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 +676 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
added_tokens.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"[service-unknown]": 30536, "[noise-Positive]": 30531, "[service-Negative]": 30534, "[noise-no majority]": 30533, "[ambiance-no majority]": 30529, "[food-unknown]": 30524, "[food-Positive]": 30523, "[noise-Negative]": 30530, "[service-Positive]": 30535, "[ambiance-Negative]": 30526, "[ambiance-Positive]": 30527, "[noise-unknown]": 30532, "[food-no majority]": 30525, "[ambiance-unknown]": 30528, "[service-no majority]": 30537, "[food-Negative]": 30522}
|
config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "CEBaB/bert-base-uncased.CEBaB.sa.5-class.exclusive.seed_77",
|
| 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 sha256:88285272ed278ace9aa481d00143ceaa2cade6ebc1dc64bdaa7777902edb7106
|
| 3 |
+
size 880829397
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:86ef92e1d6ff6f4ef26768b88c32fcb0a9ab4cae68b349e1259f6418ca67d010
|
| 3 |
+
size 440427497
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:382769948bb4f5e38fda7a3b6c56ac4788045f1ef239850a126ef46ddaced519
|
| 3 |
+
size 15523
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f678f0f7c14334d1dd104139028feff1f69bed7de6d9fa457d52ab6fbd9c49c0
|
| 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 +1,676 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.45797210931777954,
|
| 3 |
+
"best_model_checkpoint": "model_output/e2e_opentable_5_way__approximate__0-shot__seed-77__bert-base-uncased/checkpoint-1000",
|
| 4 |
+
"epoch": 1.7152658662092626,
|
| 5 |
+
"global_step": 1000,
|
| 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": 15.0971,
|
| 14 |
+
"step": 20
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.07,
|
| 18 |
+
"learning_rate": 4.956672443674177e-05,
|
| 19 |
+
"loss": 10.8654,
|
| 20 |
+
"step": 40
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.09,
|
| 24 |
+
"eval_distillation_accuracy_counterfactual": 0.512827090815803,
|
| 25 |
+
"eval_distillation_accuracy_factual": 0.8973832734735762,
|
| 26 |
+
"eval_distillation_f1_counterfactual": 0.38283775595303926,
|
| 27 |
+
"eval_distillation_f1_factual": 0.8849844553565298,
|
| 28 |
+
"eval_groundtruth_accuracy_counterfactual": 0.47998973832734737,
|
| 29 |
+
"eval_groundtruth_f1_counterfactual": 0.35799290350733415,
|
| 30 |
+
"eval_groundtruth_f1_factual": 0.6974663941237289,
|
| 31 |
+
"eval_icace_cosine": 0.4845265746116638,
|
| 32 |
+
"eval_icace_l2": 0.6881101131439209,
|
| 33 |
+
"eval_icace_normdiff": 0.4512614905834198,
|
| 34 |
+
"eval_loss": 7.50442361831665,
|
| 35 |
+
"eval_runtime": 22.2098,
|
| 36 |
+
"eval_samples_per_second": 175.508,
|
| 37 |
+
"eval_steps_per_second": 2.747,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.1,
|
| 42 |
+
"learning_rate": 4.935008665511265e-05,
|
| 43 |
+
"loss": 10.0023,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.14,
|
| 48 |
+
"learning_rate": 4.913344887348354e-05,
|
| 49 |
+
"loss": 9.8393,
|
| 50 |
+
"step": 80
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.17,
|
| 54 |
+
"learning_rate": 4.891681109185442e-05,
|
| 55 |
+
"loss": 9.2084,
|
| 56 |
+
"step": 100
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.17,
|
| 60 |
+
"eval_distillation_accuracy_counterfactual": 0.534376603386352,
|
| 61 |
+
"eval_distillation_accuracy_factual": 0.879168804515136,
|
| 62 |
+
"eval_distillation_f1_counterfactual": 0.47020198824723397,
|
| 63 |
+
