End of training
Browse files- README.md +115 -0
- all_results.json +15 -0
- datasets/all_binary_and_xe_ey_fae_counterfactual/adapter_config.json +41 -0
- datasets/all_binary_and_xe_ey_fae_counterfactual/head_config.json +14 -0
- datasets/all_binary_and_xe_ey_fae_counterfactual/pytorch_adapter.bin +3 -0
- datasets/all_binary_and_xe_ey_fae_counterfactual/pytorch_model_head.bin +3 -0
- eval_results.json +10 -0
- train_results.json +8 -0
- trainer_state.json +645 -0
README.md
ADDED
|
@@ -0,0 +1,115 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: google/electra-base-generator
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
datasets:
|
| 7 |
+
- datasets/all_binary_and_xe_ey_fae_counterfactual
|
| 8 |
+
metrics:
|
| 9 |
+
- accuracy
|
| 10 |
+
model-index:
|
| 11 |
+
- name: electra-adapter-finetuned-xe_ey_fae
|
| 12 |
+
results:
|
| 13 |
+
- task:
|
| 14 |
+
name: Masked Language Modeling
|
| 15 |
+
type: fill-mask
|
| 16 |
+
dataset:
|
| 17 |
+
name: datasets/all_binary_and_xe_ey_fae_counterfactual
|
| 18 |
+
type: datasets/all_binary_and_xe_ey_fae_counterfactual
|
| 19 |
+
metrics:
|
| 20 |
+
- name: Accuracy
|
| 21 |
+
type: accuracy
|
| 22 |
+
value: 0.6258363412553052
|
| 23 |
+
---
|
| 24 |
+
|
| 25 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 26 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 27 |
+
|
| 28 |
+
# electra-adapter-finetuned-xe_ey_fae
|
| 29 |
+
|
| 30 |
+
This model is a fine-tuned version of [google/electra-base-generator](https://huggingface.co/google/electra-base-generator) on the datasets/all_binary_and_xe_ey_fae_counterfactual dataset.
|
| 31 |
+
It achieves the following results on the evaluation set:
|
| 32 |
+
- Loss: 2.0392
|
| 33 |
+
- Accuracy: 0.6258
|
| 34 |
+
|
| 35 |
+
## Model description
|
| 36 |
+
|
| 37 |
+
More information needed
|
| 38 |
+
|
| 39 |
+
## Intended uses & limitations
|
| 40 |
+
|
| 41 |
+
More information needed
|
| 42 |
+
|
| 43 |
+
## Training and evaluation data
|
| 44 |
+
|
| 45 |
+
More information needed
|
| 46 |
+
|
| 47 |
+
## Training procedure
|
| 48 |
+
|
| 49 |
+
### Training hyperparameters
|
| 50 |
+
|
| 51 |
+
The following hyperparameters were used during training:
|
| 52 |
+
- learning_rate: 1e-05
|
| 53 |
+
- train_batch_size: 8
|
| 54 |
+
- eval_batch_size: 8
|
| 55 |
+
- seed: 100
|
| 56 |
+
- gradient_accumulation_steps: 2
|
| 57 |
+
- total_train_batch_size: 16
|
| 58 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 59 |
+
- lr_scheduler_type: linear
|
| 60 |
+
- num_epochs: 3.0
|
| 61 |
+
- mixed_precision_training: Native AMP
|
| 62 |
+
|
| 63 |
+
### Training results
|
| 64 |
+
|
| 65 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 66 |
+
|:-------------:|:-----:|:-----:|:---------------:|:--------:|
|
| 67 |
+
| 3.9488 | 0.06 | 500 | 3.1500 | 0.5509 |
|
| 68 |
+
| 2.942 | 0.13 | 1000 | 2.5844 | 0.5680 |
|
| 69 |
+
| 2.6751 | 0.19 | 1500 | 2.4443 | 0.5790 |
|
| 70 |
+
| 2.582 | 0.26 | 2000 | 2.3701 | 0.5869 |
|
| 71 |
+
| 2.5267 | 0.32 | 2500 | 2.3097 | 0.5937 |
|
| 72 |
+
| 2.4722 | 0.39 | 3000 | 2.2695 | 0.5986 |
|
| 73 |
+
| 2.4289 | 0.45 | 3500 | 2.2329 | 0.6024 |
|
| 74 |
+
| 2.404 | 0.52 | 4000 | 2.2063 | 0.6055 |
|
| 75 |
+
| 2.3826 | 0.58 | 4500 | 2.1840 | 0.6087 |
|
| 76 |
+
| 2.3633 | 0.64 | 5000 | 2.1646 | 0.6109 |
|
| 77 |
+
| 2.3425 | 0.71 | 5500 | 2.1557 | 0.6121 |
|
| 78 |
+
| 2.333 | 0.77 | 6000 | 2.1350 | 0.6141 |
|
| 79 |
+
| 2.311 | 0.84 | 6500 | 2.1292 | 0.6152 |
|
| 80 |
+
| 2.3014 | 0.9 | 7000 | 2.1182 | 0.6166 |
|
| 81 |
+
| 2.2974 | 0.97 | 7500 | 2.1121 | 0.6170 |
|
| 82 |
+
| 2.2866 | 1.03 | 8000 | 2.1079 | 0.6173 |
|
| 83 |
+
