commit files to HF hub
Browse files- added_tokens.json +1 -0
- config.json +42 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +1 -0
- tokenizer.json +0 -0
- tokenizer_config.json +1 -0
- trainer_state.json +940 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
added_tokens.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"[noise-no majority]": 30533, "[food-no majority]": 30525, "[service-unknown]": 30536, "[ambiance-Positive]": 30527, "[food-Negative]": 30522, "[service-Positive]": 30535, "[service-no majority]": 30537, "[ambiance-Negative]": 30526, "[food-Positive]": 30523, "[noise-Negative]": 30530, "[ambiance-no majority]": 30529, "[noise-unknown]": 30532, "[noise-Positive]": 30531, "[ambiance-unknown]": 30528, "[service-Negative]": 30534, "[food-unknown]": 30524}
|
config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "CEBaB/bert-base-uncased.CEBaB.sa.5-class.exclusive.seed_66",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForFactualCounterfactualSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"finetuning_task": "opentable",
|
| 9 |
+
"gradient_checkpointing": false,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 768,
|
| 13 |
+
"id2label": {
|
| 14 |
+
"0": 0,
|
| 15 |
+
"1": 1,
|
| 16 |
+
"2": 2,
|
| 17 |
+
"3": 3,
|
| 18 |
+
"4": 4
|
| 19 |
+
},
|
| 20 |
+
"initializer_range": 0.02,
|
| 21 |
+
"intermediate_size": 3072,
|
| 22 |
+
"label2id": {
|
| 23 |
+
"0": 0,
|
| 24 |
+
"1": 1,
|
| 25 |
+
"2": 2,
|
| 26 |
+
"3": 3,
|
| 27 |
+
"4": 4
|
| 28 |
+
},
|
| 29 |
+
"layer_norm_eps": 1e-12,
|
| 30 |
+
"max_position_embeddings": 512,
|
| 31 |
+
"model_type": "bert",
|
| 32 |
+
"num_attention_heads": 12,
|
| 33 |
+
"num_hidden_layers": 12,
|
| 34 |
+
"pad_token_id": 0,
|
| 35 |
+
"position_embedding_type": "absolute",
|
| 36 |
+
"problem_type": "single_label_classification",
|
| 37 |
+
"torch_dtype": "float32",
|
| 38 |
+
"transformers_version": "4.18.0",
|
| 39 |
+
"type_vocab_size": 2,
|
| 40 |
+
"use_cache": true,
|
| 41 |
+
"vocab_size": 30538
|
| 42 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6230e85082611687ca000db2643699b6f0a6c9f27967979dcc2fd9e729425519
|
| 3 |
+
size 880829397
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d6220c671d8a8c5b5a34c852bc5f399ff9c8c596e2a349e9ec07829c7d60e207
|
| 3 |
+
size 440427497
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f015fd017d4902e772b448460cefbb5e1bea5f131a0af73bd37a7f9f46fbd6bb
|
| 3 |
+
size 15523
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:205f1174c0a91dea8a7361f8887a0f25857169dbf40de2edf28a99d7bf83bf5e
|
| 3 |
+
size 623
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "additional_special_tokens": ["[food-Negative]", "[food-Positive]", "[food-unknown]", "[food-no majority]", "[ambiance-Negative]", "[ambiance-Positive]", "[ambiance-unknown]", "[ambiance-no majority]", "[noise-Negative]", "[noise-Positive]", "[noise-unknown]", "[noise-no majority]", "[service-Negative]", "[service-Positive]", "[service-unknown]", "[service-no majority]"]}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "bert-base-uncased", "tokenizer_class": "BertTokenizer"}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,940 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.4564709961414337,
|
| 3 |
+
"best_model_checkpoint": "model_output/e2e_opentable_5_way__approximate__0-shot__seed-66__bert-base-uncased/checkpoint-1400",
|
| 4 |
+
"epoch": 2.4013722126929675,
|
| 5 |
+
"global_step": 1400,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.03,
|
| 12 |
+
"learning_rate": 4.9783362218370885e-05,
|
| 13 |
+
"loss": 11.2983,
|
| 14 |
+
"step": 20
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.07,
|
| 18 |
+
"learning_rate": 4.956672443674177e-05,
|
| 19 |
+
"loss": 8.8977,
|
| 20 |
+
"step": 40
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.09,
|
| 24 |
+
"eval_distillation_accuracy_counterfactual": 0.5030785017957927,
|
| 25 |
+
"eval_distillation_accuracy_factual": 0.870702924576706,
|
| 26 |
+
"eval_distillation_f1_counterfactual": 0.36918123521791174,
|
| 27 |
+
"eval_distillation_f1_factual": 0.8518990750712729,
|
| 28 |
+
"eval_groundtruth_accuracy_counterfactual": 0.4763981528989225,
|
| 29 |
+
"eval_groundtruth_f1_counterfactual": 0.3533145590655197,
|
| 30 |
+
"eval_groundtruth_f1_factual": 0.6552258773293613,
|
| 31 |
+
"eval_icace_cosine": 0.4796179533004761,
|
| 32 |
+
"eval_icace_l2": 0.6570112705230713,
|
| 33 |
+
"eval_icace_normdiff": 0.4324696958065033,
|
| 34 |
+
"eval_loss": 6.444849967956543,
|
| 35 |
+
"eval_runtime": 22.056,
|
| 36 |
+
"eval_samples_per_second": 176.732,
|
| 37 |
+
"eval_steps_per_second": 2.766,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.1,
|
| 42 |
+
"learning_rate": 4.935008665511265e-05,
|
| 43 |
+
"loss": 8.1157,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.14,
|
| 48 |
+
"learning_rate": 4.913344887348354e-05,
|
| 49 |
+
"loss": 7.8853,
|
| 50 |
+
"step": 80
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.17,
|
| 54 |
+
"learning_rate": 4.891681109185442e-05,
|
| 55 |
+
"loss": 7.6564,
|