"eval_distillation_f1_factual": 0.8636644497351712,
|
| 64 |
+
"eval_groundtruth_accuracy_counterfactual": 0.521806054386865,
|
| 65 |
+
"eval_groundtruth_f1_counterfactual": 0.46415698862260973,
|
| 66 |
+
"eval_groundtruth_f1_factual": 0.6787169266269484,
|
| 67 |
+
"eval_icace_cosine": 0.474692165851593,
|
| 68 |
+
"eval_icace_l2": 0.6402959823608398,
|
| 69 |
+
"eval_icace_normdiff": 0.41231343150138855,
|
| 70 |
+
"eval_loss": 7.25449275970459,
|
| 71 |
+
"eval_runtime": 22.2189,
|
| 72 |
+
"eval_samples_per_second": 175.436,
|
| 73 |
+
"eval_steps_per_second": 2.745,
|
| 74 |
+
"step": 100
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.21,
|
| 78 |
+
"learning_rate": 4.8700173310225307e-05,
|
| 79 |
+
"loss": 9.2547,
|
| 80 |
+
"step": 120
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.24,
|
| 84 |
+
"learning_rate": 4.848353552859619e-05,
|
| 85 |
+
"loss": 9.2275,
|
| 86 |
+
"step": 140
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.26,
|
| 90 |
+
"eval_distillation_accuracy_counterfactual": 0.5028219599794767,
|
| 91 |
+
"eval_distillation_accuracy_factual": 0.8245253976398152,
|
| 92 |
+
"eval_distillation_f1_counterfactual": 0.41359201224181497,
|
| 93 |
+
"eval_distillation_f1_factual": 0.8013559665708911,
|
| 94 |
+
"eval_groundtruth_accuracy_counterfactual": 0.49204720369420213,
|
| 95 |
+
"eval_groundtruth_f1_counterfactual": 0.4109951641517333,
|
| 96 |
+
"eval_groundtruth_f1_factual": 0.6495764505483296,
|
| 97 |
+
"eval_icace_cosine": 0.46765828132629395,
|
| 98 |
+
"eval_icace_l2": 0.6432615518569946,
|
| 99 |
+
"eval_icace_normdiff": 0.38947969675064087,
|
| 100 |
+
"eval_loss": 7.776533126831055,
|
| 101 |
+
"eval_runtime": 22.1488,
|
| 102 |
+
"eval_samples_per_second": 175.991,
|
| 103 |
+
"eval_steps_per_second": 2.754,
|
| 104 |
+
"step": 150
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.27,
|
| 108 |
+
"learning_rate": 4.826689774696707e-05,
|
| 109 |
+
"loss": 8.9152,
|
| 110 |
+
"step": 160
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.31,
|
| 114 |
+
"learning_rate": 4.8050259965337955e-05,
|
| 115 |
+
"loss": 8.7736,
|
| 116 |
+
"step": 180
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.34,
|
| 120 |
+
"learning_rate": 4.7833622183708845e-05,
|
| 121 |
+
"loss": 8.8387,
|
| 122 |
+
"step": 200
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.34,
|
| 126 |
+
"eval_distillation_accuracy_counterfactual": 0.5177013853258081,
|
| 127 |
+
"eval_distillation_accuracy_factual": 0.8614674191893279,
|
| 128 |
+
"eval_distillation_f1_counterfactual": 0.41268689402882053,
|
| 129 |
+
"eval_distillation_f1_factual": 0.8521590725130646,
|
| 130 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5030785017957927,
|
| 131 |
+
"eval_groundtruth_f1_counterfactual": 0.40850979925255315,
|
| 132 |
+
"eval_groundtruth_f1_factual": 0.6669518573115495,
|
| 133 |
+
"eval_icace_cosine": 0.4691610634326935,
|
| 134 |
+
"eval_icace_l2": 0.6340324282646179,
|
| 135 |
+
"eval_icace_normdiff": 0.39094486832618713,
|
| 136 |
+
"eval_loss": 7.458261013031006,
|
| 137 |
+
"eval_runtime": 22.3381,
|
| 138 |
+
"eval_samples_per_second": 174.5,
|
| 139 |
+
"eval_steps_per_second": 2.731,
|
| 140 |
+
"step": 200
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.38,
|
| 144 |
+
"learning_rate": 4.761698440207972e-05,
|
| 145 |
+
"loss": 8.772,
|
| 146 |
+
"step": 220
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.41,
|
| 150 |
+
"learning_rate": 4.740034662045061e-05,
|
| 151 |
+
"loss": 8.7734,
|
| 152 |
+
