Training in progress, step 208, checkpoint
Browse files- checkpoint-208/1_Pooling/config.json +10 -0
- checkpoint-208/README.md +0 -0
- checkpoint-208/added_tokens.json +3 -0
- checkpoint-208/config.json +38 -0
- checkpoint-208/config_sentence_transformers.json +10 -0
- checkpoint-208/modules.json +14 -0
- checkpoint-208/optimizer.pt +3 -0
- checkpoint-208/pytorch_model.bin +3 -0
- checkpoint-208/rng_state.pth +3 -0
- checkpoint-208/scheduler.pt +3 -0
- checkpoint-208/sentence_bert_config.json +4 -0
- checkpoint-208/special_tokens_map.json +15 -0
- checkpoint-208/spm.model +3 -0
- checkpoint-208/tokenizer.json +0 -0
- checkpoint-208/tokenizer_config.json +58 -0
- checkpoint-208/trainer_state.json +2060 -0
- checkpoint-208/training_args.bin +3 -0
checkpoint-208/1_Pooling/config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"word_embedding_dimension": 1536,
|
| 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-208/README.md
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-208/added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"[MASK]": 128000
|
| 3 |
+
}
|
checkpoint-208/config.json
ADDED
|
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "microsoft/deberta-v2-xlarge",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"DebertaV2Model"
|
| 5 |
+
],
|
| 6 |
+
"attention_head_size": 64,
|
| 7 |
+
"attention_probs_dropout_prob": 0.1,
|
| 8 |
+
"conv_act": "gelu",
|
| 9 |
+
"conv_kernel_size": 3,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 1536,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 6144,
|
| 15 |
+
"layer_norm_eps": 1e-07,
|
| 16 |
+
"max_position_embeddings": 512,
|
| 17 |
+
"max_relative_positions": -1,
|
| 18 |
+
"model_type": "deberta-v2",
|
| 19 |
+
"norm_rel_ebd": "layer_norm",
|
| 20 |
+
"num_attention_heads": 24,
|
| 21 |
+
"num_hidden_layers": 24,
|
| 22 |
+
"pad_token_id": 0,
|
| 23 |
+
"pooler_dropout": 0,
|
| 24 |
+
"pooler_hidden_act": "gelu",
|
| 25 |
+
"pooler_hidden_size": 1536,
|
| 26 |
+
"pos_att_type": [
|
| 27 |
+
"p2c",
|
| 28 |
+
"c2p"
|
| 29 |
+
],
|
| 30 |
+
"position_biased_input": false,
|
| 31 |
+
"position_buckets": 256,
|
| 32 |
+
"relative_attention": true,
|
| 33 |
+
"share_att_key": true,
|
| 34 |
+
"torch_dtype": "float32",
|
| 35 |
+
"transformers_version": "4.42.4",
|
| 36 |
+
"type_vocab_size": 0,
|
| 37 |
+
"vocab_size": 128100
|
| 38 |
+
}
|
checkpoint-208/config_sentence_transformers.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"__version__": {
|
| 3 |
+
"sentence_transformers": "3.0.1",
|
| 4 |
+
"transformers": "4.42.4",
|
| 5 |
+
"pytorch": "2.4.0+cu121"
|
| 6 |
+
},
|
| 7 |
+
"prompts": {},
|
| 8 |
+
"default_prompt_name": null,
|
| 9 |
+
"similarity_fn_name": null
|
| 10 |
+
}
|
checkpoint-208/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-208/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f2c109680eb36bd304a09a3cec75f5d702dba62a79d3f7913a8748cf24ca85e4
|
| 3 |
+
size 7077084396
|
checkpoint-208/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d904abc02d87b068c44add959c95f723bdb089cc88fd6cdb8737a0a25ffa0b23
|
| 3 |
+
size 3538506546
|
checkpoint-208/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9a4e829db68d65871813b4e9210305eec9a0624e7c9f8b780cfb6d5af0ea4214
|
| 3 |
+
size 14244
|
checkpoint-208/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:60758a3ed3bdd9d4d71b21d70b5d02a1e7bbe140e8892b1195df357b2e78c103
|
| 3 |
+
size 1064
|
checkpoint-208/sentence_bert_config.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"max_seq_length": 512,
|
| 3 |
+
"do_lower_case": false
|
| 4 |
+
}
|
checkpoint-208/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-208/spm.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5598d5e96f339a8d980c15f9afd405a2e5e1be7db41de3ed13b0f03fac1e8c17
|
| 3 |
+
size 2447305
|
checkpoint-208/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-208/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-208/trainer_state.json
ADDED
|
@@ -0,0 +1,2060 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.40102827763496146,
|
| 5 |
+
"eval_steps": 26,
|
| 6 |
+
"global_step": 208,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.005784061696658098,
|
| 13 |
+
"grad_norm": NaN,
|
| 14 |
+
"learning_rate": 0.0,
|
| 15 |
+
"loss": 7.8503,
|
| 16 |
+
"step": 3
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.011568123393316195,
|
| 20 |
+
"grad_norm": 127.71849060058594,
|
| 21 |
+
"learning_rate": 5.791505791505792e-08,
|
| 22 |
+
"loss": 8.4022,
|
| 23 |
+
"step": 6
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.017352185089974295,
|
| 27 |
+
"grad_norm": 244.984619140625,
|
| 28 |
+
"learning_rate": 1.7374517374517373e-07,
|
| 29 |
+
"loss": 11.1776,
|
| 30 |
+
"step": 9
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.02313624678663239,
|
| 34 |
+
"grad_norm": 425.6427917480469,
|
| 35 |
+
"learning_rate": 3.4749034749034746e-07,
|
| 36 |
+
"loss": 9.7845,
|
| 37 |
+
"step": 12
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.02892030848329049,
|
| 41 |
+
"grad_norm": 238.8634490966797,
|
| 42 |
+
"learning_rate": 5.212355212355212e-07,
|
| 43 |
+
"loss": 8.9224,
|
| 44 |
+
"step": 15
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.03470437017994859,
|
| 48 |
+
"grad_norm": 173.4210968017578,
|
| 49 |
+
"learning_rate": 6.949806949806949e-07,
|
| 50 |
+
"loss": 11.1202,
|
| 51 |
+
"step": 18
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.04048843187660668,
|
| 55 |
+
"grad_norm": 77.89308166503906,
|
| 56 |
+
"learning_rate": 8.687258687258688e-07,
|
| 57 |
+
"loss": 7.413,
|
| 58 |
+
"step": 21
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.04627249357326478,
|
| 62 |
+
"grad_norm": Infinity,
|
| 63 |
+
"learning_rate": 9.845559845559844e-07,
|
| 64 |
+
"loss": 7.7803,
|
| 65 |
+
"step": 24
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.05012853470437018,
|
| 69 |
+
"eval_Qnli-dev_cosine_accuracy": 0.609375,
|
| 70 |
+
"eval_Qnli-dev_cosine_accuracy_threshold": 0.9602330923080444,
|
| 71 |
+
"eval_Qnli-dev_cosine_ap": 0.587737015984062,
|
| 72 |
+
"eval_Qnli-dev_cosine_f1": 0.6358695652173914,
|
| 73 |
+
"eval_Qnli-dev_cosine_f1_threshold": 0.7741116285324097,
|
| 74 |
+
"eval_Qnli-dev_cosine_precision": 0.468,
|
| 75 |
+
"eval_Qnli-dev_cosine_recall": 0.9915254237288136,
|
| 76 |
+
"eval_Qnli-dev_dot_accuracy": 0.56640625,
|
| 77 |
+
"eval_Qnli-dev_dot_accuracy_threshold": 750.123779296875,
|
| 78 |
+
"eval_Qnli-dev_dot_ap": 0.5198069471819926,
|
| 79 |
+
"eval_Qnli-dev_dot_f1": 0.6318607764390897,
|
| 80 |
+
"eval_Qnli-dev_dot_f1_threshold": 387.37371826171875,
|
| 81 |
+
"eval_Qnli-dev_dot_precision": 0.461839530332681,
|
| 82 |
+
"eval_Qnli-dev_dot_recall": 1.0,
|
| 83 |
+
"eval_Qnli-dev_euclidean_accuracy": 0.6171875,
|
| 84 |
+
"eval_Qnli-dev_euclidean_accuracy_threshold": 7.253280162811279,
|
| 85 |
+
"eval_Qnli-dev_euclidean_ap": 0.5993562103857255,
|
| 86 |
+
"eval_Qnli-dev_euclidean_f1": 0.6376021798365122,
|
| 87 |
+
"eval_Qnli-dev_euclidean_f1_threshold": 17.42431640625,
|
| 88 |
+
"eval_Qnli-dev_euclidean_precision": 0.46987951807228917,
|
| 89 |
+
"eval_Qnli-dev_euclidean_recall": 0.9915254237288136,
|
| 90 |
+
"eval_Qnli-dev_manhattan_accuracy": 0.625,
|
| 91 |
+
"eval_Qnli-dev_manhattan_accuracy_threshold": 212.0604705810547,
|
| 92 |
+
"eval_Qnli-dev_manhattan_ap": 0.609409682371795,
|
| 93 |
+
"eval_Qnli-dev_manhattan_f1": 0.6358695652173914,
|
| 94 |
+
"eval_Qnli-dev_manhattan_f1_threshold": 461.66192626953125,
|
| 95 |
+
"eval_Qnli-dev_manhattan_precision": 0.468,
|
| 96 |
+
"eval_Qnli-dev_manhattan_recall": 0.9915254237288136,
|
| 97 |
+
"eval_Qnli-dev_max_accuracy": 0.625,
|
| 98 |
+
"eval_Qnli-dev_max_accuracy_threshold": 750.123779296875,
|
| 99 |
+
"eval_Qnli-dev_max_ap": 0.609409682371795,
|
| 100 |
+
"eval_Qnli-dev_max_f1": 0.6376021798365122,
|
| 101 |
+
"eval_Qnli-dev_max_f1_threshold": 461.66192626953125,
|
| 102 |
+
"eval_Qnli-dev_max_precision": 0.46987951807228917,
|
| 103 |
+
"eval_Qnli-dev_max_recall": 1.0,
|
| 104 |
+
"eval_allNLI-dev_cosine_accuracy": 0.666015625,
|
| 105 |
+
"eval_allNLI-dev_cosine_accuracy_threshold": 0.9931652545928955,
|
| 106 |
+
"eval_allNLI-dev_cosine_ap": 0.3330277035108763,
|
| 107 |
+
"eval_allNLI-dev_cosine_f1": 0.5088235294117648,
|
| 108 |
+
"eval_allNLI-dev_cosine_f1_threshold": 0.40660202503204346,
|
| 109 |
+
"eval_allNLI-dev_cosine_precision": 0.34122287968441817,
|
| 110 |
+
"eval_allNLI-dev_cosine_recall": 1.0,
|
| 111 |
+
"eval_allNLI-dev_dot_accuracy": 0.662109375,
|
| 112 |
+
"eval_allNLI-dev_dot_accuracy_threshold": 810.7718505859375,
|
| 113 |
+
"eval_allNLI-dev_dot_ap": 0.2980492563498839,
|
| 114 |
+
"eval_allNLI-dev_dot_f1": 0.5088235294117648,
|
| 115 |
+
"eval_allNLI-dev_dot_f1_threshold": 294.67071533203125,
|
| 116 |
+
"eval_allNLI-dev_dot_precision": 0.34122287968441817,
|
| 117 |
+
"eval_allNLI-dev_dot_recall": 1.0,
|
| 118 |
+
"eval_allNLI-dev_euclidean_accuracy": 0.666015625,
|
| 119 |
+
"eval_allNLI-dev_euclidean_accuracy_threshold": 3.3851852416992188,
|
| 120 |
+
"eval_allNLI-dev_euclidean_ap": 0.33790858355605125,
|
| 121 |
+
"eval_allNLI-dev_euclidean_f1": 0.5095729013254787,
|
| 122 |
+
"eval_allNLI-dev_euclidean_f1_threshold": 28.78042984008789,
|
| 123 |
+
"eval_allNLI-dev_euclidean_precision": 0.34189723320158105,
|
| 124 |
+
"eval_allNLI-dev_euclidean_recall": 1.0,
|
| 125 |
+
"eval_allNLI-dev_manhattan_accuracy": 0.666015625,
|
| 126 |
+
"eval_allNLI-dev_manhattan_accuracy_threshold": 98.2369613647461,
|
| 127 |
+
"eval_allNLI-dev_manhattan_ap": 0.3452833632811151,
|
| 128 |
+
"eval_allNLI-dev_manhattan_f1": 0.5091463414634146,
|
| 129 |
+
"eval_allNLI-dev_manhattan_f1_threshold": 626.5145874023438,
|
| 130 |
+
"eval_allNLI-dev_manhattan_precision": 0.34575569358178054,
|
| 131 |
+
"eval_allNLI-dev_manhattan_recall": 0.9653179190751445,
|
| 132 |
+
"eval_allNLI-dev_max_accuracy": 0.666015625,
|
| 133 |
+
"eval_allNLI-dev_max_accuracy_threshold": 810.7718505859375,
|
| 134 |
+
"eval_allNLI-dev_max_ap": 0.3452833632811151,
|
| 135 |
+
"eval_allNLI-dev_max_f1": 0.5095729013254787,
|
| 136 |
+
"eval_allNLI-dev_max_f1_threshold": 626.5145874023438,
|
| 137 |
+
"eval_allNLI-dev_max_precision": 0.34575569358178054,
|
| 138 |
+
"eval_allNLI-dev_max_recall": 1.0,
|
| 139 |
+
"eval_sequential_score": 0.609409682371795,
|
| 140 |
+
"eval_sts-test_pearson_cosine": 0.24255956652335486,
|
| 141 |
+
"eval_sts-test_pearson_dot": 0.18022025813169615,
|
| 142 |
+
"eval_sts-test_pearson_euclidean": 0.28657386934462287,
|
| 143 |
+
"eval_sts-test_pearson_manhattan": 0.32587411772454034,
|
| 144 |
+
"eval_sts-test_pearson_max": 0.32587411772454034,
|
| 145 |
+
"eval_sts-test_spearman_cosine": 0.32225123770630715,
|
| 146 |
+
"eval_sts-test_spearman_dot": 0.17184726031108927,
|
| 147 |
+
"eval_sts-test_spearman_euclidean": 0.3279675400801376,
|
| 148 |
+
"eval_sts-test_spearman_manhattan": 0.3635547276663291,
|
| 149 |
+
"eval_sts-test_spearman_max": 0.3635547276663291,
|
| 150 |
+
"eval_vitaminc-pairs_loss": 4.122440338134766,
|
| 151 |
+
"eval_vitaminc-pairs_runtime": 4.7337,
|
| 152 |
+
"eval_vitaminc-pairs_samples_per_second": 27.04,
|
| 153 |
+
"eval_vitaminc-pairs_steps_per_second": 0.211,
|
| 154 |
+
"step": 26
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.05012853470437018,
|
| 158 |
+
"eval_negation-triplets_loss": 3.917590379714966,
|
| 159 |
+
"eval_negation-triplets_runtime": 3.3679,
|
| 160 |
+
"eval_negation-triplets_samples_per_second": 38.006,
|
| 161 |
+
"eval_negation-triplets_steps_per_second": 0.297,
|
| 162 |
+
"step": 26
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 0.05012853470437018,
|
| 166 |
+
"eval_scitail-pairs-pos_loss": 2.5062460899353027,
|
| 167 |
+
"eval_scitail-pairs-pos_runtime": 2.7274,
|
| 168 |
+
"eval_scitail-pairs-pos_samples_per_second": 46.931,
|
| 169 |
+
"eval_scitail-pairs-pos_steps_per_second": 0.367,
|
| 170 |
+
"step": 26
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.05012853470437018,
|
| 174 |
+
"eval_scitail-pairs-qa_loss": 2.978612184524536,
|
| 175 |
+
"eval_scitail-pairs-qa_runtime": 2.2452,
|
| 176 |
+
"eval_scitail-pairs-qa_samples_per_second": 57.011,
|
| 177 |
+
"eval_scitail-pairs-qa_steps_per_second": 0.445,
|
| 178 |
+
"step": 26
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.05012853470437018,
|
| 182 |
+
"eval_xsum-pairs_loss": 5.394700050354004,
|
| 183 |