| 56 |
+
"step": 100
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.17,
|
| 60 |
+
"eval_distillation_accuracy_counterfactual": 0.5312981015905592,
|
| 61 |
+
"eval_distillation_accuracy_factual": 0.8776295536172396,
|
| 62 |
+
"eval_distillation_f1_counterfactual": 0.4177716134310396,
|
| 63 |
+
"eval_distillation_f1_factual": 0.8606344639589892,
|
| 64 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5,
|
| 65 |
+
"eval_groundtruth_f1_counterfactual": 0.39828651286680133,
|
| 66 |
+
"eval_groundtruth_f1_factual": 0.6882338232741171,
|
| 67 |
+
"eval_icace_cosine": 0.4674900770187378,
|
| 68 |
+
"eval_icace_l2": 0.6195012331008911,
|
| 69 |
+
"eval_icace_normdiff": 0.3801978528499603,
|
| 70 |
+
"eval_loss": 6.348506450653076,
|
| 71 |
+
"eval_runtime": 22.6417,
|
| 72 |
+
"eval_samples_per_second": 172.16,
|
| 73 |
+
"eval_steps_per_second": 2.694,
|
| 74 |
+
"step": 100
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.21,
|
| 78 |
+
"learning_rate": 4.8700173310225307e-05,
|
| 79 |
+
"loss": 7.5574,
|
| 80 |
+
"step": 120
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.24,
|
| 84 |
+
"learning_rate": 4.848353552859619e-05,
|
| 85 |
+
"loss": 7.2643,
|
| 86 |
+
"step": 140
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.26,
|
| 90 |
+
"eval_distillation_accuracy_counterfactual": 0.5287326834273987,
|
| 91 |
+
"eval_distillation_accuracy_factual": 0.8694202154951257,
|
| 92 |
+
"eval_distillation_f1_counterfactual": 0.4118396244303133,
|
| 93 |
+
"eval_distillation_f1_factual": 0.8599084137519469,
|
| 94 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5141097998973833,
|
| 95 |
+
"eval_groundtruth_f1_counterfactual": 0.4079251742601512,
|
| 96 |
+
"eval_groundtruth_f1_factual": 0.6810018997954198,
|
| 97 |
+
"eval_icace_cosine": 0.46454283595085144,
|
| 98 |
+
"eval_icace_l2": 0.6308194398880005,
|
| 99 |
+
"eval_icace_normdiff": 0.40617796778678894,
|
| 100 |
+
"eval_loss": 6.1462531089782715,
|
| 101 |
+
"eval_runtime": 22.1649,
|
| 102 |
+
"eval_samples_per_second": 175.863,
|
| 103 |
+
"eval_steps_per_second": 2.752,
|
| 104 |
+
"step": 150
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.27,
|
| 108 |
+
"learning_rate": 4.826689774696707e-05,
|
| 109 |
+
"loss": 7.3808,
|
| 110 |
+
"step": 160
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.31,
|
| 114 |
+
"learning_rate": 4.8050259965337955e-05,
|
| 115 |
+
"loss": 7.1507,
|
| 116 |
+
"step": 180
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.34,
|
| 120 |
+
"learning_rate": 4.7833622183708845e-05,
|
| 121 |
+
"loss": 7.0277,
|
| 122 |
+
"step": 200
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.34,
|
| 126 |
+
"eval_distillation_accuracy_counterfactual": 0.521806054386865,
|
| 127 |
+
"eval_distillation_accuracy_factual": 0.8594150846587993,
|
| 128 |
+
"eval_distillation_f1_counterfactual": 0.43996859803992977,
|
| 129 |
+
"eval_distillation_f1_factual": 0.8377228086695208,
|
| 130 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5048742945100051,
|
| 131 |
+
"eval_groundtruth_f1_counterfactual": 0.42823671858777795,
|
| 132 |
+
"eval_groundtruth_f1_factual": 0.6661417838076318,
|
| 133 |
+
"eval_icace_cosine": 0.4762553572654724,
|
| 134 |
+
"eval_icace_l2": 0.6346240043640137,
|
| 135 |
+
"eval_icace_normdiff": 0.38476699590682983,
|
| 136 |
+
"eval_loss": 6.443660259246826,
|
| 137 |
+
"eval_runtime": 22.3822,
|
| 138 |
+
"eval_samples_per_second": 174.156,
|
| 139 |
+
"eval_steps_per_second": 2.725,
|
| 140 |
+
"step": 200
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.38,
|
| 144 |
+
"learning_rate": 4.761698440207972e-05,
|
| 145 |
+
"loss": 6.9694,
|
| 146 |
+
"step": 220
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.41,
|
| 150 |
+
"learning_rate": 4.740034662045061e-05,
|
| 151 |
+
"loss": 7.0702,
|
| 152 |
+
"step": 240
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.43,
|
| 156 |
+
"eval_distillation_accuracy_counterfactual": 0.5307850179579271,
|
| 157 |
+
"eval_distillation_accuracy_factual": 0.8640328373524885,
|
| 158 |
+
"eval_distillation_f1_counterfactual": 0.43581597774498526,
|
| 159 |
+
"eval_distillation_f1_factual": 0.8511773812631599,
|
| 160 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5082093381221139,
|
| 161 |
+
"eval_groundtruth_f1_counterfactual": 0.4226816426807245,
|
| 162 |
+
"eval_groundtruth_f1_factual": 0.6714755980104481,
|
| 163 |
+
"eval_icace_cosine": 0.4607926905155182,
|
| 164 |
+
"eval_icace_l2": 0.6025251150131226,
|
| 165 |
+
"eval_icace_normdiff": 0.366618812084198,
|
| 166 |
+
"eval_loss": 6.052988052368164,
|
| 167 |
+
"eval_runtime": 22.0965,
|
| 168 |
+
"eval_samples_per_second": 176.408,
|
| 169 |
+
"eval_steps_per_second": 2.761,
|
| 170 |
+
"step": 250
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.45,
|
| 174 |
+
"learning_rate": 4.7183708838821494e-05,
|
| 175 |
+
"loss": 6.8827,
|
| 176 |
+
"step": 260
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.48,
|
| 180 |
+
"learning_rate": 4.6967071057192376e-05,
|
| 181 |
+
"loss": 6.8895,
|
| 182 |
+
"step": 280