"step": 240
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.43,
|
| 156 |
+
"eval_distillation_accuracy_counterfactual": 0.4671626475115444,
|
| 157 |
+
"eval_distillation_accuracy_factual": 0.8671113391482812,
|
| 158 |
+
"eval_distillation_f1_counterfactual": 0.38582609157213904,
|
| 159 |
+
"eval_distillation_f1_factual": 0.8563293543368673,
|
| 160 |
+
"eval_groundtruth_accuracy_counterfactual": 0.46151872755259105,
|
| 161 |
+
"eval_groundtruth_f1_counterfactual": 0.39182214143822086,
|
| 162 |
+
"eval_groundtruth_f1_factual": 0.6883791067687154,
|
| 163 |
+
"eval_icace_cosine": 0.4946799874305725,
|
| 164 |
+
"eval_icace_l2": 0.6939188241958618,
|
| 165 |
+
"eval_icace_normdiff": 0.43225187063217163,
|
| 166 |
+
"eval_loss": 8.251187324523926,
|
| 167 |
+
"eval_runtime": 22.1682,
|
| 168 |
+
"eval_samples_per_second": 175.838,
|
| 169 |
+
"eval_steps_per_second": 2.752,
|
| 170 |
+
"step": 250
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.45,
|
| 174 |
+
"learning_rate": 4.7183708838821494e-05,
|
| 175 |
+
"loss": 8.5326,
|
| 176 |
+
"step": 260
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.48,
|
| 180 |
+
"learning_rate": 4.6967071057192376e-05,
|
| 181 |
+
"loss": 8.593,
|
| 182 |
+
"step": 280
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.51,
|
| 186 |
+
"learning_rate": 4.675043327556326e-05,
|
| 187 |
+
"loss": 8.5207,
|
| 188 |
+
"step": 300
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.51,
|
| 192 |
+
"eval_distillation_accuracy_counterfactual": 0.5187275525910724,
|
| 193 |
+
"eval_distillation_accuracy_factual": 0.8471010774756286,
|
| 194 |
+
"eval_distillation_f1_counterfactual": 0.4809461537601309,
|
| 195 |
+
"eval_distillation_f1_factual": 0.8415144400606737,
|
| 196 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5023088763468445,
|
| 197 |
+
"eval_groundtruth_f1_counterfactual": 0.46717757401427457,
|
| 198 |
+
"eval_groundtruth_f1_factual": 0.6817982025512722,
|
| 199 |
+
"eval_icace_cosine": 0.48118895292282104,
|
| 200 |
+
"eval_icace_l2": 0.6521334052085876,
|
| 201 |
+
"eval_icace_normdiff": 0.4162002503871918,
|
| 202 |
+
"eval_loss": 7.6342902183532715,
|
| 203 |
+
"eval_runtime": 22.1084,
|
| 204 |
+
"eval_samples_per_second": 176.313,
|
| 205 |
+
"eval_steps_per_second": 2.759,
|
| 206 |
+
"step": 300
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.55,
|
| 210 |
+
"learning_rate": 4.653379549393415e-05,
|
| 211 |
+
"loss": 8.3762,
|
| 212 |
+
"step": 320
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.58,
|
| 216 |
+
"learning_rate": 4.6317157712305025e-05,
|
| 217 |
+
"loss": 8.3251,
|
| 218 |
+
"step": 340
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.6,
|
| 222 |
+
"eval_distillation_accuracy_counterfactual": 0.5295023088763469,
|
| 223 |
+
"eval_distillation_accuracy_factual": 0.8568496664956388,
|
| 224 |
+
"eval_distillation_f1_counterfactual": 0.4641469168961435,
|
| 225 |
+
"eval_distillation_f1_factual": 0.8494977579970042,
|
| 226 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5166752180605438,
|
| 227 |
+
"eval_groundtruth_f1_counterfactual": 0.4567649352125045,
|
| 228 |
+
"eval_groundtruth_f1_factual": 0.6694324318001126,
|
| 229 |
+
"eval_icace_cosine": 0.46943578124046326,
|
| 230 |
+
"eval_icace_l2": 0.6349437236785889,
|
| 231 |
+
"eval_icace_normdiff": 0.39818716049194336,
|
| 232 |
+
"eval_loss": 7.341701507568359,
|
| 233 |
+
"eval_runtime": 22.1905,
|
| 234 |
+
"eval_samples_per_second": 175.66,
|
| 235 |
+
"eval_steps_per_second": 2.749,
|