+
"eval_xsum-pairs_runtime": 3.0886,
|
| 184 |
+
"eval_xsum-pairs_samples_per_second": 41.442,
|
| 185 |
+
"eval_xsum-pairs_steps_per_second": 0.324,
|
| 186 |
+
"step": 26
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 0.05012853470437018,
|
| 190 |
+
"eval_sciq_pairs_loss": 0.8977569937705994,
|
| 191 |
+
"eval_sciq_pairs_runtime": 4.7516,
|
| 192 |
+
"eval_sciq_pairs_samples_per_second": 26.938,
|
| 193 |
+
"eval_sciq_pairs_steps_per_second": 0.21,
|
| 194 |
+
"step": 26
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.05012853470437018,
|
| 198 |
+
"eval_qasc_pairs_loss": 6.6808977127075195,
|
| 199 |
+
"eval_qasc_pairs_runtime": 2.1087,
|
| 200 |
+
"eval_qasc_pairs_samples_per_second": 60.701,
|
| 201 |
+
"eval_qasc_pairs_steps_per_second": 0.474,
|
| 202 |
+
"step": 26
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.05012853470437018,
|
| 206 |
+
"eval_openbookqa_pairs_loss": 6.402599334716797,
|
| 207 |
+
"eval_openbookqa_pairs_runtime": 2.2222,
|
| 208 |
+
"eval_openbookqa_pairs_samples_per_second": 57.601,
|
| 209 |
+
"eval_openbookqa_pairs_steps_per_second": 0.45,
|
| 210 |
+
"step": 26
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 0.05012853470437018,
|
| 214 |
+
"eval_msmarco_pairs_loss": 15.782674789428711,
|
| 215 |
+
"eval_msmarco_pairs_runtime": 2.307,
|
| 216 |
+
"eval_msmarco_pairs_samples_per_second": 55.483,
|
| 217 |
+
"eval_msmarco_pairs_steps_per_second": 0.433,
|
| 218 |
+
"step": 26
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.05012853470437018,
|
| 222 |
+
"eval_nq_pairs_loss": 10.447425842285156,
|
| 223 |
+
"eval_nq_pairs_runtime": 3.6239,
|
| 224 |
+
"eval_nq_pairs_samples_per_second": 35.321,
|
| 225 |
+
"eval_nq_pairs_steps_per_second": 0.276,
|
| 226 |
+
"step": 26
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.05012853470437018,
|
| 230 |
+
"eval_trivia_pairs_loss": 6.107333183288574,
|
| 231 |
+
"eval_trivia_pairs_runtime": 3.1299,
|
| 232 |
+
"eval_trivia_pairs_samples_per_second": 40.895,
|
| 233 |
+
"eval_trivia_pairs_steps_per_second": 0.319,
|
| 234 |
+
"step": 26
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.05012853470437018,
|
| 238 |
+
"eval_gooaq_pairs_loss": 8.095273971557617,
|
| 239 |
+
"eval_gooaq_pairs_runtime": 2.1098,
|
| 240 |
+
"eval_gooaq_pairs_samples_per_second": 60.669,
|
| 241 |
+
"eval_gooaq_pairs_steps_per_second": 0.474,
|
| 242 |
+
"step": 26
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.05012853470437018,
|
| 246 |
+
"eval_paws-pos_loss": 1.6077361106872559,
|
| 247 |
+
"eval_paws-pos_runtime": 2.2412,
|
| 248 |
+
"eval_paws-pos_samples_per_second": 57.112,
|
| 249 |
+
"eval_paws-pos_steps_per_second": 0.446,
|
| 250 |
+
"step": 26
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.05012853470437018,
|
| 254 |
+
"eval_global_dataset_loss": 6.348417282104492,
|
| 255 |
+
"eval_global_dataset_runtime": 8.3683,
|
| 256 |
+
"eval_global_dataset_samples_per_second": 38.837,
|
| 257 |
+
"eval_global_dataset_steps_per_second": 0.358,
|
| 258 |
+
"step": 26
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 0.05205655526992288,
|
| 262 |
+
"grad_norm": 172.1571807861328,
|
| 263 |
+
"learning_rate": 1.1583011583011583e-06,
|
| 264 |
+
"loss": 7.9729,
|
| 265 |
+
"step": 27
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.05784061696658098,
|
| 269 |
+
"grad_norm": 106.302001953125,
|
| 270 |
+
"learning_rate": 1.3320463320463321e-06,
|
| 271 |
+
"loss": 6.0587,
|
| 272 |
+
"step": 30
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.06362467866323908,
|
| 276 |
+
"grad_norm": 72.31057739257812,
|
| 277 |
+
"learning_rate": 1.5057915057915057e-06,
|
| 278 |
+
"loss": 5.6742,
|
| 279 |
+
"step": 33
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.06940874035989718,
|
| 283 |
+
"grad_norm": 258.2577209472656,
|
| 284 |
+
"learning_rate": 1.6795366795366796e-06,
|
| 285 |
+
"loss": 6.5406,
|
| 286 |
+
"step": 36
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 0.07519280205655526,
|
| 290 |
+
"grad_norm": 156.0019073486328,
|
| 291 |
+
"learning_rate": 1.7953667953667953e-06,
|
| 292 |
+
"loss": 5.4429,
|
| 293 |
+
"step": 39
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 0.08097686375321336,
|
| 297 |
+
"grad_norm": 227.6062469482422,
|
| 298 |
+
"learning_rate": 1.969111969111969e-06,
|
| 299 |
+
"loss": 6.7855,
|
| 300 |
+
"step": 42
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 0.08676092544987146,
|
| 304 |
+
"grad_norm": 221.12347412109375,
|
| 305 |
+
"learning_rate": 2.1428571428571427e-06,
|
| 306 |
+
"loss": 5.3403,
|
| 307 |
+
"step": 45
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 0.09254498714652956,
|
| 311 |
+
"grad_norm": 363.6319274902344,
|
| 312 |
+
"learning_rate": 2.3166023166023166e-06,
|
| 313 |
+
"loss": 4.2282,
|
| 314 |
+
"step": 48
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.09832904884318766,
|
| 318 |
+
"grad_norm": 119.22310638427734,
|
| 319 |
+
"learning_rate": 2.4903474903474904e-06,
|
| 320 |
+
"loss": 4.7411,
|
| 321 |
+
"step": 51
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.10025706940874037,
|
| 325 |
+
"eval_Qnli-dev_cosine_accuracy": 0.58984375,
|
| 326 |
+
"eval_Qnli-dev_cosine_accuracy_threshold": 0.9407284259796143,
|
| 327 |
+
"eval_Qnli-dev_cosine_ap": 0.5957496162735323,
|
| 328 |
+
"eval_Qnli-dev_cosine_f1": 0.6314363143631436,
|
| 329 |
+
"eval_Qnli-dev_cosine_f1_threshold": 0.7475543022155762,
|
| 330 |
+
"eval_Qnli-dev_cosine_precision": 0.4641434262948207,
|
| 331 |
+
"eval_Qnli-dev_cosine_recall": 0.9872881355932204,
|
| 332 |
+
"eval_Qnli-dev_dot_accuracy": 0.591796875,
|
| 333 |
+
"eval_Qnli-dev_dot_accuracy_threshold": 787.171630859375,
|
| 334 |
+
"eval_Qnli-dev_dot_ap": 0.556540262374452,
|
| 335 |
+
"eval_Qnli-dev_dot_f1": 0.6328767123287672,
|
| 336 |
+
"eval_Qnli-dev_dot_f1_threshold": 576.0565795898438,
|
| 337 |
+
"eval_Qnli-dev_dot_precision": 0.4676113360323887,
|
| 338 |
+
"eval_Qnli-dev_dot_recall": 0.9788135593220338,
|
| 339 |
+
"eval_Qnli-dev_euclidean_accuracy": 0.587890625,
|
| 340 |
+
"eval_Qnli-dev_euclidean_accuracy_threshold": 9.92269515991211,
|
| 341 |
+
"eval_Qnli-dev_euclidean_ap": 0.5913529071853763,
|
| 342 |
+
"eval_Qnli-dev_euclidean_f1": 0.6322930800542741,
|
| 343 |
+
"eval_Qnli-dev_euclidean_f1_threshold": 19.546676635742188,
|
| 344 |
+
"eval_Qnli-dev_euclidean_precision": 0.46506986027944114,
|
| 345 |
+
"eval_Qnli-dev_euclidean_recall": 0.9872881355932204,
|
| 346 |
+
"eval_Qnli-dev_manhattan_accuracy": 0.591796875,
|
| 347 |
+
"eval_Qnli-dev_manhattan_accuracy_threshold": 314.9248352050781,
|
| 348 |
+
"eval_Qnli-dev_manhattan_ap": 0.5944730102616272,
|
| 349 |
+
"eval_Qnli-dev_manhattan_f1": 0.632952691680261,
|
| 350 |
+
"eval_Qnli-dev_manhattan_f1_threshold": 424.9676818847656,
|
| 351 |
+
"eval_Qnli-dev_manhattan_precision": 0.5145888594164456,
|
| 352 |
+
"eval_Qnli-dev_manhattan_recall": 0.8220338983050848,
|
| 353 |
+
"eval_Qnli-dev_max_accuracy": 0.591796875,
|
| 354 |
+
"eval_Qnli-dev_max_accuracy_threshold": 787.171630859375,
|
| 355 |
+
"eval_Qnli-dev_max_ap": 0.5957496162735323,
|
| 356 |
+
"eval_Qnli-dev_max_f1": 0.632952691680261,
|
| 357 |
+
"eval_Qnli-dev_max_f1_threshold": 576.0565795898438,
|
| 358 |
+
"eval_Qnli-dev_max_precision": 0.5145888594164456,
|
| 359 |
+
"eval_Qnli-dev_max_recall": 0.9872881355932204,
|
| 360 |
+
"eval_allNLI-dev_cosine_accuracy": 0.666015625,
|
| 361 |
+
"eval_allNLI-dev_cosine_accuracy_threshold": 0.9878466129302979,
|
| 362 |
+
"eval_allNLI-dev_cosine_ap": 0.3898337405414008,
|
| 363 |
+
"eval_allNLI-dev_cosine_f1": 0.5065885797950219,
|
| 364 |
+
"eval_allNLI-dev_cosine_f1_threshold": 0.6534674167633057,
|
| 365 |
+
"eval_allNLI-dev_cosine_precision": 0.3392156862745098,
|
| 366 |
+
"eval_allNLI-dev_cosine_recall": 1.0,
|
| 367 |
+
"eval_allNLI-dev_dot_accuracy": 0.666015625,
|
| 368 |
+
"eval_allNLI-dev_dot_accuracy_threshold": 925.5989990234375,
|
| 369 |
+
"eval_allNLI-dev_dot_ap": 0.3391870470132937,
|
| 370 |
+
"eval_allNLI-dev_dot_f1": 0.5046153846153847,
|
| 371 |
+
"eval_allNLI-dev_dot_f1_threshold": 608.7012939453125,
|
| 372 |
+
"eval_allNLI-dev_dot_precision": 0.3438155136268344,
|
| 373 |
+
"eval_allNLI-dev_dot_recall": 0.9479768786127167,
|
| 374 |
+
"eval_allNLI-dev_euclidean_accuracy": 0.666015625,
|
| 375 |
+
"eval_allNLI-dev_euclidean_accuracy_threshold": 4.708977699279785,
|
| 376 |
+
"eval_allNLI-dev_euclidean_ap": 0.39556533875401867,
|
| 377 |
+
"eval_allNLI-dev_euclidean_f1": 0.5080763582966226,
|
| 378 |
+
"eval_allNLI-dev_euclidean_f1_threshold": 23.160057067871094,
|
| 379 |
+
"eval_allNLI-dev_euclidean_precision": 0.3405511811023622,
|
| 380 |
+
"eval_allNLI-dev_euclidean_recall": 1.0,
|
| 381 |
+
"eval_allNLI-dev_manhattan_accuracy": 0.666015625,
|
| 382 |
+
"eval_allNLI-dev_manhattan_accuracy_threshold": 142.7260284423828,
|
| 383 |
+
"eval_allNLI-dev_manhattan_ap": 0.4050661514301679,
|
| 384 |
+
"eval_allNLI-dev_manhattan_f1": 0.5118483412322274,
|
| 385 |
+
"eval_allNLI-dev_manhattan_f1_threshold": 504.7154846191406,
|
| 386 |
+
"eval_allNLI-dev_manhattan_precision": 0.3521739130434783,
|
| 387 |
+
"eval_allNLI-dev_manhattan_recall": 0.9364161849710982,
|
| 388 |
+
"eval_allNLI-dev_max_accuracy": 0.666015625,
|
| 389 |
+
"eval_allNLI-dev_max_accuracy_threshold": 925.5989990234375,
|
| 390 |
+
"eval_allNLI-dev_max_ap": 0.4050661514301679,
|
| 391 |
+
"eval_allNLI-dev_max_f1": 0.5118483412322274,
|
| 392 |
+
"eval_allNLI-dev_max_f1_threshold": 608.7012939453125,
|
| 393 |
+
"eval_allNLI-dev_max_precision": 0.3521739130434783,
|
| 394 |
+
"eval_allNLI-dev_max_recall": 1.0,
|
| 395 |
+
"eval_sequential_score": 0.5957496162735323,
|
| 396 |
+
"eval_sts-test_pearson_cosine": 0.5067228548513025,
|
| 397 |
+
"eval_sts-test_pearson_dot": 0.28867418819353924,
|
| 398 |
+
"eval_sts-test_pearson_euclidean": 0.5315956249457565,
|
| 399 |
+
"eval_sts-test_pearson_manhattan": 0.5620907203006935,
|
| 400 |
+
"eval_sts-test_pearson_max": 0.5620907203006935,
|
| 401 |
+
"eval_sts-test_spearman_cosine": 0.5257409961657202,
|
| 402 |
+
"eval_sts-test_spearman_dot": 0.280153994622986,
|
| 403 |
+
"eval_sts-test_spearman_euclidean": 0.5306475244646732,
|
| 404 |
+
"eval_sts-test_spearman_manhattan": 0.5605804796765843,
|
| 405 |
+
"eval_sts-test_spearman_max": 0.5605804796765843,
|
| 406 |
+
"eval_vitaminc-pairs_loss": 3.929797887802124,
|
| 407 |
+
"eval_vitaminc-pairs_runtime": 4.6018,
|
| 408 |
+
"eval_vitaminc-pairs_samples_per_second": 27.815,
|
| 409 |
+
"eval_vitaminc-pairs_steps_per_second": 0.217,
|
| 410 |
+
"step": 52
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.10025706940874037,
|
| 414 |
+
"eval_negation-triplets_loss": 3.267047166824341,
|
| 415 |
+
"eval_negation-triplets_runtime": 3.2813,
|
| 416 |
+
"eval_negation-triplets_samples_per_second": 39.009,
|
| 417 |
+
"eval_negation-triplets_steps_per_second": 0.305,
|
| 418 |
+
"step": 52
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 0.10025706940874037,
|
| 422 |
+
"eval_scitail-pairs-pos_loss": 1.134178876876831,
|
| 423 |
+
"eval_scitail-pairs-pos_runtime": 2.6261,
|
| 424 |
+
"eval_scitail-pairs-pos_samples_per_second": 48.742,
|
| 425 |
+
"eval_scitail-pairs-pos_steps_per_second": 0.381,
|
| 426 |
+
"step": 52
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 0.10025706940874037,
|
| 430 |
+
"eval_scitail-pairs-qa_loss": 2.0085601806640625,
|
| 431 |
+
"eval_scitail-pairs-qa_runtime": 2.2161,
|
| 432 |
+
"eval_scitail-pairs-qa_samples_per_second": 57.759,
|
| 433 |
+
"eval_scitail-pairs-qa_steps_per_second": 0.451,
|
| 434 |
+
"step": 52
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.10025706940874037,
|
| 438 |
+
"eval_xsum-pairs_loss": 4.462350845336914,
|
| 439 |
+
"eval_xsum-pairs_runtime": 3.1126,
|
| 440 |
+
"eval_xsum-pairs_samples_per_second": 41.124,
|
| 441 |
+
"eval_xsum-pairs_steps_per_second": 0.321,
|
| 442 |
+
"step": 52
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 0.10025706940874037,
|
| 446 |
+
"eval_sciq_pairs_loss": 0.71834796667099,
|
| 447 |
+
"eval_sciq_pairs_runtime": 4.7505,
|
| 448 |
+
"eval_sciq_pairs_samples_per_second": 26.944,
|
| 449 |
+
"eval_sciq_pairs_steps_per_second": 0.211,
|
| 450 |
+
"step": 52
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.10025706940874037,
|
| 454 |
+
"eval_qasc_pairs_loss": 4.889453887939453,
|
| 455 |
+
"eval_qasc_pairs_runtime": 2.1101,