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.51,
|
| 186 |
+
"learning_rate": 4.675043327556326e-05,
|
| 187 |
+
"loss": 6.744,
|
| 188 |
+
"step": 300
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.51,
|
| 192 |
+
"eval_distillation_accuracy_counterfactual": 0.5377116469984607,
|
| 193 |
+
"eval_distillation_accuracy_factual": 0.8612108773730118,
|
| 194 |
+
"eval_distillation_f1_counterfactual": 0.4223090065654679,
|
| 195 |
+
"eval_distillation_f1_factual": 0.8456621535049905,
|
| 196 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5143663417136993,
|
| 197 |
+
"eval_groundtruth_f1_counterfactual": 0.40663553727210966,
|
| 198 |
+
"eval_groundtruth_f1_factual": 0.6775861228321817,
|
| 199 |
+
"eval_icace_cosine": 0.46002256870269775,
|
| 200 |
+
"eval_icace_l2": 0.5889779329299927,
|
| 201 |
+
"eval_icace_normdiff": 0.3519256114959717,
|
| 202 |
+
"eval_loss": 5.902462005615234,
|
| 203 |
+
"eval_runtime": 22.3125,
|
| 204 |
+
"eval_samples_per_second": 174.7,
|
| 205 |
+
"eval_steps_per_second": 2.734,
|
| 206 |
+
"step": 300
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.55,
|
| 210 |
+
"learning_rate": 4.653379549393415e-05,
|
| 211 |
+
"loss": 6.6897,
|
| 212 |
+
"step": 320
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.58,
|
| 216 |
+
"learning_rate": 4.6317157712305025e-05,
|
| 217 |
+
"loss": 6.7666,
|
| 218 |
+
"step": 340
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.6,
|
| 222 |
+
"eval_distillation_accuracy_counterfactual": 0.5548999486916367,
|
| 223 |
+
"eval_distillation_accuracy_factual": 0.8499230374551052,
|
| 224 |
+
"eval_distillation_f1_counterfactual": 0.49580886864510487,
|
| 225 |
+
"eval_distillation_f1_factual": 0.8361446774720787,
|
| 226 |
+
"eval_groundtruth_accuracy_counterfactual": 0.538737814263725,
|
| 227 |
+
"eval_groundtruth_f1_counterfactual": 0.4813622017124305,
|
| 228 |
+
"eval_groundtruth_f1_factual": 0.674748187131342,
|
| 229 |
+
"eval_icace_cosine": 0.46281909942626953,
|
| 230 |
+
"eval_icace_l2": 0.5994406342506409,
|
| 231 |
+
"eval_icace_normdiff": 0.3744681477546692,
|
| 232 |
+
"eval_loss": 5.933077335357666,
|
| 233 |
+
"eval_runtime": 22.1654,
|
| 234 |
+
"eval_samples_per_second": 175.86,
|
| 235 |
+
"eval_steps_per_second": 2.752,
|
| 236 |
+
"step": 350
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.62,
|
| 240 |
+
"learning_rate": 4.6100519930675915e-05,
|
| 241 |
+
"loss": 6.5721,
|
| 242 |
+
"step": 360
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.65,
|
| 246 |
+
"learning_rate": 4.58838821490468e-05,
|
| 247 |
+
"loss": 6.7086,
|
| 248 |
+
"step": 380
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.69,
|
| 252 |
+
"learning_rate": 4.566724436741768e-05,
|
| 253 |
+
"loss": 6.5311,
|
| 254 |
+
"step": 400
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.69,
|
| 258 |
+
"eval_distillation_accuracy_counterfactual": 0.5556695741405849,
|
| 259 |
+
"eval_distillation_accuracy_factual": 0.8724987172909184,
|
| 260 |
+
"eval_distillation_f1_counterfactual": 0.456210753567529,
|
| 261 |
+
"eval_distillation_f1_factual": 0.8597691779856307,
|
| 262 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5333504361210878,
|
| 263 |
+
"eval_groundtruth_f1_counterfactual": 0.444792193959889,
|
| 264 |
+
"eval_groundtruth_f1_factual": 0.6708274089754076,
|
| 265 |
+
"eval_icace_cosine": 0.45892879366874695,
|
| 266 |
+
"eval_icace_l2": 0.5904973745346069,
|
| 267 |
+
"eval_icace_normdiff": 0.3636997938156128,
|
| 268 |
+
"eval_loss": 5.868222236633301,
|
| 269 |
+
"eval_runtime": 22.3242,
|
| 270 |
+
"eval_samples_per_second": 174.609,
|
| 271 |
+
"eval_steps_per_second": 2.732,
|
| 272 |
+
"step": 400
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.72,
|
| 276 |
+
"learning_rate": 4.5450606585788563e-05,
|
| 277 |
+
"loss": 6.5486,
|
| 278 |
+
"step": 420
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.75,
|
| 282 |
+
"learning_rate": 4.5233968804159446e-05,
|
| 283 |
+
"loss": 6.4615,
|
| 284 |
+
"step": 440
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.77,
|
| 288 |
+
"eval_distillation_accuracy_counterfactual": 0.5428424833247819,
|
| 289 |
+
"eval_distillation_accuracy_factual": 0.8601847101077476,
|
| 290 |
+
"eval_distillation_f1_counterfactual": 0.45475006277273583,
|
| 291 |
+
"eval_distillation_f1_factual": 0.8490794790111732,
|
| 292 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5223191380194971,
|
| 293 |
+
"eval_groundtruth_f1_counterfactual": 0.44181890062982204,
|
| 294 |
+
"eval_groundtruth_f1_factual": 0.6911106055165974,
|
| 295 |
+
"eval_icace_cosine": 0.4593462646007538,
|
| 296 |
+
"eval_icace_l2": 0.5996768474578857,
|
| 297 |
+
"eval_icace_normdiff": 0.3729654550552368,
|
| 298 |
+
"eval_loss": 5.918853282928467,
|
| 299 |
+
"eval_runtime": 22.1691,
|
| 300 |
+
"eval_samples_per_second": 175.83,
|
| 301 |
+
"eval_steps_per_second": 2.752,
|
| 302 |
+
"step": 450
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.79,
|
| 306 |
+
"learning_rate": 4.501733102253033e-05,
|
| 307 |
+
"loss": 6.4495,
|
| 308 |
+
"step": 460