| 236 |
+
"step": 350
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.62,
|
| 240 |
+
"learning_rate": 4.6100519930675915e-05,
|
| 241 |
+
"loss": 8.2535,
|
| 242 |
+
"step": 360
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.65,
|
| 246 |
+
"learning_rate": 4.58838821490468e-05,
|
| 247 |
+
"loss": 8.3502,
|
| 248 |
+
"step": 380
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.69,
|
| 252 |
+
"learning_rate": 4.566724436741768e-05,
|
| 253 |
+
"loss": 8.1628,
|
| 254 |
+
"step": 400
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.69,
|
| 258 |
+
"eval_distillation_accuracy_counterfactual": 0.5397639815289892,
|
| 259 |
+
"eval_distillation_accuracy_factual": 0.8355566957414059,
|
| 260 |
+
"eval_distillation_f1_counterfactual": 0.46033001340715457,
|
| 261 |
+
"eval_distillation_f1_factual": 0.8292453735319636,
|
| 262 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5223191380194971,
|
| 263 |
+
"eval_groundtruth_f1_counterfactual": 0.44829969536055286,
|
| 264 |
+
"eval_groundtruth_f1_factual": 0.6718756022768833,
|
| 265 |
+
"eval_icace_cosine": 0.460631787776947,
|
| 266 |
+
"eval_icace_l2": 0.6126735210418701,
|
| 267 |
+
"eval_icace_normdiff": 0.3810984492301941,
|
| 268 |
+
"eval_loss": 7.215612411499023,
|
| 269 |
+
"eval_runtime": 22.2088,
|
| 270 |
+
"eval_samples_per_second": 175.516,
|
| 271 |
+
"eval_steps_per_second": 2.747,
|
| 272 |
+
"step": 400
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.72,
|
| 276 |
+
"learning_rate": 4.5450606585788563e-05,
|
| 277 |
+
"loss": 8.1842,
|
| 278 |
+
"step": 420
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.75,
|
| 282 |
+
"learning_rate": 4.5233968804159446e-05,
|
| 283 |
+
"loss": 8.2209,
|
| 284 |
+
"step": 440
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.77,
|
| 288 |
+
"eval_distillation_accuracy_counterfactual": 0.5256541816316059,
|
| 289 |
+
"eval_distillation_accuracy_factual": 0.8045151359671626,
|
| 290 |
+
"eval_distillation_f1_counterfactual": 0.41138103040788077,
|
| 291 |
+
"eval_distillation_f1_factual": 0.798165576043234,
|
| 292 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5248845561826577,
|
| 293 |
+
"eval_groundtruth_f1_counterfactual": 0.4184296633356827,
|
| 294 |
+
"eval_groundtruth_f1_factual": 0.66434863984353,
|
| 295 |
+
"eval_icace_cosine": 0.47930216789245605,
|
| 296 |
+
"eval_icace_l2": 0.6362362504005432,
|
| 297 |
+
"eval_icace_normdiff": 0.4014618396759033,
|
| 298 |
+
"eval_loss": 7.654078483581543,
|
| 299 |
+
"eval_runtime": 22.2042,
|
| 300 |
+
"eval_samples_per_second": 175.552,
|
| 301 |
+
"eval_steps_per_second": 2.747,
|
| 302 |
+
"step": 450
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.79,
|
| 306 |
+
"learning_rate": 4.501733102253033e-05,
|
| 307 |
+
"loss": 8.2117,
|
| 308 |
+
"step": 460
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.82,
|
| 312 |
+
"learning_rate": 4.480069324090121e-05,
|
| 313 |
+
"loss": 8.246,
|
| 314 |
+
"step": 480
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.86,
|
| 318 |
+
"learning_rate": 4.45840554592721e-05,
|
| 319 |
+
"loss": 8.1357,
|
| 320 |
+
"step": 500
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.86,
|
| 324 |
+
"eval_distillation_accuracy_counterfactual": 0.5497691123653156,
|
| 325 |
+
"eval_distillation_accuracy_factual": 0.8237557721908672,
|
| 326 |
+
"eval_distillation_f1_counterfactual": 0.45037064661810805,
|
| 327 |
+
"eval_distillation_f1_factual": 0.8056668211328342,
|
| 328 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5402770651616213,