|
| 456 |
+
"eval_qasc_pairs_samples_per_second": 60.662,
|
| 457 |
+
"eval_qasc_pairs_steps_per_second": 0.474,
|
| 458 |
+
"step": 52
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.10025706940874037,
|
| 462 |
+
"eval_openbookqa_pairs_loss": 3.4273369312286377,
|
| 463 |
+
"eval_openbookqa_pairs_runtime": 2.2311,
|
| 464 |
+
"eval_openbookqa_pairs_samples_per_second": 57.371,
|
| 465 |
+
"eval_openbookqa_pairs_steps_per_second": 0.448,
|
| 466 |
+
"step": 52
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.10025706940874037,
|
| 470 |
+
"eval_msmarco_pairs_loss": 7.532540321350098,
|
| 471 |
+
"eval_msmarco_pairs_runtime": 2.2959,
|
| 472 |
+
"eval_msmarco_pairs_samples_per_second": 55.751,
|
| 473 |
+
"eval_msmarco_pairs_steps_per_second": 0.436,
|
| 474 |
+
"step": 52
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.10025706940874037,
|
| 478 |
+
"eval_nq_pairs_loss": 6.29541015625,
|
| 479 |
+
"eval_nq_pairs_runtime": 3.661,
|
| 480 |
+
"eval_nq_pairs_samples_per_second": 34.963,
|
| 481 |
+
"eval_nq_pairs_steps_per_second": 0.273,
|
| 482 |
+
"step": 52
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.10025706940874037,
|
| 486 |
+
"eval_trivia_pairs_loss": 5.789977073669434,
|
| 487 |
+
"eval_trivia_pairs_runtime": 3.0826,
|
| 488 |
+
"eval_trivia_pairs_samples_per_second": 41.523,
|
| 489 |
+
"eval_trivia_pairs_steps_per_second": 0.324,
|
| 490 |
+
"step": 52
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 0.10025706940874037,
|
| 494 |
+
"eval_gooaq_pairs_loss": 5.785163879394531,
|
| 495 |
+
"eval_gooaq_pairs_runtime": 2.1382,
|
| 496 |
+
"eval_gooaq_pairs_samples_per_second": 59.863,
|
| 497 |
+
"eval_gooaq_pairs_steps_per_second": 0.468,
|
| 498 |
+
"step": 52
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 0.10025706940874037,
|
| 502 |
+
"eval_paws-pos_loss": 0.47830095887184143,
|
| 503 |
+
"eval_paws-pos_runtime": 2.3113,
|
| 504 |
+
"eval_paws-pos_samples_per_second": 55.379,
|
| 505 |
+
"eval_paws-pos_steps_per_second": 0.433,
|
| 506 |
+
"step": 52
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.10025706940874037,
|
| 510 |
+
"eval_global_dataset_loss": 3.8913991451263428,
|
| 511 |
+
"eval_global_dataset_runtime": 8.3623,
|
| 512 |
+
"eval_global_dataset_samples_per_second": 38.865,
|
| 513 |
+
"eval_global_dataset_steps_per_second": 0.359,
|
| 514 |
+
"step": 52
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.10411311053984576,
|
| 518 |
+
"grad_norm": 55.83478546142578,
|
| 519 |
+
"learning_rate": 2.6640926640926642e-06,
|
| 520 |
+
"loss": 3.9082,
|
| 521 |
+
"step": 54
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.10989717223650386,
|
| 525 |
+
"grad_norm": 57.4188117980957,
|
| 526 |
+
"learning_rate": 2.7799227799227797e-06,
|
| 527 |
+
"loss": 4.3922,
|
| 528 |
+
"step": 57
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.11568123393316196,
|
| 532 |
+
"grad_norm": 116.89701843261719,
|
| 533 |
+
"learning_rate": 2.9536679536679535e-06,
|
| 534 |
+
"loss": 3.2655,
|
| 535 |
+
"step": 60
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.12146529562982006,
|
| 539 |
+
"grad_norm": 53.49580001831055,
|
| 540 |
+
"learning_rate": 3.1274131274131274e-06,
|
| 541 |
+
"loss": 3.1043,
|
| 542 |
+
"step": 63
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.12724935732647816,
|
| 546 |
+
"grad_norm": 51.044410705566406,
|
| 547 |
+
"learning_rate": 3.3011583011583012e-06,
|
| 548 |
+
"loss": 2.2074,
|
| 549 |
+
"step": 66
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.13303341902313626,
|
| 553 |
+
"grad_norm": 139.9288787841797,
|
| 554 |
+
"learning_rate": 3.474903474903475e-06,
|
| 555 |
+
"loss": 1.4414,
|
| 556 |
+
"step": 69
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.13881748071979436,
|
| 560 |
+
"grad_norm": 37.221168518066406,
|
| 561 |
+
"learning_rate": 3.5907335907335905e-06,
|
| 562 |
+
"loss": 1.5937,
|
| 563 |
+
"step": 72
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.14460154241645246,
|
| 567 |
+
"grad_norm": 25.855661392211914,
|
| 568 |
+
"learning_rate": 3.7644787644787644e-06,
|
| 569 |
+
"loss": 1.0306,
|
| 570 |
+
"step": 75
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.15038560411311053,
|
| 574 |
+
"grad_norm": 29.79730796813965,
|
| 575 |
+
"learning_rate": 3.938223938223938e-06,
|
| 576 |
+
"loss": 1.0784,
|
| 577 |
+
"step": 78
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.15038560411311053,
|
| 581 |
+
"eval_Qnli-dev_cosine_accuracy": 0.638671875,
|
| 582 |
+
"eval_Qnli-dev_cosine_accuracy_threshold": 0.856202244758606,
|
| 583 |
+
"eval_Qnli-dev_cosine_ap": 0.6292803059732273,
|
| 584 |
+
"eval_Qnli-dev_cosine_f1": 0.6408839779005525,
|
| 585 |
+
"eval_Qnli-dev_cosine_f1_threshold": 0.6206527352333069,
|
| 586 |
+
"eval_Qnli-dev_cosine_precision": 0.47540983606557374,
|
| 587 |
+
"eval_Qnli-dev_cosine_recall": 0.9830508474576272,
|
| 588 |
+
"eval_Qnli-dev_dot_accuracy": 0.626953125,
|
| 589 |
+
"eval_Qnli-dev_dot_accuracy_threshold": 887.830810546875,
|
| 590 |
+
"eval_Qnli-dev_dot_ap": 0.592075047542057,
|
| 591 |
+
"eval_Qnli-dev_dot_f1": 0.640990371389271,
|
| 592 |
+
"eval_Qnli-dev_dot_f1_threshold": 601.355224609375,
|
| 593 |
+
"eval_Qnli-dev_dot_precision": 0.4745417515274949,
|
| 594 |
+
"eval_Qnli-dev_dot_recall": 0.9872881355932204,
|
| 595 |
+
"eval_Qnli-dev_euclidean_accuracy": 0.638671875,
|
| 596 |
+
"eval_Qnli-dev_euclidean_accuracy_threshold": 17.094371795654297,
|
| 597 |
+
"eval_Qnli-dev_euclidean_ap": 0.6289192660680781,
|
| 598 |
+
"eval_Qnli-dev_euclidean_f1": 0.6423357664233577,
|
| 599 |
+
"eval_Qnli-dev_euclidean_f1_threshold": 24.894683837890625,
|
| 600 |
+
"eval_Qnli-dev_euclidean_precision": 0.48997772828507796,
|
| 601 |
+
"eval_Qnli-dev_euclidean_recall": 0.9322033898305084,
|
| 602 |
+
"eval_Qnli-dev_manhattan_accuracy": 0.640625,
|
| 603 |
+
"eval_Qnli-dev_manhattan_accuracy_threshold": 526.0111083984375,
|
| 604 |
+
"eval_Qnli-dev_manhattan_ap": 0.6256143574081243,
|
| 605 |
+
"eval_Qnli-dev_manhattan_f1": 0.6417704011065007,
|
| 606 |
+
"eval_Qnli-dev_manhattan_f1_threshold": 853.242431640625,
|
| 607 |
+
"eval_Qnli-dev_manhattan_precision": 0.47638603696098564,
|
| 608 |
+
"eval_Qnli-dev_manhattan_recall": 0.9830508474576272,
|
| 609 |
+
"eval_Qnli-dev_max_accuracy": 0.640625,
|
| 610 |
+
"eval_Qnli-dev_max_accuracy_threshold": 887.830810546875,
|
| 611 |
+
"eval_Qnli-dev_max_ap": 0.6292803059732273,
|
| 612 |
+
"eval_Qnli-dev_max_f1": 0.6423357664233577,
|
| 613 |
+
"eval_Qnli-dev_max_f1_threshold": 853.242431640625,
|
| 614 |
+
"eval_Qnli-dev_max_precision": 0.48997772828507796,
|
| 615 |
+
"eval_Qnli-dev_max_recall": 0.9872881355932204,
|
| 616 |
+
"eval_allNLI-dev_cosine_accuracy": 0.6796875,
|
| 617 |
+
"eval_allNLI-dev_cosine_accuracy_threshold": 0.9426460862159729,
|
| 618 |
+
"eval_allNLI-dev_cosine_ap": 0.473508692504868,
|
| 619 |
+
"eval_allNLI-dev_cosine_f1": 0.5328947368421053,
|
| 620 |
+
"eval_allNLI-dev_cosine_f1_threshold": 0.6884603500366211,
|
| 621 |
+
"eval_allNLI-dev_cosine_precision": 0.3724137931034483,
|
| 622 |
+
"eval_allNLI-dev_cosine_recall": 0.9364161849710982,
|
| 623 |
+
"eval_allNLI-dev_dot_accuracy": 0.673828125,
|
| 624 |
+
"eval_allNLI-dev_dot_accuracy_threshold": 980.9593505859375,
|
| 625 |
+
"eval_allNLI-dev_dot_ap": 0.44188020452374555,
|
| 626 |
+
"eval_allNLI-dev_dot_f1": 0.53156146179402,
|
| 627 |
+
"eval_allNLI-dev_dot_f1_threshold": 683.7774658203125,
|
| 628 |
+
"eval_allNLI-dev_dot_precision": 0.372960372960373,
|
| 629 |
+
"eval_allNLI-dev_dot_recall": 0.9248554913294798,
|
| 630 |
+
"eval_allNLI-dev_euclidean_accuracy": 0.6796875,
|
| 631 |
+
"eval_allNLI-dev_euclidean_accuracy_threshold": 10.647405624389648,
|
| 632 |
+
"eval_allNLI-dev_euclidean_ap": 0.47440611686548606,
|
| 633 |
+
"eval_allNLI-dev_euclidean_f1": 0.5292207792207793,
|
| 634 |
+
"eval_allNLI-dev_euclidean_f1_threshold": 25.08370590209961,
|
| 635 |
+
"eval_allNLI-dev_euclidean_precision": 0.36794582392776526,
|
| 636 |
+
"eval_allNLI-dev_euclidean_recall": 0.9421965317919075,
|
| 637 |
+
"eval_allNLI-dev_manhattan_accuracy": 0.6796875,
|
| 638 |
+
"eval_allNLI-dev_manhattan_accuracy_threshold": 318.408203125,
|
| 639 |
+
"eval_allNLI-dev_manhattan_ap": 0.4747137405292771,
|
| 640 |
+
"eval_allNLI-dev_manhattan_f1": 0.5305343511450381,
|
| 641 |
+
"eval_allNLI-dev_manhattan_f1_threshold": 697.8466796875,
|
| 642 |
+
"eval_allNLI-dev_manhattan_precision": 0.396011396011396,
|
| 643 |
+
"eval_allNLI-dev_manhattan_recall": 0.8034682080924855,
|
| 644 |
+
"eval_allNLI-dev_max_accuracy": 0.6796875,
|
| 645 |
+
"eval_allNLI-dev_max_accuracy_threshold": 980.9593505859375,
|
| 646 |
+
"eval_allNLI-dev_max_ap": 0.4747137405292771,
|
| 647 |
+
"eval_allNLI-dev_max_f1": 0.5328947368421053,
|
| 648 |
+
"eval_allNLI-dev_max_f1_threshold": 697.8466796875,
|
| 649 |
+
"eval_allNLI-dev_max_precision": 0.396011396011396,
|
| 650 |
+
"eval_allNLI-dev_max_recall": 0.9421965317919075,
|
| 651 |
+
"eval_sequential_score": 0.6292803059732273,
|
| 652 |
+
"eval_sts-test_pearson_cosine": 0.8216305666941104,
|
| 653 |
+
"eval_sts-test_pearson_dot": 0.7893058557928705,
|
| 654 |
+
"eval_sts-test_pearson_euclidean": 0.841701309677291,
|
| 655 |
+
"eval_sts-test_pearson_manhattan": 0.8438559225680149,
|
| 656 |
+
"eval_sts-test_pearson_max": 0.8438559225680149,
|
| 657 |
+
"eval_sts-test_spearman_cosine": 0.8291168678373173,
|
| 658 |
+
"eval_sts-test_spearman_dot": 0.7782024179758329,
|
| 659 |
+
"eval_sts-test_spearman_euclidean": 0.829029110613729,
|
| 660 |
+
"eval_sts-test_spearman_manhattan": 0.8308224279114665,
|
| 661 |
+
"eval_sts-test_spearman_max": 0.8308224279114665,
|
| 662 |
+
"eval_vitaminc-pairs_loss": 3.763850212097168,
|
| 663 |
+
"eval_vitaminc-pairs_runtime": 4.6142,
|
| 664 |
+
"eval_vitaminc-pairs_samples_per_second": 27.741,
|
| 665 |
+
"eval_vitaminc-pairs_steps_per_second": 0.217,
|
| 666 |
+
"step": 78
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 0.15038560411311053,
|
| 670 |
+
"eval_negation-triplets_loss": 1.4772971868515015,
|
| 671 |
+
"eval_negation-triplets_runtime": 3.4145,
|
| 672 |
+
"eval_negation-triplets_samples_per_second": 37.488,
|
| 673 |
+
"eval_negation-triplets_steps_per_second": 0.293,
|
| 674 |
+
"step": 78
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.15038560411311053,
|
| 678 |
+
"eval_scitail-pairs-pos_loss": 0.11086193472146988,
|
| 679 |
+
"eval_scitail-pairs-pos_runtime": 2.8091,
|
| 680 |
+
"eval_scitail-pairs-pos_samples_per_second": 45.565,
|
| 681 |
+
"eval_scitail-pairs-pos_steps_per_second": 0.356,
|
| 682 |
+
"step": 78
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.15038560411311053,
|
| 686 |
+
"eval_scitail-pairs-qa_loss": 0.12822097539901733,
|
| 687 |
+
"eval_scitail-pairs-qa_runtime": 2.312,
|
| 688 |
+
"eval_scitail-pairs-qa_samples_per_second": 55.364,
|
| 689 |
+
"eval_scitail-pairs-qa_steps_per_second": 0.433,
|
| 690 |
+
"step": 78
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 0.15038560411311053,
|
| 694 |
+
"eval_xsum-pairs_loss": 0.43598121404647827,
|
| 695 |
+
"eval_xsum-pairs_runtime": 3.1471,
|
| 696 |
+
"eval_xsum-pairs_samples_per_second": 40.673,
|
| 697 |
+
"eval_xsum-pairs_steps_per_second": 0.318,
|
| 698 |
+
"step": 78
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 0.15038560411311053,
|
| 702 |
+
"eval_sciq_pairs_loss": 0.20338551700115204,
|
| 703 |
+
"eval_sciq_pairs_runtime": 4.8592,
|
| 704 |
+
"eval_sciq_pairs_samples_per_second": 26.342,
|
| 705 |
+
"eval_sciq_pairs_steps_per_second": 0.206,
|
| 706 |
+
"step": 78
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 0.15038560411311053,
|
| 710 |
+
"eval_qasc_pairs_loss": 1.3059492111206055,
|
| 711 |
+
"eval_qasc_pairs_runtime": 2.1191,
|
| 712 |
+
"eval_qasc_pairs_samples_per_second": 60.403,
|
| 713 |
+
"eval_qasc_pairs_steps_per_second": 0.472,
|
| 714 |
+
"step": 78
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 0.15038560411311053,
|
| 718 |
+
"eval_openbookqa_pairs_loss": 1.1711227893829346,
|
| 719 |
+
"eval_openbookqa_pairs_runtime": 2.2105,
|
| 720 |
+
"eval_openbookqa_pairs_samples_per_second": 57.907,
|
| 721 |
+
"eval_openbookqa_pairs_steps_per_second": 0.452,
|
| 722 |
+
"step": 78
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 0.15038560411311053,
|
| 726 |
+
"eval_msmarco_pairs_loss": 1.3856205940246582,
|
| 727 |
+