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.82,
|
| 312 |
+
"learning_rate": 4.480069324090121e-05,
|
| 313 |
+
"loss": 6.6214,
|
| 314 |
+
"step": 480
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.86,
|
| 318 |
+
"learning_rate": 4.45840554592721e-05,
|
| 319 |
+
"loss": 6.5091,
|
| 320 |
+
"step": 500
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.86,
|
| 324 |
+
"eval_distillation_accuracy_counterfactual": 0.5528476141611083,
|
| 325 |
+
"eval_distillation_accuracy_factual": 0.8606977937403797,
|
| 326 |
+
"eval_distillation_f1_counterfactual": 0.42193207949843314,
|
| 327 |
+
"eval_distillation_f1_factual": 0.8449657823376556,
|
| 328 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5318111852231914,
|
| 329 |
+
"eval_groundtruth_f1_counterfactual": 0.4123797131290104,
|
| 330 |
+
"eval_groundtruth_f1_factual": 0.6754423477994138,
|
| 331 |
+
"eval_icace_cosine": 0.4626261293888092,
|
| 332 |
+
"eval_icace_l2": 0.6023271083831787,
|
| 333 |
+
"eval_icace_normdiff": 0.37480491399765015,
|
| 334 |
+
"eval_loss": 5.962100028991699,
|
| 335 |
+
"eval_runtime": 22.2411,
|
| 336 |
+
"eval_samples_per_second": 175.261,
|
| 337 |
+
"eval_steps_per_second": 2.743,
|
| 338 |
+
"step": 500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.89,
|
| 342 |
+
"learning_rate": 4.436741767764298e-05,
|
| 343 |
+
"loss": 6.5299,
|
| 344 |
+
"step": 520
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.93,
|
| 348 |
+
"learning_rate": 4.415077989601387e-05,
|
| 349 |
+
"loss": 6.5688,
|
| 350 |
+
"step": 540
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.94,
|
| 354 |
+
"eval_distillation_accuracy_counterfactual": 0.5377116469984607,
|
| 355 |
+
"eval_distillation_accuracy_factual": 0.8583889173935352,
|
| 356 |
+
"eval_distillation_f1_counterfactual": 0.46729782076302406,
|
| 357 |
+
"eval_distillation_f1_factual": 0.8469926857732132,
|
| 358 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5105182144689584,
|
| 359 |
+
"eval_groundtruth_f1_counterfactual": 0.44493052070683115,
|
| 360 |
+
"eval_groundtruth_f1_factual": 0.6879462817233734,
|
| 361 |
+
"eval_icace_cosine": 0.4675835072994232,
|
| 362 |
+
"eval_icace_l2": 0.6114413738250732,
|
| 363 |
+
"eval_icace_normdiff": 0.376953661441803,
|
| 364 |
+
"eval_loss": 6.076689720153809,
|
| 365 |
+
"eval_runtime": 22.1309,
|
| 366 |
+
"eval_samples_per_second": 176.134,
|
| 367 |
+
"eval_steps_per_second": 2.756,
|
| 368 |
+
"step": 550
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.96,
|
| 372 |
+
"learning_rate": 4.393414211438475e-05,
|
| 373 |
+
"loss": 6.5147,
|
| 374 |
+
"step": 560
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.99,
|
| 378 |
+
"learning_rate": 4.371750433275563e-05,
|
| 379 |
+
"loss": 6.5651,
|
| 380 |
+
"step": 580
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 1.03,
|
| 384 |
+
"learning_rate": 4.3500866551126516e-05,
|
| 385 |
+
"loss": 6.2884,
|
| 386 |
+
"step": 600
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 1.03,
|
| 390 |
+
"eval_distillation_accuracy_counterfactual": 0.5315546434068753,
|
| 391 |
+
"eval_distillation_accuracy_factual": 0.8199076449461262,
|
| 392 |
+
"eval_distillation_f1_counterfactual": 0.4240354476459813,
|
| 393 |
+
"eval_distillation_f1_factual": 0.8110143323147028,
|
| 394 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5141097998973833,
|
| 395 |
+
"eval_groundtruth_f1_counterfactual": 0.4161445570845627,
|
| 396 |
+
"eval_groundtruth_f1_factual": 0.6701739679915534,
|
| 397 |
+
"eval_icace_cosine": 0.4710710346698761,
|
| 398 |
+
"eval_icace_l2": 0.6157806515693665,
|
| 399 |
+
"eval_icace_normdiff": 0.36949479579925537,
|
| 400 |
+
"eval_loss": 6.440975666046143,
|
| 401 |
+
"eval_runtime": 22.0879,
|
| 402 |
+
"eval_samples_per_second": 176.477,
|
| 403 |
+
"eval_steps_per_second": 2.762,
|
| 404 |
+
"step": 600
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 1.06,
|
| 408 |
+
"learning_rate": 4.3284228769497406e-05,
|
| 409 |
+
"loss": 6.3874,
|
| 410 |
+
"step": 620
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 1.1,
|
| 414 |
+
"learning_rate": 4.306759098786828e-05,
|
| 415 |
+
"loss": 6.2439,
|
| 416 |
+
"step": 640
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.11,
|
| 420 |
+
"eval_distillation_accuracy_counterfactual": 0.5572088250384812,
|
| 421 |
+
"eval_distillation_accuracy_factual": 0.8345305284761416,
|
| 422 |
+
"eval_distillation_f1_counterfactual": 0.4681040280988178,
|
| 423 |
+
"eval_distillation_f1_factual": 0.8289204998070453,
|
| 424 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5428424833247819,
|
| 425 |
+
"eval_groundtruth_f1_counterfactual": 0.4604017658322358,
|
| 426 |
+
"eval_groundtruth_f1_factual": 0.6775906371456756,
|
| 427 |
+
"eval_icace_cosine": 0.4579065144062042,
|
| 428 |
+
"eval_icace_l2": 0.5784377455711365,
|
| 429 |
+
"eval_icace_normdiff": 0.3504405617713928,
|
| 430 |
+
"eval_loss": 5.936356544494629,
|
| 431 |
+
"eval_runtime": 22.1397,
|
| 432 |
+
"eval_samples_per_second": 176.064,
|
| 433 |
+
"eval_steps_per_second": 2.755,
|