|
| 329 |
+
"eval_groundtruth_f1_counterfactual": 0.44977490137963505,
|
| 330 |
+
"eval_groundtruth_f1_factual": 0.6736698728564816,
|
| 331 |
+
"eval_icace_cosine": 0.4668065309524536,
|
| 332 |
+
"eval_icace_l2": 0.6126391887664795,
|
| 333 |
+
"eval_icace_normdiff": 0.39072343707084656,
|
| 334 |
+
"eval_loss": 7.208556175231934,
|
| 335 |
+
"eval_runtime": 22.0856,
|
| 336 |
+
"eval_samples_per_second": 176.495,
|
| 337 |
+
"eval_steps_per_second": 2.762,
|
| 338 |
+
"step": 500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.89,
|
| 342 |
+
"learning_rate": 4.436741767764298e-05,
|
| 343 |
+
"loss": 8.0972,
|
| 344 |
+
"step": 520
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.93,
|
| 348 |
+
"learning_rate": 4.415077989601387e-05,
|
| 349 |
+
"loss": 7.9764,
|
| 350 |
+
"step": 540
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.94,
|
| 354 |
+
"eval_distillation_accuracy_counterfactual": 0.5413032324268856,
|
| 355 |
+
"eval_distillation_accuracy_factual": 0.8263211903540277,
|
| 356 |
+
"eval_distillation_f1_counterfactual": 0.4579261774469936,
|
| 357 |
+
"eval_distillation_f1_factual": 0.8169241589118235,
|
| 358 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5338635197537198,
|
| 359 |
+
"eval_groundtruth_f1_counterfactual": 0.45648415392539654,
|
| 360 |
+
"eval_groundtruth_f1_factual": 0.6666874640710526,
|
| 361 |
+
"eval_icace_cosine": 0.4673744738101959,
|
| 362 |
+
"eval_icace_l2": 0.6225017309188843,
|
| 363 |
+
"eval_icace_normdiff": 0.3859357535839081,
|
| 364 |
+
"eval_loss": 7.50295877456665,
|
| 365 |
+
"eval_runtime": 22.0967,
|
| 366 |
+
"eval_samples_per_second": 176.406,
|
| 367 |
+
"eval_steps_per_second": 2.761,
|
| 368 |
+
"step": 550
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.96,
|
| 372 |
+
"learning_rate": 4.393414211438475e-05,
|
| 373 |
+
"loss": 8.0277,
|
| 374 |
+
"step": 560
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.99,
|
| 378 |
+
"learning_rate": 4.371750433275563e-05,
|
| 379 |
+
"loss": 7.9231,
|
| 380 |
+
"step": 580
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 1.03,
|
| 384 |
+
"learning_rate": 4.3500866551126516e-05,
|
| 385 |
+
"loss": 8.1605,
|
| 386 |
+
"step": 600
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 1.03,
|
| 390 |
+
"eval_distillation_accuracy_counterfactual": 0.538737814263725,
|
| 391 |
+
"eval_distillation_accuracy_factual": 0.8424833247819394,
|
| 392 |
+
"eval_distillation_f1_counterfactual": 0.4424763530876773,
|
| 393 |
+
"eval_distillation_f1_factual": 0.8305147218419746,
|
| 394 |
+
"eval_groundtruth_accuracy_counterfactual": 0.530528476141611,
|
| 395 |
+
"eval_groundtruth_f1_counterfactual": 0.44459191212630245,
|
| 396 |
+
"eval_groundtruth_f1_factual": 0.675893177241911,
|
| 397 |
+
"eval_icace_cosine": 0.47036415338516235,
|
| 398 |
+
"eval_icace_l2": 0.6303676962852478,
|
| 399 |
+
"eval_icace_normdiff": 0.40152838826179504,
|
| 400 |
+
"eval_loss": 7.446086406707764,
|
| 401 |
+
"eval_runtime": 22.193,
|
| 402 |
+
"eval_samples_per_second": 175.641,
|
| 403 |
+
"eval_steps_per_second": 2.749,
|
| 404 |
+
"step": 600
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 1.06,
|
| 408 |
+
"learning_rate": 4.3284228769497406e-05,
|
| 409 |
+
"loss": 8.0369,
|
| 410 |
+
"step": 620
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 1.1,
|
| 414 |
+
"learning_rate": 4.306759098786828e-05,
|
| 415 |
+
"loss": 7.8522,
|
| 416 |
+
"step": 640
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.11,
