 |
+
"eval_nli-pairs_samples_per_second": 46.01,
|
| 79 |
+
"eval_nli-pairs_steps_per_second": 1.15,
|
| 80 |
+
"eval_sts-test_pearson_cosine": 0.39488461174644296,
|
| 81 |
+
"eval_sts-test_pearson_dot": 0.15593446481859455,
|
| 82 |
+
"eval_sts-test_pearson_euclidean": 0.39975070029693277,
|
| 83 |
+
"eval_sts-test_pearson_manhattan": 0.4314268556737928,
|
| 84 |
+
"eval_sts-test_pearson_max": 0.4314268556737928,
|
| 85 |
+
"eval_sts-test_spearman_cosine": 0.3997824055251076,
|
| 86 |
+
"eval_sts-test_spearman_dot": 0.14324216739430146,
|
| 87 |
+
"eval_sts-test_spearman_euclidean": 0.40262274612650517,
|
| 88 |
+
"eval_sts-test_spearman_manhattan": 0.42925492969387746,
|
| 89 |
+
"eval_sts-test_spearman_max": 0.42925492969387746,
|
| 90 |
+
"step": 479
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.07512547051442911,
|
| 94 |
+
"eval_vitaminc-pairs_loss": 6.247874736785889,
|
| 95 |
+
"eval_vitaminc-pairs_runtime": 1.4382,
|
| 96 |
+
"eval_vitaminc-pairs_samples_per_second": 115.426,
|
| 97 |
+
"eval_vitaminc-pairs_steps_per_second": 2.781,
|
| 98 |
+
"step": 479
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.07512547051442911,
|
| 102 |
+
"eval_sts-label_loss": 3.371708869934082,
|
| 103 |
+
"eval_sts-label_runtime": 0.3925,
|
| 104 |
+
"eval_sts-label_samples_per_second": 509.603,
|
| 105 |
+
"eval_sts-label_steps_per_second": 12.74,
|
| 106 |
+
"step": 479
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.07512547051442911,
|
| 110 |
+
"eval_qnli-contrastive_loss": 3.5311310291290283,
|
| 111 |
+
"eval_qnli-contrastive_runtime": 0.2814,
|
| 112 |
+
"eval_qnli-contrastive_samples_per_second": 710.798,
|
| 113 |
+
"eval_qnli-contrastive_steps_per_second": 17.77,
|
| 114 |
+
"step": 479
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.07512547051442911,
|
| 118 |
+
"eval_scitail-pairs-qa_loss": 5.4017333984375,
|
| 119 |
+
"eval_scitail-pairs-qa_runtime": 1.1087,
|
| 120 |
+
"eval_scitail-pairs-qa_samples_per_second": 180.387,
|
| 121 |
+
"eval_scitail-pairs-qa_steps_per_second": 4.51,
|
| 122 |
+
"step": 479
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.07512547051442911,
|
| 126 |
+
"eval_scitail-pairs-pos_loss": 4.29502534866333,
|
| 127 |
+
"eval_scitail-pairs-pos_runtime": 2.4269,
|
| 128 |
+
"eval_scitail-pairs-pos_samples_per_second": 82.41,
|
| 129 |
+
"eval_scitail-pairs-pos_steps_per_second": 2.06,
|
| 130 |
+
"step": 479
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.07512547051442911,
|
| 134 |
+
"eval_xsum-pairs_loss": 3.540722370147705,
|
| 135 |
+
"eval_xsum-pairs_runtime": 1.0447,
|
| 136 |
+
"eval_xsum-pairs_samples_per_second": 191.444,
|
| 137 |
+
"eval_xsum-pairs_steps_per_second": 4.786,
|
| 138 |
+
"step": 479
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 0.07512547051442911,
|
| 142 |
+
"eval_compression-pairs_loss": 2.8060033321380615,
|
| 143 |
+
"eval_compression-pairs_runtime": 0.2399,
|
| 144 |
+
"eval_compression-pairs_samples_per_second": 833.621,
|
| 145 |
+
"eval_compression-pairs_steps_per_second": 20.841,
|
| 146 |
+
"step": 479
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.07512547051442911,
|
| 150 |
+
"eval_sciq_pairs_loss": 10.621454238891602,
|
| 151 |
+
"eval_sciq_pairs_runtime": 9.0638,
|
| 152 |
+
"eval_sciq_pairs_samples_per_second": 22.066,
|
| 153 |
+
"eval_sciq_pairs_steps_per_second": 0.552,
|
| 154 |
+
"step": 479
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.07512547051442911,
|
| 158 |
+
"eval_qasc_pairs_loss": 7.7197771072387695,
|
| 159 |
+
"eval_qasc_pairs_runtime": 1.2078,
|
| 160 |
+
"eval_qasc_pairs_samples_per_second": 165.595,
|
| 161 |
+
"eval_qasc_pairs_steps_per_second": 4.14,
|
| 162 |
+
"step": 479
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 0.07512547051442911,
|
| 166 |
+
"eval_openbookqa_pairs_loss": 7.620975494384766,
|
| 167 |
+
"eval_openbookqa_pairs_runtime": 1.053,
|
| 168 |
+
"eval_openbookqa_pairs_samples_per_second": 189.941,
|
| 169 |
+
"eval_openbookqa_pairs_steps_per_second": 4.749,
|
| 170 |
+
"step": 479
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.07512547051442911,
|
| 174 |
+
"eval_msmarco_pairs_loss": 8.353594779968262,
|
| 175 |
+
"eval_msmarco_pairs_runtime": 2.5338,
|
| 176 |
+
"eval_msmarco_pairs_samples_per_second": 78.932,
|
| 177 |
+
"eval_msmarco_pairs_steps_per_second": 1.973,
|
| 178 |
+
"step": 479
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.07512547051442911,
|
| 182 |
+
"eval_nq_pairs_loss": 7.995354652404785,
|
| 183 |
+
"eval_nq_pairs_runtime": 5.7107,
|
| 184 |
+
"eval_nq_pairs_samples_per_second": 35.022,
|
| 185 |
+
"eval_nq_pairs_steps_per_second": 0.876,
|
| 186 |
+
"step": 479
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 0.07512547051442911,
|
| 190 |
+
"eval_trivia_pairs_loss": 8.177907943725586,
|
| 191 |
+
"eval_trivia_pairs_runtime": 9.1824,
|
| 192 |
+
"eval_trivia_pairs_samples_per_second": 21.781,
|
| 193 |
+
"eval_trivia_pairs_steps_per_second": 0.545,
|
| 194 |
+
"step": 479
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.07512547051442911,
|
| 198 |
+
"eval_quora_pairs_loss": 1.1983369588851929,
|
| 199 |
+
"eval_quora_pairs_runtime": 0.659,
|
| 200 |
+
"eval_quora_pairs_samples_per_second": 303.48,
|
| 201 |
+
"eval_quora_pairs_steps_per_second": 7.587,
|
| 202 |
+
"step": 479
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.07512547051442911,
|
| 206 |
+
"eval_gooaq_pairs_loss": 7.4573974609375,
|
| 207 |
+
"eval_gooaq_pairs_runtime": 1.5917,
|
| 208 |
+
"eval_gooaq_pairs_samples_per_second": 125.65,
|
| 209 |
+
"eval_gooaq_pairs_steps_per_second": 3.141,
|
| 210 |
+
"step": 479
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 0.07512547051442911,
|
| 214 |
+
"eval_mrpc_pairs_loss": 2.1152825355529785,
|
| 215 |
+
"eval_mrpc_pairs_runtime": 0.241,
|
| 216 |
+
"eval_mrpc_pairs_samples_per_second": 829.751,
|
| 217 |
+
"eval_mrpc_pairs_steps_per_second": 20.744,
|
| 218 |
+
"step": 479
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.07528230865746549,
|
| 222 |
+
"grad_norm": 17.753456115722656,
|
| 223 |
+
"learning_rate": 3.469259723964868e-06,
|
| 224 |
+
"loss": 8.0369,
|
| 225 |
+
"step": 480
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.08281053952321205,
|
| 229 |
+
"grad_norm": 12.012594223022461,
|
| 230 |
+
"learning_rate": 3.820577164366374e-06,
|
| 231 |
+
"loss": 6.2732,
|
| 232 |
+
"step": 528
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.0903387703889586,
|
| 236 |
+
"grad_norm": 34.92698287963867,
|
| 237 |
+
"learning_rate": 4.1718946047678796e-06,
|
| 238 |
+
"loss": 7.8529,
|
| 239 |
+
"step": 576
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 0.09786700125470514,
|
| 243 |
+
"grad_norm": 23.564632415771484,
|
| 244 |
+
"learning_rate": 4.523212045169385e-06,
|
| 245 |
+
"loss": 5.8643,
|
| 246 |
+
"step": 624
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 0.1053952321204517,
|
| 250 |
+
"grad_norm": 22.126293182373047,
|
| 251 |
+
"learning_rate": 4.874529485570891e-06,
|
| 252 |
+
"loss": 6.3179,
|
| 253 |
+
"step": 672
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 0.11292346298619825,
|
| 257 |
+
"grad_norm": 25.067686080932617,
|
| 258 |
+
"learning_rate": 5.225846925972396e-06,
|
| 259 |
+
"loss": 6.1175,
|
| 260 |
+
"step": 720
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.12045169385194479,
|
| 264 |
+
"grad_norm": 29.170730590820312,
|
| 265 |
+
"learning_rate": 5.577164366373902e-06,
