gpt2-xl-summarization_reward_model

Tristan/gpt2-xl-summarization_reward_model

gpt2-xl-summarization_reward_model is a machine learning model from Tristan. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as mit.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

Access

Public

Updated Jan 26, 2023

Repo size

31.4 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin6.3 GB · 100%

At a glance

License: mit
Access: Public
Created: Jan 26, 2023
Updated: Jan 26, 2023
SHA: a3cbdbd4

Model card

From the Hugging Face model README

gpt2-xl-summarization_reward_model

This model is a fine-tuned version of gpt2-xl on the None dataset. It achieves the following results on the evaluation set:

Loss: 1.2875
Accuracy: 0.6157

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: 2e-05
train_batch_size: 1
eval_batch_size: 1
seed: 42
distributed_type: multi-GPU
num_devices: 16
gradient_accumulation_steps: 4
total_train_batch_size: 64
total_eval_batch_size: 16
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.5856	1.0	1451	0.6854	0.6218
0.4314	2.0	2902	0.8053	0.6133
0.3166	3.0	4353	0.8060	0.6146
0.2625	4.0	5804	0.9857	0.6162
0.2279	5.0	7255	1.2875	0.6157

Framework versions

Transformers 4.26.0
Pytorch 1.13.1+cu117
Datasets 2.8.0
Tokenizers 0.13.2

License: mit
Created: Jan 26, 2023
Updated: Jan 26, 2023