| 2.2675 | 1.1 | 8500 | 2.0940 | 0.6192 |
|
| 84 |
+
| 2.2789 | 1.16 | 9000 | 2.0882 | 0.6201 |
|
| 85 |
+
| 2.2684 | 1.22 | 9500 | 2.0873 | 0.6200 |
|
| 86 |
+
| 2.2608 | 1.29 | 10000 | 2.0796 | 0.6209 |
|
| 87 |
+
| 2.2478 | 1.35 | 10500 | 2.0827 | 0.6204 |
|
| 88 |
+
| 2.2524 | 1.42 | 11000 | 2.0741 | 0.6215 |
|
| 89 |
+
| 2.2502 | 1.48 | 11500 | 2.0685 | 0.6220 |
|
| 90 |
+
| 2.243 | 1.55 | 12000 | 2.0665 | 0.6228 |
|
| 91 |
+
| 2.2417 | 1.61 | 12500 | 2.0632 | 0.6229 |
|
| 92 |
+
| 2.2398 | 1.68 | 13000 | 2.0593 | 0.6232 |
|
| 93 |
+
| 2.2233 | 1.74 | 13500 | 2.0600 | 0.6232 |
|
| 94 |
+
| 2.2277 | 1.8 | 14000 | 2.0535 | 0.6236 |
|
| 95 |
+
| 2.2344 | 1.87 | 14500 | 2.0485 | 0.6248 |
|
| 96 |
+
| 2.2274 | 1.93 | 15000 | 2.0507 | 0.6245 |
|
| 97 |
+
| 2.2212 | 2.0 | 15500 | 2.0428 | 0.6256 |
|
| 98 |
+
| 2.214 | 2.06 | 16000 | 2.0464 | 0.6244 |
|
| 99 |
+
| 2.2104 | 2.13 | 16500 | 2.0477 | 0.6250 |
|
| 100 |
+
| 2.2185 | 2.19 | 17000 | 2.0397 | 0.6257 |
|
| 101 |
+
| 2.2157 | 2.26 | 17500 | 2.0419 | 0.6257 |
|
| 102 |
+
| 2.2128 | 2.32 | 18000 | 2.0439 | 0.6255 |
|
| 103 |
+
| 2.2154 | 2.38 | 18500 | 2.0372 | 0.6259 |
|
| 104 |
+
| 2.2099 | 2.45 | 19000 | 2.0337 | 0.6263 |
|
| 105 |
+
| 2.2045 | 2.51 | 19500 | 2.0396 | 0.6259 |
|
| 106 |
+
| 2.2138 | 2.58 | 20000 | 2.0390 | 0.6262 |
|
| 107 |
+
| 2.2103 | 2.64 | 20500 | 2.0339 | 0.6263 |
|
| 108 |
+
|
| 109 |
+
|
| 110 |
+
### Framework versions
|
| 111 |
+
|
| 112 |
+
- Transformers 4.36.2
|
| 113 |
+
- Pytorch 2.2.0+cu121
|
| 114 |
+
- Datasets 2.17.0
|
| 115 |
+
- Tokenizers 0.15.2
|
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.64,
|
| 3 |
+
"eval_accuracy": 0.6258363412553052,
|
| 4 |
+
"eval_loss": 2.0392136573791504,
|
| 5 |
+
"eval_runtime": 87.4708,
|
| 6 |
+
"eval_samples": 15525,
|
| 7 |
+
"eval_samples_per_second": 177.488,
|
| 8 |
+
"eval_steps_per_second": 22.19,
|
| 9 |
+
"perplexity": 7.684564122147852,
|
| 10 |
+
"train_loss": 2.351401915015244,
|
| 11 |
+
"train_runtime": 7059.8447,
|
| 12 |
+
"train_samples": 124124,
|
| 13 |
+
"train_samples_per_second": 52.745,
|
| 14 |
+
"train_steps_per_second": 3.297
|
| 15 |
+
}
|
datasets/all_binary_and_xe_ey_fae_counterfactual/adapter_config.json
ADDED
|
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"config": {
|
| 3 |
+
"adapter_residual_before_ln": false,
|
| 4 |
+
"cross_adapter": false,
|
| 5 |
+
"factorized_phm_W": true,
|
| 6 |
+
"factorized_phm_rule": false,
|
| 7 |
+
"hypercomplex_nonlinearity": "glorot-uniform",
|
| 8 |
+
"init_weights": "bert",
|
| 9 |
+
"inv_adapter": null,
|
| 10 |
+
"inv_adapter_reduction_factor": null,
|
| 11 |
+
"is_parallel": false,
|
| 12 |
+
"learn_phm": true,
|
| 13 |
+
"leave_out": [],
|
| 14 |
+
"ln_after": false,
|
| 15 |
+
"ln_before": false,
|
| 16 |
+
"mh_adapter": false,
|
| 17 |
+
"non_linearity": "relu",
|
| 18 |
+
"original_ln_after": true,
|
| 19 |
+
"original_ln_before": true,
|
| 20 |
+
"output_adapter": true,
|
| 21 |
+
"phm_bias": true,
|
| 22 |
+
"phm_c_init": "normal",
|
| 23 |
+
"phm_dim": 4,
|
| 24 |
+
"phm_init_range": 0.0001,
|
| 25 |
+
"phm_layer": false,
|
| 26 |
+
"phm_rank": 1,
|
| 27 |
+
"reduction_factor": 16,
|
| 28 |
+
"residual_before_ln": true,
|
| 29 |
+
"scaling": 1.0,
|
| 30 |
+
"shared_W_phm": false,
|
| 31 |
+
"shared_phm_rule": true,
|
| 32 |
+
"use_gating": false
|
| 33 |
+
},
|
| 34 |
+
"config_id": "9076f36a74755ac4",
|
| 35 |
+
"hidden_size": 256,
|
| 36 |
+
"model_class": "ElectraForMaskedLM",
|
| 37 |
+
"model_name": "google/electra-base-generator",
|
| 38 |
+
"model_type": "electra",
|
| 39 |
+
"name": "datasets/all_binary_and_xe_ey_fae_counterfactual",
|
| 40 |
+
