Model Card for Model ID

Summary Length PPO experiment #7
No KL divergence in loss

Model Details

Dataset size: 16384
Epochs: 1
Batch Size: 16 * 4 (w/ 4 GPUs)

Optimizer args: Torch AdamW default, except

LR = 0.00001

Downloads last month: 94

Safetensors

Model size

1.24B params

Tensor type

BF16

Inference Providers NEW

Text Generation

This model is not currently available via any of the supported third-party Inference Providers, and the model is not deployed on the HF Inference API.

Collection including amang1802/Llama3.2-1B-summary-length-exp7

PPO experiments

Collection

Using PPO with simpler reward functions • 8 items • Updated 12 days ago