|
| 420 |
+
"eval_distillation_accuracy_counterfactual": 0.5325808106721396,
|
| 421 |
+
"eval_distillation_accuracy_factual": 0.7945100051308364,
|
| 422 |
+
"eval_distillation_f1_counterfactual": 0.45243712925320734,
|
| 423 |
+
"eval_distillation_f1_factual": 0.7760743993737376,
|
| 424 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5248845561826577,
|
| 425 |
+
"eval_groundtruth_f1_counterfactual": 0.4539872598778271,
|
| 426 |
+
"eval_groundtruth_f1_factual": 0.6485178712630223,
|
| 427 |
+
"eval_icace_cosine": 0.4656234085559845,
|
| 428 |
+
"eval_icace_l2": 0.6243467330932617,
|
| 429 |
+
"eval_icace_normdiff": 0.38494807481765747,
|
| 430 |
+
"eval_loss": 7.678111553192139,
|
| 431 |
+
"eval_runtime": 22.2461,
|
| 432 |
+
"eval_samples_per_second": 175.222,
|
| 433 |
+
"eval_steps_per_second": 2.742,
|
| 434 |
+
"step": 650
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 1.13,
|
| 438 |
+
"learning_rate": 4.285095320623917e-05,
|
| 439 |
+
"loss": 7.7876,
|
| 440 |
+
"step": 660
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 1.17,
|
| 444 |
+
"learning_rate": 4.2634315424610055e-05,
|
| 445 |
+
"loss": 7.8585,
|
| 446 |
+
"step": 680
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 1.2,
|
| 450 |
+
"learning_rate": 4.241767764298094e-05,
|
| 451 |
+
"loss": 7.9963,
|
| 452 |
+
"step": 700
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 1.2,
|
| 456 |
+
"eval_distillation_accuracy_counterfactual": 0.5220625962031811,
|
| 457 |
+
"eval_distillation_accuracy_factual": 0.8532580810672139,
|
| 458 |
+
"eval_distillation_f1_counterfactual": 0.421394437601727,
|
| 459 |
+
"eval_distillation_f1_factual": 0.843810107668179,
|
| 460 |
+
"eval_groundtruth_accuracy_counterfactual": 0.513083632632119,
|
| 461 |
+
"eval_groundtruth_f1_counterfactual": 0.41873912487772846,
|
| 462 |
+
"eval_groundtruth_f1_factual": 0.6890304825574103,
|
| 463 |
+
"eval_icace_cosine": 0.47455719113349915,
|
| 464 |
+
"eval_icace_l2": 0.6358718276023865,
|
| 465 |
+
"eval_icace_normdiff": 0.3861025869846344,
|
| 466 |
+
"eval_loss": 7.682065010070801,
|
| 467 |
+
"eval_runtime": 22.4604,
|
| 468 |
+
"eval_samples_per_second": 173.55,
|
| 469 |
+
"eval_steps_per_second": 2.716,
|
| 470 |
+
"step": 700
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 1.23,
|
| 474 |
+
"learning_rate": 4.220103986135182e-05,
|
| 475 |
+
"loss": 7.9938,
|
| 476 |
+
"step": 720
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 1.27,
|
| 480 |
+
"learning_rate": 4.198440207972271e-05,
|
| 481 |
+
"loss": 7.7139,
|
| 482 |
+
"step": 740
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 1.29,
|
| 486 |
+
"eval_distillation_accuracy_counterfactual": 0.5405336069779374,
|
| 487 |
+
"eval_distillation_accuracy_factual": 0.8509492047203694,
|
| 488 |
+
"eval_distillation_f1_counterfactual": 0.47182248378312613,
|
| 489 |
+
"eval_distillation_f1_factual": 0.8355752615379481,
|
| 490 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5269368907131863,
|
| 491 |
+
"eval_groundtruth_f1_counterfactual": 0.4620646268597837,
|
| 492 |
+
"eval_groundtruth_f1_factual": 0.6925217992694058,
|
| 493 |
+
"eval_icace_cosine": 0.45927247405052185,
|
| 494 |
+
"eval_icace_l2": 0.6215088367462158,
|
| 495 |
+
"eval_icace_normdiff": 0.39000242948532104,
|
| 496 |
+
"eval_loss": 7.237956523895264,
|
| 497 |
+
"eval_runtime": 22.1166,
|
| 498 |
+
"eval_samples_per_second": 176.248,
|
| 499 |
+
"eval_steps_per_second": 2.758,
|
| 500 |
+
"step": 750