"version": "0.1.2"
|
| 41 |
+
}
|
datasets/all_binary_and_xe_ey_fae_counterfactual/head_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"config": null,
|
| 3 |
+
"hidden_size": 256,
|
| 4 |
+
"label2id": {
|
| 5 |
+
"LABEL_0": 0,
|
| 6 |
+
"LABEL_1": 1
|
| 7 |
+
},
|
| 8 |
+
"model_class": "ElectraForMaskedLM",
|
| 9 |
+
"model_name": "google/electra-base-generator",
|
| 10 |
+
"model_type": "electra",
|
| 11 |
+
"name": null,
|
| 12 |
+
"num_labels": 2,
|
| 13 |
+
"version": "0.1.2"
|
| 14 |
+
}
|
datasets/all_binary_and_xe_ey_fae_counterfactual/pytorch_adapter.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:48344746d047a38e3f7159048f13f2271dd6ce2f8ee3e79f55ed1043d9b44f21
|
| 3 |
+
size 425830
|
datasets/all_binary_and_xe_ey_fae_counterfactual/pytorch_model_head.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2c0648fbb6aa6e8b4fd9494a9da8a1ff89ca22bc8ff2ead6f16d707f696a993c
|
| 3 |
+
size 94684086
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.64,
|
| 3 |
+
"eval_accuracy": 0.6258363412553052,
|
| 4 |
+
"eval_loss": 2.0392136573791504,
|
| 5 |
+
"eval_runtime": 87.4708,
|
| 6 |
+
"eval_samples": 15525,
|
| 7 |
+
"eval_samples_per_second": 177.488,
|
| 8 |
+
"eval_steps_per_second": 22.19,
|
| 9 |
+
"perplexity": 7.684564122147852
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.64,
|
| 3 |
+
"train_loss": 2.351401915015244,
|
| 4 |
+
"train_runtime": 7059.8447,
|
| 5 |
+
"train_samples": 124124,
|
| 6 |
+
"train_samples_per_second": 52.745,
|
| 7 |
+
"train_steps_per_second": 3.297
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,645 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 2.0336806774139404,
|
| 3 |
+
"best_model_checkpoint": "finetuning/output/electra-adapter-finetuned_xe_ey_fae/checkpoint-19000",
|
| 4 |
+
"epoch": 2.642433616911575,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 20500,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.06,
|
| 13 |
+
"learning_rate": 9.785167998625076e-06,
|
| 14 |
+
"loss": 3.9488,
|
| 15 |
+
"step": 500
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.06,
|
| 19 |
+
"eval_accuracy": 0.5508950432882589,
|
| 20 |
+
"eval_loss": 3.1499977111816406,
|
| 21 |
+
"eval_runtime": 85.1217,
|
| 22 |
+
"eval_samples_per_second": 182.386,
|
| 23 |
+
"eval_steps_per_second": 22.803,
|
| 24 |
+
"step": 500
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.13,
|
| 28 |
+
"learning_rate": 9.57033599725015e-06,
|
| 29 |
+
"loss": 2.942,
|
| 30 |
+
"step": 1000
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.13,
|
| 34 |
+
"eval_accuracy": 0.5680209177510359,
|
| 35 |
+
"eval_loss": 2.584392547607422,
|
| 36 |
+
"eval_runtime": 79.4716,
|
| 37 |
+
"eval_samples_per_second": 195.353,
|
| 38 |
+
"eval_steps_per_second": 24.424,
|
| 39 |
+
"step": 1000
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.19,
|
| 43 |
+
"learning_rate": 9.355503995875225e-06,
|
| 44 |
+
"loss": 2.6751,
|
| 45 |
+
"step": 1500
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.19,
|
| 49 |
+
"eval_accuracy": 0.578970986434046,
|
| 50 |
+
"eval_loss": 2.444335699081421,
|
| 51 |
+
"eval_runtime": 87.5675,
|
| 52 |
+
"eval_samples_per_second": 177.292,
|
| 53 |
+
"eval_steps_per_second": 22.166,
|
| 54 |
+
"step": 1500
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.26,
|
| 58 |
+
"learning_rate": 9.140671994500302e-06,
|
| 59 |
+
"loss": 2.582,
|
| 60 |
+
"step": 2000
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.26,
|
| 64 |
+
"eval_accuracy": 0.5868782143731802,
|
| 65 |
+
"eval_loss": 2.3700673580169678,
|
| 66 |
+
"eval_runtime": 83.7436,
|
| 67 |
+
"eval_samples_per_second": 185.387,