| 434 |
+
"step": 650
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 1.13,
|
| 438 |
+
"learning_rate": 4.285095320623917e-05,
|
| 439 |
+
"loss": 6.2876,
|
| 440 |
+
"step": 660
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 1.17,
|
| 444 |
+
"learning_rate": 4.2634315424610055e-05,
|
| 445 |
+
"loss": 6.3896,
|
| 446 |
+
"step": 680
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 1.2,
|
| 450 |
+
"learning_rate": 4.241767764298094e-05,
|
| 451 |
+
"loss": 6.1169,
|
| 452 |
+
"step": 700
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 1.2,
|
| 456 |
+
"eval_distillation_accuracy_counterfactual": 0.5479733196511031,
|
| 457 |
+
"eval_distillation_accuracy_factual": 0.8129810159055926,
|
| 458 |
+
"eval_distillation_f1_counterfactual": 0.48600933269771956,
|
| 459 |
+
"eval_distillation_f1_factual": 0.7860512885448971,
|
| 460 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5336069779374037,
|
| 461 |
+
"eval_groundtruth_f1_counterfactual": 0.47412523543027546,
|
| 462 |
+
"eval_groundtruth_f1_factual": 0.6485794965038972,
|
| 463 |
+
"eval_icace_cosine": 0.4659746587276459,
|
| 464 |
+
"eval_icace_l2": 0.5945326089859009,
|
| 465 |
+
"eval_icace_normdiff": 0.36559173464775085,
|
| 466 |
+
"eval_loss": 6.143641948699951,
|
| 467 |
+
"eval_runtime": 22.3066,
|
| 468 |
+
"eval_samples_per_second": 174.747,
|
| 469 |
+
"eval_steps_per_second": 2.735,
|
| 470 |
+
"step": 700
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 1.23,
|
| 474 |
+
"learning_rate": 4.220103986135182e-05,
|
| 475 |
+
"loss": 6.1743,
|
| 476 |
+
"step": 720
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 1.27,
|
| 480 |
+
"learning_rate": 4.198440207972271e-05,
|
| 481 |
+
"loss": 6.4308,
|
| 482 |
+
"step": 740
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 1.29,
|
| 486 |
+
"eval_distillation_accuracy_counterfactual": 0.5569522832221652,
|
| 487 |
+
"eval_distillation_accuracy_factual": 0.8193945613134941,
|
| 488 |
+
"eval_distillation_f1_counterfactual": 0.4527655231915041,
|
| 489 |
+
"eval_distillation_f1_factual": 0.8096871483467666,
|
| 490 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5551564905079528,
|
| 491 |
+
"eval_groundtruth_f1_counterfactual": 0.45780758221444573,
|
| 492 |
+
"eval_groundtruth_f1_factual": 0.6628929940851196,
|
| 493 |
+
"eval_icace_cosine": 0.4578511118888855,
|
| 494 |
+
"eval_icace_l2": 0.5670422911643982,
|
| 495 |
+
"eval_icace_normdiff": 0.35481762886047363,
|
| 496 |
+
"eval_loss": 5.8051252365112305,
|
| 497 |
+
"eval_runtime": 22.076,
|
| 498 |
+
"eval_samples_per_second": 176.572,
|
| 499 |
+
"eval_steps_per_second": 2.763,
|
| 500 |
+
"step": 750
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 1.3,
|
| 504 |
+
"learning_rate": 4.1767764298093586e-05,
|
| 505 |
+
"loss": 6.1878,
|
| 506 |
+
"step": 760
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.34,
|
| 510 |
+
"learning_rate": 4.1551126516464476e-05,
|
| 511 |
+
"loss": 6.2818,
|
| 512 |
+
"step": 780
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 1.37,
|
| 516 |
+
"learning_rate": 4.133448873483536e-05,
|
| 517 |
+
"loss": 6.1387,
|
| 518 |
+
"step": 800
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 1.37,
|
| 522 |
+
"eval_distillation_accuracy_counterfactual": 0.5515649050795279,
|
| 523 |
+
"eval_distillation_accuracy_factual": 0.853771164699846,
|
| 524 |
+
"eval_distillation_f1_counterfactual": 0.46108787809547397,
|
| 525 |
+
"eval_distillation_f1_factual": 0.8408838366147668,
|
| 526 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5328373524884557,
|
| 527 |
+
"eval_groundtruth_f1_counterfactual": 0.4439112446666075,
|
| 528 |
+
"eval_groundtruth_f1_factual": 0.6747449056620686,
|
| 529 |
+
"eval_icace_cosine": 0.45763716101646423,
|
| 530 |
+
"eval_icace_l2": 0.5753128528594971,
|
| 531 |
+
"eval_icace_normdiff": 0.3466692864894867,
|
| 532 |
+
"eval_loss": 5.853026866912842,
|
| 533 |
+
"eval_runtime": 22.2637,
|
| 534 |
+
"eval_samples_per_second": 175.083,
|
| 535 |
+
"eval_steps_per_second": 2.74,
|
| 536 |
+
"step": 800
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 1.41,
|
| 540 |
+
"learning_rate": 4.111785095320624e-05,
|
| 541 |
+
"loss": 6.3818,
|
| 542 |
+
"step": 820
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.44,
|
| 546 |
+
"learning_rate": 4.0901213171577124e-05,
|
| 547 |
+
"loss": 6.3433,
|
| 548 |
+
"step": 840
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.46,
|
| 552 |
+
"eval_distillation_accuracy_counterfactual": 0.5448948178553105,
|
| 553 |
+
"eval_distillation_accuracy_factual": 0.8476141611082606,
|
| 554 |
+
"eval_distillation_f1_counterfactual": 0.4361535399476864,
|
| 555 |
+
"eval_distillation_f1_factual": 0.8329766547479427,
|
| 556 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5351462288353002,
|
| 557 |
+
"eval_groundtruth_f1_counterfactual": 0.42831401054324375,
|
| 558 |
+
"eval_groundtruth_f1_factual": 0.6756139171118364,
|
| 559 |
+
"eval_icace_cosine": 0.4597473442554474,
|
| 560 |
+
"eval_icace_l2": 0.583962619304657,
|
| 561 |
+