|
| 266 |
+
"loss": 5.2392,
|
| 267 |
+
"step": 768
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.12797992471769135,
|
| 271 |
+
"grad_norm": 29.377540588378906,
|
| 272 |
+
"learning_rate": 5.928481806775407e-06,
|
| 273 |
+
"loss": 5.8324,
|
| 274 |
+
"step": 816
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 0.1355081555834379,
|
| 278 |
+
"grad_norm": 33.512088775634766,
|
| 279 |
+
"learning_rate": 6.279799247176913e-06,
|
| 280 |
+
"loss": 5.1523,
|
| 281 |
+
"step": 864
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.14303638644918445,
|
| 285 |
+
"grad_norm": 32.54931640625,
|
| 286 |
+
"learning_rate": 6.6311166875784185e-06,
|
| 287 |
+
"loss": 6.0303,
|
| 288 |
+
"step": 912
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.15025094102885822,
|
| 292 |
+
"eval_nli-pairs_loss": 4.317643165588379,
|
| 293 |
+
"eval_nli-pairs_runtime": 4.0158,
|
| 294 |
+
"eval_nli-pairs_samples_per_second": 49.803,
|
| 295 |
+
"eval_nli-pairs_steps_per_second": 1.245,
|
| 296 |
+
"eval_sts-test_pearson_cosine": 0.6751726661173544,
|
| 297 |
+
"eval_sts-test_pearson_dot": 0.5308173325280101,
|
| 298 |
+
"eval_sts-test_pearson_euclidean": 0.6660519042507951,
|
| 299 |
+
"eval_sts-test_pearson_manhattan": 0.6723411683739887,
|
| 300 |
+
"eval_sts-test_pearson_max": 0.6751726661173544,
|
| 301 |
+
"eval_sts-test_spearman_cosine": 0.6458966208807124,
|
| 302 |
+
"eval_sts-test_spearman_dot": 0.5040208096497271,
|
| 303 |
+
"eval_sts-test_spearman_euclidean": 0.6447205374312966,
|
| 304 |
+
"eval_sts-test_spearman_manhattan": 0.6498560301461127,
|
| 305 |
+
"eval_sts-test_spearman_max": 0.6498560301461127,
|
| 306 |
+
"step": 958
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 0.15025094102885822,
|
| 310 |
+
"eval_vitaminc-pairs_loss": 6.306981563568115,
|
| 311 |
+
"eval_vitaminc-pairs_runtime": 1.4737,
|
| 312 |
+
"eval_vitaminc-pairs_samples_per_second": 112.643,
|
| 313 |
+
"eval_vitaminc-pairs_steps_per_second": 2.714,
|
| 314 |
+
"step": 958
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.15025094102885822,
|
| 318 |
+
"eval_sts-label_loss": 3.7677345275878906,
|
| 319 |
+
"eval_sts-label_runtime": 0.4136,
|
| 320 |
+
"eval_sts-label_samples_per_second": 483.59,
|
| 321 |
+
"eval_sts-label_steps_per_second": 12.09,
|
| 322 |
+
"step": 958
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 0.15025094102885822,
|
| 326 |
+
"eval_qnli-contrastive_loss": 2.8633975982666016,
|
| 327 |
+
"eval_qnli-contrastive_runtime": 0.2816,
|
| 328 |
+
"eval_qnli-contrastive_samples_per_second": 710.123,
|
| 329 |
+
"eval_qnli-contrastive_steps_per_second": 17.753,
|
| 330 |
+
"step": 958
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 0.15025094102885822,
|
| 334 |
+
"eval_scitail-pairs-qa_loss": 1.4301409721374512,
|
| 335 |
+
"eval_scitail-pairs-qa_runtime": 1.0526,
|
| 336 |
+
"eval_scitail-pairs-qa_samples_per_second": 190.003,
|
| 337 |
+
"eval_scitail-pairs-qa_steps_per_second": 4.75,
|
| 338 |
+
"step": 958
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.15025094102885822,
|
| 342 |
+
"eval_scitail-pairs-pos_loss": 2.468087673187256,
|
| 343 |
+
"eval_scitail-pairs-pos_runtime": 2.3275,
|
| 344 |
+
"eval_scitail-pairs-pos_samples_per_second": 85.928,
|
| 345 |
+
"eval_scitail-pairs-pos_steps_per_second": 2.148,
|
| 346 |
+
"step": 958
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.15025094102885822,
|
| 350 |
+
"eval_xsum-pairs_loss": 2.153658151626587,
|
| 351 |
+
"eval_xsum-pairs_runtime": 1.0409,
|
| 352 |
+
"eval_xsum-pairs_samples_per_second": 192.151,
|
| 353 |
+
"eval_xsum-pairs_steps_per_second": 4.804,
|
| 354 |
+
"step": 958
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.15025094102885822,
|
| 358 |
+
"eval_compression-pairs_loss": 1.6288033723831177,
|
| 359 |
+
"eval_compression-pairs_runtime": 0.2383,
|
| 360 |
+
"eval_compression-pairs_samples_per_second": 839.253,
|
| 361 |
+
"eval_compression-pairs_steps_per_second": 20.981,
|
| 362 |
+
"step": 958
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.15025094102885822,
|
| 366 |
+
"eval_sciq_pairs_loss": 9.78779411315918,
|
| 367 |
+
"eval_sciq_pairs_runtime": 9.0233,
|
| 368 |
+
"eval_sciq_pairs_samples_per_second": 22.165,
|
| 369 |
+
"eval_sciq_pairs_steps_per_second": 0.554,
|
| 370 |
+
"step": 958
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 0.15025094102885822,
|
| 374 |
+
"eval_qasc_pairs_loss": 3.7814972400665283,
|
| 375 |
+
"eval_qasc_pairs_runtime": 1.2108,
|
| 376 |
+
"eval_qasc_pairs_samples_per_second": 165.179,
|
| 377 |
+
"eval_qasc_pairs_steps_per_second": 4.129,
|
| 378 |
+
"step": 958
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 0.15025094102885822,
|
| 382 |
+
"eval_openbookqa_pairs_loss": 5.005772590637207,
|
| 383 |
+
"eval_openbookqa_pairs_runtime": 1.0415,
|
| 384 |
+
"eval_openbookqa_pairs_samples_per_second": 192.037,
|
| 385 |
+
"eval_openbookqa_pairs_steps_per_second": 4.801,
|
| 386 |
+
"step": 958
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.15025094102885822,
|
| 390 |
+
"eval_msmarco_pairs_loss": 4.574879169464111,
|
| 391 |
+
"eval_msmarco_pairs_runtime": 2.527,
|
| 392 |
+
"eval_msmarco_pairs_samples_per_second": 79.146,
|
| 393 |
+
"eval_msmarco_pairs_steps_per_second": 1.979,
|
| 394 |
+
"step": 958
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.15025094102885822,
|
| 398 |
+
"eval_nq_pairs_loss": 5.281248569488525,
|
| 399 |
+
"eval_nq_pairs_runtime": 5.6503,
|
| 400 |
+
"eval_nq_pairs_samples_per_second": 35.397,
|
| 401 |
+
"eval_nq_pairs_steps_per_second": 0.885,
|
| 402 |
+
"step": 958
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.15025094102885822,
|
| 406 |
+
"eval_trivia_pairs_loss": 4.913428783416748,
|
| 407 |
+
"eval_trivia_pairs_runtime": 9.0564,
|
| 408 |
+
"eval_trivia_pairs_samples_per_second": 22.084,
|
| 409 |
+
"eval_trivia_pairs_steps_per_second": 0.552,
|
| 410 |
+
"step": 958
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.15025094102885822,
|
| 414 |
+
"eval_quora_pairs_loss": 0.9212128520011902,
|
| 415 |
+
"eval_quora_pairs_runtime": 0.6293,
|
| 416 |
+
"eval_quora_pairs_samples_per_second": 317.839,
|
| 417 |
+
"eval_quora_pairs_steps_per_second": 7.946,
|
| 418 |
+
"step": 958
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 0.15025094102885822,
|
| 422 |
+
"eval_gooaq_pairs_loss": 3.932173490524292,
|
| 423 |
+
"eval_gooaq_pairs_runtime": 1.6066,
|
| 424 |
+
"eval_gooaq_pairs_samples_per_second": 124.483,
|
| 425 |
+
"eval_gooaq_pairs_steps_per_second": 3.112,
|
| 426 |
+
"step": 958
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 0.15025094102885822,
|
| 430 |
+
"eval_mrpc_pairs_loss": 1.0853501558303833,
|
| 431 |
+
"eval_mrpc_pairs_runtime": 0.2402,
|
| 432 |
+
"eval_mrpc_pairs_samples_per_second": 832.56,
|
| 433 |
+
"eval_mrpc_pairs_steps_per_second": 20.814,
|
| 434 |
+
"step": 958
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.15056461731493098,
|
| 438 |
+
"grad_norm": 28.37123680114746,
|
| 439 |
+
"learning_rate": 6.982434127979924e-06,
|
| 440 |
+
"loss": 5.7748,
|
| 441 |
+
"step": 960
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 0.15809284818067754,
|
| 445 |
+
"grad_norm": 26.708221435546875,
|
| 446 |
+
"learning_rate": 7.33375156838143e-06,
|
| 447 |
+
"loss": 4.8728,
|
| 448 |
+
"step": 1008
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 0.1656210790464241,
|
| 452 |
+
"grad_norm": 26.786447525024414,