|
| 68 |
+
"eval_steps_per_second": 23.178,
|
| 69 |
+
"step": 2000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.32,
|
| 73 |
+
"learning_rate": 8.926269657128126e-06,
|
| 74 |
+
"loss": 2.5267,
|
| 75 |
+
"step": 2500
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.32,
|
| 79 |
+
"eval_accuracy": 0.5937291646823507,
|
| 80 |
+
"eval_loss": 2.309689998626709,
|
| 81 |
+
"eval_runtime": 81.3517,
|
| 82 |
+
"eval_samples_per_second": 190.838,
|
| 83 |
+
"eval_steps_per_second": 23.859,
|
| 84 |
+
"step": 2500
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.39,
|
| 88 |
+
"learning_rate": 8.711437655753203e-06,
|
| 89 |
+
"loss": 2.4722,
|
| 90 |
+
"step": 3000
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.39,
|
| 94 |
+
"eval_accuracy": 0.5985969269659629,
|
| 95 |
+
"eval_loss": 2.2695114612579346,
|
| 96 |
+
"eval_runtime": 87.4381,
|
| 97 |
+
"eval_samples_per_second": 177.554,
|
| 98 |
+
"eval_steps_per_second": 22.199,
|
| 99 |
+
"step": 3000
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.45,
|
| 103 |
+
"learning_rate": 8.497035318381027e-06,
|
| 104 |
+
"loss": 2.4289,
|
| 105 |
+
"step": 3500
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.45,
|
| 109 |
+
"eval_accuracy": 0.602404170197503,
|
| 110 |
+
"eval_loss": 2.2328779697418213,
|
| 111 |
+
"eval_runtime": 83.7759,
|
| 112 |
+
"eval_samples_per_second": 185.316,
|
| 113 |
+
"eval_steps_per_second": 23.169,
|
| 114 |
+
"step": 3500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.52,
|
| 118 |
+
"learning_rate": 8.282203317006102e-06,
|
| 119 |
+
"loss": 2.404,
|
| 120 |
+
"step": 4000
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.52,
|
| 124 |
+
"eval_accuracy": 0.6055254061674608,
|
| 125 |
+
"eval_loss": 2.206317901611328,
|
| 126 |
+
"eval_runtime": 87.3965,
|
| 127 |
+
"eval_samples_per_second": 177.639,
|
| 128 |
+
"eval_steps_per_second": 22.209,
|
| 129 |
+
"step": 4000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.58,
|
| 133 |
+
"learning_rate": 8.067371315631177e-06,
|
| 134 |
+
"loss": 2.3826,
|
| 135 |
+
"step": 4500
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.58,
|
| 139 |
+
"eval_accuracy": 0.6086694296803393,
|
| 140 |
+
"eval_loss": 2.183983087539673,
|
| 141 |
+
"eval_runtime": 87.0314,
|
| 142 |
+
"eval_samples_per_second": 178.384,
|
| 143 |
+
"eval_steps_per_second": 22.302,
|
| 144 |
+
"step": 4500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.64,
|
| 148 |
+
"learning_rate": 7.852539314256252e-06,
|
| 149 |
+
"loss": 2.3633,
|
| 150 |
+
"step": 5000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.64,
|
| 154 |
+
"eval_accuracy": 0.6108753723178051,
|
| 155 |
+
"eval_loss": 2.1645586490631104,
|
| 156 |
+
"eval_runtime": 83.9383,
|
| 157 |
+
"eval_samples_per_second": 184.957,
|
| 158 |
+
"eval_steps_per_second": 23.124,
|
| 159 |
+
"step": 5000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.71,
|
| 163 |
+
"learning_rate": 7.637707312881327e-06,
|
| 164 |
+
"loss": 2.3425,
|
| 165 |
+
"step": 5500
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.71,
|
| 169 |
+
"eval_accuracy": 0.6121162378522405,
|
| 170 |
+
"eval_loss": 2.155695676803589,
|
| 171 |
+
"eval_runtime": 87.4417,
|
| 172 |
+
"eval_samples_per_second": 177.547,
|
| 173 |
+
"eval_steps_per_second": 22.198,
|
| 174 |
+
"step": 5500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.77,
|
| 178 |
+
"learning_rate": 7.4228753115064025e-06,
|
| 179 |
+
"loss": 2.333,
|
| 180 |
+
"step": 6000
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 0.77,
|
| 184 |
+
"eval_accuracy": 0.6140775893820937,
|
| 185 |
+
"eval_loss": 2.1349785327911377,
