metadata

license: mit
base_model: gpt2-xl
tags:
  - generated_from_trainer
datasets:
  - tyzhu/lmind_hotpot_train500_eval300_v1_recite_qa
metrics:
  - accuracy
model-index:
  - name: lmind_hotpot_train500_eval300_v1_recite_qa_gpt2-xl
    results:
      - task:
          name: Causal Language Modeling
          type: text-generation
        dataset:
          name: tyzhu/lmind_hotpot_train500_eval300_v1_recite_qa
          type: tyzhu/lmind_hotpot_train500_eval300_v1_recite_qa
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.7413294517224648

lmind_hotpot_train500_eval300_v1_recite_qa_gpt2-xl

This model is a fine-tuned version of gpt2-xl on the tyzhu/lmind_hotpot_train500_eval300_v1_recite_qa dataset. It achieves the following results on the evaluation set:

Loss: 0.4030
Accuracy: 0.7413

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 3e-05
train_batch_size: 16
eval_batch_size: 16
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: constant
num_epochs: 10.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.0739	0.5	66	1.8405	0.6206
1.8806	1.01	132	1.5749	0.6365
1.3619	1.51	198	1.3425	0.6533
1.283	2.02	264	1.1253	0.6685
0.8433	2.52	330	0.9735	0.6825
0.7629	3.02	396	0.7874	0.6982
0.5058	3.53	462	0.6921	0.7086
0.4593	4.03	528	0.5641	0.7197
0.3064	4.53	594	0.5348	0.7245
0.2967	5.04	660	0.4770	0.7304
0.2167	5.54	726	0.4582	0.7324
0.2157	6.05	792	0.4308	0.7358
0.1597	6.55	858	0.4301	0.7373
0.1481	7.05	924	0.4224	0.7385
0.1293	7.56	990	0.4125	0.7394
0.125	8.06	1056	0.4122	0.7400
0.1139	8.56	1122	0.4069	0.7407
0.1141	9.07	1188	0.4082	0.7409
0.0994	9.57	1254	0.4065	0.7412

Framework versions

Transformers 4.34.0
Pytorch 2.1.0+cu121
Datasets 2.14.5
Tokenizers 0.14.1