lm-human-preference-details

lm-human-preference-details's activity

• All
• Models
• Datasets
• Spaces
• Papers
• Collections
• Community
• Posts

vwxyzjn 

authored 5 papers 21 days ago

The N+ Implementation Details of RLHF with PPO: A Case Study on TL;DR Summarization

Paper • 2403.17031 • Published Mar 24, 2024 • 4

A2C is a special case of PPO

Paper • 2205.09123 • Published May 18, 2022

Asynchronous RLHF: Faster and More Efficient Off-Policy RL for Language Models

Paper • 2410.18252 • Published Oct 23, 2024 • 5

TÜLU 3: Pushing Frontiers in Open Language Model Post-Training

Paper • 2411.15124 • Published Nov 22, 2024 • 58

2 OLMo 2 Furious

Paper • 2501.00656 • Published 26 days ago • 15

vwxyzjn 

authored a paper 12 months ago

Open RL Benchmark: Comprehensive Tracked Experiments for Reinforcement Learning

Paper • 2402.03046 • Published Feb 5, 2024 • 6

vwxyzjn 

authored a paper about 1 year ago

Zephyr: Direct Distillation of LM Alignment

Paper • 2310.16944 • Published Oct 25, 2023 • 123

vwxyzjn 

authored a paper over 1 year ago

Cleanba: A Reproducible and Efficient Distributed Reinforcement Learning Platform

Paper • 2310.00036 • Published Sep 29, 2023 • 2

vwxyzjn 

updated a Space over 1 year ago

Running

3

💻

Rlhf Demo

vwxyzjn 

updated 11 models over 1 year ago

lm-human-preference-details/train_policy_accelerate_tf_adam_gpt2_xl_grad_accu__sentiment_offline_5k.json__seed1

Text Generation • Updated Oct 6, 2023 • 87

lm-human-preference-details/train_policy_accelerate_tf_adam_gpt2_xl_grad_accu__sentiment_offline_5k.json__seed5

Text Generation • Updated Oct 6, 2023 • 88

lm-human-preference-details/train_policy_accelerate_tf_adam_gpt2_xl_grad_accu__sentiment_offline_5k.json__seed3

Text Generation • Updated Oct 6, 2023 • 88

lm-human-preference-details/train_policy_accelerate_tf_adam_gpt2_xl_grad_accu__sentiment_offline_5k.json__seed4

Text Generation • Updated Oct 6, 2023 • 87

lm-human-preference-details/train_policy_accelerate_tf_adam_gpt2_xl_grad_accu__sentiment_offline_5k.json__seed2

Text Generation • Updated Oct 6, 2023 • 88

lm-human-preference-details/train_policy_accelerate_pt_adam_gpt2_xl_grad_accu__sentiment_offline_5k.json__seed3

Text Generation • Updated Oct 6, 2023 • 91

lm-human-preference-details/train_policy_accelerate_pt_adam_gpt2_xl_grad_accu__sentiment_offline_5k.json__seed5

Text Generation • Updated Oct 6, 2023 • 88

lm-human-preference-details/train_policy_accelerate_pt_adam_gpt2_xl_grad_accu__sentiment_offline_5k.json__seed2

Text Generation • Updated Oct 6, 2023 • 87

lm-human-preference-details/train_policy_accelerate_pt_adam_gpt2_xl_grad_accu__sentiment_offline_5k.json__seed4

Text Generation • Updated Oct 6, 2023 • 86

lm-human-preference-details/train_policy_accelerate_pt_adam_gpt2__sentiment_offline_5k.json__seed5

Text Generation • Updated Oct 6, 2023 • 122

lm-human-preference-details/train_policy_accelerate_pt_adam_gpt2__sentiment_offline_5k.json__seed2

Text Generation • Updated Oct 5, 2023 • 122