|
| 186 |
+
"eval_runtime": 85.1022,
|
| 187 |
+
"eval_samples_per_second": 182.428,
|
| 188 |
+
"eval_steps_per_second": 22.808,
|
| 189 |
+
"step": 6000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.84,
|
| 193 |
+
"learning_rate": 7.208472974134228e-06,
|
| 194 |
+
"loss": 2.311,
|
| 195 |
+
"step": 6500
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.84,
|
| 199 |
+
"eval_accuracy": 0.6151508455851109,
|
| 200 |
+
"eval_loss": 2.1292011737823486,
|
| 201 |
+
"eval_runtime": 79.4597,
|
| 202 |
+
"eval_samples_per_second": 195.382,
|
| 203 |
+
"eval_steps_per_second": 24.427,
|
| 204 |
+
"step": 6500
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.9,
|
| 208 |
+
"learning_rate": 6.993640972759303e-06,
|
| 209 |
+
"loss": 2.3014,
|
| 210 |
+
"step": 7000
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 0.9,
|
| 214 |
+
"eval_accuracy": 0.6166432908599604,
|
| 215 |
+
"eval_loss": 2.1181797981262207,
|
| 216 |
+
"eval_runtime": 87.6275,
|
| 217 |
+
"eval_samples_per_second": 177.17,
|
| 218 |
+
"eval_steps_per_second": 22.151,
|
| 219 |
+
"step": 7000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.97,
|
| 223 |
+
"learning_rate": 6.7788089713843775e-06,
|
| 224 |
+
"loss": 2.2974,
|
| 225 |
+
"step": 7500
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.97,
|
| 229 |
+
"eval_accuracy": 0.6169897785349233,
|
| 230 |
+
"eval_loss": 2.112070083618164,
|
| 231 |
+
"eval_runtime": 83.9336,
|
| 232 |
+
"eval_samples_per_second": 184.968,
|
| 233 |
+
"eval_steps_per_second": 23.125,
|
| 234 |
+
"step": 7500
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.03,
|
| 238 |
+
"learning_rate": 6.563976970009453e-06,
|
| 239 |
+
"loss": 2.2866,
|
| 240 |
+
"step": 8000
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 1.03,
|
| 244 |
+
"eval_accuracy": 0.6173022781800038,
|
| 245 |
+
"eval_loss": 2.107919454574585,
|
| 246 |
+
"eval_runtime": 82.2636,
|
| 247 |
+
"eval_samples_per_second": 188.723,
|
| 248 |
+
"eval_steps_per_second": 23.595,
|
| 249 |
+
"step": 8000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 1.1,
|
| 253 |
+
"learning_rate": 6.349574632637278e-06,
|
| 254 |
+
"loss": 2.2675,
|
| 255 |
+
"step": 8500
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.1,
|
| 259 |
+
"eval_accuracy": 0.6191927234863566,
|
| 260 |
+
"eval_loss": 2.0939817428588867,
|
| 261 |
+
"eval_runtime": 87.5998,
|
| 262 |
+
"eval_samples_per_second": 177.226,
|
| 263 |
+
"eval_steps_per_second": 22.158,
|
| 264 |
+
"step": 8500
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 1.16,
|
| 268 |
+
"learning_rate": 6.134742631262354e-06,
|
| 269 |
+
"loss": 2.2789,
|
| 270 |
+
"step": 9000
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 1.16,
|
| 274 |
+
"eval_accuracy": 0.6201220093575694,
|
| 275 |
+
"eval_loss": 2.088168144226074,
|
| 276 |
+
"eval_runtime": 83.772,
|
| 277 |
+
"eval_samples_per_second": 185.324,
|
| 278 |
+
"eval_steps_per_second": 23.17,
|
| 279 |
+
"step": 9000
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 1.22,
|
| 283 |
+
"learning_rate": 5.919910629887429e-06,
|
| 284 |
+
"loss": 2.2684,
|
| 285 |
+
"step": 9500
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 1.22,
|
| 289 |
+
"eval_accuracy": 0.6199849943877651,
|
| 290 |
+
"eval_loss": 2.0872652530670166,
|
| 291 |
+
"eval_runtime": 87.4418,
|
| 292 |
+
"eval_samples_per_second": 177.547,
|
| 293 |
+
"eval_steps_per_second": 22.198,
|
| 294 |
+
"step": 9500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 1.29,
|
| 298 |
+
"learning_rate": 5.705078628512504e-06,
|
| 299 |
+
"loss": 2.2608,
|
| 300 |
+
"step": 10000