|
| 453 |
+
"learning_rate": 7.685069008782934e-06,
|
| 454 |
+
"loss": 4.7375,
|
| 455 |
+
"step": 1056
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 0.17314930991217065,
|
| 459 |
+
"grad_norm": 26.215879440307617,
|
| 460 |
+
"learning_rate": 8.03638644918444e-06,
|
| 461 |
+
"loss": 4.6766,
|
| 462 |
+
"step": 1104
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 0.1806775407779172,
|
| 466 |
+
"grad_norm": 35.618831634521484,
|
| 467 |
+
"learning_rate": 8.387703889585947e-06,
|
| 468 |
+
"loss": 4.3209,
|
| 469 |
+
"step": 1152
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 0.18820577164366373,
|
| 473 |
+
"grad_norm": 37.166072845458984,
|
| 474 |
+
"learning_rate": 8.739021329987453e-06,
|
| 475 |
+
"loss": 3.7761,
|
| 476 |
+
"step": 1200
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.19573400250941028,
|
| 480 |
+
"grad_norm": 35.78367233276367,
|
| 481 |
+
"learning_rate": 9.090338770388957e-06,
|
| 482 |
+
"loss": 4.2161,
|
| 483 |
+
"step": 1248
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 0.20326223337515684,
|
| 487 |
+
"grad_norm": 36.299678802490234,
|
| 488 |
+
"learning_rate": 9.441656210790464e-06,
|
| 489 |
+
"loss": 4.9089,
|
| 490 |
+
"step": 1296
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 0.2107904642409034,
|
| 494 |
+
"grad_norm": 18.610933303833008,
|
| 495 |
+
"learning_rate": 9.792973651191968e-06,
|
| 496 |
+
"loss": 4.3406,
|
| 497 |
+
"step": 1344
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 0.21831869510664995,
|
| 501 |
+
"grad_norm": 9.592538833618164,
|
| 502 |
+
"learning_rate": 1.0144291091593475e-05,
|
| 503 |
+
"loss": 3.5664,
|
| 504 |
+
"step": 1392
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.22537641154328733,
|
| 508 |
+
"eval_nli-pairs_loss": 3.2245519161224365,
|
| 509 |
+
"eval_nli-pairs_runtime": 4.0436,
|
| 510 |
+
"eval_nli-pairs_samples_per_second": 49.461,
|
| 511 |
+
"eval_nli-pairs_steps_per_second": 1.237,
|
| 512 |
+
"eval_sts-test_pearson_cosine": 0.7002978854888552,
|
| 513 |
+
"eval_sts-test_pearson_dot": 0.5685392445320393,
|
| 514 |
+
"eval_sts-test_pearson_euclidean": 0.6963744527231541,
|
| 515 |
+
"eval_sts-test_pearson_manhattan": 0.7050517306003169,
|
| 516 |
+
"eval_sts-test_pearson_max": 0.7050517306003169,
|
| 517 |
+
"eval_sts-test_spearman_cosine": 0.6718756239728468,
|
| 518 |
+
"eval_sts-test_spearman_dot": 0.5416448961602434,
|
| 519 |
+
"eval_sts-test_spearman_euclidean": 0.6742379556154348,
|
| 520 |
+
"eval_sts-test_spearman_manhattan": 0.6824201536078427,
|
| 521 |
+
"eval_sts-test_spearman_max": 0.6824201536078427,
|
| 522 |
+
"step": 1437
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 0.22537641154328733,
|
| 526 |
+
"eval_vitaminc-pairs_loss": 6.715206623077393,
|
| 527 |
+
"eval_vitaminc-pairs_runtime": 1.4251,
|
| 528 |
+
"eval_vitaminc-pairs_samples_per_second": 116.481,
|
| 529 |
+
"eval_vitaminc-pairs_steps_per_second": 2.807,
|
| 530 |
+
"step": 1437
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 0.22537641154328733,
|
| 534 |
+
"eval_sts-label_loss": 4.016364097595215,
|
| 535 |
+
"eval_sts-label_runtime": 0.4049,
|
| 536 |
+
"eval_sts-label_samples_per_second": 493.95,
|
| 537 |
+
"eval_sts-label_steps_per_second": 12.349,
|
| 538 |
+
"step": 1437
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 0.22537641154328733,
|
| 542 |
+
"eval_qnli-contrastive_loss": 1.999517560005188,
|
| 543 |
+
"eval_qnli-contrastive_runtime": 0.2804,
|
| 544 |
+
"eval_qnli-contrastive_samples_per_second": 713.282,
|
| 545 |
+
"eval_qnli-contrastive_steps_per_second": 17.832,
|
| 546 |
+
"step": 1437
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 0.22537641154328733,
|
| 550 |
+
"eval_scitail-pairs-qa_loss": 1.0403239727020264,
|
| 551 |
+
"eval_scitail-pairs-qa_runtime": 1.0483,
|
| 552 |
+
"eval_scitail-pairs-qa_samples_per_second": 190.793,
|
| 553 |
+
"eval_scitail-pairs-qa_steps_per_second": 4.77,
|
| 554 |
+
"step": 1437
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.22537641154328733,
|
| 558 |
+
"eval_scitail-pairs-pos_loss": 1.9232473373413086,
|
| 559 |
+
"eval_scitail-pairs-pos_runtime": 2.3447,
|
| 560 |
+
"eval_scitail-pairs-pos_samples_per_second": 85.298,
|
| 561 |
+
"eval_scitail-pairs-pos_steps_per_second": 2.132,
|
| 562 |
+
"step": 1437
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.22537641154328733,
|
| 566 |
+
"eval_xsum-pairs_loss": 1.6821197271347046,
|
| 567 |
+
"eval_xsum-pairs_runtime": 1.0422,
|
| 568 |
+
"eval_xsum-pairs_samples_per_second": 191.901,
|
| 569 |
+
"eval_xsum-pairs_steps_per_second": 4.798,
|
| 570 |
+
"step": 1437
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.22537641154328733,
|
| 574 |
+
"eval_compression-pairs_loss": 1.1713249683380127,
|
| 575 |
+
"eval_compression-pairs_runtime": 0.2392,
|
| 576 |
+
"eval_compression-pairs_samples_per_second": 836.05,
|
| 577 |
+
"eval_compression-pairs_steps_per_second": 20.901,
|
| 578 |
+
"step": 1437
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.22537641154328733,
|
| 582 |
+
"eval_sciq_pairs_loss": 9.443825721740723,
|
| 583 |
+
"eval_sciq_pairs_runtime": 8.9916,
|
| 584 |
+
"eval_sciq_pairs_samples_per_second": 22.243,
|
| 585 |
+
"eval_sciq_pairs_steps_per_second": 0.556,
|
| 586 |
+
"step": 1437
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 0.22537641154328733,
|
| 590 |
+
"eval_qasc_pairs_loss": 2.9044029712677,
|
| 591 |
+
"eval_qasc_pairs_runtime": 1.2182,
|
| 592 |
+
"eval_qasc_pairs_samples_per_second": 164.182,
|
| 593 |
+
"eval_qasc_pairs_steps_per_second": 4.105,
|
| 594 |
+
"step": 1437
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 0.22537641154328733,
|
| 598 |
+
"eval_openbookqa_pairs_loss": 4.360418796539307,
|
| 599 |
+
"eval_openbookqa_pairs_runtime": 1.0522,
|
| 600 |
+
"eval_openbookqa_pairs_samples_per_second": 190.077,
|
| 601 |
+
"eval_openbookqa_pairs_steps_per_second": 4.752,
|
| 602 |
+
"step": 1437
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.22537641154328733,
|
| 606 |
+
"eval_msmarco_pairs_loss": 3.516049861907959,
|
| 607 |
+
"eval_msmarco_pairs_runtime": 2.5595,
|
| 608 |
+
"eval_msmarco_pairs_samples_per_second": 78.139,
|
| 609 |
+
"eval_msmarco_pairs_steps_per_second": 1.953,
|
| 610 |
+
"step": 1437
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 0.22537641154328733,
|
| 614 |
+
"eval_nq_pairs_loss": 4.016308784484863,
|
| 615 |
+
"eval_nq_pairs_runtime": 5.6561,
|
| 616 |
+
"eval_nq_pairs_samples_per_second": 35.36,
|
| 617 |
+
"eval_nq_pairs_steps_per_second": 0.884,
|
| 618 |
+
"step": 1437
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.22537641154328733,
|
| 622 |
+
"eval_trivia_pairs_loss": 3.781872272491455,
|
| 623 |
+
"eval_trivia_pairs_runtime": 9.0801,
|
| 624 |
+
"eval_trivia_pairs_samples_per_second": 22.026,
|
| 625 |
+
"eval_trivia_pairs_steps_per_second": 0.551,
|
| 626 |
+
"step": 1437
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.22537641154328733,
|
| 630 |
+
"eval_quora_pairs_loss": 0.8747495412826538,
|
| 631 |
+
"eval_quora_pairs_runtime": 0.6229,
|
| 632 |
+
"eval_quora_pairs_samples_per_second": 321.076,
|
| 633 |
+
"eval_quora_pairs_steps_per_second": 8.027,
|
| 634 |
+
"step": 1437
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 0.22537641154328733,
|
| 638 |
+
"eval_gooaq_pairs_loss": 3.0769765377044678,
|
| 639 |
+
"eval_gooaq_pairs_runtime": 1.552,
|
| 640 |
+
"eval_gooaq_pairs_samples_per_second": 128.863,