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 1.3,
|
| 504 |
+
"learning_rate": 4.1767764298093586e-05,
|
| 505 |
+
"loss": 7.8577,
|
| 506 |
+
"step": 760
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.34,
|
| 510 |
+
"learning_rate": 4.1551126516464476e-05,
|
| 511 |
+
"loss": 7.8288,
|
| 512 |
+
"step": 780
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 1.37,
|
| 516 |
+
"learning_rate": 4.133448873483536e-05,
|
| 517 |
+
"loss": 7.6519,
|
| 518 |
+
"step": 800
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 1.37,
|
| 522 |
+
"eval_distillation_accuracy_counterfactual": 0.5279630579784504,
|
| 523 |
+
"eval_distillation_accuracy_factual": 0.805797845048743,
|
| 524 |
+
"eval_distillation_f1_counterfactual": 0.44398440406113127,
|
| 525 |
+
"eval_distillation_f1_factual": 0.7889017139693454,
|
| 526 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5207798871216008,
|
| 527 |
+
"eval_groundtruth_f1_counterfactual": 0.4413428310825916,
|
| 528 |
+
"eval_groundtruth_f1_factual": 0.6594914555824491,
|
| 529 |
+
"eval_icace_cosine": 0.47146424651145935,
|
| 530 |
+
"eval_icace_l2": 0.6373152732849121,
|
| 531 |
+
"eval_icace_normdiff": 0.40072518587112427,
|
| 532 |
+
"eval_loss": 7.812716960906982,
|
| 533 |
+
"eval_runtime": 22.1689,
|
| 534 |
+
"eval_samples_per_second": 175.832,
|
| 535 |
+
"eval_steps_per_second": 2.752,
|
| 536 |
+
"step": 800
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 1.41,
|
| 540 |
+
"learning_rate": 4.111785095320624e-05,
|
| 541 |
+
"loss": 7.8938,
|
| 542 |
+
"step": 820
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.44,
|
| 546 |
+
"learning_rate": 4.0901213171577124e-05,
|
| 547 |
+
"loss": 7.7746,
|
| 548 |
+
"step": 840
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.46,
|
| 552 |
+
"eval_distillation_accuracy_counterfactual": 0.5243714725500257,
|
| 553 |
+
"eval_distillation_accuracy_factual": 0.836326321190354,
|
| 554 |
+
"eval_distillation_f1_counterfactual": 0.4465427270486524,
|
| 555 |
+
"eval_distillation_f1_factual": 0.82730940599887,
|
| 556 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5179579271421242,
|
| 557 |
+
"eval_groundtruth_f1_counterfactual": 0.4470807155891924,
|
| 558 |
+
"eval_groundtruth_f1_factual": 0.6724916960045889,
|
| 559 |
+
"eval_icace_cosine": 0.4680478870868683,
|
| 560 |
+
"eval_icace_l2": 0.6270192861557007,
|
| 561 |
+
"eval_icace_normdiff": 0.3876318037509918,
|
| 562 |
+
"eval_loss": 7.512824535369873,
|
| 563 |
+
"eval_runtime": 22.2529,
|
| 564 |
+
"eval_samples_per_second": 175.168,
|
| 565 |
+
"eval_steps_per_second": 2.741,
|
| 566 |
+
"step": 850
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 1.48,
|
| 570 |
+
"learning_rate": 4.068457538994801e-05,
|
| 571 |
+
"loss": 7.6544,
|
| 572 |
+
"step": 860
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 1.51,
|
| 576 |
+
"learning_rate": 4.04679376083189e-05,
|
| 577 |
+
"loss": 7.7708,
|
| 578 |
+
"step": 880
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 1.54,
|
| 582 |
+
"learning_rate": 4.025129982668977e-05,
|
| 583 |
+
"loss": 7.6759,
|
| 584 |
+
"step": 900
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 1.54,
|
| 588 |
+
"eval_distillation_accuracy_counterfactual": 0.5194971780400205,
|
| 589 |
+
"eval_distillation_accuracy_factual": 0.814520266803489,
|
| 590 |
+
"eval_distillation_f1_counterfactual": 0.45570094642434356,
|
| 591 |
+
"eval_distillation_f1_factual": 0.805867060477024,
|
| 592 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5007696254489482,
|