"eval_msmarco_pairs_runtime": 2.3065,
|
| 728 |
+
"eval_msmarco_pairs_samples_per_second": 55.494,
|
| 729 |
+
"eval_msmarco_pairs_steps_per_second": 0.434,
|
| 730 |
+
"step": 78
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.15038560411311053,
|
| 734 |
+
"eval_nq_pairs_loss": 1.2713654041290283,
|
| 735 |
+
"eval_nq_pairs_runtime": 3.6729,
|
| 736 |
+
"eval_nq_pairs_samples_per_second": 34.85,
|
| 737 |
+
"eval_nq_pairs_steps_per_second": 0.272,
|
| 738 |
+
"step": 78
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.15038560411311053,
|
| 742 |
+
"eval_trivia_pairs_loss": 1.082784652709961,
|
| 743 |
+
"eval_trivia_pairs_runtime": 3.145,
|
| 744 |
+
"eval_trivia_pairs_samples_per_second": 40.7,
|
| 745 |
+
"eval_trivia_pairs_steps_per_second": 0.318,
|
| 746 |
+
"step": 78
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.15038560411311053,
|
| 750 |
+
"eval_gooaq_pairs_loss": 0.703183650970459,
|
| 751 |
+
"eval_gooaq_pairs_runtime": 2.1975,
|
| 752 |
+
"eval_gooaq_pairs_samples_per_second": 58.248,
|
| 753 |
+
"eval_gooaq_pairs_steps_per_second": 0.455,
|
| 754 |
+
"step": 78
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 0.15038560411311053,
|
| 758 |
+
"eval_paws-pos_loss": 0.04489962384104729,
|
| 759 |
+
"eval_paws-pos_runtime": 2.3636,
|
| 760 |
+
"eval_paws-pos_samples_per_second": 54.155,
|
| 761 |
+
"eval_paws-pos_steps_per_second": 0.423,
|
| 762 |
+
"step": 78
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 0.15038560411311053,
|
| 766 |
+
"eval_global_dataset_loss": 0.6713590621948242,
|
| 767 |
+
"eval_global_dataset_runtime": 8.4648,
|
| 768 |
+
"eval_global_dataset_samples_per_second": 38.394,
|
| 769 |
+
"eval_global_dataset_steps_per_second": 0.354,
|
| 770 |
+
"step": 78
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 0.15616966580976863,
|
| 774 |
+
"grad_norm": 22.163389205932617,
|
| 775 |
+
"learning_rate": 4.1119691119691125e-06,
|
| 776 |
+
"loss": 0.9674,
|
| 777 |
+
"step": 81
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 0.16195372750642673,
|
| 781 |
+
"grad_norm": 25.121131896972656,
|
| 782 |
+
"learning_rate": 4.2857142857142855e-06,
|
| 783 |
+
"loss": 0.9335,
|
| 784 |
+
"step": 84
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 0.16773778920308482,
|
| 788 |
+
"grad_norm": 44.308528900146484,
|
| 789 |
+
"learning_rate": 4.45945945945946e-06,
|
| 790 |
+
"loss": 0.8806,
|
| 791 |
+
"step": 87
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 0.17352185089974292,
|
| 795 |
+
"grad_norm": 9.972362518310547,
|
| 796 |
+
"learning_rate": 4.633204633204633e-06,
|
| 797 |
+
"loss": 0.631,
|
| 798 |
+
"step": 90
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 0.17930591259640102,
|
| 802 |
+
"grad_norm": 13.609597206115723,
|
| 803 |
+
"learning_rate": 4.806949806949808e-06,
|
| 804 |
+
"loss": 0.3384,
|
| 805 |
+
"step": 93
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"epoch": 0.18508997429305912,
|
| 809 |
+
"grad_norm": 6.892751693725586,
|
| 810 |
+
"learning_rate": 4.980694980694981e-06,
|
| 811 |
+
"loss": 0.404,
|
| 812 |
+
"step": 96
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 0.19087403598971722,
|
| 816 |
+
"grad_norm": 7.414526462554932,
|
| 817 |
+
"learning_rate": 5.154440154440155e-06,
|
| 818 |
+
"loss": 0.6488,
|
| 819 |
+
"step": 99
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 0.19665809768637532,
|
| 823 |
+
"grad_norm": 23.75099754333496,
|
| 824 |
+
"learning_rate": 5.3281853281853285e-06,
|
| 825 |
+
"loss": 0.4728,
|
| 826 |
+
"step": 102
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 0.20051413881748073,
|
| 830 |
+
"eval_Qnli-dev_cosine_accuracy": 0.693359375,
|
| 831 |
+
"eval_Qnli-dev_cosine_accuracy_threshold": 0.8276629447937012,
|
| 832 |
+
"eval_Qnli-dev_cosine_ap": 0.7127503402574797,
|
| 833 |
+
"eval_Qnli-dev_cosine_f1": 0.6730038022813688,
|
| 834 |
+
"eval_Qnli-dev_cosine_f1_threshold": 0.7761298418045044,
|
| 835 |
+
"eval_Qnli-dev_cosine_precision": 0.6103448275862069,
|
| 836 |
+
"eval_Qnli-dev_cosine_recall": 0.75,
|
| 837 |
+
"eval_Qnli-dev_dot_accuracy": 0.681640625,
|
| 838 |
+
"eval_Qnli-dev_dot_accuracy_threshold": 857.5540771484375,
|
| 839 |
+
"eval_Qnli-dev_dot_ap": 0.6617256136301952,
|
| 840 |
+
"eval_Qnli-dev_dot_f1": 0.6678141135972461,
|
| 841 |
+
"eval_Qnli-dev_dot_f1_threshold": 756.1309204101562,
|
| 842 |
+
"eval_Qnli-dev_dot_precision": 0.5623188405797102,
|
| 843 |
+
"eval_Qnli-dev_dot_recall": 0.8220338983050848,
|
| 844 |
+
"eval_Qnli-dev_euclidean_accuracy": 0.69140625,
|
| 845 |
+
"eval_Qnli-dev_euclidean_accuracy_threshold": 18.45541763305664,
|
| 846 |
+
"eval_Qnli-dev_euclidean_ap": 0.7154480594709788,
|
| 847 |
+
"eval_Qnli-dev_euclidean_f1": 0.6725043782837128,
|
| 848 |
+
"eval_Qnli-dev_euclidean_f1_threshold": 22.723068237304688,
|
| 849 |
+
"eval_Qnli-dev_euclidean_precision": 0.573134328358209,
|
| 850 |
+
"eval_Qnli-dev_euclidean_recall": 0.8135593220338984,
|
| 851 |
+
"eval_Qnli-dev_manhattan_accuracy": 0.69140625,
|
| 852 |
+
"eval_Qnli-dev_manhattan_accuracy_threshold": 585.3442993164062,
|
| 853 |
+
"eval_Qnli-dev_manhattan_ap": 0.713289230749026,
|
| 854 |
+
"eval_Qnli-dev_manhattan_f1": 0.6733333333333333,
|
| 855 |
+
"eval_Qnli-dev_manhattan_f1_threshold": 733.2077026367188,
|
| 856 |
+
"eval_Qnli-dev_manhattan_precision": 0.554945054945055,
|
| 857 |
+
"eval_Qnli-dev_manhattan_recall": 0.8559322033898306,
|
| 858 |
+
"eval_Qnli-dev_max_accuracy": 0.693359375,
|
| 859 |
+
"eval_Qnli-dev_max_accuracy_threshold": 857.5540771484375,
|
| 860 |
+
"eval_Qnli-dev_max_ap": 0.7154480594709788,
|
| 861 |
+
"eval_Qnli-dev_max_f1": 0.6733333333333333,
|
| 862 |
+
"eval_Qnli-dev_max_f1_threshold": 756.1309204101562,
|
| 863 |
+
"eval_Qnli-dev_max_precision": 0.6103448275862069,
|
| 864 |
+
"eval_Qnli-dev_max_recall": 0.8559322033898306,
|
| 865 |
+
"eval_allNLI-dev_cosine_accuracy": 0.693359375,
|
| 866 |
+
"eval_allNLI-dev_cosine_accuracy_threshold": 0.8820043802261353,
|
| 867 |
+
"eval_allNLI-dev_cosine_ap": 0.5380653098565689,
|
| 868 |
+
"eval_allNLI-dev_cosine_f1": 0.5689655172413792,
|
| 869 |
+
"eval_allNLI-dev_cosine_f1_threshold": 0.7308224439620972,
|
| 870 |
+
"eval_allNLI-dev_cosine_precision": 0.4536082474226804,
|
| 871 |
+
"eval_allNLI-dev_cosine_recall": 0.7630057803468208,
|
| 872 |
+
"eval_allNLI-dev_dot_accuracy": 0.6875,
|
| 873 |
+
"eval_allNLI-dev_dot_accuracy_threshold": 904.0134887695312,
|
| 874 |
+
"eval_allNLI-dev_dot_ap": 0.5175334662627952,
|
| 875 |
+
"eval_allNLI-dev_dot_f1": 0.5675675675675675,
|
| 876 |
+
"eval_allNLI-dev_dot_f1_threshold": 762.652587890625,
|
| 877 |
+
"eval_allNLI-dev_dot_precision": 0.46494464944649444,
|
| 878 |
+
"eval_allNLI-dev_dot_recall": 0.7283236994219653,
|
| 879 |
+
"eval_allNLI-dev_euclidean_accuracy": 0.6953125,
|
| 880 |
+
"eval_allNLI-dev_euclidean_accuracy_threshold": 15.50227165222168,
|
| 881 |
+
"eval_allNLI-dev_euclidean_ap": 0.5410854453600531,
|
| 882 |
+
"eval_allNLI-dev_euclidean_f1": 0.5645514223194749,
|
| 883 |
+
"eval_allNLI-dev_euclidean_f1_threshold": 23.451454162597656,
|
| 884 |
+
"eval_allNLI-dev_euclidean_precision": 0.45422535211267606,
|
| 885 |
+
"eval_allNLI-dev_euclidean_recall": 0.7456647398843931,
|
| 886 |
+
"eval_allNLI-dev_manhattan_accuracy": 0.6953125,
|
| 887 |
+
"eval_allNLI-dev_manhattan_accuracy_threshold": 518.75146484375,
|
| 888 |
+
"eval_allNLI-dev_manhattan_ap": 0.5430149701961973,
|
| 889 |
+
"eval_allNLI-dev_manhattan_f1": 0.5639913232104121,
|
| 890 |
+
"eval_allNLI-dev_manhattan_f1_threshold": 733.8897705078125,
|
| 891 |
+
"eval_allNLI-dev_manhattan_precision": 0.4513888888888889,
|
| 892 |
+
"eval_allNLI-dev_manhattan_recall": 0.7514450867052023,
|
| 893 |
+
"eval_allNLI-dev_max_accuracy": 0.6953125,
|
| 894 |
+
"eval_allNLI-dev_max_accuracy_threshold": 904.0134887695312,
|
| 895 |
+
"eval_allNLI-dev_max_ap": 0.5430149701961973,
|
| 896 |
+
"eval_allNLI-dev_max_f1": 0.5689655172413792,
|
| 897 |
+
"eval_allNLI-dev_max_f1_threshold": 762.652587890625,
|
| 898 |
+
"eval_allNLI-dev_max_precision": 0.46494464944649444,
|
| 899 |
+
"eval_allNLI-dev_max_recall": 0.7630057803468208,
|
| 900 |
+
"eval_sequential_score": 0.7154480594709788,
|
| 901 |
+
"eval_sts-test_pearson_cosine": 0.8841456597076958,
|
| 902 |
+
"eval_sts-test_pearson_dot": 0.8597263949983345,
|
| 903 |
+
"eval_sts-test_pearson_euclidean": 0.8977156894521804,
|
| 904 |
+
"eval_sts-test_pearson_manhattan": 0.8973099980705126,
|
| 905 |
+
"eval_sts-test_pearson_max": 0.8977156894521804,
|
| 906 |
+
"eval_sts-test_spearman_cosine": 0.8862653474156692,
|
| 907 |
+
"eval_sts-test_spearman_dot": 0.8451249310521576,
|
| 908 |
+
"eval_sts-test_spearman_euclidean": 0.8866065310325173,
|
| 909 |
+
"eval_sts-test_spearman_manhattan": 0.8860841943095571,
|
| 910 |
+
"eval_sts-test_spearman_max": 0.8866065310325173,
|
| 911 |
+
"eval_vitaminc-pairs_loss": 3.314167022705078,
|
| 912 |
+
"eval_vitaminc-pairs_runtime": 4.4886,
|
| 913 |
+
"eval_vitaminc-pairs_samples_per_second": 28.517,
|
| 914 |
+
"eval_vitaminc-pairs_steps_per_second": 0.223,
|
| 915 |
+
"step": 104
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 0.20051413881748073,
|
| 919 |
+
"eval_negation-triplets_loss": 1.3400739431381226,
|
| 920 |
+
"eval_negation-triplets_runtime": 3.3068,
|
| 921 |
+
"eval_negation-triplets_samples_per_second": 38.708,
|
| 922 |
+
"eval_negation-triplets_steps_per_second": 0.302,
|
| 923 |
+
"step": 104
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 0.20051413881748073,
|
| 927 |
+
"eval_scitail-pairs-pos_loss": 0.07263079285621643,
|
| 928 |
+
"eval_scitail-pairs-pos_runtime": 2.6213,
|
| 929 |
+
"eval_scitail-pairs-pos_samples_per_second": 48.83,
|
| 930 |
+
"eval_scitail-pairs-pos_steps_per_second": 0.381,
|
| 931 |
+
"step": 104
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 0.20051413881748073,
|
| 935 |
+
"eval_scitail-pairs-qa_loss": 0.048478204756975174,
|
| 936 |
+
"eval_scitail-pairs-qa_runtime": 2.202,
|
| 937 |
+
"eval_scitail-pairs-qa_samples_per_second": 58.13,
|
| 938 |
+
"eval_scitail-pairs-qa_steps_per_second": 0.454,
|
| 939 |
+
"step": 104
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 0.20051413881748073,
|
| 943 |
+
"eval_xsum-pairs_loss": 0.14091075956821442,
|
| 944 |
+
"eval_xsum-pairs_runtime": 3.0891,
|
| 945 |
+
"eval_xsum-pairs_samples_per_second": 41.436,
|
| 946 |
+
"eval_xsum-pairs_steps_per_second": 0.324,
|
| 947 |
+
"step": 104
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.20051413881748073,
|
| 951 |
+
"eval_sciq_pairs_loss": 0.1505264937877655,
|
| 952 |
+
"eval_sciq_pairs_runtime": 4.762,
|
| 953 |
+
"eval_sciq_pairs_samples_per_second": 26.879,
|
| 954 |
+
"eval_sciq_pairs_steps_per_second": 0.21,
|
| 955 |
+
"step": 104
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.20051413881748073,
|
| 959 |
+
"eval_qasc_pairs_loss": 0.4244489073753357,
|
| 960 |
+
"eval_qasc_pairs_runtime": 2.0783,
|
| 961 |
+
"eval_qasc_pairs_samples_per_second": 61.589,
|
| 962 |
+
"eval_qasc_pairs_steps_per_second": 0.481,
|
| 963 |
+
"step": 104
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 0.20051413881748073,
|
| 967 |
+
"eval_openbookqa_pairs_loss": 0.5923758745193481,
|
| 968 |
+
"eval_openbookqa_pairs_runtime": 2.1938,
|
| 969 |
+
"eval_openbookqa_pairs_samples_per_second": 58.345,
|
| 970 |
+
"eval_openbookqa_pairs_steps_per_second": 0.456,
|
| 971 |
+
"step": 104
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 0.20051413881748073,
|
| 975 |
+
"eval_msmarco_pairs_loss": 0.5882836580276489,
|
| 976 |
+
"eval_msmarco_pairs_runtime": 2.2483,
|
| 977 |
+
"eval_msmarco_pairs_samples_per_second": 56.933,
|
| 978 |
+
"eval_msmarco_pairs_steps_per_second": 0.445,
|
| 979 |
+
"step": 104
|
| 980 |
+
},
|
| 981 |
+
{
|
| 982 |
+
"epoch": 0.20051413881748073,
|
| 983 |
+
"eval_nq_pairs_loss": 0.5424538254737854,
|
| 984 |
+
"eval_nq_pairs_runtime": 3.6424,
|
| 985 |
+
"eval_nq_pairs_samples_per_second": 35.142,
|
| 986 |
+
"eval_nq_pairs_steps_per_second": 0.275,
|
| 987 |
+
"step": 104
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 0.20051413881748073,
|
| 991 |
+
"eval_trivia_pairs_loss": 0.6437886953353882,
|
| 992 |
+
"eval_trivia_pairs_runtime": 3.0809,
|
| 993 |
+
"eval_trivia_pairs_samples_per_second": 41.546,
|
| 994 |
+
"eval_trivia_pairs_steps_per_second": 0.325,
|
| 995 |
+
"step": 104