|
| 641 |
+
"eval_gooaq_pairs_steps_per_second": 3.222,
|
| 642 |
+
"step": 1437
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 0.22537641154328733,
|
| 646 |
+
"eval_mrpc_pairs_loss": 0.7370794415473938,
|
| 647 |
+
"eval_mrpc_pairs_runtime": 0.2401,
|
| 648 |
+
"eval_mrpc_pairs_samples_per_second": 832.867,
|
| 649 |
+
"eval_mrpc_pairs_steps_per_second": 20.822,
|
| 650 |
+
"step": 1437
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 0.2258469259723965,
|
| 654 |
+
"grad_norm": 29.73522186279297,
|
| 655 |
+
"learning_rate": 1.049560853199498e-05,
|
| 656 |
+
"loss": 4.7194,
|
| 657 |
+
"step": 1440
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 0.23337515683814303,
|
| 661 |
+
"grad_norm": 30.467117309570312,
|
| 662 |
+
"learning_rate": 1.0846925972396486e-05,
|
| 663 |
+
"loss": 3.6345,
|
| 664 |
+
"step": 1488
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 0.24090338770388958,
|
| 668 |
+
"grad_norm": 24.454021453857422,
|
| 669 |
+
"learning_rate": 1.1198243412797992e-05,
|
| 670 |
+
"loss": 3.5947,
|
| 671 |
+
"step": 1536
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 0.24843161856963614,
|
| 675 |
+
"grad_norm": 20.165475845336914,
|
| 676 |
+
"learning_rate": 1.1549560853199497e-05,
|
| 677 |
+
"loss": 4.0526,
|
| 678 |
+
"step": 1584
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 0.2559598494353827,
|
| 682 |
+
"grad_norm": 34.79319381713867,
|
| 683 |
+
"learning_rate": 1.1900878293601003e-05,
|
| 684 |
+
"loss": 3.7962,
|
| 685 |
+
"step": 1632
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 0.26348808030112925,
|
| 689 |
+
"grad_norm": 127.97925567626953,
|
| 690 |
+
"learning_rate": 1.2252195734002508e-05,
|
| 691 |
+
"loss": 4.1927,
|
| 692 |
+
"step": 1680
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 0.2710163111668758,
|
| 696 |
+
"grad_norm": 27.80243682861328,
|
| 697 |
+
"learning_rate": 1.2603513174404014e-05,
|
| 698 |
+
"loss": 3.6351,
|
| 699 |
+
"step": 1728
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 0.27854454203262236,
|
| 703 |
+
"grad_norm": 31.81105613708496,
|
| 704 |
+
"learning_rate": 1.295483061480552e-05,
|
| 705 |
+
"loss": 3.4256,
|
| 706 |
+
"step": 1776
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 0.2860727728983689,
|
| 710 |
+
"grad_norm": 32.932865142822266,
|
| 711 |
+
"learning_rate": 1.3306148055207025e-05,
|
| 712 |
+
"loss": 3.3175,
|
| 713 |
+
"step": 1824
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 0.2936010037641154,
|
| 717 |
+
"grad_norm": 31.197385787963867,
|
| 718 |
+
"learning_rate": 1.365746549560853e-05,
|
| 719 |
+
"loss": 3.4984,
|
| 720 |
+
"step": 1872
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 0.30050188205771644,
|
| 724 |
+
"eval_nli-pairs_loss": 2.852742910385132,
|
| 725 |
+
"eval_nli-pairs_runtime": 4.1529,
|
| 726 |
+
"eval_nli-pairs_samples_per_second": 48.16,
|
| 727 |
+
"eval_nli-pairs_steps_per_second": 1.204,
|
| 728 |
+
"eval_sts-test_pearson_cosine": 0.7132313507241694,
|
| 729 |
+
"eval_sts-test_pearson_dot": 0.559846529627866,
|
| 730 |
+
"eval_sts-test_pearson_euclidean": 0.7145939583366395,
|
| 731 |
+
"eval_sts-test_pearson_manhattan": 0.724552982808093,
|
| 732 |
+
"eval_sts-test_pearson_max": 0.724552982808093,
|
| 733 |
+
"eval_sts-test_spearman_cosine": 0.6912239915389706,
|
| 734 |
+
"eval_sts-test_spearman_dot": 0.5394217029355446,
|
| 735 |
+
"eval_sts-test_spearman_euclidean": 0.6946616748545426,
|
| 736 |
+
"eval_sts-test_spearman_manhattan": 0.70491424059339,
|
| 737 |
+
"eval_sts-test_spearman_max": 0.70491424059339,
|
| 738 |
+
"step": 1916
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.30050188205771644,
|
| 742 |
+
"eval_vitaminc-pairs_loss": 6.633151054382324,
|
| 743 |
+
"eval_vitaminc-pairs_runtime": 1.4454,
|
| 744 |
+
"eval_vitaminc-pairs_samples_per_second": 114.849,
|
| 745 |
+
"eval_vitaminc-pairs_steps_per_second": 2.767,
|
| 746 |
+
"step": 1916
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.30050188205771644,
|
| 750 |
+
"eval_sts-label_loss": 3.8717281818389893,
|
| 751 |
+
"eval_sts-label_runtime": 0.4016,
|
| 752 |
+
"eval_sts-label_samples_per_second": 498.049,
|
| 753 |
+
"eval_sts-label_steps_per_second": 12.451,
|
| 754 |
+
"step": 1916
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 0.30050188205771644,
|
| 758 |
+
"eval_qnli-contrastive_loss": 1.4170150756835938,
|
| 759 |
+
"eval_qnli-contrastive_runtime": 0.2814,
|
| 760 |
+
"eval_qnli-contrastive_samples_per_second": 710.85,
|
| 761 |
+
"eval_qnli-contrastive_steps_per_second": 17.771,
|
| 762 |
+
"step": 1916
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 0.30050188205771644,
|
| 766 |
+
"eval_scitail-pairs-qa_loss": 0.6900365948677063,
|
| 767 |
+
"eval_scitail-pairs-qa_runtime": 1.0611,
|
| 768 |
+
"eval_scitail-pairs-qa_samples_per_second": 188.492,
|
| 769 |
+
"eval_scitail-pairs-qa_steps_per_second": 4.712,
|
| 770 |
+
"step": 1916
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 0.30050188205771644,
|
| 774 |
+
"eval_scitail-pairs-pos_loss": 1.352358102798462,
|
| 775 |
+
"eval_scitail-pairs-pos_runtime": 2.3596,
|
| 776 |
+
"eval_scitail-pairs-pos_samples_per_second": 84.761,
|
| 777 |
+
"eval_scitail-pairs-pos_steps_per_second": 2.119,
|
| 778 |
+
"step": 1916
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 0.30050188205771644,
|
| 782 |
+
"eval_xsum-pairs_loss": 1.376610517501831,
|
| 783 |
+
"eval_xsum-pairs_runtime": 1.0393,
|
| 784 |
+
"eval_xsum-pairs_samples_per_second": 192.443,
|
| 785 |
+
"eval_xsum-pairs_steps_per_second": 4.811,
|
| 786 |
+
"step": 1916
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.30050188205771644,
|
| 790 |
+
"eval_compression-pairs_loss": 0.871735692024231,
|
| 791 |
+
"eval_compression-pairs_runtime": 0.2351,
|
| 792 |
+
"eval_compression-pairs_samples_per_second": 850.579,
|
| 793 |
+
"eval_compression-pairs_steps_per_second": 21.264,
|
| 794 |
+
"step": 1916
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.30050188205771644,
|
| 798 |
+
"eval_sciq_pairs_loss": 9.191713333129883,
|
| 799 |
+
"eval_sciq_pairs_runtime": 9.1572,
|
| 800 |
+
"eval_sciq_pairs_samples_per_second": 21.841,
|
| 801 |
+
"eval_sciq_pairs_steps_per_second": 0.546,
|
| 802 |
+
"step": 1916
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 0.30050188205771644,
|
| 806 |
+
"eval_qasc_pairs_loss": 2.369694709777832,
|
| 807 |
+
"eval_qasc_pairs_runtime": 1.2239,
|
| 808 |
+
"eval_qasc_pairs_samples_per_second": 163.415,
|
| 809 |
+
"eval_qasc_pairs_steps_per_second": 4.085,
|
| 810 |
+
"step": 1916
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 0.30050188205771644,
|
| 814 |
+
"eval_openbookqa_pairs_loss": 3.9601967334747314,
|
| 815 |
+
"eval_openbookqa_pairs_runtime": 1.0681,
|
| 816 |
+
"eval_openbookqa_pairs_samples_per_second": 187.247,
|
| 817 |
+
"eval_openbookqa_pairs_steps_per_second": 4.681,
|
| 818 |
+
"step": 1916
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 0.30050188205771644,
|
| 822 |
+
"eval_msmarco_pairs_loss": 3.0808801651000977,
|
| 823 |
+
"eval_msmarco_pairs_runtime": 2.5507,
|
| 824 |
+
"eval_msmarco_pairs_samples_per_second": 78.409,
|
| 825 |
+
"eval_msmarco_pairs_steps_per_second": 1.96,
|
| 826 |
+
"step": 1916
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 0.30050188205771644,
|
| 830 |
+
"eval_nq_pairs_loss": 3.4922549724578857,
|
| 831 |
+
"eval_nq_pairs_runtime": 5.7154,
|