| 593 |
+
"eval_groundtruth_f1_counterfactual": 0.43858879322504424,
|
| 594 |
+
"eval_groundtruth_f1_factual": 0.657928782779729,
|
| 595 |
+
"eval_icace_cosine": 0.4692392945289612,
|
| 596 |
+
"eval_icace_l2": 0.6410536766052246,
|
| 597 |
+
"eval_icace_normdiff": 0.4014453887939453,
|
| 598 |
+
"eval_loss": 7.68869161605835,
|
| 599 |
+
"eval_runtime": 22.1915,
|
| 600 |
+
"eval_samples_per_second": 175.653,
|
| 601 |
+
"eval_steps_per_second": 2.749,
|
| 602 |
+
"step": 900
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 1.58,
|
| 606 |
+
"learning_rate": 4.003466204506066e-05,
|
| 607 |
+
"loss": 7.8545,
|
| 608 |
+
"step": 920
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 1.61,
|
| 612 |
+
"learning_rate": 3.9818024263431546e-05,
|
| 613 |
+
"loss": 7.5503,
|
| 614 |
+
"step": 940
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 1.63,
|
| 618 |
+
"eval_distillation_accuracy_counterfactual": 0.5402770651616213,
|
| 619 |
+
"eval_distillation_accuracy_factual": 0.8211903540277066,
|
| 620 |
+
"eval_distillation_f1_counterfactual": 0.453286607279591,
|
| 621 |
+
"eval_distillation_f1_factual": 0.8132738069526344,
|
| 622 |
+
"eval_groundtruth_accuracy_counterfactual": 0.530528476141611,
|
| 623 |
+
"eval_groundtruth_f1_counterfactual": 0.4485755018961964,
|
| 624 |
+
"eval_groundtruth_f1_factual": 0.6565776752764576,
|
| 625 |
+
"eval_icace_cosine": 0.4694054126739502,
|
| 626 |
+
"eval_icace_l2": 0.625873863697052,
|
| 627 |
+
"eval_icace_normdiff": 0.39317741990089417,
|
| 628 |
+
"eval_loss": 7.5347418785095215,
|
| 629 |
+
"eval_runtime": 22.1059,
|
| 630 |
+
"eval_samples_per_second": 176.333,
|
| 631 |
+
"eval_steps_per_second": 2.759,
|
| 632 |
+
"step": 950
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.65,
|
| 636 |
+
"learning_rate": 3.960138648180243e-05,
|
| 637 |
+
"loss": 7.7698,
|
| 638 |
+
"step": 960
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 1.68,
|
| 642 |
+
"learning_rate": 3.938474870017331e-05,
|
| 643 |
+
"loss": 7.7985,
|
| 644 |
+
"step": 980
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 1.72,
|
| 648 |
+
"learning_rate": 3.91681109185442e-05,
|
| 649 |
+
"loss": 7.686,
|
| 650 |
+
"step": 1000
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 1.72,
|
| 654 |
+
"eval_distillation_accuracy_counterfactual": 0.5420728578758338,
|
| 655 |
+
"eval_distillation_accuracy_factual": 0.8396613648024628,
|
| 656 |
+
"eval_distillation_f1_counterfactual": 0.456055619550645,
|
| 657 |
+
"eval_distillation_f1_factual": 0.827631226039748,
|
| 658 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5292457670600308,
|
| 659 |
+
"eval_groundtruth_f1_counterfactual": 0.45013798447650544,
|
| 660 |
+
"eval_groundtruth_f1_factual": 0.6763146243844917,
|
| 661 |
+
"eval_icace_cosine": 0.45797210931777954,
|
| 662 |
+
"eval_icace_l2": 0.611527144908905,
|
| 663 |
+
"eval_icace_normdiff": 0.37649333477020264,
|
| 664 |
+
"eval_loss": 7.2710371017456055,
|
| 665 |
+
"eval_runtime": 22.2409,
|
| 666 |
+
"eval_samples_per_second": 175.262,
|
| 667 |
+
"eval_steps_per_second": 2.743,
|
| 668 |
+
"step": 1000
|
| 669 |
+
}
|
| 670 |
+
],
|
| 671 |
+
"max_steps": 4616,
|
| 672 |
+
"num_train_epochs": 8,
|
| 673 |
+
"total_flos": 8474526459302400.0,
|
| 674 |
+
"trial_name": null,
|
| 675 |
+
"trial_params": null
|
| 676 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d0ba90754c31fbbdd65794ee39c1c11a5cee019a6d04af812f5552c2adfb5c3f
|
| 3 |
+
size 3183
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|