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 0.20051413881748073,
|
| 999 |
+
"eval_gooaq_pairs_loss": 0.22378715872764587,
|
| 1000 |
+
"eval_gooaq_pairs_runtime": 2.101,
|
| 1001 |
+
"eval_gooaq_pairs_samples_per_second": 60.924,
|
| 1002 |
+
"eval_gooaq_pairs_steps_per_second": 0.476,
|
| 1003 |
+
"step": 104
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.20051413881748073,
|
| 1007 |
+
"eval_paws-pos_loss": 0.03581737354397774,
|
| 1008 |
+
"eval_paws-pos_runtime": 2.2447,
|
| 1009 |
+
"eval_paws-pos_samples_per_second": 57.024,
|
| 1010 |
+
"eval_paws-pos_steps_per_second": 0.445,
|
| 1011 |
+
"step": 104
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.20051413881748073,
|
| 1015 |
+
"eval_global_dataset_loss": 0.3989756405353546,
|
| 1016 |
+
"eval_global_dataset_runtime": 8.402,
|
| 1017 |
+
"eval_global_dataset_samples_per_second": 38.681,
|
| 1018 |
+
"eval_global_dataset_steps_per_second": 0.357,
|
| 1019 |
+
"step": 104
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 0.20244215938303342,
|
| 1023 |
+
"grad_norm": 96.23986053466797,
|
| 1024 |
+
"learning_rate": 5.5019305019305015e-06,
|
| 1025 |
+
"loss": 0.5094,
|
| 1026 |
+
"step": 105
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 0.20822622107969152,
|
| 1030 |
+
"grad_norm": 33.88426208496094,
|
| 1031 |
+
"learning_rate": 5.675675675675676e-06,
|
| 1032 |
+
"loss": 0.8002,
|
| 1033 |
+
"step": 108
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 0.21401028277634962,
|
| 1037 |
+
"grad_norm": 7.319549560546875,
|
| 1038 |
+
"learning_rate": 5.849420849420849e-06,
|
| 1039 |
+
"loss": 0.3886,
|
| 1040 |
+
"step": 111
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 0.21979434447300772,
|
| 1044 |
+
"grad_norm": 8.65904712677002,
|
| 1045 |
+
"learning_rate": 6.023166023166023e-06,
|
| 1046 |
+
"loss": 0.6937,
|
| 1047 |
+
"step": 114
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 0.22557840616966582,
|
| 1051 |
+
"grad_norm": 376.61663818359375,
|
| 1052 |
+
"learning_rate": 6.196911196911197e-06,
|
| 1053 |
+
"loss": 0.2909,
|
| 1054 |
+
"step": 117
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 0.23136246786632392,
|
| 1058 |
+
"grad_norm": 10.809043884277344,
|
| 1059 |
+
"learning_rate": 6.370656370656371e-06,
|
| 1060 |
+
"loss": 0.3885,
|
| 1061 |
+
"step": 120
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 0.23714652956298202,
|
| 1065 |
+
"grad_norm": 8.564398765563965,
|
| 1066 |
+
"learning_rate": 6.5444015444015445e-06,
|
| 1067 |
+
"loss": 0.29,
|
| 1068 |
+
"step": 123
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 0.24293059125964012,
|
| 1072 |
+
"grad_norm": 1.5215222835540771,
|
| 1073 |
+
"learning_rate": 6.718146718146718e-06,
|
| 1074 |
+
"loss": 0.3485,
|
| 1075 |
+
"step": 126
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 0.2487146529562982,
|
| 1079 |
+
"grad_norm": 181.20809936523438,
|
| 1080 |
+
"learning_rate": 6.891891891891892e-06,
|
| 1081 |
+
"loss": 0.3931,
|
| 1082 |
+
"step": 129
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 0.2506426735218509,
|
| 1086 |
+
"eval_Qnli-dev_cosine_accuracy": 0.67578125,
|
| 1087 |
+
"eval_Qnli-dev_cosine_accuracy_threshold": 0.7986456155776978,
|
| 1088 |
+
"eval_Qnli-dev_cosine_ap": 0.6985736393833883,
|
| 1089 |
+
"eval_Qnli-dev_cosine_f1": 0.6777777777777777,
|
| 1090 |
+
"eval_Qnli-dev_cosine_f1_threshold": 0.726032018661499,
|
| 1091 |
+
"eval_Qnli-dev_cosine_precision": 0.6019736842105263,
|
| 1092 |
+
"eval_Qnli-dev_cosine_recall": 0.7754237288135594,
|
| 1093 |
+
"eval_Qnli-dev_dot_accuracy": 0.66796875,
|
| 1094 |
+
"eval_Qnli-dev_dot_accuracy_threshold": 840.5162353515625,
|
| 1095 |
+
"eval_Qnli-dev_dot_ap": 0.6656432640480556,
|
| 1096 |
+
"eval_Qnli-dev_dot_f1": 0.662020905923345,
|
| 1097 |
+
"eval_Qnli-dev_dot_f1_threshold": 715.1839599609375,
|
| 1098 |
+
"eval_Qnli-dev_dot_precision": 0.5621301775147929,
|
| 1099 |
+
"eval_Qnli-dev_dot_recall": 0.8050847457627118,
|
| 1100 |
+
"eval_Qnli-dev_euclidean_accuracy": 0.677734375,
|
| 1101 |
+
"eval_Qnli-dev_euclidean_accuracy_threshold": 20.397323608398438,
|
| 1102 |
+
"eval_Qnli-dev_euclidean_ap": 0.6992667316315335,
|
| 1103 |
+
"eval_Qnli-dev_euclidean_f1": 0.67296786389414,
|
| 1104 |
+
"eval_Qnli-dev_euclidean_f1_threshold": 23.44580841064453,
|
| 1105 |
+
"eval_Qnli-dev_euclidean_precision": 0.6075085324232082,
|
| 1106 |
+
"eval_Qnli-dev_euclidean_recall": 0.7542372881355932,
|
| 1107 |
+
"eval_Qnli-dev_manhattan_accuracy": 0.671875,
|
| 1108 |
+
"eval_Qnli-dev_manhattan_accuracy_threshold": 628.4102172851562,
|
| 1109 |
+
"eval_Qnli-dev_manhattan_ap": 0.6982133105013253,
|
| 1110 |
+
"eval_Qnli-dev_manhattan_f1": 0.673992673992674,
|
| 1111 |
+
"eval_Qnli-dev_manhattan_f1_threshold": 746.4862060546875,
|
| 1112 |
+
"eval_Qnli-dev_manhattan_precision": 0.5935483870967742,
|
| 1113 |
+
"eval_Qnli-dev_manhattan_recall": 0.7796610169491526,
|
| 1114 |
+
"eval_Qnli-dev_max_accuracy": 0.677734375,
|
| 1115 |
+
"eval_Qnli-dev_max_accuracy_threshold": 840.5162353515625,
|
| 1116 |
+
"eval_Qnli-dev_max_ap": 0.6992667316315335,
|
| 1117 |
+
"eval_Qnli-dev_max_f1": 0.6777777777777777,
|
| 1118 |
+
"eval_Qnli-dev_max_f1_threshold": 746.4862060546875,
|
| 1119 |
+
"eval_Qnli-dev_max_precision": 0.6075085324232082,
|
| 1120 |
+
"eval_Qnli-dev_max_recall": 0.8050847457627118,
|
| 1121 |
+
"eval_allNLI-dev_cosine_accuracy": 0.7109375,
|
| 1122 |
+
"eval_allNLI-dev_cosine_accuracy_threshold": 0.8466876745223999,
|
| 1123 |
+
"eval_allNLI-dev_cosine_ap": 0.5614885136859779,
|
| 1124 |
+
"eval_allNLI-dev_cosine_f1": 0.5809128630705395,
|
| 1125 |
+
"eval_allNLI-dev_cosine_f1_threshold": 0.6695160269737244,
|
| 1126 |
+
"eval_allNLI-dev_cosine_precision": 0.45307443365695793,
|
| 1127 |
+
"eval_allNLI-dev_cosine_recall": 0.8092485549132948,
|
| 1128 |
+
"eval_allNLI-dev_dot_accuracy": 0.69921875,
|
| 1129 |
+
"eval_allNLI-dev_dot_accuracy_threshold": 901.9229736328125,
|
| 1130 |
+
"eval_allNLI-dev_dot_ap": 0.5637771052921187,
|
| 1131 |
+
"eval_allNLI-dev_dot_f1": 0.5812807881773399,
|
| 1132 |
+
"eval_allNLI-dev_dot_f1_threshold": 752.6895141601562,
|
| 1133 |
+
"eval_allNLI-dev_dot_precision": 0.5064377682403434,
|
| 1134 |
+
"eval_allNLI-dev_dot_recall": 0.6820809248554913,
|
| 1135 |
+
"eval_allNLI-dev_euclidean_accuracy": 0.708984375,
|
| 1136 |
+
"eval_allNLI-dev_euclidean_accuracy_threshold": 17.844379425048828,
|
| 1137 |
+
"eval_allNLI-dev_euclidean_ap": 0.5616160818796474,
|
| 1138 |
+
"eval_allNLI-dev_euclidean_f1": 0.5838509316770187,
|
| 1139 |
+
"eval_allNLI-dev_euclidean_f1_threshold": 26.068946838378906,
|
| 1140 |
+
"eval_allNLI-dev_euclidean_precision": 0.45483870967741935,
|
| 1141 |
+
"eval_allNLI-dev_euclidean_recall": 0.815028901734104,
|
| 1142 |
+
"eval_allNLI-dev_manhattan_accuracy": 0.7109375,
|
| 1143 |
+
"eval_allNLI-dev_manhattan_accuracy_threshold": 553.34228515625,
|
| 1144 |
+
"eval_allNLI-dev_manhattan_ap": 0.5624349287984178,
|
| 1145 |
+
"eval_allNLI-dev_manhattan_f1": 0.5778688524590163,
|
| 1146 |
+
"eval_allNLI-dev_manhattan_f1_threshold": 816.2254028320312,
|
| 1147 |
+
"eval_allNLI-dev_manhattan_precision": 0.44761904761904764,
|
| 1148 |
+
"eval_allNLI-dev_manhattan_recall": 0.815028901734104,
|
| 1149 |
+
"eval_allNLI-dev_max_accuracy": 0.7109375,
|
| 1150 |
+
"eval_allNLI-dev_max_accuracy_threshold": 901.9229736328125,
|
| 1151 |
+
"eval_allNLI-dev_max_ap": 0.5637771052921187,
|
| 1152 |
+
"eval_allNLI-dev_max_f1": 0.5838509316770187,
|
| 1153 |
+
"eval_allNLI-dev_max_f1_threshold": 816.2254028320312,
|
| 1154 |
+
"eval_allNLI-dev_max_precision": 0.5064377682403434,
|
| 1155 |
+
"eval_allNLI-dev_max_recall": 0.815028901734104,
|
| 1156 |
+
"eval_sequential_score": 0.6992667316315335,
|
| 1157 |
+
"eval_sts-test_pearson_cosine": 0.9024404439516205,
|
| 1158 |
+
"eval_sts-test_pearson_dot": 0.8941554024395801,
|
| 1159 |
+
"eval_sts-test_pearson_euclidean": 0.9164387414940136,
|
| 1160 |
+
"eval_sts-test_pearson_manhattan": 0.9158406136217865,
|
| 1161 |
+
"eval_sts-test_pearson_max": 0.9164387414940136,
|
| 1162 |
+
"eval_sts-test_spearman_cosine": 0.9081672412582797,
|
| 1163 |
+
"eval_sts-test_spearman_dot": 0.8855668249766109,
|
| 1164 |
+
"eval_sts-test_spearman_euclidean": 0.9076990535616029,
|
| 1165 |
+
"eval_sts-test_spearman_manhattan": 0.907316833038059,
|
| 1166 |
+
"eval_sts-test_spearman_max": 0.9081672412582797,
|
| 1167 |
+
"eval_vitaminc-pairs_loss": 3.8795158863067627,
|
| 1168 |
+
"eval_vitaminc-pairs_runtime": 4.6143,
|
| 1169 |
+
"eval_vitaminc-pairs_samples_per_second": 27.74,
|
| 1170 |
+
"eval_vitaminc-pairs_steps_per_second": 0.217,
|
| 1171 |
+
"step": 130
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.2506426735218509,
|
| 1175 |
+
"eval_negation-triplets_loss": 1.1134737730026245,
|
| 1176 |
+
"eval_negation-triplets_runtime": 3.4103,
|
| 1177 |
+
"eval_negation-triplets_samples_per_second": 37.533,
|
| 1178 |
+
"eval_negation-triplets_steps_per_second": 0.293,
|
| 1179 |
+
"step": 130
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.2506426735218509,
|
| 1183 |
+
"eval_scitail-pairs-pos_loss": 0.045735035091638565,
|
| 1184 |
+
"eval_scitail-pairs-pos_runtime": 2.7664,
|
| 1185 |
+
"eval_scitail-pairs-pos_samples_per_second": 46.27,
|
| 1186 |
+
"eval_scitail-pairs-pos_steps_per_second": 0.361,
|
| 1187 |
+
"step": 130
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 0.2506426735218509,
|
| 1191 |
+
"eval_scitail-pairs-qa_loss": 0.036478620022535324,
|
| 1192 |
+
"eval_scitail-pairs-qa_runtime": 2.2905,
|
| 1193 |
+
"eval_scitail-pairs-qa_samples_per_second": 55.884,
|
| 1194 |
+
"eval_scitail-pairs-qa_steps_per_second": 0.437,
|
| 1195 |
+
"step": 130
|
| 1196 |
+
},
|
| 1197 |
+
{
|
| 1198 |
+
"epoch": 0.2506426735218509,
|
| 1199 |
+
"eval_xsum-pairs_loss": 0.05133158713579178,
|
| 1200 |
+
"eval_xsum-pairs_runtime": 3.1323,
|
| 1201 |
+
"eval_xsum-pairs_samples_per_second": 40.865,
|
| 1202 |
+
"eval_xsum-pairs_steps_per_second": 0.319,
|
| 1203 |
+
"step": 130
|
| 1204 |
+
},
|
| 1205 |
+
{
|
| 1206 |
+
"epoch": 0.2506426735218509,
|
| 1207 |
+
"eval_sciq_pairs_loss": 0.12314368784427643,
|
| 1208 |
+
"eval_sciq_pairs_runtime": 4.8477,
|
| 1209 |
+
"eval_sciq_pairs_samples_per_second": 26.404,
|
| 1210 |
+
"eval_sciq_pairs_steps_per_second": 0.206,
|
| 1211 |
+
"step": 130
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 0.2506426735218509,
|
| 1215 |
+
"eval_qasc_pairs_loss": 0.22055241465568542,
|
| 1216 |
+
"eval_qasc_pairs_runtime": 2.1243,
|
| 1217 |
+
"eval_qasc_pairs_samples_per_second": 60.255,
|
| 1218 |
+
"eval_qasc_pairs_steps_per_second": 0.471,
|
| 1219 |
+
"step": 130
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 0.2506426735218509,
|
| 1223 |
+
"eval_openbookqa_pairs_loss": 0.4567987024784088,
|
| 1224 |
+
"eval_openbookqa_pairs_runtime": 2.3078,
|
| 1225 |
+
"eval_openbookqa_pairs_samples_per_second": 55.465,
|
| 1226 |
+
"eval_openbookqa_pairs_steps_per_second": 0.433,
|
| 1227 |
+
"step": 130
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.2506426735218509,
|
| 1231 |
+
"eval_msmarco_pairs_loss": 0.21620944142341614,
|
| 1232 |
+
"eval_msmarco_pairs_runtime": 2.3399,
|
| 1233 |
+
"eval_msmarco_pairs_samples_per_second": 54.703,
|
| 1234 |
+
"eval_msmarco_pairs_steps_per_second": 0.427,
|
| 1235 |
+
"step": 130
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.2506426735218509,
|
| 1239 |
+
"eval_nq_pairs_loss": 0.26776355504989624,
|
| 1240 |
+
"eval_nq_pairs_runtime": 3.7038,
|
| 1241 |
+
"eval_nq_pairs_samples_per_second": 34.559,
|
| 1242 |
+
"eval_nq_pairs_steps_per_second": 0.27,
|
| 1243 |
+
"step": 130
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 0.2506426735218509,
|
| 1247 |
+
"eval_trivia_pairs_loss": 0.15397313237190247,
|
| 1248 |
+
"eval_trivia_pairs_runtime": 3.1256,
|
| 1249 |
+
"eval_trivia_pairs_samples_per_second": 40.952,
|
| 1250 |
+
"eval_trivia_pairs_steps_per_second": 0.32,
|
| 1251 |
+
"step": 130
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"epoch": 0.2506426735218509,
|
| 1255 |
+
"eval_gooaq_pairs_loss": 0.12057099491357803,
|
| 1256 |
+
"eval_gooaq_pairs_runtime": 2.1825,
|
| 1257 |
+
"eval_gooaq_pairs_samples_per_second": 58.648,
|
| 1258 |
+
"eval_gooaq_pairs_steps_per_second": 0.458,
|
| 1259 |
+
"step": 130
|
| 1260 |
+
},
|
| 1261 |
+
{
|
| 1262 |
+
"epoch": 0.2506426735218509,
|