| 832 |
+
"eval_nq_pairs_samples_per_second": 34.993,
|
| 833 |
+
"eval_nq_pairs_steps_per_second": 0.875,
|
| 834 |
+
"step": 1916
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 0.30050188205771644,
|
| 838 |
+
"eval_trivia_pairs_loss": 3.4910638332366943,
|
| 839 |
+
"eval_trivia_pairs_runtime": 9.1195,
|
| 840 |
+
"eval_trivia_pairs_samples_per_second": 21.931,
|
| 841 |
+
"eval_trivia_pairs_steps_per_second": 0.548,
|
| 842 |
+
"step": 1916
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.30050188205771644,
|
| 846 |
+
"eval_quora_pairs_loss": 0.833874523639679,
|
| 847 |
+
"eval_quora_pairs_runtime": 0.6419,
|
| 848 |
+
"eval_quora_pairs_samples_per_second": 311.554,
|
| 849 |
+
"eval_quora_pairs_steps_per_second": 7.789,
|
| 850 |
+
"step": 1916
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.30050188205771644,
|
| 854 |
+
"eval_gooaq_pairs_loss": 2.622526168823242,
|
| 855 |
+
"eval_gooaq_pairs_runtime": 1.5751,
|
| 856 |
+
"eval_gooaq_pairs_samples_per_second": 126.977,
|
| 857 |
+
"eval_gooaq_pairs_steps_per_second": 3.174,
|
| 858 |
+
"step": 1916
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 0.30050188205771644,
|
| 862 |
+
"eval_mrpc_pairs_loss": 0.4888114929199219,
|
| 863 |
+
"eval_mrpc_pairs_runtime": 0.2398,
|
| 864 |
+
"eval_mrpc_pairs_samples_per_second": 833.994,
|
| 865 |
+
"eval_mrpc_pairs_steps_per_second": 20.85,
|
| 866 |
+
"step": 1916
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 0.30112923462986196,
|
| 870 |
+
"grad_norm": 6.916851997375488,
|
| 871 |
+
"learning_rate": 1.4008782936010036e-05,
|
| 872 |
+
"loss": 2.7233,
|
| 873 |
+
"step": 1920
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 0.3086574654956085,
|
| 877 |
+
"grad_norm": 17.678085327148438,
|
| 878 |
+
"learning_rate": 1.4360100376411543e-05,
|
| 879 |
+
"loss": 3.6816,
|
| 880 |
+
"step": 1968
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 0.3161856963613551,
|
| 884 |
+
"grad_norm": 92.62138366699219,
|
| 885 |
+
"learning_rate": 1.4711417816813047e-05,
|
| 886 |
+
"loss": 3.3232,
|
| 887 |
+
"step": 2016
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 0.3237139272271016,
|
| 891 |
+
"grad_norm": 27.2542781829834,
|
| 892 |
+
"learning_rate": 1.5062735257214554e-05,
|
| 893 |
+
"loss": 3.3469,
|
| 894 |
+
"step": 2064
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 0.3312421580928482,
|
| 898 |
+
"grad_norm": 6.005978584289551,
|
| 899 |
+
"learning_rate": 1.5414052697616058e-05,
|
| 900 |
+
"loss": 3.7509,
|
| 901 |
+
"step": 2112
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 0.33877038895859474,
|
| 905 |
+
"grad_norm": 16.488624572753906,
|
| 906 |
+
"learning_rate": 1.5765370138017566e-05,
|
| 907 |
+
"loss": 3.1811,
|
| 908 |
+
"step": 2160
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 0.3462986198243413,
|
| 912 |
+
"grad_norm": 11.462204933166504,
|
| 913 |
+
"learning_rate": 1.611668757841907e-05,
|
| 914 |
+
"loss": 3.3341,
|
| 915 |
+
"step": 2208
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 0.35382685069008785,
|
| 919 |
+
"grad_norm": 33.485206604003906,
|
| 920 |
+
"learning_rate": 1.6468005018820577e-05,
|
| 921 |
+
"loss": 2.764,
|
| 922 |
+
"step": 2256
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 0.3613550815558344,
|
| 926 |
+
"grad_norm": 25.066240310668945,
|
| 927 |
+
"learning_rate": 1.681932245922208e-05,
|
| 928 |
+
"loss": 3.6488,
|
| 929 |
+
"step": 2304
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 0.36888331242158096,
|
| 933 |
+
"grad_norm": 28.305265426635742,
|
| 934 |
+
"learning_rate": 1.7170639899623588e-05,
|
| 935 |
+
"loss": 2.721,
|
| 936 |
+
"step": 2352
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 0.3756273525721455,
|
| 940 |
+
"eval_nli-pairs_loss": 2.527458667755127,
|
| 941 |
+
"eval_nli-pairs_runtime": 4.1153,
|
| 942 |
+
"eval_nli-pairs_samples_per_second": 48.599,
|
| 943 |
+
"eval_nli-pairs_steps_per_second": 1.215,
|
| 944 |
+
"eval_sts-test_pearson_cosine": 0.7258900302408404,
|
| 945 |
+
"eval_sts-test_pearson_dot": 0.5655223839113195,
|
| 946 |
+
"eval_sts-test_pearson_euclidean": 0.7228747263710285,
|
| 947 |
+
"eval_sts-test_pearson_manhattan": 0.732591374373909,
|
| 948 |
+
"eval_sts-test_pearson_max": 0.732591374373909,
|
| 949 |
+
"eval_sts-test_spearman_cosine": 0.707910346125958,
|
| 950 |
+
"eval_sts-test_spearman_dot": 0.5482635095738919,
|
| 951 |
+
"eval_sts-test_spearman_euclidean": 0.7064759533156177,
|
| 952 |
+
"eval_sts-test_spearman_manhattan": 0.7166423493246757,
|
| 953 |
+
"eval_sts-test_spearman_max": 0.7166423493246757,
|
| 954 |
+
"step": 2395
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.3756273525721455,
|
| 958 |
+
"eval_vitaminc-pairs_loss": 6.437549114227295,
|
| 959 |
+
"eval_vitaminc-pairs_runtime": 1.4278,
|
| 960 |
+
"eval_vitaminc-pairs_samples_per_second": 116.261,
|
| 961 |
+
"eval_vitaminc-pairs_steps_per_second": 2.801,
|
| 962 |
+
"step": 2395
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.3756273525721455,
|
| 966 |
+
"eval_sts-label_loss": 4.1980671882629395,
|
| 967 |
+
"eval_sts-label_runtime": 0.3956,
|
| 968 |
+
"eval_sts-label_samples_per_second": 505.555,
|
| 969 |
+
"eval_sts-label_steps_per_second": 12.639,
|
| 970 |
+
"step": 2395
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 0.3756273525721455,
|
| 974 |
+
"eval_qnli-contrastive_loss": 1.0682133436203003,
|
| 975 |
+
"eval_qnli-contrastive_runtime": 0.2789,
|
| 976 |
+
"eval_qnli-contrastive_samples_per_second": 717.152,
|
| 977 |
+
"eval_qnli-contrastive_steps_per_second": 17.929,
|
| 978 |
+
"step": 2395
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 0.3756273525721455,
|
| 982 |
+
"eval_scitail-pairs-qa_loss": 0.5046552419662476,
|
| 983 |
+
"eval_scitail-pairs-qa_runtime": 1.0451,
|
| 984 |
+
"eval_scitail-pairs-qa_samples_per_second": 191.365,
|
| 985 |
+
"eval_scitail-pairs-qa_steps_per_second": 4.784,
|
| 986 |
+
"step": 2395
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 0.3756273525721455,
|
| 990 |
+
"eval_scitail-pairs-pos_loss": 1.1998459100723267,
|
| 991 |
+
"eval_scitail-pairs-pos_runtime": 2.3442,
|
| 992 |
+
"eval_scitail-pairs-pos_samples_per_second": 85.316,
|
| 993 |
+
"eval_scitail-pairs-pos_steps_per_second": 2.133,
|
| 994 |
+
"step": 2395
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 0.3756273525721455,
|
| 998 |
+
"eval_xsum-pairs_loss": 1.1817097663879395,
|
| 999 |
+
"eval_xsum-pairs_runtime": 1.0372,
|
| 1000 |
+
"eval_xsum-pairs_samples_per_second": 192.835,
|
| 1001 |
+
"eval_xsum-pairs_steps_per_second": 4.821,
|
| 1002 |
+
"step": 2395
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 0.3756273525721455,
|
| 1006 |
+
"eval_compression-pairs_loss": 0.6974765062332153,
|
| 1007 |
+
"eval_compression-pairs_runtime": 0.2369,
|
| 1008 |
+
"eval_compression-pairs_samples_per_second": 844.401,
|
| 1009 |
+
"eval_compression-pairs_steps_per_second": 21.11,
|
| 1010 |
+
"step": 2395
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.3756273525721455,
|
| 1014 |
+
"eval_sciq_pairs_loss": 8.970888137817383,
|
| 1015 |
+
"eval_sciq_pairs_runtime": 9.0441,
|
| 1016 |
+
"eval_sciq_pairs_samples_per_second": 22.114,
|
| 1017 |
+
"eval_sciq_pairs_steps_per_second": 0.553,
|
| 1018 |
+
"step": 2395
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.3756273525721455,
|
| 1022 |
+
"eval_qasc_pairs_loss": 1.9235339164733887,
|
| 1023 |
+
"eval_qasc_pairs_runtime": 1.2061,