"eval_icace_normdiff": 0.3553682863712311,
|
| 562 |
+
"eval_loss": 6.0229010581970215,
|
| 563 |
+
"eval_runtime": 22.1925,
|
| 564 |
+
"eval_samples_per_second": 175.645,
|
| 565 |
+
"eval_steps_per_second": 2.749,
|
| 566 |
+
"step": 850
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 1.48,
|
| 570 |
+
"learning_rate": 4.068457538994801e-05,
|
| 571 |
+
"loss": 6.2674,
|
| 572 |
+
"step": 860
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 1.51,
|
| 576 |
+
"learning_rate": 4.04679376083189e-05,
|
| 577 |
+
"loss": 6.5452,
|
| 578 |
+
"step": 880
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 1.54,
|
| 582 |
+
"learning_rate": 4.025129982668977e-05,
|
| 583 |
+
"loss": 6.2361,
|
| 584 |
+
"step": 900
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 1.54,
|
| 588 |
+
"eval_distillation_accuracy_counterfactual": 0.5454079014879425,
|
| 589 |
+
"eval_distillation_accuracy_factual": 0.8186249358645459,
|
| 590 |
+
"eval_distillation_f1_counterfactual": 0.4448454069324789,
|
| 591 |
+
"eval_distillation_f1_factual": 0.8057071203974917,
|
| 592 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5287326834273987,
|
| 593 |
+
"eval_groundtruth_f1_counterfactual": 0.42960759215150884,
|
| 594 |
+
"eval_groundtruth_f1_factual": 0.6670500492904399,
|
| 595 |
+
"eval_icace_cosine": 0.46398869156837463,
|
| 596 |
+
"eval_icace_l2": 0.5982965230941772,
|
| 597 |
+
"eval_icace_normdiff": 0.3660006821155548,
|
| 598 |
+
"eval_loss": 6.118747234344482,
|
| 599 |
+
"eval_runtime": 22.1991,
|
| 600 |
+
"eval_samples_per_second": 175.593,
|
| 601 |
+
"eval_steps_per_second": 2.748,
|
| 602 |
+
"step": 900
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 1.58,
|
| 606 |
+
"learning_rate": 4.003466204506066e-05,
|
| 607 |
+
"loss": 6.3136,
|
| 608 |
+
"step": 920
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 1.61,
|
| 612 |
+
"learning_rate": 3.9818024263431546e-05,
|
| 613 |
+
"loss": 6.3279,
|
| 614 |
+
"step": 940
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 1.63,
|
| 618 |
+
"eval_distillation_accuracy_counterfactual": 0.5451513596716264,
|
| 619 |
+
"eval_distillation_accuracy_factual": 0.8355566957414059,
|
| 620 |
+
"eval_distillation_f1_counterfactual": 0.4168175870039869,
|
| 621 |
+
"eval_distillation_f1_factual": 0.8267012706170487,
|
| 622 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5274499743458184,
|
| 623 |
+
"eval_groundtruth_f1_counterfactual": 0.41075982637956343,
|
| 624 |
+
"eval_groundtruth_f1_factual": 0.6781751045050392,
|
| 625 |
+
"eval_icace_cosine": 0.45960184931755066,
|
| 626 |
+
"eval_icace_l2": 0.5789852142333984,
|
| 627 |
+
"eval_icace_normdiff": 0.35161492228507996,
|
| 628 |
+
"eval_loss": 5.898173809051514,
|
| 629 |
+
"eval_runtime": 22.1328,
|
| 630 |
+
"eval_samples_per_second": 176.119,
|
| 631 |
+
"eval_steps_per_second": 2.756,
|
| 632 |
+
"step": 950
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.65,
|
| 636 |
+
"learning_rate": 3.960138648180243e-05,
|
| 637 |
+
"loss": 6.349,
|
| 638 |
+
"step": 960
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 1.68,
|
| 642 |
+
"learning_rate": 3.938474870017331e-05,
|
| 643 |
+
"loss": 6.2648,
|
| 644 |
+
"step": 980
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 1.72,
|
| 648 |
+
"learning_rate": 3.91681109185442e-05,
|
| 649 |
+
"loss": 6.307,
|
| 650 |
+
"step": 1000
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 1.72,
|
| 654 |
+
"eval_distillation_accuracy_counterfactual": 0.547460236018471,
|
| 655 |
+
"eval_distillation_accuracy_factual": 0.8311954848640328,
|
| 656 |
+
"eval_distillation_f1_counterfactual": 0.4400284488718956,
|
| 657 |
+
"eval_distillation_f1_factual": 0.819341898820874,
|
| 658 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5328373524884557,
|
| 659 |
+
"eval_groundtruth_f1_counterfactual": 0.431004456933043,
|
| 660 |
+
"eval_groundtruth_f1_factual": 0.6724529205225047,
|
| 661 |
+
"eval_icace_cosine": 0.4630897343158722,
|
| 662 |
+
"eval_icace_l2": 0.5932038426399231,
|
| 663 |
+
"eval_icace_normdiff": 0.3649563491344452,
|
| 664 |
+
"eval_loss": 5.96807861328125,
|
| 665 |
+
"eval_runtime": 22.2743,
|
| 666 |
+
"eval_samples_per_second": 175.0,
|
| 667 |
+
"eval_steps_per_second": 2.739,
|
| 668 |
+
"step": 1000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.75,
|
| 672 |
+
"learning_rate": 3.895147313691508e-05,
|
| 673 |
+
"loss": 6.3078,
|
| 674 |
+
"step": 1020
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.78,
|
| 678 |
+
"learning_rate": 3.873483535528597e-05,
|
| 679 |
+
"loss": 6.299,
|
| 680 |
+
"step": 1040
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 1.8,
|
| 684 |
+
"eval_distillation_accuracy_counterfactual": 0.538481272447409,
|
| 685 |
+
"eval_distillation_accuracy_factual": 0.8358132375577219,
|
| 686 |
+
"eval_distillation_f1_counterfactual": 0.43944572973337176,
|
| 687 |
+
"eval_distillation_f1_factual": 0.8137606391644893,
|
| 688 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5210364289379169,
|
| 689 |
+
"eval_groundtruth_f1_counterfactual": 0.4266673104908225,
|