| 1263 |
+
"eval_paws-pos_loss": 0.026105860248208046,
|
| 1264 |
+
"eval_paws-pos_runtime": 2.2995,
|
| 1265 |
+
"eval_paws-pos_samples_per_second": 55.665,
|
| 1266 |
+
"eval_paws-pos_steps_per_second": 0.435,
|
| 1267 |
+
"step": 130
|
| 1268 |
+
},
|
| 1269 |
+
{
|
| 1270 |
+
"epoch": 0.2506426735218509,
|
| 1271 |
+
"eval_global_dataset_loss": 0.3591004014015198,
|
| 1272 |
+
"eval_global_dataset_runtime": 8.4153,
|
| 1273 |
+
"eval_global_dataset_samples_per_second": 38.62,
|
| 1274 |
+
"eval_global_dataset_steps_per_second": 0.356,
|
| 1275 |
+
"step": 130
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 0.2544987146529563,
|
| 1279 |
+
"grad_norm": 11.167648315429688,
|
| 1280 |
+
"learning_rate": 7.065637065637066e-06,
|
| 1281 |
+
"loss": 0.3394,
|
| 1282 |
+
"step": 132
|
| 1283 |
+
},
|
| 1284 |
+
{
|
| 1285 |
+
"epoch": 0.2602827763496144,
|
| 1286 |
+
"grad_norm": 3.4576871395111084,
|
| 1287 |
+
"learning_rate": 7.239382239382239e-06,
|
| 1288 |
+
"loss": 0.1276,
|
| 1289 |
+
"step": 135
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 0.2660668380462725,
|
| 1293 |
+
"grad_norm": 9.594191551208496,
|
| 1294 |
+
"learning_rate": 7.413127413127414e-06,
|
| 1295 |
+
"loss": 0.3569,
|
| 1296 |
+
"step": 138
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 0.2718508997429306,
|
| 1300 |
+
"grad_norm": 4.255222320556641,
|
| 1301 |
+
"learning_rate": 7.586872586872587e-06,
|
| 1302 |
+
"loss": 0.1231,
|
| 1303 |
+
"step": 141
|
| 1304 |
+
},
|
| 1305 |
+
{
|
| 1306 |
+
"epoch": 0.2776349614395887,
|
| 1307 |
+
"grad_norm": 6.074405670166016,
|
| 1308 |
+
"learning_rate": 7.76061776061776e-06,
|
| 1309 |
+
"loss": 0.3086,
|
| 1310 |
+
"step": 144
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 0.2834190231362468,
|
| 1314 |
+
"grad_norm": 19.442323684692383,
|
| 1315 |
+
"learning_rate": 7.934362934362936e-06,
|
| 1316 |
+
"loss": 0.3541,
|
| 1317 |
+
"step": 147
|
| 1318 |
+
},
|
| 1319 |
+
{
|
| 1320 |
+
"epoch": 0.2892030848329049,
|
| 1321 |
+
"grad_norm": 5.141862392425537,
|
| 1322 |
+
"learning_rate": 8.108108108108109e-06,
|
| 1323 |
+
"loss": 0.2597,
|
| 1324 |
+
"step": 150
|
| 1325 |
+
},
|
| 1326 |
+
{
|
| 1327 |
+
"epoch": 0.294987146529563,
|
| 1328 |
+
"grad_norm": 5.942401885986328,
|
| 1329 |
+
"learning_rate": 8.281853281853282e-06,
|
| 1330 |
+
"loss": 0.1585,
|
| 1331 |
+
"step": 153
|
| 1332 |
+
},
|
| 1333 |
+
{
|
| 1334 |
+
"epoch": 0.30077120822622105,
|
| 1335 |
+
"grad_norm": 97.85575103759766,
|
| 1336 |
+
"learning_rate": 8.455598455598455e-06,
|
| 1337 |
+
"loss": 0.1436,
|
| 1338 |
+
"step": 156
|
| 1339 |
+
},
|
| 1340 |
+
{
|
| 1341 |
+
"epoch": 0.30077120822622105,
|
| 1342 |
+
"eval_Qnli-dev_cosine_accuracy": 0.693359375,
|
| 1343 |
+
"eval_Qnli-dev_cosine_accuracy_threshold": 0.7395063638687134,
|
| 1344 |
+
"eval_Qnli-dev_cosine_ap": 0.7187155891006661,
|
| 1345 |
+
"eval_Qnli-dev_cosine_f1": 0.6837606837606838,
|
| 1346 |
+
"eval_Qnli-dev_cosine_f1_threshold": 0.660994291305542,
|
| 1347 |
+
"eval_Qnli-dev_cosine_precision": 0.5730659025787965,
|
| 1348 |
+
"eval_Qnli-dev_cosine_recall": 0.847457627118644,
|
| 1349 |
+
"eval_Qnli-dev_dot_accuracy": 0.669921875,
|
| 1350 |
+
"eval_Qnli-dev_dot_accuracy_threshold": 701.7633056640625,
|
| 1351 |
+
"eval_Qnli-dev_dot_ap": 0.6888124099944863,
|
| 1352 |
+
"eval_Qnli-dev_dot_f1": 0.6812816188870151,
|
| 1353 |
+
"eval_Qnli-dev_dot_f1_threshold": 633.31884765625,
|
| 1354 |
+
"eval_Qnli-dev_dot_precision": 0.5658263305322129,
|
| 1355 |
+
"eval_Qnli-dev_dot_recall": 0.8559322033898306,
|
| 1356 |
+
"eval_Qnli-dev_euclidean_accuracy": 0.697265625,
|
| 1357 |
+
"eval_Qnli-dev_euclidean_accuracy_threshold": 22.954120635986328,
|
| 1358 |
+
"eval_Qnli-dev_euclidean_ap": 0.721593279891293,
|
| 1359 |
+
"eval_Qnli-dev_euclidean_f1": 0.6762075134168157,
|
| 1360 |
+
"eval_Qnli-dev_euclidean_f1_threshold": 24.89653968811035,
|
| 1361 |
+
"eval_Qnli-dev_euclidean_precision": 0.5851393188854489,
|
| 1362 |
+
"eval_Qnli-dev_euclidean_recall": 0.8008474576271186,
|
| 1363 |
+
"eval_Qnli-dev_manhattan_accuracy": 0.6953125,
|
| 1364 |
+
"eval_Qnli-dev_manhattan_accuracy_threshold": 718.2297973632812,
|
| 1365 |
+
"eval_Qnli-dev_manhattan_ap": 0.7239405100010182,
|
| 1366 |
+
"eval_Qnli-dev_manhattan_f1": 0.6807017543859649,
|
| 1367 |
+
"eval_Qnli-dev_manhattan_f1_threshold": 784.701171875,
|
| 1368 |
+
"eval_Qnli-dev_manhattan_precision": 0.5808383233532934,
|
| 1369 |
+
"eval_Qnli-dev_manhattan_recall": 0.8220338983050848,
|
| 1370 |
+
"eval_Qnli-dev_max_accuracy": 0.697265625,
|
| 1371 |
+
"eval_Qnli-dev_max_accuracy_threshold": 718.2297973632812,
|
| 1372 |
+
"eval_Qnli-dev_max_ap": 0.7239405100010182,
|
| 1373 |
+
"eval_Qnli-dev_max_f1": 0.6837606837606838,
|
| 1374 |
+
"eval_Qnli-dev_max_f1_threshold": 784.701171875,
|
| 1375 |
+
"eval_Qnli-dev_max_precision": 0.5851393188854489,
|
| 1376 |
+
"eval_Qnli-dev_max_recall": 0.8559322033898306,
|
| 1377 |
+
"eval_allNLI-dev_cosine_accuracy": 0.71875,
|
| 1378 |
+
"eval_allNLI-dev_cosine_accuracy_threshold": 0.8460206985473633,
|
| 1379 |
+
"eval_allNLI-dev_cosine_ap": 0.5796621470925665,
|
| 1380 |
+
"eval_allNLI-dev_cosine_f1": 0.6139954853273137,
|
| 1381 |
+
"eval_allNLI-dev_cosine_f1_threshold": 0.7097446322441101,
|
| 1382 |
+
"eval_allNLI-dev_cosine_precision": 0.5037037037037037,
|
| 1383 |
+
"eval_allNLI-dev_cosine_recall": 0.7861271676300579,
|
| 1384 |
+
"eval_allNLI-dev_dot_accuracy": 0.7109375,
|
| 1385 |
+
"eval_allNLI-dev_dot_accuracy_threshold": 837.8741455078125,
|
| 1386 |
+
"eval_allNLI-dev_dot_ap": 0.5763352242724786,
|
| 1387 |
+
"eval_allNLI-dev_dot_f1": 0.6179245283018867,
|
| 1388 |
+
"eval_allNLI-dev_dot_f1_threshold": 712.9136352539062,
|
| 1389 |
+
"eval_allNLI-dev_dot_precision": 0.5219123505976095,
|
| 1390 |
+
"eval_allNLI-dev_dot_recall": 0.7572254335260116,
|
| 1391 |
+
"eval_allNLI-dev_euclidean_accuracy": 0.724609375,
|
| 1392 |
+
"eval_allNLI-dev_euclidean_accuracy_threshold": 18.82543182373047,
|
| 1393 |
+
"eval_allNLI-dev_euclidean_ap": 0.5803015452539461,
|
| 1394 |
+
"eval_allNLI-dev_euclidean_f1": 0.6085011185682326,
|
| 1395 |
+
"eval_allNLI-dev_euclidean_f1_threshold": 24.0869197845459,
|
| 1396 |
+
"eval_allNLI-dev_euclidean_precision": 0.49635036496350365,
|
| 1397 |
+
"eval_allNLI-dev_euclidean_recall": 0.7861271676300579,
|
| 1398 |
+
"eval_allNLI-dev_manhattan_accuracy": 0.720703125,
|
| 1399 |
+
"eval_allNLI-dev_manhattan_accuracy_threshold": 586.8082275390625,
|
| 1400 |
+
"eval_allNLI-dev_manhattan_ap": 0.5806697040885319,
|
| 1401 |
+
"eval_allNLI-dev_manhattan_f1": 0.6035242290748899,
|
| 1402 |
+
"eval_allNLI-dev_manhattan_f1_threshold": 756.8704833984375,
|
| 1403 |
+
"eval_allNLI-dev_manhattan_precision": 0.4875444839857651,
|
| 1404 |
+
"eval_allNLI-dev_manhattan_recall": 0.791907514450867,
|
| 1405 |
+
"eval_allNLI-dev_max_accuracy": 0.724609375,
|
| 1406 |
+
"eval_allNLI-dev_max_accuracy_threshold": 837.8741455078125,
|
| 1407 |
+
"eval_allNLI-dev_max_ap": 0.5806697040885319,
|
| 1408 |
+
"eval_allNLI-dev_max_f1": 0.6179245283018867,
|
| 1409 |
+
"eval_allNLI-dev_max_f1_threshold": 756.8704833984375,
|
| 1410 |
+
"eval_allNLI-dev_max_precision": 0.5219123505976095,
|
| 1411 |
+
"eval_allNLI-dev_max_recall": 0.791907514450867,
|
| 1412 |
+
"eval_sequential_score": 0.7239405100010182,
|
| 1413 |
+
"eval_sts-test_pearson_cosine": 0.9086022204177822,
|
| 1414 |
+
"eval_sts-test_pearson_dot": 0.8963707820565703,
|
| 1415 |
+
"eval_sts-test_pearson_euclidean": 0.9226263845809013,
|
| 1416 |
+
"eval_sts-test_pearson_manhattan": 0.9216624653503607,
|
| 1417 |
+
"eval_sts-test_pearson_max": 0.9226263845809013,
|
| 1418 |
+
"eval_sts-test_spearman_cosine": 0.9176886091344368,
|
| 1419 |
+
"eval_sts-test_spearman_dot": 0.8941249219513181,
|
| 1420 |
+
"eval_sts-test_spearman_euclidean": 0.9176283739366038,
|
| 1421 |
+
"eval_sts-test_spearman_manhattan": 0.9167815458172944,
|
| 1422 |
+
"eval_sts-test_spearman_max": 0.9176886091344368,
|
| 1423 |
+
"eval_vitaminc-pairs_loss": 3.8346316814422607,
|
| 1424 |
+
"eval_vitaminc-pairs_runtime": 4.4216,
|
| 1425 |
+
"eval_vitaminc-pairs_samples_per_second": 28.949,
|
| 1426 |
+
"eval_vitaminc-pairs_steps_per_second": 0.226,
|
| 1427 |
+
"step": 156
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 0.30077120822622105,
|
| 1431 |
+
"eval_negation-triplets_loss": 0.9441184997558594,
|
| 1432 |
+
"eval_negation-triplets_runtime": 3.3236,
|
| 1433 |
+
"eval_negation-triplets_samples_per_second": 38.512,
|
| 1434 |
+
"eval_negation-triplets_steps_per_second": 0.301,
|
| 1435 |
+
"step": 156
|
| 1436 |
+
},
|
| 1437 |
+
{
|
| 1438 |
+
"epoch": 0.30077120822622105,
|
| 1439 |
+
"eval_scitail-pairs-pos_loss": 0.03759249672293663,
|
| 1440 |
+
"eval_scitail-pairs-pos_runtime": 2.6436,
|
| 1441 |
+
"eval_scitail-pairs-pos_samples_per_second": 48.418,
|
| 1442 |
+
"eval_scitail-pairs-pos_steps_per_second": 0.378,
|
| 1443 |
+
"step": 156
|
| 1444 |
+
},
|
| 1445 |
+
{
|
| 1446 |
+
"epoch": 0.30077120822622105,
|
| 1447 |
+
"eval_scitail-pairs-qa_loss": 0.04104204475879669,
|
| 1448 |
+
"eval_scitail-pairs-qa_runtime": 2.205,
|
| 1449 |
+
"eval_scitail-pairs-qa_samples_per_second": 58.05,
|
| 1450 |
+
"eval_scitail-pairs-qa_steps_per_second": 0.454,
|
| 1451 |
+
"step": 156
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.30077120822622105,
|
| 1455 |
+
"eval_xsum-pairs_loss": 0.05032652989029884,
|
| 1456 |
+
"eval_xsum-pairs_runtime": 3.0619,
|
| 1457 |
+
"eval_xsum-pairs_samples_per_second": 41.805,
|
| 1458 |
+
"eval_xsum-pairs_steps_per_second": 0.327,
|
| 1459 |
+
"step": 156
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.30077120822622105,
|
| 1463 |
+
"eval_sciq_pairs_loss": 0.08709195256233215,
|
| 1464 |
+
"eval_sciq_pairs_runtime": 4.8405,
|
| 1465 |
+
"eval_sciq_pairs_samples_per_second": 26.444,
|
| 1466 |
+
"eval_sciq_pairs_steps_per_second": 0.207,
|
| 1467 |
+
"step": 156
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 0.30077120822622105,
|
| 1471 |
+
"eval_qasc_pairs_loss": 0.17168982326984406,
|
| 1472 |
+
"eval_qasc_pairs_runtime": 2.078,
|
| 1473 |
+
"eval_qasc_pairs_samples_per_second": 61.599,
|
| 1474 |
+
"eval_qasc_pairs_steps_per_second": 0.481,
|
| 1475 |
+
"step": 156
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 0.30077120822622105,
|
| 1479 |
+
"eval_openbookqa_pairs_loss": 0.4304167926311493,
|
| 1480 |
+
"eval_openbookqa_pairs_runtime": 2.1991,
|
| 1481 |
+
"eval_openbookqa_pairs_samples_per_second": 58.205,
|
| 1482 |
+
"eval_openbookqa_pairs_steps_per_second": 0.455,
|
| 1483 |
+
"step": 156
|
| 1484 |
+
},
|
| 1485 |
+
{
|
| 1486 |
+
"epoch": 0.30077120822622105,
|
| 1487 |
+
"eval_msmarco_pairs_loss": 0.13501481711864471,
|
| 1488 |
+
"eval_msmarco_pairs_runtime": 2.2186,
|
| 1489 |
+
"eval_msmarco_pairs_samples_per_second": 57.694,
|
| 1490 |
+
"eval_msmarco_pairs_steps_per_second": 0.451,
|
| 1491 |
+
"step": 156
|
| 1492 |
+
},
|
| 1493 |
+
{
|
| 1494 |
+
"epoch": 0.30077120822622105,
|
| 1495 |
+
"eval_nq_pairs_loss": 0.20035718381404877,
|
| 1496 |
+
"eval_nq_pairs_runtime": 3.6367,
|
| 1497 |
+
"eval_nq_pairs_samples_per_second": 35.197,
|
| 1498 |
+
"eval_nq_pairs_steps_per_second": 0.275,
|
| 1499 |
+
"step": 156
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 0.30077120822622105,
|
| 1503 |
+
"eval_trivia_pairs_loss": 0.08228898048400879,
|
| 1504 |
+
"eval_trivia_pairs_runtime": 3.086,
|
| 1505 |
+
"eval_trivia_pairs_samples_per_second": 41.477,
|
| 1506 |
+
"eval_trivia_pairs_steps_per_second": 0.324,
|
| 1507 |
+
"step": 156
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.30077120822622105,
|
| 1511 |
+
"eval_gooaq_pairs_loss": 0.0767039954662323,
|
| 1512 |
+
"eval_gooaq_pairs_runtime": 2.1123,
|
| 1513 |
+
"eval_gooaq_pairs_samples_per_second": 60.599,
|
| 1514 |
+
"eval_gooaq_pairs_steps_per_second": 0.473,
|
| 1515 |
+
"step": 156
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.30077120822622105,
|
| 1519 |
+
"eval_paws-pos_loss": 0.02134825475513935,
|
| 1520 |
+
"eval_paws-pos_runtime": 2.2234,
|
| 1521 |
+
"eval_paws-pos_samples_per_second": 57.571,
|
| 1522 |
+
"eval_paws-pos_steps_per_second": 0.45,
|
| 1523 |
+
"step": 156
|
| 1524 |
+
},
|
| 1525 |
+
{
|
| 1526 |
+
"epoch": 0.30077120822622105,
|
| 1527 |
+
"eval_global_dataset_loss": 0.2958550453186035,
|
| 1528 |
+
"eval_global_dataset_runtime": 8.3542,