|
| 1024 |
+
"eval_qasc_pairs_samples_per_second": 165.828,
|
| 1025 |
+
"eval_qasc_pairs_steps_per_second": 4.146,
|
| 1026 |
+
"step": 2395
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 0.3756273525721455,
|
| 1030 |
+
"eval_openbookqa_pairs_loss": 3.6225194931030273,
|
| 1031 |
+
"eval_openbookqa_pairs_runtime": 1.0455,
|
| 1032 |
+
"eval_openbookqa_pairs_samples_per_second": 191.296,
|
| 1033 |
+
"eval_openbookqa_pairs_steps_per_second": 4.782,
|
| 1034 |
+
"step": 2395
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 0.3756273525721455,
|
| 1038 |
+
"eval_msmarco_pairs_loss": 2.664341926574707,
|
| 1039 |
+
"eval_msmarco_pairs_runtime": 2.5305,
|
| 1040 |
+
"eval_msmarco_pairs_samples_per_second": 79.036,
|
| 1041 |
+
"eval_msmarco_pairs_steps_per_second": 1.976,
|
| 1042 |
+
"step": 2395
|
| 1043 |
+
},
|
| 1044 |
+
{
|
| 1045 |
+
"epoch": 0.3756273525721455,
|
| 1046 |
+
"eval_nq_pairs_loss": 3.055206298828125,
|
| 1047 |
+
"eval_nq_pairs_runtime": 5.6527,
|
| 1048 |
+
"eval_nq_pairs_samples_per_second": 35.381,
|
| 1049 |
+
"eval_nq_pairs_steps_per_second": 0.885,
|
| 1050 |
+
"step": 2395
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 0.3756273525721455,
|
| 1054 |
+
"eval_trivia_pairs_loss": 2.9497525691986084,
|
| 1055 |
+
"eval_trivia_pairs_runtime": 9.0334,
|
| 1056 |
+
"eval_trivia_pairs_samples_per_second": 22.14,
|
| 1057 |
+
"eval_trivia_pairs_steps_per_second": 0.554,
|
| 1058 |
+
"step": 2395
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 0.3756273525721455,
|
| 1062 |
+
"eval_quora_pairs_loss": 0.7771684527397156,
|
| 1063 |
+
"eval_quora_pairs_runtime": 0.626,
|
| 1064 |
+
"eval_quora_pairs_samples_per_second": 319.495,
|
| 1065 |
+
"eval_quora_pairs_steps_per_second": 7.987,
|
| 1066 |
+
"step": 2395
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.3756273525721455,
|
| 1070 |
+
"eval_gooaq_pairs_loss": 2.266879081726074,
|
| 1071 |
+
"eval_gooaq_pairs_runtime": 1.5425,
|
| 1072 |
+
"eval_gooaq_pairs_samples_per_second": 129.664,
|
| 1073 |
+
"eval_gooaq_pairs_steps_per_second": 3.242,
|
| 1074 |
+
"step": 2395
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.3756273525721455,
|
| 1078 |
+
"eval_mrpc_pairs_loss": 0.36913084983825684,
|
| 1079 |
+
"eval_mrpc_pairs_runtime": 0.2383,
|
| 1080 |
+
"eval_mrpc_pairs_samples_per_second": 839.153,
|
| 1081 |
+
"eval_mrpc_pairs_steps_per_second": 20.979,
|
| 1082 |
+
"step": 2395
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 0.37641154328732745,
|
| 1086 |
+
"grad_norm": 36.60768127441406,
|
| 1087 |
+
"learning_rate": 1.752195734002509e-05,
|
| 1088 |
+
"loss": 3.3609,
|
| 1089 |
+
"step": 2400
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 0.383939774153074,
|
| 1093 |
+
"grad_norm": 24.15782928466797,
|
| 1094 |
+
"learning_rate": 1.7873274780426595e-05,
|
| 1095 |
+
"loss": 2.6252,
|
| 1096 |
+
"step": 2448
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 0.39146800501882056,
|
| 1100 |
+
"grad_norm": 7.509932041168213,
|
| 1101 |
+
"learning_rate": 1.8224592220828106e-05,
|
| 1102 |
+
"loss": 3.5142,
|
| 1103 |
+
"step": 2496
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 0.3989962358845671,
|
| 1107 |
+
"grad_norm": 29.380950927734375,
|
| 1108 |
+
"learning_rate": 1.857590966122961e-05,
|
| 1109 |
+
"loss": 4.0597,
|
| 1110 |
+
"step": 2544
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 0.4065244667503137,
|
| 1114 |
+
"grad_norm": 28.593975067138672,
|
| 1115 |
+
"learning_rate": 1.8927227101631114e-05,
|
| 1116 |
+
"loss": 2.8512,
|
| 1117 |
+
"step": 2592
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 0.41405269761606023,
|
| 1121 |
+
"grad_norm": 21.228628158569336,
|
| 1122 |
+
"learning_rate": 1.927854454203262e-05,
|
| 1123 |
+
"loss": 2.717,
|
| 1124 |
+
"step": 2640
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 0.4215809284818068,
|
| 1128 |
+
"grad_norm": 43.00386047363281,
|
| 1129 |
+
"learning_rate": 1.962986198243413e-05,
|
| 1130 |
+
"loss": 3.4717,
|
| 1131 |
+
"step": 2688
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 0.42910915934755334,
|
| 1135 |
+
"grad_norm": 25.004785537719727,
|
| 1136 |
+
"learning_rate": 1.9981179422835632e-05,
|
| 1137 |
+
"loss": 3.1105,
|
| 1138 |
+
"step": 2736
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 0.4366373902132999,
|
| 1142 |
+
"grad_norm": 7.555154323577881,
|
| 1143 |
+
"learning_rate": 2.0332496863237136e-05,
|
| 1144 |
+
"loss": 2.7798,
|
| 1145 |
+
"step": 2784
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 0.44416562107904645,
|
| 1149 |
+
"grad_norm": 30.839733123779297,
|
| 1150 |
+
"learning_rate": 2.0683814303638643e-05,
|
| 1151 |
+
"loss": 3.3606,
|
| 1152 |
+
"step": 2832
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 0.45075282308657466,
|
| 1156 |
+
"eval_nli-pairs_loss": 2.284590721130371,
|
| 1157 |
+
"eval_nli-pairs_runtime": 4.0714,
|
| 1158 |
+
"eval_nli-pairs_samples_per_second": 49.123,
|
| 1159 |
+
"eval_nli-pairs_steps_per_second": 1.228,
|
| 1160 |
+
"eval_sts-test_pearson_cosine": 0.7382507781851606,
|
| 1161 |
+
"eval_sts-test_pearson_dot": 0.5710221319397019,
|
| 1162 |
+
"eval_sts-test_pearson_euclidean": 0.7307583601561211,
|
| 1163 |
+
"eval_sts-test_pearson_manhattan": 0.7394202696141936,
|
| 1164 |
+
"eval_sts-test_pearson_max": 0.7394202696141936,
|
| 1165 |
+
"eval_sts-test_spearman_cosine": 0.7211579109789371,
|
| 1166 |
+
"eval_sts-test_spearman_dot": 0.5515579746967598,
|
| 1167 |
+
"eval_sts-test_spearman_euclidean": 0.7142073811971875,
|
| 1168 |
+
"eval_sts-test_spearman_manhattan": 0.7240537218564107,
|
| 1169 |
+
"eval_sts-test_spearman_max": 0.7240537218564107,
|
| 1170 |
+
"step": 2874
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 0.45075282308657466,
|
| 1174 |
+
"eval_vitaminc-pairs_loss": 6.35264253616333,
|
| 1175 |
+
"eval_vitaminc-pairs_runtime": 1.4349,
|
| 1176 |
+
"eval_vitaminc-pairs_samples_per_second": 115.687,
|
| 1177 |
+
"eval_vitaminc-pairs_steps_per_second": 2.788,
|
| 1178 |
+
"step": 2874
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.45075282308657466,
|
| 1182 |
+
"eval_sts-label_loss": 4.186042308807373,
|
| 1183 |
+
"eval_sts-label_runtime": 0.3983,
|
| 1184 |
+
"eval_sts-label_samples_per_second": 502.129,
|
| 1185 |
+
"eval_sts-label_steps_per_second": 12.553,
|
| 1186 |
+
"step": 2874
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.45075282308657466,
|
| 1190 |
+
"eval_qnli-contrastive_loss": 0.781445324420929,
|
| 1191 |
+
"eval_qnli-contrastive_runtime": 0.2765,
|
| 1192 |
+
"eval_qnli-contrastive_samples_per_second": 723.448,
|
| 1193 |
+
"eval_qnli-contrastive_steps_per_second": 18.086,
|
| 1194 |
+
"step": 2874
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 0.45075282308657466,
|
| 1198 |
+
"eval_scitail-pairs-qa_loss": 0.4217279851436615,
|
| 1199 |
+
"eval_scitail-pairs-qa_runtime": 1.0438,
|
| 1200 |
+
"eval_scitail-pairs-qa_samples_per_second": 191.612,
|
| 1201 |
+
"eval_scitail-pairs-qa_steps_per_second": 4.79,
|
| 1202 |
+
"step": 2874
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 0.45075282308657466,
|
| 1206 |
+
"eval_scitail-pairs-pos_loss": 1.051362156867981,
|
| 1207 |
+
"eval_scitail-pairs-pos_runtime": 2.3425,
|
| 1208 |
+
"eval_scitail-pairs-pos_samples_per_second": 85.379,
|
| 1209 |
+
"eval_scitail-pairs-pos_steps_per_second": 2.134,
|
| 1210 |
+
"step": 2874
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 0.45075282308657466,