| 690 |
+
"eval_groundtruth_f1_factual": 0.6679874290517764,
|
| 691 |
+
"eval_icace_cosine": 0.4646293520927429,
|
| 692 |
+
"eval_icace_l2": 0.5929473042488098,
|
| 693 |
+
"eval_icace_normdiff": 0.35142672061920166,
|
| 694 |
+
"eval_loss": 6.2012529373168945,
|
| 695 |
+
"eval_runtime": 22.1799,
|
| 696 |
+
"eval_samples_per_second": 175.745,
|
| 697 |
+
"eval_steps_per_second": 2.75,
|
| 698 |
+
"step": 1050
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 1.82,
|
| 702 |
+
"learning_rate": 3.851819757365685e-05,
|
| 703 |
+
"loss": 6.1563,
|
| 704 |
+
"step": 1060
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 1.85,
|
| 708 |
+
"learning_rate": 3.830155979202773e-05,
|
| 709 |
+
"loss": 6.2483,
|
| 710 |
+
"step": 1080
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 1.89,
|
| 714 |
+
"learning_rate": 3.8084922010398616e-05,
|
| 715 |
+
"loss": 6.2315,
|
| 716 |
+
"step": 1100
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 1.89,
|
| 720 |
+
"eval_distillation_accuracy_counterfactual": 0.5502821959979477,
|
| 721 |
+
"eval_distillation_accuracy_factual": 0.8283735248845562,
|
| 722 |
+
"eval_distillation_f1_counterfactual": 0.4492053235764312,
|
| 723 |
+
"eval_distillation_f1_factual": 0.8130071646506035,
|
| 724 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5438686505900462,
|
| 725 |
+
"eval_groundtruth_f1_counterfactual": 0.4459935654992118,
|
| 726 |
+
"eval_groundtruth_f1_factual": 0.6759199159298241,
|
| 727 |
+
"eval_icace_cosine": 0.4584900140762329,
|
| 728 |
+
"eval_icace_l2": 0.5902418494224548,
|
| 729 |
+
"eval_icace_normdiff": 0.3675350546836853,
|
| 730 |
+
"eval_loss": 6.030585765838623,
|
| 731 |
+
"eval_runtime": 22.2302,
|
| 732 |
+
"eval_samples_per_second": 175.347,
|
| 733 |
+
"eval_steps_per_second": 2.744,
|
| 734 |
+
"step": 1100
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 1.92,
|
| 738 |
+
"learning_rate": 3.78682842287695e-05,
|
| 739 |
+
"loss": 6.0839,
|
| 740 |
+
"step": 1120
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 1.96,
|
| 744 |
+
"learning_rate": 3.765164644714038e-05,
|
| 745 |
+
"loss": 6.4395,
|
| 746 |
+
"step": 1140
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 1.97,
|
| 750 |
+
"eval_distillation_accuracy_counterfactual": 0.5420728578758338,
|
| 751 |
+
"eval_distillation_accuracy_factual": 0.8350436121087738,
|
| 752 |
+
"eval_distillation_f1_counterfactual": 0.44687319238313783,
|
| 753 |
+
"eval_distillation_f1_factual": 0.8227278561647481,
|
| 754 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5279630579784504,
|
| 755 |
+
"eval_groundtruth_f1_counterfactual": 0.4362421678695392,
|
| 756 |
+
"eval_groundtruth_f1_factual": 0.6689379744755434,
|
| 757 |
+
"eval_icace_cosine": 0.4657202959060669,
|
| 758 |
+
"eval_icace_l2": 0.5986185073852539,
|
| 759 |
+
"eval_icace_normdiff": 0.37084171175956726,
|
| 760 |
+
"eval_loss": 6.077243328094482,
|
| 761 |
+
"eval_runtime": 22.1,
|
| 762 |
+
"eval_samples_per_second": 176.38,
|
| 763 |
+
"eval_steps_per_second": 2.76,
|
| 764 |
+
"step": 1150
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 1.99,
|
| 768 |
+
"learning_rate": 3.7435008665511264e-05,
|
| 769 |
+
"loss": 6.4188,
|
| 770 |
+
"step": 1160
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 2.02,
|
| 774 |
+
"learning_rate": 3.7218370883882154e-05,
|
| 775 |
+
"loss": 6.2773,
|
| 776 |
+
"step": 1180
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 2.06,
|
| 780 |
+
"learning_rate": 3.700173310225303e-05,
|
| 781 |
+
"loss": 6.0979,
|
| 782 |
+
"step": 1200
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 2.06,
|
| 786 |
+
"eval_distillation_accuracy_counterfactual": 0.5554130323242689,
|
| 787 |
+
"eval_distillation_accuracy_factual": 0.8304258594150846,
|
| 788 |
+
"eval_distillation_f1_counterfactual": 0.4772175429149318,
|
| 789 |
+
"eval_distillation_f1_factual": 0.808134058844989,
|
| 790 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5389943560800411,
|
| 791 |
+
"eval_groundtruth_f1_counterfactual": 0.4650970297326843,
|
| 792 |
+
"eval_groundtruth_f1_factual": 0.6534003932896711,
|
| 793 |
+
"eval_icace_cosine": 0.4595852196216583,
|
| 794 |
+
"eval_icace_l2": 0.5800026059150696,
|
| 795 |
+
"eval_icace_normdiff": 0.3539329171180725,
|
| 796 |
+
"eval_loss": 5.965435981750488,
|
| 797 |
+
"eval_runtime": 22.2847,
|
| 798 |
+
"eval_samples_per_second": 174.919,
|
| 799 |
+
"eval_steps_per_second": 2.737,
|
| 800 |
+
"step": 1200
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 2.09,
|
| 804 |
+
"learning_rate": 3.678509532062392e-05,
|
| 805 |
+
"loss": 6.067,
|
| 806 |
+
"step": 1220
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 2.13,
|
| 810 |
+
"learning_rate": 3.65684575389948e-05,
|
| 811 |
+
"loss": 6.2014,
|
| 812 |
+
"step": 1240
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 2.14,
|
| 816 |
+
"eval_distillation_accuracy_counterfactual": 0.5477167778347871,
|
| 817 |
+
"eval_distillation_accuracy_factual": 0.827860441251924,
|
| 818 |
+
"eval_distillation_f1_counterfactual": 0.45873512192151045,
|
| 819 |