|
| 1529 |
+
"eval_global_dataset_samples_per_second": 38.903,
|
| 1530 |
+
"eval_global_dataset_steps_per_second": 0.359,
|
| 1531 |
+
"step": 156
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"epoch": 0.3065552699228792,
|
| 1535 |
+
"grad_norm": 47.52979278564453,
|
| 1536 |
+
"learning_rate": 8.62934362934363e-06,
|
| 1537 |
+
"loss": 0.1941,
|
| 1538 |
+
"step": 159
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 0.31233933161953725,
|
| 1542 |
+
"grad_norm": 1234.25927734375,
|
| 1543 |
+
"learning_rate": 8.803088803088804e-06,
|
| 1544 |
+
"loss": 0.3041,
|
| 1545 |
+
"step": 162
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 0.3181233933161954,
|
| 1549 |
+
"grad_norm": 185.80238342285156,
|
| 1550 |
+
"learning_rate": 8.976833976833977e-06,
|
| 1551 |
+
"loss": 0.2358,
|
| 1552 |
+
"step": 165
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 0.32390745501285345,
|
| 1556 |
+
"grad_norm": 7.233102798461914,
|
| 1557 |
+
"learning_rate": 9.15057915057915e-06,
|
| 1558 |
+
"loss": 0.2148,
|
| 1559 |
+
"step": 168
|
| 1560 |
+
},
|
| 1561 |
+
{
|
| 1562 |
+
"epoch": 0.3296915167095116,
|
| 1563 |
+
"grad_norm": 342.77685546875,
|
| 1564 |
+
"learning_rate": 9.324324324324325e-06,
|
| 1565 |
+
"loss": 0.8567,
|
| 1566 |
+
"step": 171
|
| 1567 |
+
},
|
| 1568 |
+
{
|
| 1569 |
+
"epoch": 0.33547557840616965,
|
| 1570 |
+
"grad_norm": 181.07650756835938,
|
| 1571 |
+
"learning_rate": 9.498069498069498e-06,
|
| 1572 |
+
"loss": 0.3668,
|
| 1573 |
+
"step": 174
|
| 1574 |
+
},
|
| 1575 |
+
{
|
| 1576 |
+
"epoch": 0.3412596401028278,
|
| 1577 |
+
"grad_norm": 198.4124755859375,
|
| 1578 |
+
"learning_rate": 9.671814671814673e-06,
|
| 1579 |
+
"loss": 0.3278,
|
| 1580 |
+
"step": 177
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 0.34704370179948585,
|
| 1584 |
+
"grad_norm": 91.80472564697266,
|
| 1585 |
+
"learning_rate": 9.845559845559846e-06,
|
| 1586 |
+
"loss": 0.474,
|
| 1587 |
+
"step": 180
|
| 1588 |
+
},
|
| 1589 |
+
{
|
| 1590 |
+
"epoch": 0.3508997429305913,
|
| 1591 |
+
"eval_Qnli-dev_cosine_accuracy": 0.6953125,
|
| 1592 |
+
"eval_Qnli-dev_cosine_accuracy_threshold": 0.7320406436920166,
|
| 1593 |
+
"eval_Qnli-dev_cosine_ap": 0.7163901550352688,
|
| 1594 |
+
"eval_Qnli-dev_cosine_f1": 0.6857142857142857,
|
| 1595 |
+
"eval_Qnli-dev_cosine_f1_threshold": 0.6526604890823364,
|
| 1596 |
+
"eval_Qnli-dev_cosine_precision": 0.5682451253481894,
|
| 1597 |
+
"eval_Qnli-dev_cosine_recall": 0.864406779661017,
|
| 1598 |
+
"eval_Qnli-dev_dot_accuracy": 0.67578125,
|
| 1599 |
+
"eval_Qnli-dev_dot_accuracy_threshold": 759.5008544921875,
|
| 1600 |
+
"eval_Qnli-dev_dot_ap": 0.6857309318375049,
|
| 1601 |
+
"eval_Qnli-dev_dot_f1": 0.6857142857142857,
|
| 1602 |
+
"eval_Qnli-dev_dot_f1_threshold": 653.0701904296875,
|
| 1603 |
+
"eval_Qnli-dev_dot_precision": 0.5682451253481894,
|
| 1604 |
+
"eval_Qnli-dev_dot_recall": 0.864406779661017,
|
| 1605 |
+
"eval_Qnli-dev_euclidean_accuracy": 0.689453125,
|
| 1606 |
+
"eval_Qnli-dev_euclidean_accuracy_threshold": 23.348251342773438,
|
| 1607 |
+
"eval_Qnli-dev_euclidean_ap": 0.7186338110228241,
|
| 1608 |
+
"eval_Qnli-dev_euclidean_f1": 0.6819672131147542,
|
| 1609 |
+
"eval_Qnli-dev_euclidean_f1_threshold": 26.874767303466797,
|
| 1610 |
+
"eval_Qnli-dev_euclidean_precision": 0.5561497326203209,
|
| 1611 |
+
"eval_Qnli-dev_euclidean_recall": 0.8813559322033898,
|
| 1612 |
+
"eval_Qnli-dev_manhattan_accuracy": 0.6875,
|
| 1613 |
+
"eval_Qnli-dev_manhattan_accuracy_threshold": 695.7578125,
|
| 1614 |
+
"eval_Qnli-dev_manhattan_ap": 0.7187237394606566,
|
| 1615 |
+
"eval_Qnli-dev_manhattan_f1": 0.6815286624203821,
|
| 1616 |
+
"eval_Qnli-dev_manhattan_f1_threshold": 856.7666015625,
|
| 1617 |
+
"eval_Qnli-dev_manhattan_precision": 0.5459183673469388,
|
| 1618 |
+
"eval_Qnli-dev_manhattan_recall": 0.9067796610169492,
|
| 1619 |
+
"eval_Qnli-dev_max_accuracy": 0.6953125,
|
| 1620 |
+
"eval_Qnli-dev_max_accuracy_threshold": 759.5008544921875,
|
| 1621 |
+
"eval_Qnli-dev_max_ap": 0.7187237394606566,
|
| 1622 |
+
"eval_Qnli-dev_max_f1": 0.6857142857142857,
|
| 1623 |
+
"eval_Qnli-dev_max_f1_threshold": 856.7666015625,
|
| 1624 |
+
"eval_Qnli-dev_max_precision": 0.5682451253481894,
|
| 1625 |
+
"eval_Qnli-dev_max_recall": 0.9067796610169492,
|
| 1626 |
+
"eval_allNLI-dev_cosine_accuracy": 0.705078125,
|
| 1627 |
+
"eval_allNLI-dev_cosine_accuracy_threshold": 0.8531240224838257,
|
| 1628 |
+
"eval_allNLI-dev_cosine_ap": 0.5555367525973895,
|
| 1629 |
+
"eval_allNLI-dev_cosine_f1": 0.5972222222222222,
|
| 1630 |
+
"eval_allNLI-dev_cosine_f1_threshold": 0.7072098255157471,
|
| 1631 |
+
"eval_allNLI-dev_cosine_precision": 0.4980694980694981,
|
| 1632 |
+
"eval_allNLI-dev_cosine_recall": 0.7456647398843931,
|
| 1633 |
+
"eval_allNLI-dev_dot_accuracy": 0.705078125,
|
| 1634 |
+
"eval_allNLI-dev_dot_accuracy_threshold": 858.6700439453125,
|
| 1635 |
+
"eval_allNLI-dev_dot_ap": 0.5501313597611072,
|
| 1636 |
+
"eval_allNLI-dev_dot_f1": 0.5992217898832685,
|
| 1637 |
+
"eval_allNLI-dev_dot_f1_threshold": 657.9625244140625,
|
| 1638 |
+
"eval_allNLI-dev_dot_precision": 0.45161290322580644,
|
| 1639 |
+
"eval_allNLI-dev_dot_recall": 0.8901734104046243,
|
| 1640 |
+
"eval_allNLI-dev_euclidean_accuracy": 0.705078125,
|
| 1641 |
+
"eval_allNLI-dev_euclidean_accuracy_threshold": 16.42839813232422,
|
| 1642 |
+
"eval_allNLI-dev_euclidean_ap": 0.5519081828104085,
|
| 1643 |
+
"eval_allNLI-dev_euclidean_f1": 0.5919282511210763,
|
| 1644 |
+
"eval_allNLI-dev_euclidean_f1_threshold": 24.696884155273438,
|
| 1645 |
+
"eval_allNLI-dev_euclidean_precision": 0.4835164835164835,
|
| 1646 |
+
"eval_allNLI-dev_euclidean_recall": 0.7630057803468208,
|
| 1647 |
+
"eval_allNLI-dev_manhattan_accuracy": 0.705078125,
|
| 1648 |
+
"eval_allNLI-dev_manhattan_accuracy_threshold": 517.3240356445312,
|
| 1649 |
+
"eval_allNLI-dev_manhattan_ap": 0.5527648330520126,
|
| 1650 |
+
"eval_allNLI-dev_manhattan_f1": 0.5900900900900902,
|
| 1651 |
+
"eval_allNLI-dev_manhattan_f1_threshold": 765.703857421875,
|
| 1652 |
+
"eval_allNLI-dev_manhattan_precision": 0.4833948339483395,
|
| 1653 |
+
"eval_allNLI-dev_manhattan_recall": 0.7572254335260116,
|
| 1654 |
+
"eval_allNLI-dev_max_accuracy": 0.705078125,
|
| 1655 |
+
"eval_allNLI-dev_max_accuracy_threshold": 858.6700439453125,
|
| 1656 |
+
"eval_allNLI-dev_max_ap": 0.5555367525973895,
|
| 1657 |
+
"eval_allNLI-dev_max_f1": 0.5992217898832685,
|
| 1658 |
+
"eval_allNLI-dev_max_f1_threshold": 765.703857421875,
|
| 1659 |
+
"eval_allNLI-dev_max_precision": 0.4980694980694981,
|
| 1660 |
+
"eval_allNLI-dev_max_recall": 0.8901734104046243,
|
| 1661 |
+
"eval_sequential_score": 0.7187237394606566,
|
| 1662 |
+
"eval_sts-test_pearson_cosine": 0.8999110842263669,
|
| 1663 |
+
"eval_sts-test_pearson_dot": 0.8830335713257453,
|
| 1664 |
+
"eval_sts-test_pearson_euclidean": 0.9123289195839425,
|
| 1665 |
+
"eval_sts-test_pearson_manhattan": 0.913131760355957,
|
| 1666 |
+
"eval_sts-test_pearson_max": 0.913131760355957,
|
| 1667 |
+
"eval_sts-test_spearman_cosine": 0.9102556931248936,
|
| 1668 |
+
"eval_sts-test_spearman_dot": 0.8782251567890449,
|
| 1669 |
+
"eval_sts-test_spearman_euclidean": 0.9078544746924153,
|
| 1670 |
+
"eval_sts-test_spearman_manhattan": 0.9084604962741777,
|
| 1671 |
+
"eval_sts-test_spearman_max": 0.9102556931248936,
|
| 1672 |
+
"eval_vitaminc-pairs_loss": 3.8443844318389893,
|
| 1673 |
+
"eval_vitaminc-pairs_runtime": 4.62,
|
| 1674 |
+
"eval_vitaminc-pairs_samples_per_second": 27.706,
|
| 1675 |
+
"eval_vitaminc-pairs_steps_per_second": 0.216,
|
| 1676 |
+
"step": 182
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.3508997429305913,
|
| 1680 |
+
"eval_negation-triplets_loss": 0.8706199526786804,
|
| 1681 |
+
"eval_negation-triplets_runtime": 3.4865,
|
| 1682 |
+
"eval_negation-triplets_samples_per_second": 36.714,
|
| 1683 |
+
"eval_negation-triplets_steps_per_second": 0.287,
|
| 1684 |
+
"step": 182
|
| 1685 |
+
},
|
| 1686 |
+
{
|
| 1687 |
+
"epoch": 0.3508997429305913,
|
| 1688 |
+
"eval_scitail-pairs-pos_loss": 0.05496061593294144,
|
| 1689 |
+
"eval_scitail-pairs-pos_runtime": 2.8238,
|
| 1690 |
+
"eval_scitail-pairs-pos_samples_per_second": 45.328,
|
| 1691 |
+
"eval_scitail-pairs-pos_steps_per_second": 0.354,
|
| 1692 |
+
"step": 182
|
| 1693 |
+
},
|
| 1694 |
+
{
|
| 1695 |
+
"epoch": 0.3508997429305913,
|
| 1696 |
+
"eval_scitail-pairs-qa_loss": 0.12511999905109406,
|
| 1697 |
+
"eval_scitail-pairs-qa_runtime": 2.2972,
|
| 1698 |
+
"eval_scitail-pairs-qa_samples_per_second": 55.721,
|
| 1699 |
+
"eval_scitail-pairs-qa_steps_per_second": 0.435,
|
| 1700 |
+
"step": 182
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 0.3508997429305913,
|
| 1704 |
+
"eval_xsum-pairs_loss": 0.03821578621864319,
|
| 1705 |
+
"eval_xsum-pairs_runtime": 3.1015,
|
| 1706 |
+
"eval_xsum-pairs_samples_per_second": 41.27,
|
| 1707 |
+
"eval_xsum-pairs_steps_per_second": 0.322,
|
| 1708 |
+
"step": 182
|
| 1709 |
+
},
|
| 1710 |
+
{
|
| 1711 |
+
"epoch": 0.3508997429305913,
|
| 1712 |
+
"eval_sciq_pairs_loss": 0.08246304094791412,
|
| 1713 |
+
"eval_sciq_pairs_runtime": 4.8955,
|
| 1714 |
+
"eval_sciq_pairs_samples_per_second": 26.147,
|
| 1715 |
+
"eval_sciq_pairs_steps_per_second": 0.204,
|
| 1716 |
+
"step": 182
|
| 1717 |
+
},
|
| 1718 |
+
{
|
| 1719 |
+
"epoch": 0.3508997429305913,
|
| 1720 |
+
"eval_qasc_pairs_loss": 0.15105530619621277,
|
| 1721 |
+
"eval_qasc_pairs_runtime": 2.2079,
|
| 1722 |
+
"eval_qasc_pairs_samples_per_second": 57.972,
|
| 1723 |
+
"eval_qasc_pairs_steps_per_second": 0.453,
|
| 1724 |
+
"step": 182
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.3508997429305913,
|
| 1728 |
+
"eval_openbookqa_pairs_loss": 0.7893790602684021,
|
| 1729 |
+
"eval_openbookqa_pairs_runtime": 2.2579,
|
| 1730 |
+
"eval_openbookqa_pairs_samples_per_second": 56.69,
|
| 1731 |
+
"eval_openbookqa_pairs_steps_per_second": 0.443,
|
| 1732 |
+
"step": 182
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.3508997429305913,
|
| 1736 |
+
"eval_msmarco_pairs_loss": 0.41756388545036316,
|
| 1737 |
+
"eval_msmarco_pairs_runtime": 2.2908,
|
| 1738 |
+
"eval_msmarco_pairs_samples_per_second": 55.877,
|
| 1739 |
+
"eval_msmarco_pairs_steps_per_second": 0.437,
|
| 1740 |
+
"step": 182
|
| 1741 |
+
},
|
| 1742 |
+
{
|
| 1743 |
+
"epoch": 0.3508997429305913,
|
| 1744 |
+
"eval_nq_pairs_loss": 0.2524656355381012,
|
| 1745 |
+
"eval_nq_pairs_runtime": 3.6631,
|
| 1746 |
+
"eval_nq_pairs_samples_per_second": 34.943,
|
| 1747 |
+
"eval_nq_pairs_steps_per_second": 0.273,
|
| 1748 |
+
"step": 182
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"epoch": 0.3508997429305913,
|
| 1752 |
+
"eval_trivia_pairs_loss": 0.062444861978292465,
|
| 1753 |
+
"eval_trivia_pairs_runtime": 3.115,
|
| 1754 |
+
"eval_trivia_pairs_samples_per_second": 41.091,
|
| 1755 |
+
"eval_trivia_pairs_steps_per_second": 0.321,
|
| 1756 |
+
"step": 182
|
| 1757 |
+
},
|
| 1758 |
+
{
|
| 1759 |
+
"epoch": 0.3508997429305913,
|
| 1760 |
+
"eval_gooaq_pairs_loss": 0.12118146568536758,
|
| 1761 |
+
"eval_gooaq_pairs_runtime": 2.236,
|
| 1762 |
+
"eval_gooaq_pairs_samples_per_second": 57.245,
|
| 1763 |
+
"eval_gooaq_pairs_steps_per_second": 0.447,
|
| 1764 |
+
"step": 182
|
| 1765 |
+
},
|
| 1766 |
+
{
|
| 1767 |
+
"epoch": 0.3508997429305913,
|
| 1768 |
+
"eval_paws-pos_loss": 0.020900897681713104,
|
| 1769 |
+
"eval_paws-pos_runtime": 2.2981,
|
| 1770 |
+
"eval_paws-pos_samples_per_second": 55.698,
|
| 1771 |
+
"eval_paws-pos_steps_per_second": 0.435,
|
| 1772 |
+
"step": 182
|
| 1773 |
+
},
|
| 1774 |
+
{
|
| 1775 |
+
"epoch": 0.3508997429305913,
|
| 1776 |
+
"eval_global_dataset_loss": 0.38996464014053345,
|
| 1777 |
+
"eval_global_dataset_runtime": 8.423,
|
| 1778 |
+
"eval_global_dataset_samples_per_second": 38.585,
|
| 1779 |
+
"eval_global_dataset_steps_per_second": 0.356,
|
| 1780 |
+
"step": 182
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.352827763496144,
|
| 1784 |
+
"grad_norm": 412.62738037109375,
|
| 1785 |
+
"learning_rate": 1.0019305019305019e-05,
|
| 1786 |
+
"loss": 0.5365,
|
| 1787 |
+
"step": 183
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.35861182519280205,
|
| 1791 |
+
"grad_norm": 693.9551391601562,
|
| 1792 |
+
"learning_rate": 1.0193050193050193e-05,
|
| 1793 |
+
"loss": 0.6902,
|
| 1794 |
+
"step": 186