|
| 1214 |
+
"eval_xsum-pairs_loss": 1.0554753541946411,
|
| 1215 |
+
"eval_xsum-pairs_runtime": 1.044,
|
| 1216 |
+
"eval_xsum-pairs_samples_per_second": 191.573,
|
| 1217 |
+
"eval_xsum-pairs_steps_per_second": 4.789,
|
| 1218 |
+
"step": 2874
|
| 1219 |
+
},
|
| 1220 |
+
{
|
| 1221 |
+
"epoch": 0.45075282308657466,
|
| 1222 |
+
"eval_compression-pairs_loss": 0.6035106778144836,
|
| 1223 |
+
"eval_compression-pairs_runtime": 0.241,
|
| 1224 |
+
"eval_compression-pairs_samples_per_second": 830.038,
|
| 1225 |
+
"eval_compression-pairs_steps_per_second": 20.751,
|
| 1226 |
+
"step": 2874
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 0.45075282308657466,
|
| 1230 |
+
"eval_sciq_pairs_loss": 8.811105728149414,
|
| 1231 |
+
"eval_sciq_pairs_runtime": 9.0357,
|
| 1232 |
+
"eval_sciq_pairs_samples_per_second": 22.134,
|
| 1233 |
+
"eval_sciq_pairs_steps_per_second": 0.553,
|
| 1234 |
+
"step": 2874
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.45075282308657466,
|
| 1238 |
+
"eval_qasc_pairs_loss": 1.615903377532959,
|
| 1239 |
+
"eval_qasc_pairs_runtime": 1.214,
|
| 1240 |
+
"eval_qasc_pairs_samples_per_second": 164.746,
|
| 1241 |
+
"eval_qasc_pairs_steps_per_second": 4.119,
|
| 1242 |
+
"step": 2874
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.45075282308657466,
|
| 1246 |
+
"eval_openbookqa_pairs_loss": 3.4049320220947266,
|
| 1247 |
+
"eval_openbookqa_pairs_runtime": 1.0554,
|
| 1248 |
+
"eval_openbookqa_pairs_samples_per_second": 189.509,
|
| 1249 |
+
"eval_openbookqa_pairs_steps_per_second": 4.738,
|
| 1250 |
+
"step": 2874
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 0.45075282308657466,
|
| 1254 |
+
"eval_msmarco_pairs_loss": 2.3909060955047607,
|
| 1255 |
+
"eval_msmarco_pairs_runtime": 2.5301,
|
| 1256 |
+
"eval_msmarco_pairs_samples_per_second": 79.048,
|
| 1257 |
+
"eval_msmarco_pairs_steps_per_second": 1.976,
|
| 1258 |
+
"step": 2874
|
| 1259 |
+
},
|
| 1260 |
+
{
|
| 1261 |
+
"epoch": 0.45075282308657466,
|
| 1262 |
+
"eval_nq_pairs_loss": 2.794445753097534,
|
| 1263 |
+
"eval_nq_pairs_runtime": 5.6752,
|
| 1264 |
+
"eval_nq_pairs_samples_per_second": 35.241,
|
| 1265 |
+
"eval_nq_pairs_steps_per_second": 0.881,
|
| 1266 |
+
"step": 2874
|
| 1267 |
+
},
|
| 1268 |
+
{
|
| 1269 |
+
"epoch": 0.45075282308657466,
|
| 1270 |
+
"eval_trivia_pairs_loss": 2.753361701965332,
|
| 1271 |
+
"eval_trivia_pairs_runtime": 9.0766,
|
| 1272 |
+
"eval_trivia_pairs_samples_per_second": 22.035,
|
| 1273 |
+
"eval_trivia_pairs_steps_per_second": 0.551,
|
| 1274 |
+
"step": 2874
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 0.45075282308657466,
|
| 1278 |
+
"eval_quora_pairs_loss": 0.205492302775383,
|
| 1279 |
+
"eval_quora_pairs_runtime": 0.6182,
|
| 1280 |
+
"eval_quora_pairs_samples_per_second": 323.536,
|
| 1281 |
+
"eval_quora_pairs_steps_per_second": 8.088,
|
| 1282 |
+
"step": 2874
|
| 1283 |
+
},
|
| 1284 |
+
{
|
| 1285 |
+
"epoch": 0.45075282308657466,
|
| 1286 |
+
"eval_gooaq_pairs_loss": 2.038878917694092,
|
| 1287 |
+
"eval_gooaq_pairs_runtime": 1.5488,
|
| 1288 |
+
"eval_gooaq_pairs_samples_per_second": 129.134,
|
| 1289 |
+
"eval_gooaq_pairs_steps_per_second": 3.228,
|
| 1290 |
+
"step": 2874
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.45075282308657466,
|
| 1294 |
+
"eval_mrpc_pairs_loss": 0.3230588436126709,
|
| 1295 |
+
"eval_mrpc_pairs_runtime": 0.2358,
|
| 1296 |
+
"eval_mrpc_pairs_samples_per_second": 848.229,
|
| 1297 |
+
"eval_mrpc_pairs_steps_per_second": 21.206,
|
| 1298 |
+
"step": 2874
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.451693851944793,
|
| 1302 |
+
"grad_norm": 5.271574020385742,
|
| 1303 |
+
"learning_rate": 2.103513174404015e-05,
|
| 1304 |
+
"loss": 2.6918,
|
| 1305 |
+
"step": 2880
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.4592220828105395,
|
| 1309 |
+
"grad_norm": 21.954103469848633,
|
| 1310 |
+
"learning_rate": 2.1386449184441654e-05,
|
| 1311 |
+
"loss": 2.8354,
|
| 1312 |
+
"step": 2928
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.46675031367628605,
|
| 1316 |
+
"grad_norm": 28.671293258666992,
|
| 1317 |
+
"learning_rate": 2.173776662484316e-05,
|
| 1318 |
+
"loss": 2.9499,
|
| 1319 |
+
"step": 2976
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.4742785445420326,
|
| 1323 |
+
"grad_norm": 26.562397003173828,
|
| 1324 |
+
"learning_rate": 2.2089084065244666e-05,
|
| 1325 |
+
"loss": 2.6211,
|
| 1326 |
+
"step": 3024
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.48180677540777916,
|
| 1330 |
+
"grad_norm": 48.511756896972656,
|
| 1331 |
+
"learning_rate": 2.2440401505646173e-05,
|
| 1332 |
+
"loss": 3.3356,
|
| 1333 |
+
"step": 3072
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.4893350062735257,
|
| 1337 |
+
"grad_norm": 46.71563720703125,
|
| 1338 |
+
"learning_rate": 2.2791718946047677e-05,
|
| 1339 |
+
"loss": 2.846,
|
| 1340 |
+
"step": 3120
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.4968632371392723,
|
| 1344 |
+
"grad_norm": 24.524322509765625,
|
| 1345 |
+
"learning_rate": 2.3143036386449184e-05,
|
| 1346 |
+
"loss": 2.4866,
|
| 1347 |
+
"step": 3168
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.5043914680050188,
|
| 1351 |
+
"grad_norm": 85.22843933105469,
|
| 1352 |
+
"learning_rate": 2.3494353826850688e-05,
|
| 1353 |
+
"loss": 2.6334,
|
| 1354 |
+
"step": 3216
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.5119196988707654,
|
| 1358 |
+
"grad_norm": 28.435443878173828,
|
| 1359 |
+
"learning_rate": 2.384567126725219e-05,
|
| 1360 |
+
"loss": 2.6118,
|
| 1361 |
+
"step": 3264
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.5194479297365119,
|
| 1365 |
+
"grad_norm": 21.590103149414062,
|
| 1366 |
+
"learning_rate": 2.41969887076537e-05,
|
| 1367 |
+
"loss": 2.5833,
|
| 1368 |
+
"step": 3312
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.5258782936010038,
|
| 1372 |
+
"eval_nli-pairs_loss": 2.0752949714660645,
|
| 1373 |
+
"eval_nli-pairs_runtime": 4.0304,
|
| 1374 |
+
"eval_nli-pairs_samples_per_second": 49.623,
|
| 1375 |
+
"eval_nli-pairs_steps_per_second": 1.241,
|
| 1376 |
+
"eval_sts-test_pearson_cosine": 0.7401847199967786,
|
| 1377 |
+
"eval_sts-test_pearson_dot": 0.5441501995975192,
|
| 1378 |
+
"eval_sts-test_pearson_euclidean": 0.7344996320188322,
|
| 1379 |
+
"eval_sts-test_pearson_manhattan": 0.7394640598472787,
|
| 1380 |
+
"eval_sts-test_pearson_max": 0.7401847199967786,
|
| 1381 |
+
"eval_sts-test_spearman_cosine": 0.7300085598018916,
|
| 1382 |
+
"eval_sts-test_spearman_dot": 0.5241747185593542,
|
| 1383 |
+
"eval_sts-test_spearman_euclidean": 0.7194131601167465,
|
| 1384 |
+
"eval_sts-test_spearman_manhattan": 0.726961581928453,
|
| 1385 |
+
"eval_sts-test_spearman_max": 0.7300085598018916,
|
| 1386 |
+
"step": 3353
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"epoch": 0.5258782936010038,
|
| 1390 |
+
"eval_vitaminc-pairs_loss": 6.441956996917725,
|
| 1391 |
+
"eval_vitaminc-pairs_runtime": 1.4416,
|
| 1392 |
+
"eval_vitaminc-pairs_samples_per_second": 115.149,
|
| 1393 |
+
"eval_vitaminc-pairs_steps_per_second": 2.775,
|
| 1394 |
+
"step": 3353
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 0.5258782936010038,
|
| 1398 |
+
"eval_sts-label_loss": 4.200085639953613,
|
| 1399 |
+
"eval_sts-label_runtime": 0.3949,
|
| 1400 |
+
"eval_sts-label_samples_per_second": 506.436,
|
| 1401 |
+
"eval_sts-label_steps_per_second": 12.661,
|