+
"eval_distillation_f1_factual": 0.8182707512505438,
|
| 820 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5371985633658286,
|
| 821 |
+
"eval_groundtruth_f1_counterfactual": 0.45174422882383675,
|
| 822 |
+
"eval_groundtruth_f1_factual": 0.6686107185288309,
|
| 823 |
+
"eval_icace_cosine": 0.4629949629306793,
|
| 824 |
+
"eval_icace_l2": 0.5902252197265625,
|
| 825 |
+
"eval_icace_normdiff": 0.36104583740234375,
|
| 826 |
+
"eval_loss": 6.057925224304199,
|
| 827 |
+
"eval_runtime": 22.207,
|
| 828 |
+
"eval_samples_per_second": 175.53,
|
| 829 |
+
"eval_steps_per_second": 2.747,
|
| 830 |
+
"step": 1250
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 2.16,
|
| 834 |
+
"learning_rate": 3.6351819757365686e-05,
|
| 835 |
+
"loss": 6.0228,
|
| 836 |
+
"step": 1260
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 2.2,
|
| 840 |
+
"learning_rate": 3.613518197573657e-05,
|
| 841 |
+
"loss": 6.077,
|
| 842 |
+
"step": 1280
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 2.23,
|
| 846 |
+
"learning_rate": 3.591854419410746e-05,
|
| 847 |
+
"loss": 6.0282,
|
| 848 |
+
"step": 1300
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 2.23,
|
| 852 |
+
"eval_distillation_accuracy_counterfactual": 0.5507952796305798,
|
| 853 |
+
"eval_distillation_accuracy_factual": 0.8165726013340174,
|
| 854 |
+
"eval_distillation_f1_counterfactual": 0.4631377949704046,
|
| 855 |
+
"eval_distillation_f1_factual": 0.8093111197445927,
|
| 856 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5284761416110826,
|
| 857 |
+
"eval_groundtruth_f1_counterfactual": 0.4474398103025621,
|
| 858 |
+
"eval_groundtruth_f1_factual": 0.669141158800666,
|
| 859 |
+
"eval_icace_cosine": 0.4620840549468994,
|
| 860 |
+
"eval_icace_l2": 0.5971253514289856,
|
| 861 |
+
"eval_icace_normdiff": 0.3704546391963959,
|
| 862 |
+
"eval_loss": 6.050764083862305,
|
| 863 |
+
"eval_runtime": 22.2574,
|
| 864 |
+
"eval_samples_per_second": 175.133,
|
| 865 |
+
"eval_steps_per_second": 2.741,
|
| 866 |
+
"step": 1300
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 2.26,
|
| 870 |
+
"learning_rate": 3.5701906412478334e-05,
|
| 871 |
+
"loss": 6.1973,
|
| 872 |
+
"step": 1320
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 2.3,
|
| 876 |
+
"learning_rate": 3.5485268630849224e-05,
|
| 877 |
+
"loss": 6.0877,
|
| 878 |
+
"step": 1340
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 2.32,
|
| 882 |
+
"eval_distillation_accuracy_counterfactual": 0.5064135454079015,
|
| 883 |
+
"eval_distillation_accuracy_factual": 0.8309389430477168,
|
| 884 |
+
"eval_distillation_f1_counterfactual": 0.4306802598777498,
|
| 885 |
+
"eval_distillation_f1_factual": 0.8171042534137397,
|
| 886 |
+
"eval_groundtruth_accuracy_counterfactual": 0.48460749102103645,
|
| 887 |
+
"eval_groundtruth_f1_counterfactual": 0.4162277530517664,
|
| 888 |
+
"eval_groundtruth_f1_factual": 0.6732278788009838,
|
| 889 |
+
"eval_icace_cosine": 0.47389736771583557,
|
| 890 |
+
"eval_icace_l2": 0.6207495331764221,
|
| 891 |
+
"eval_icace_normdiff": 0.3774127960205078,
|
| 892 |
+
"eval_loss": 6.461216926574707,
|
| 893 |
+
"eval_runtime": 22.063,
|
| 894 |
+
"eval_samples_per_second": 176.676,
|
| 895 |
+
"eval_steps_per_second": 2.765,
|
| 896 |
+
"step": 1350
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 2.33,
|
| 900 |
+
"learning_rate": 3.526863084922011e-05,
|
| 901 |
+
"loss": 6.0128,
|
| 902 |
+
"step": 1360
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 2.37,
|
| 906 |
+
"learning_rate": 3.505199306759099e-05,
|
| 907 |
+
"loss": 6.0307,
|
| 908 |
+
"step": 1380
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 2.4,
|
| 912 |
+
"learning_rate": 3.483535528596187e-05,
|
| 913 |
+
"loss": 6.1284,
|
| 914 |
+
"step": 1400
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 2.4,
|
| 918 |
+
"eval_distillation_accuracy_counterfactual": 0.5510518214468958,
|
| 919 |
+
"eval_distillation_accuracy_factual": 0.8055413032324269,
|
| 920 |
+
"eval_distillation_f1_counterfactual": 0.4448362647873446,
|
| 921 |
+
"eval_distillation_f1_factual": 0.7931010998998913,
|
| 922 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5266803488968702,
|
| 923 |
+
"eval_groundtruth_f1_counterfactual": 0.4296197215462649,
|
| 924 |
+
"eval_groundtruth_f1_factual": 0.6675776163244647,
|
| 925 |
+
"eval_icace_cosine": 0.4564709961414337,
|
| 926 |
+
"eval_icace_l2": 0.5935506820678711,
|
| 927 |
+
"eval_icace_normdiff": 0.36554762721061707,
|
| 928 |
+
"eval_loss": 6.071511268615723,
|
| 929 |
+
"eval_runtime": 22.0977,
|
| 930 |
+
"eval_samples_per_second": 176.398,
|
| 931 |
+
"eval_steps_per_second": 2.76,
|
| 932 |
+
"step": 1400
|
| 933 |
+
}
|
| 934 |
+
],
|
| 935 |
+
"max_steps": 4616,
|
| 936 |
+
"num_train_epochs": 8,
|
| 937 |
+
"total_flos": 1.18623500497152e+16,
|
| 938 |
+
"trial_name": null,
|
| 939 |
+
"trial_params": null
|
| 940 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d4b389da751526709c1a17f652a207554bb46c2a02d4ab278a2f3a47dc5280bc
|
| 3 |
+
size 3183
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|