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.3643958868894602,
|
| 1798 |
+
"grad_norm": 739.5757446289062,
|
| 1799 |
+
"learning_rate": 1.0366795366795366e-05,
|
| 1800 |
+
"loss": 0.4105,
|
| 1801 |
+
"step": 189
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.37017994858611825,
|
| 1805 |
+
"grad_norm": 174.94882202148438,
|
| 1806 |
+
"learning_rate": 1.0540540540540541e-05,
|
| 1807 |
+
"loss": 0.2434,
|
| 1808 |
+
"step": 192
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.3759640102827764,
|
| 1812 |
+
"grad_norm": 5.913711071014404,
|
| 1813 |
+
"learning_rate": 1.0714285714285714e-05,
|
| 1814 |
+
"loss": 0.1521,
|
| 1815 |
+
"step": 195
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.38174807197943444,
|
| 1819 |
+
"grad_norm": 34.247581481933594,
|
| 1820 |
+
"learning_rate": 1.0888030888030889e-05,
|
| 1821 |
+
"loss": 0.1878,
|
| 1822 |
+
"step": 198
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.38753213367609257,
|
| 1826 |
+
"grad_norm": 50.26247787475586,
|
| 1827 |
+
"learning_rate": 1.1061776061776062e-05,
|
| 1828 |
+
"loss": 0.3544,
|
| 1829 |
+
"step": 201
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.39331619537275064,
|
| 1833 |
+
"grad_norm": 227.15286254882812,
|
| 1834 |
+
"learning_rate": 1.1235521235521235e-05,
|
| 1835 |
+
"loss": 0.1397,
|
| 1836 |
+
"step": 204
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.3991002570694087,
|
| 1840 |
+
"grad_norm": 95.1203384399414,
|
| 1841 |
+
"learning_rate": 1.140926640926641e-05,
|
| 1842 |
+
"loss": 0.2982,
|
| 1843 |
+
"step": 207
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.40102827763496146,
|
| 1847 |
+
"eval_Qnli-dev_cosine_accuracy": 0.685546875,
|
| 1848 |
+
"eval_Qnli-dev_cosine_accuracy_threshold": 0.715149462223053,
|
| 1849 |
+
"eval_Qnli-dev_cosine_ap": 0.7217325275419479,
|
| 1850 |
+
"eval_Qnli-dev_cosine_f1": 0.6854130052724077,
|
| 1851 |
+
"eval_Qnli-dev_cosine_f1_threshold": 0.658507227897644,
|
| 1852 |
+
"eval_Qnli-dev_cosine_precision": 0.5855855855855856,
|
| 1853 |
+
"eval_Qnli-dev_cosine_recall": 0.826271186440678,
|
| 1854 |
+
"eval_Qnli-dev_dot_accuracy": 0.669921875,
|
| 1855 |
+
"eval_Qnli-dev_dot_accuracy_threshold": 717.3741455078125,
|
| 1856 |
+
"eval_Qnli-dev_dot_ap": 0.6857348319823914,
|
| 1857 |
+
"eval_Qnli-dev_dot_f1": 0.6757679180887372,
|
| 1858 |
+
"eval_Qnli-dev_dot_f1_threshold": 627.569580078125,
|
| 1859 |
+
"eval_Qnli-dev_dot_precision": 0.5657142857142857,
|
| 1860 |
+
"eval_Qnli-dev_dot_recall": 0.8389830508474576,
|
| 1861 |
+
"eval_Qnli-dev_euclidean_accuracy": 0.68359375,
|
| 1862 |
+
"eval_Qnli-dev_euclidean_accuracy_threshold": 23.550647735595703,
|
| 1863 |
+
"eval_Qnli-dev_euclidean_ap": 0.7266186529674616,
|
| 1864 |
+
"eval_Qnli-dev_euclidean_f1": 0.6959578207381371,
|
| 1865 |
+
"eval_Qnli-dev_euclidean_f1_threshold": 25.634620666503906,
|
| 1866 |
+
"eval_Qnli-dev_euclidean_precision": 0.5945945945945946,
|
| 1867 |
+
"eval_Qnli-dev_euclidean_recall": 0.8389830508474576,
|
| 1868 |
+
"eval_Qnli-dev_manhattan_accuracy": 0.6796875,
|
| 1869 |
+
"eval_Qnli-dev_manhattan_accuracy_threshold": 681.6762084960938,
|
| 1870 |
+
"eval_Qnli-dev_manhattan_ap": 0.7264245248641615,
|
| 1871 |
+
"eval_Qnli-dev_manhattan_f1": 0.6938053097345133,
|
| 1872 |
+
"eval_Qnli-dev_manhattan_f1_threshold": 793.83837890625,
|
| 1873 |
+
"eval_Qnli-dev_manhattan_precision": 0.5957446808510638,
|
| 1874 |
+
"eval_Qnli-dev_manhattan_recall": 0.8305084745762712,
|
| 1875 |
+
"eval_Qnli-dev_max_accuracy": 0.685546875,
|
| 1876 |
+
"eval_Qnli-dev_max_accuracy_threshold": 717.3741455078125,
|
| 1877 |
+
"eval_Qnli-dev_max_ap": 0.7266186529674616,
|
| 1878 |
+
"eval_Qnli-dev_max_f1": 0.6959578207381371,
|
| 1879 |
+
"eval_Qnli-dev_max_f1_threshold": 793.83837890625,
|
| 1880 |
+
"eval_Qnli-dev_max_precision": 0.5957446808510638,
|
| 1881 |
+
"eval_Qnli-dev_max_recall": 0.8389830508474576,
|
| 1882 |
+
"eval_allNLI-dev_cosine_accuracy": 0.720703125,
|
| 1883 |
+
"eval_allNLI-dev_cosine_accuracy_threshold": 0.8279609680175781,
|
| 1884 |
+
"eval_allNLI-dev_cosine_ap": 0.5853174176053627,
|
| 1885 |
+
"eval_allNLI-dev_cosine_f1": 0.6189376443418013,
|
| 1886 |
+
"eval_allNLI-dev_cosine_f1_threshold": 0.7112143039703369,
|
| 1887 |
+
"eval_allNLI-dev_cosine_precision": 0.5153846153846153,
|
| 1888 |
+
"eval_allNLI-dev_cosine_recall": 0.7745664739884393,
|
| 1889 |
+
"eval_allNLI-dev_dot_accuracy": 0.708984375,
|
| 1890 |
+
"eval_allNLI-dev_dot_accuracy_threshold": 844.9638061523438,
|
| 1891 |
+
"eval_allNLI-dev_dot_ap": 0.5767480495013521,
|
| 1892 |
+
"eval_allNLI-dev_dot_f1": 0.620525059665871,
|
| 1893 |
+
"eval_allNLI-dev_dot_f1_threshold": 698.743408203125,
|
| 1894 |
+
"eval_allNLI-dev_dot_precision": 0.5284552845528455,
|
| 1895 |
+
"eval_allNLI-dev_dot_recall": 0.7514450867052023,
|
| 1896 |
+
"eval_allNLI-dev_euclidean_accuracy": 0.71875,
|
| 1897 |
+
"eval_allNLI-dev_euclidean_accuracy_threshold": 17.55453109741211,
|
| 1898 |
+
"eval_allNLI-dev_euclidean_ap": 0.5835774601767848,
|
| 1899 |
+
"eval_allNLI-dev_euclidean_f1": 0.6063569682151589,
|
| 1900 |
+
"eval_allNLI-dev_euclidean_f1_threshold": 22.958927154541016,
|
| 1901 |
+
"eval_allNLI-dev_euclidean_precision": 0.5254237288135594,
|
| 1902 |
+
"eval_allNLI-dev_euclidean_recall": 0.7167630057803468,
|
| 1903 |
+
"eval_allNLI-dev_manhattan_accuracy": 0.71875,
|
| 1904 |
+
"eval_allNLI-dev_manhattan_accuracy_threshold": 546.237060546875,
|
| 1905 |
+
"eval_allNLI-dev_manhattan_ap": 0.5834282573106326,
|
| 1906 |
+
"eval_allNLI-dev_manhattan_f1": 0.6093023255813954,
|
| 1907 |
+
"eval_allNLI-dev_manhattan_f1_threshold": 738.0977172851562,
|
| 1908 |
+
"eval_allNLI-dev_manhattan_precision": 0.5097276264591439,
|
| 1909 |
+
"eval_allNLI-dev_manhattan_recall": 0.7572254335260116,
|
| 1910 |
+
"eval_allNLI-dev_max_accuracy": 0.720703125,
|
| 1911 |
+
"eval_allNLI-dev_max_accuracy_threshold": 844.9638061523438,
|
| 1912 |
+
"eval_allNLI-dev_max_ap": 0.5853174176053627,
|
| 1913 |
+
"eval_allNLI-dev_max_f1": 0.620525059665871,
|
| 1914 |
+
"eval_allNLI-dev_max_f1_threshold": 738.0977172851562,
|
| 1915 |
+
"eval_allNLI-dev_max_precision": 0.5284552845528455,
|
| 1916 |
+
"eval_allNLI-dev_max_recall": 0.7745664739884393,
|
| 1917 |
+
"eval_sequential_score": 0.7266186529674616,
|
| 1918 |
+
"eval_sts-test_pearson_cosine": 0.9110444991658042,
|
| 1919 |
+
"eval_sts-test_pearson_dot": 0.9027007693117498,
|
| 1920 |
+
"eval_sts-test_pearson_euclidean": 0.9255983263878029,
|
| 1921 |
+
"eval_sts-test_pearson_manhattan": 0.925028098980626,
|
| 1922 |
+
"eval_sts-test_pearson_max": 0.9255983263878029,
|
| 1923 |
+
"eval_sts-test_spearman_cosine": 0.9211899030744385,
|
| 1924 |
+
"eval_sts-test_spearman_dot": 0.9034949858127107,
|
| 1925 |
+
"eval_sts-test_spearman_euclidean": 0.9209135640353466,
|
| 1926 |
+
"eval_sts-test_spearman_manhattan": 0.9206623304538951,
|
| 1927 |
+
"eval_sts-test_spearman_max": 0.9211899030744385,
|
| 1928 |
+
"eval_vitaminc-pairs_loss": 3.210468292236328,
|
| 1929 |
+
"eval_vitaminc-pairs_runtime": 4.4941,
|
| 1930 |
+
"eval_vitaminc-pairs_samples_per_second": 28.482,
|
| 1931 |
+
"eval_vitaminc-pairs_steps_per_second": 0.223,
|
| 1932 |
+
"step": 208
|
| 1933 |
+
},
|
| 1934 |
+
{
|
| 1935 |
+
"epoch": 0.40102827763496146,
|
| 1936 |
+
"eval_negation-triplets_loss": 0.6793166995048523,
|
| 1937 |
+
"eval_negation-triplets_runtime": 3.2613,
|
| 1938 |
+
"eval_negation-triplets_samples_per_second": 39.249,
|
| 1939 |
+
"eval_negation-triplets_steps_per_second": 0.307,
|
| 1940 |
+
"step": 208
|
| 1941 |
+
},
|
| 1942 |
+
{
|
| 1943 |
+
"epoch": 0.40102827763496146,
|
| 1944 |
+
"eval_scitail-pairs-pos_loss": 0.020910607650876045,
|
| 1945 |
+
"eval_scitail-pairs-pos_runtime": 2.6259,
|
| 1946 |
+
"eval_scitail-pairs-pos_samples_per_second": 48.745,
|
| 1947 |
+
"eval_scitail-pairs-pos_steps_per_second": 0.381,
|
| 1948 |
+
"step": 208
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.40102827763496146,
|
| 1952 |
+
"eval_scitail-pairs-qa_loss": 0.024287551641464233,
|
| 1953 |
+
"eval_scitail-pairs-qa_runtime": 2.2326,
|
| 1954 |
+
"eval_scitail-pairs-qa_samples_per_second": 57.331,
|
| 1955 |
+
"eval_scitail-pairs-qa_steps_per_second": 0.448,
|
| 1956 |
+
"step": 208
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 0.40102827763496146,
|
| 1960 |
+
"eval_xsum-pairs_loss": 0.028975818306207657,
|
| 1961 |
+
"eval_xsum-pairs_runtime": 3.0795,
|
| 1962 |
+
"eval_xsum-pairs_samples_per_second": 41.565,
|
| 1963 |
+
"eval_xsum-pairs_steps_per_second": 0.325,
|
| 1964 |
+
"step": 208
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 0.40102827763496146,
|
| 1968 |
+
"eval_sciq_pairs_loss": 0.08505217730998993,
|
| 1969 |
+
"eval_sciq_pairs_runtime": 4.7361,
|
| 1970 |
+
"eval_sciq_pairs_samples_per_second": 27.026,
|
| 1971 |
+
"eval_sciq_pairs_steps_per_second": 0.211,
|
| 1972 |
+
"step": 208
|
| 1973 |
+
},
|
| 1974 |
+
{
|
| 1975 |
+
"epoch": 0.40102827763496146,
|
| 1976 |
+
"eval_qasc_pairs_loss": 0.1334802210330963,
|
| 1977 |
+
"eval_qasc_pairs_runtime": 2.0711,
|
| 1978 |
+
"eval_qasc_pairs_samples_per_second": 61.804,
|
| 1979 |
+
"eval_qasc_pairs_steps_per_second": 0.483,
|
| 1980 |
+
"step": 208
|
| 1981 |
+
},
|
| 1982 |
+
{
|
| 1983 |
+
"epoch": 0.40102827763496146,
|
| 1984 |
+
"eval_openbookqa_pairs_loss": 0.4361065924167633,
|
| 1985 |
+
"eval_openbookqa_pairs_runtime": 2.2402,
|
| 1986 |
+
"eval_openbookqa_pairs_samples_per_second": 57.137,
|
| 1987 |
+
"eval_openbookqa_pairs_steps_per_second": 0.446,
|
| 1988 |
+
"step": 208
|
| 1989 |
+
},
|
| 1990 |
+
{
|
| 1991 |
+
"epoch": 0.40102827763496146,
|
| 1992 |
+
"eval_msmarco_pairs_loss": 0.09543590992689133,
|
| 1993 |
+
"eval_msmarco_pairs_runtime": 2.2223,
|
| 1994 |
+
"eval_msmarco_pairs_samples_per_second": 57.599,
|
| 1995 |
+
"eval_msmarco_pairs_steps_per_second": 0.45,
|
| 1996 |
+
"step": 208
|
| 1997 |
+
},
|
| 1998 |
+
{
|
| 1999 |
+
"epoch": 0.40102827763496146,
|
| 2000 |
+
"eval_nq_pairs_loss": 0.24286797642707825,
|
| 2001 |
+
"eval_nq_pairs_runtime": 3.617,
|
| 2002 |
+
"eval_nq_pairs_samples_per_second": 35.389,
|
| 2003 |
+
"eval_nq_pairs_steps_per_second": 0.276,
|
| 2004 |
+
"step": 208
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.40102827763496146,
|
| 2008 |
+
"eval_trivia_pairs_loss": 0.053676724433898926,
|
| 2009 |
+
"eval_trivia_pairs_runtime": 3.0737,
|
| 2010 |
+
"eval_trivia_pairs_samples_per_second": 41.644,
|
| 2011 |
+
"eval_trivia_pairs_steps_per_second": 0.325,
|
| 2012 |
+
"step": 208
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 0.40102827763496146,
|
| 2016 |
+
"eval_gooaq_pairs_loss": 0.07278460264205933,
|
| 2017 |
+
"eval_gooaq_pairs_runtime": 2.1378,
|
| 2018 |
+
"eval_gooaq_pairs_samples_per_second": 59.874,
|
| 2019 |
+
"eval_gooaq_pairs_steps_per_second": 0.468,
|
| 2020 |
+
"step": 208
|
| 2021 |
+
},
|
| 2022 |
+
{
|
| 2023 |
+
"epoch": 0.40102827763496146,
|
| 2024 |
+
"eval_paws-pos_loss": 0.022397885099053383,
|
| 2025 |
+
"eval_paws-pos_runtime": 2.253,
|
| 2026 |
+
"eval_paws-pos_samples_per_second": 56.812,
|
| 2027 |
+
"eval_paws-pos_steps_per_second": 0.444,
|
| 2028 |
+
"step": 208
|
| 2029 |
+
},
|
| 2030 |
+
{
|
| 2031 |
+
"epoch": 0.40102827763496146,
|
| 2032 |
+
"eval_global_dataset_loss": 0.250370055437088,
|
| 2033 |
+
"eval_global_dataset_runtime": 8.321,
|
| 2034 |
+
"eval_global_dataset_samples_per_second": 39.058,
|
| 2035 |
+
"eval_global_dataset_steps_per_second": 0.361,
|
| 2036 |
+
"step": 208
|
| 2037 |
+
}
|
| 2038 |
+
],
|
| 2039 |
+
"logging_steps": 3,
|
| 2040 |
+
"max_steps": 1036,
|
| 2041 |
+
"num_input_tokens_seen": 0,
|
| 2042 |
+
"num_train_epochs": 2,
|
| 2043 |
+
"save_steps": 52,
|
| 2044 |
+
"stateful_callbacks": {
|
| 2045 |
+
"TrainerControl": {
|
| 2046 |
+
"args": {
|
| 2047 |
+
"should_epoch_stop": false,
|
| 2048 |
+
"should_evaluate": false,
|
| 2049 |
+
"should_log": false,
|
| 2050 |
+
"should_save": true,
|
| 2051 |
+
"should_training_stop": false
|
| 2052 |
+
},
|
| 2053 |
+
"attributes": {}
|
| 2054 |
+
}
|
| 2055 |
+
},
|
| 2056 |
+
"total_flos": 0.0,
|
| 2057 |
+
"train_batch_size": 64,
|
| 2058 |
+
"trial_name": null,
|
| 2059 |
+
"trial_params": null
|
| 2060 |
+
}
|
checkpoint-208/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:73b62473318b3b28c705f94b87acfb0646732bae15a12e95a938b998dc89c2e7
|
| 3 |
+
size 5560
|