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 1.29,
|
| 304 |
+
"eval_accuracy": 0.6208952330586832,
|
| 305 |
+
"eval_loss": 2.0795998573303223,
|
| 306 |
+
"eval_runtime": 86.9343,
|
| 307 |
+
"eval_samples_per_second": 178.583,
|
| 308 |
+
"eval_steps_per_second": 22.327,
|
| 309 |
+
"step": 10000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.35,
|
| 313 |
+
"learning_rate": 5.490246627137579e-06,
|
| 314 |
+
"loss": 2.2478,
|
| 315 |
+
"step": 10500
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 1.35,
|
| 319 |
+
"eval_accuracy": 0.620409766315376,
|
| 320 |
+
"eval_loss": 2.082674503326416,
|
| 321 |
+
"eval_runtime": 84.0547,
|
| 322 |
+
"eval_samples_per_second": 184.701,
|
| 323 |
+
"eval_steps_per_second": 23.092,
|
| 324 |
+
"step": 10500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.42,
|
| 328 |
+
"learning_rate": 5.275844289765404e-06,
|
| 329 |
+
"loss": 2.2524,
|
| 330 |
+
"step": 11000
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 1.42,
|
| 334 |
+
"eval_accuracy": 0.6214935816878795,
|
| 335 |
+
"eval_loss": 2.074056386947632,
|
| 336 |
+
"eval_runtime": 87.5237,
|
| 337 |
+
"eval_samples_per_second": 177.381,
|
| 338 |
+
"eval_steps_per_second": 22.177,
|
| 339 |
+
"step": 11000
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.48,
|
| 343 |
+
"learning_rate": 5.061012288390479e-06,
|
| 344 |
+
"loss": 2.2502,
|
| 345 |
+
"step": 11500
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 1.48,
|
| 349 |
+
"eval_accuracy": 0.6220323169678965,
|
| 350 |
+
"eval_loss": 2.068490505218506,
|
| 351 |
+
"eval_runtime": 84.958,
|
| 352 |
+
"eval_samples_per_second": 182.737,
|
| 353 |
+
"eval_steps_per_second": 22.847,
|
| 354 |
+
"step": 11500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 1.55,
|
| 358 |
+
"learning_rate": 4.8461802870155545e-06,
|
| 359 |
+
"loss": 2.243,
|
| 360 |
+
"step": 12000
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.55,
|
| 364 |
+
"eval_accuracy": 0.622761702720804,
|
| 365 |
+
"eval_loss": 2.0664761066436768,
|
| 366 |
+
"eval_runtime": 79.0021,
|
| 367 |
+
"eval_samples_per_second": 196.514,
|
| 368 |
+
"eval_steps_per_second": 24.569,
|
| 369 |
+
"step": 12000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 1.61,
|
| 373 |
+
"learning_rate": 4.631348285640629e-06,
|
| 374 |
+
"loss": 2.2417,
|
| 375 |
+
"step": 12500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 1.61,
|
| 379 |
+
"eval_accuracy": 0.6228723852166125,
|
| 380 |
+
"eval_loss": 2.0631983280181885,
|
| 381 |
+
"eval_runtime": 87.1566,
|
| 382 |
+
"eval_samples_per_second": 178.128,
|
| 383 |
+
"eval_steps_per_second": 22.27,
|
| 384 |
+
"step": 12500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.68,
|
| 388 |
+
"learning_rate": 4.416516284265704e-06,
|
| 389 |
+
"loss": 2.2398,
|
| 390 |
+
"step": 13000
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 1.68,
|
| 394 |
+
"eval_accuracy": 0.6232123058100858,
|
| 395 |
+
"eval_loss": 2.0592522621154785,
|
| 396 |
+
"eval_runtime": 83.668,
|
| 397 |
+
"eval_samples_per_second": 185.555,
|
| 398 |
+
"eval_steps_per_second": 23.199,
|
| 399 |
+
"step": 13000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 1.74,
|
| 403 |
+
"learning_rate": 4.20168428289078e-06,
|
| 404 |
+
"loss": 2.2233,
|
| 405 |
+
"step": 13500
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 1.74,
|
| 409 |
+
"eval_accuracy": 0.6232258668129607,
|
| 410 |
+
"eval_loss": 2.060002326965332,
|
| 411 |
+
"eval_runtime": 80.0466,
|
| 412 |
+
"eval_samples_per_second": 193.95,
|
| 413 |
+
"eval_steps_per_second": 24.248,
|
| 414 |
+
"step": 13500
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 1.8,