| 1402 |
+
"step": 3353
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.5258782936010038,
|
| 1406 |
+
"eval_qnli-contrastive_loss": 0.5195684432983398,
|
| 1407 |
+
"eval_qnli-contrastive_runtime": 0.2809,
|
| 1408 |
+
"eval_qnli-contrastive_samples_per_second": 712.107,
|
| 1409 |
+
"eval_qnli-contrastive_steps_per_second": 17.803,
|
| 1410 |
+
"step": 3353
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.5258782936010038,
|
| 1414 |
+
"eval_scitail-pairs-qa_loss": 0.35189124941825867,
|
| 1415 |
+
"eval_scitail-pairs-qa_runtime": 1.0578,
|
| 1416 |
+
"eval_scitail-pairs-qa_samples_per_second": 189.064,
|
| 1417 |
+
"eval_scitail-pairs-qa_steps_per_second": 4.727,
|
| 1418 |
+
"step": 3353
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 0.5258782936010038,
|
| 1422 |
+
"eval_scitail-pairs-pos_loss": 0.8873756527900696,
|
| 1423 |
+
"eval_scitail-pairs-pos_runtime": 2.4029,
|
| 1424 |
+
"eval_scitail-pairs-pos_samples_per_second": 83.232,
|
| 1425 |
+
"eval_scitail-pairs-pos_steps_per_second": 2.081,
|
| 1426 |
+
"step": 3353
|
| 1427 |
+
},
|
| 1428 |
+
{
|
| 1429 |
+
"epoch": 0.5258782936010038,
|
| 1430 |
+
"eval_xsum-pairs_loss": 0.939339280128479,
|
| 1431 |
+
"eval_xsum-pairs_runtime": 1.041,
|
| 1432 |
+
"eval_xsum-pairs_samples_per_second": 192.121,
|
| 1433 |
+
"eval_xsum-pairs_steps_per_second": 4.803,
|
| 1434 |
+
"step": 3353
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 0.5258782936010038,
|
| 1438 |
+
"eval_compression-pairs_loss": 0.5007131695747375,
|
| 1439 |
+
"eval_compression-pairs_runtime": 0.2338,
|
| 1440 |
+
"eval_compression-pairs_samples_per_second": 855.479,
|
| 1441 |
+
"eval_compression-pairs_steps_per_second": 21.387,
|
| 1442 |
+
"step": 3353
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 0.5258782936010038,
|
| 1446 |
+
"eval_sciq_pairs_loss": 8.558987617492676,
|
| 1447 |
+
"eval_sciq_pairs_runtime": 9.0984,
|
| 1448 |
+
"eval_sciq_pairs_samples_per_second": 21.982,
|
| 1449 |
+
"eval_sciq_pairs_steps_per_second": 0.55,
|
| 1450 |
+
"step": 3353
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"epoch": 0.5258782936010038,
|
| 1454 |
+
"eval_qasc_pairs_loss": 1.4318852424621582,
|
| 1455 |
+
"eval_qasc_pairs_runtime": 1.2286,
|
| 1456 |
+
"eval_qasc_pairs_samples_per_second": 162.79,
|
| 1457 |
+
"eval_qasc_pairs_steps_per_second": 4.07,
|
| 1458 |
+
"step": 3353
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.5258782936010038,
|
| 1462 |
+
"eval_openbookqa_pairs_loss": 3.1973114013671875,
|
| 1463 |
+
"eval_openbookqa_pairs_runtime": 1.0491,
|
| 1464 |
+
"eval_openbookqa_pairs_samples_per_second": 190.633,
|
| 1465 |
+
"eval_openbookqa_pairs_steps_per_second": 4.766,
|
| 1466 |
+
"step": 3353
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.5258782936010038,
|
| 1470 |
+
"eval_msmarco_pairs_loss": 2.2080254554748535,
|
| 1471 |
+
"eval_msmarco_pairs_runtime": 2.5223,
|
| 1472 |
+
"eval_msmarco_pairs_samples_per_second": 79.294,
|
| 1473 |
+
"eval_msmarco_pairs_steps_per_second": 1.982,
|
| 1474 |
+
"step": 3353
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 0.5258782936010038,
|
| 1478 |
+
"eval_nq_pairs_loss": 2.5810558795928955,
|
| 1479 |
+
"eval_nq_pairs_runtime": 5.6341,
|
| 1480 |
+
"eval_nq_pairs_samples_per_second": 35.498,
|
| 1481 |
+
"eval_nq_pairs_steps_per_second": 0.887,
|
| 1482 |
+
"step": 3353
|
| 1483 |
+
},
|
| 1484 |
+
{
|
| 1485 |
+
"epoch": 0.5258782936010038,
|
| 1486 |
+
"eval_trivia_pairs_loss": 2.655771255493164,
|
| 1487 |
+
"eval_trivia_pairs_runtime": 9.0716,
|
| 1488 |
+
"eval_trivia_pairs_samples_per_second": 22.047,
|
| 1489 |
+
"eval_trivia_pairs_steps_per_second": 0.551,
|
| 1490 |
+
"step": 3353
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 0.5258782936010038,
|
| 1494 |
+
"eval_quora_pairs_loss": 0.5028819441795349,
|
| 1495 |
+
"eval_quora_pairs_runtime": 0.6144,
|
| 1496 |
+
"eval_quora_pairs_samples_per_second": 325.522,
|
| 1497 |
+
"eval_quora_pairs_steps_per_second": 8.138,
|
| 1498 |
+
"step": 3353
|
| 1499 |
+
},
|
| 1500 |
+
{
|
| 1501 |
+
"epoch": 0.5258782936010038,
|
| 1502 |
+
"eval_gooaq_pairs_loss": 1.8867437839508057,
|
| 1503 |
+
"eval_gooaq_pairs_runtime": 1.5505,
|
| 1504 |
+
"eval_gooaq_pairs_samples_per_second": 128.994,
|
| 1505 |
+
"eval_gooaq_pairs_steps_per_second": 3.225,
|
| 1506 |
+
"step": 3353
|
| 1507 |
+
},
|
| 1508 |
+
{
|
| 1509 |
+
"epoch": 0.5258782936010038,
|
| 1510 |
+
"eval_mrpc_pairs_loss": 0.2580638825893402,
|
| 1511 |
+
"eval_mrpc_pairs_runtime": 0.2364,
|
| 1512 |
+
"eval_mrpc_pairs_samples_per_second": 846.008,
|
| 1513 |
+
"eval_mrpc_pairs_steps_per_second": 21.15,
|
| 1514 |
+
"step": 3353
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.5269761606022585,
|
| 1518 |
+
"grad_norm": 2.6962711811065674,
|
| 1519 |
+
"learning_rate": 2.4548306148055206e-05,
|
| 1520 |
+
"loss": 2.3251,
|
| 1521 |
+
"step": 3360
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.534504391468005,
|
| 1525 |
+
"grad_norm": 35.47948455810547,
|
| 1526 |
+
"learning_rate": 2.489962358845671e-05,
|
| 1527 |
+
"loss": 2.8494,
|
| 1528 |
+
"step": 3408
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.5420326223337516,
|
| 1532 |
+
"grad_norm": 8.13453483581543,
|
| 1533 |
+
"learning_rate": 2.5250941028858214e-05,
|
| 1534 |
+
"loss": 2.4009,
|
| 1535 |
+
"step": 3456
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.5495608531994981,
|
| 1539 |
+
"grad_norm": 20.041057586669922,
|
| 1540 |
+
"learning_rate": 2.560225846925972e-05,
|
| 1541 |
+
"loss": 2.5952,
|
| 1542 |
+
"step": 3504
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.5570890840652447,
|
| 1546 |
+
"grad_norm": 23.942073822021484,
|
| 1547 |
+
"learning_rate": 2.595357590966123e-05,
|
| 1548 |
+
"loss": 2.2798,
|
| 1549 |
+
"step": 3552
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.5646173149309912,
|
| 1553 |
+
"grad_norm": 17.675006866455078,
|
| 1554 |
+
"learning_rate": 2.6304893350062732e-05,
|
| 1555 |
+
"loss": 2.308,
|
| 1556 |
+
"step": 3600
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.5721455457967378,
|
| 1560 |
+
"grad_norm": 24.20000457763672,
|
| 1561 |
+
"learning_rate": 2.6656210790464236e-05,
|
| 1562 |
+
"loss": 2.122,
|
| 1563 |
+
"step": 3648
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.5796737766624843,
|
| 1567 |
+
"grad_norm": 30.06256866455078,
|
| 1568 |
+
"learning_rate": 2.7007528230865747e-05,
|
| 1569 |
+
"loss": 2.7901,
|
| 1570 |
+
"step": 3696
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.5872020075282308,
|
| 1574 |
+
"grad_norm": 22.547115325927734,
|
| 1575 |
+
"learning_rate": 2.735884567126725e-05,
|
| 1576 |
+
"loss": 2.0671,
|
| 1577 |
+
"step": 3744
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.5947302383939774,
|
| 1581 |
+
"grad_norm": 34.11716079711914,
|
| 1582 |
+
"learning_rate": 2.7710163111668754e-05,
|
| 1583 |
+
"loss": 2.366,
|
| 1584 |
+
"step": 3792
|
| 1585 |
+
}
|
| 1586 |
+
],
|
| 1587 |
+
"logging_steps": 48,
|
| 1588 |
+
"max_steps": 19128,
|
| 1589 |
+
"num_input_tokens_seen": 0,
|
| 1590 |
+
"num_train_epochs": 3,
|
| 1591 |
+
"save_steps": 1913,
|
| 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": 42,
|
| 1606 |
+
"trial_name": null,
|
| 1607 |
+
"trial_params": null
|
| 1608 |
+
}
|
checkpoint-3826/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8f0d30e9d834ad3faa1c6260f078cc6b8b6620d2f13869b6672c84d4621cb729
|
| 3 |
+
size 5624
|