|
| 418 |
+
"learning_rate": 3.987281945518604e-06,
|
| 419 |
+
"loss": 2.2277,
|
| 420 |
+
"step": 14000
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 1.8,
|
| 424 |
+
"eval_accuracy": 0.623606800420627,
|
| 425 |
+
"eval_loss": 2.0534963607788086,
|
| 426 |
+
"eval_runtime": 87.4565,
|
| 427 |
+
"eval_samples_per_second": 177.517,
|
| 428 |
+
"eval_steps_per_second": 22.194,
|
| 429 |
+
"step": 14000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.87,
|
| 433 |
+
"learning_rate": 3.77244994414368e-06,
|
| 434 |
+
"loss": 2.2344,
|
| 435 |
+
"step": 14500
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 1.87,
|
| 439 |
+
"eval_accuracy": 0.6247527084114421,
|
| 440 |
+
"eval_loss": 2.0484962463378906,
|
| 441 |
+
"eval_runtime": 83.8183,
|
| 442 |
+
"eval_samples_per_second": 185.222,
|
| 443 |
+
"eval_steps_per_second": 23.157,
|
| 444 |
+
"step": 14500
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.93,
|
| 448 |
+
"learning_rate": 3.5576179427687554e-06,
|
| 449 |
+
"loss": 2.2274,
|
| 450 |
+
"step": 15000
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.93,
|
| 454 |
+
"eval_accuracy": 0.6244717527399175,
|
| 455 |
+
"eval_loss": 2.050738573074341,
|
| 456 |
+
"eval_runtime": 87.5865,
|
| 457 |
+
"eval_samples_per_second": 177.253,
|
| 458 |
+
"eval_steps_per_second": 22.161,
|
| 459 |
+
"step": 15000
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.0,
|
| 463 |
+
"learning_rate": 3.34321560539658e-06,
|
| 464 |
+
"loss": 2.2212,
|
| 465 |
+
"step": 15500
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 2.0,
|
| 469 |
+
"eval_accuracy": 0.6256074101917349,
|
| 470 |
+
"eval_loss": 2.0428130626678467,
|
| 471 |
+
"eval_runtime": 86.8032,
|
| 472 |
+
"eval_samples_per_second": 178.853,
|
| 473 |
+
"eval_steps_per_second": 22.361,
|
| 474 |
+
"step": 15500
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 2.06,
|
| 478 |
+
"learning_rate": 3.1283836040216555e-06,
|
| 479 |
+
"loss": 2.214,
|
| 480 |
+
"step": 16000
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 2.06,
|
| 484 |
+
"eval_accuracy": 0.6244417876710062,
|
| 485 |
+
"eval_loss": 2.0463979244232178,
|
| 486 |
+
"eval_runtime": 84.1399,
|
| 487 |
+
"eval_samples_per_second": 184.514,
|
| 488 |
+
"eval_steps_per_second": 23.069,
|
| 489 |
+
"step": 16000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 2.13,
|
| 493 |
+
"learning_rate": 2.9135516026467303e-06,
|
| 494 |
+
"loss": 2.2104,
|
| 495 |
+
"step": 16500
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 2.13,
|
| 499 |
+
"eval_accuracy": 0.6249873550076295,
|
| 500 |
+
"eval_loss": 2.0476861000061035,
|
| 501 |
+
"eval_runtime": 87.5417,
|
| 502 |
+
"eval_samples_per_second": 177.344,
|
| 503 |
+
"eval_steps_per_second": 22.172,
|
| 504 |
+
"step": 16500
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 2.19,
|
| 508 |
+
"learning_rate": 2.698719601271806e-06,
|
| 509 |
+
"loss": 2.2185,
|
| 510 |
+
"step": 17000
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 2.19,
|
| 514 |
+
"eval_accuracy": 0.6257313721221357,
|
| 515 |
+
"eval_loss": 2.039674758911133,
|
| 516 |
+
"eval_runtime": 84.986,
|
| 517 |
+
"eval_samples_per_second": 182.677,
|
| 518 |
+
"eval_steps_per_second": 22.839,
|
| 519 |
+
"step": 17000
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 2.26,
|
| 523 |
+
"learning_rate": 2.483887599896881e-06,
|
| 524 |
+
"loss": 2.2157,
|
| 525 |
+
"step": 17500
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 2.26,
|
| 529 |
+
"eval_accuracy": 0.6257406865679764,
|
| 530 |
+
"eval_loss": 2.041879177093506,
|
| 531 |
+
"eval_runtime": 79.7413,
|
| 532 |
+
"eval_samples_per_second": 194.692,
|
| 533 |
+
"eval_steps_per_second": 24.341,
|
| 534 |
+
"step": 17500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 2.32,
|
| 538 |
+
"learning_rate": 2.2690555985219558e-06,
|
| 539 |
+
"loss": 2.2128,
|
| 540 |
+
"step": 18000
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 2.32,
|
| 544 |
+
"eval_accuracy": 0.6254893845927666,
|
| 545 |
+
"eval_loss": 2.043928623199463,
|
| 546 |
+
"eval_runtime": 87.45,
|
| 547 |
+
"eval_samples_per_second": 177.53,
|
| 548 |
+
"eval_steps_per_second": 22.196,
|
| 549 |
+
"step": 18000
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 2.38,
|
| 553 |
+
"learning_rate": 2.054223597147031e-06,
|
| 554 |
+
"loss": 2.2154,
|
| 555 |
+
"step": 18500
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 2.38,
|
| 559 |
+
"eval_accuracy": 0.6259225237275015,
|
| 560 |
+
"eval_loss": 2.037231683731079,
|
| 561 |
+
"eval_runtime": 83.6819,
|
| 562 |
+
"eval_samples_per_second": 185.524,
|
| 563 |
+
"eval_steps_per_second": 23.195,
|
| 564 |
+
"step": 18500
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 2.45,
|
| 568 |
+
"learning_rate": 1.8393915957721066e-06,
|
| 569 |
+
"loss": 2.2099,
|
| 570 |
+
"step": 19000
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 2.45,
|
| 574 |
+
"eval_accuracy": 0.62631184758297,
|
| 575 |
+
"eval_loss": 2.0336806774139404,
|
| 576 |
+
"eval_runtime": 81.3506,
|
| 577 |
+
"eval_samples_per_second": 190.841,
|
| 578 |
+
"eval_steps_per_second": 23.86,
|
| 579 |
+
"step": 19000
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 2.51,
|
| 583 |
+
"learning_rate": 1.6245595943971814e-06,
|
| 584 |
+
"loss": 2.2045,
|
| 585 |
+
"step": 19500
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 2.51,
|
| 589 |
+
"eval_accuracy": 0.6258799592390727,
|
| 590 |
+
"eval_loss": 2.039562225341797,
|
| 591 |
+
"eval_runtime": 87.4501,
|
| 592 |
+
"eval_samples_per_second": 177.53,
|
| 593 |
+
"eval_steps_per_second": 22.196,
|
| 594 |
+
"step": 19500
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 2.58,
|
| 598 |
+
"learning_rate": 1.4097275930222567e-06,
|
| 599 |
+
"loss": 2.2138,
|
| 600 |
+
"step": 20000
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 2.58,
|
| 604 |
+
"eval_accuracy": 0.6261649440028011,
|
| 605 |
+
"eval_loss": 2.0390186309814453,
|
| 606 |
+
"eval_runtime": 83.8434,
|
| 607 |
+
"eval_samples_per_second": 185.167,
|
| 608 |
+
"eval_steps_per_second": 23.15,
|
| 609 |
+
"step": 20000
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 2.64,
|
| 613 |
+
"learning_rate": 1.194895591647332e-06,
|
| 614 |
+
"loss": 2.2103,
|
| 615 |
+
"step": 20500
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 2.64,
|
| 619 |
+
"eval_accuracy": 0.6262993215315168,
|
| 620 |
+
"eval_loss": 2.03385329246521,
|
| 621 |
+
"eval_runtime": 87.3376,
|
| 622 |
+
"eval_samples_per_second": 177.759,
|
| 623 |
+
"eval_steps_per_second": 22.224,
|
| 624 |
+
"step": 20500
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 2.64,
|
| 628 |
+
"step": 20500,
|
| 629 |
+
"total_flos": 1.0082485751267328e+16,
|
| 630 |
+
"train_loss": 2.351401915015244,
|
| 631 |
+
"train_runtime": 7059.8447,
|
| 632 |
+
"train_samples_per_second": 52.745,
|
| 633 |
+
"train_steps_per_second": 3.297
|
| 634 |
+
}
|
| 635 |
+
],
|
| 636 |
+
"logging_steps": 500,
|
| 637 |
+
"max_steps": 23274,
|
| 638 |
+
"num_input_tokens_seen": 0,
|
| 639 |
+
"num_train_epochs": 3,
|
| 640 |
+
"save_steps": 500,
|
| 641 |
+
"total_flos": 1.0082485751267328e+16,
|
| 642 |
+
"train_batch_size": 8,
|
| 643 |
+
"trial_name": null,
|
| 644 |
+
"trial_params": null
|
| 645 |
+
}
|