iqwiki-kor
/

Llama3.2-3B-MP-RM

Generated from Trainer

Model card Files Files and versions Metrics Training metrics Community

Llama3.2-3B-MP-RM

This model is a fine-tuned version of meta-llama/Llama-3.2-3B-Instruct on the iqwiki-kor/MP-86k dataset.

RewardBench Evaluation

Model	Chat	Chat-Hard	Safety	Reasoning	Avg.
iqwiki-kor/Llama3.2-3B-MP-RM	92.5	81.8	90.2	95.5	90.0
RLHFlow/ArmoRM-Llama3-8B-v0.1	96.9	76.8	90.5	97.3	90.4

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 1e-05
train_batch_size: 16
eval_batch_size: 8
seed: 42
distributed_type: multi-GPU
num_devices: 4
gradient_accumulation_steps: 2
total_train_batch_size: 128
total_eval_batch_size: 32
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.1
num_epochs: 1

Downloads last month: 5

Safetensors

Model size

3.21B params

Tensor type

BF16

·

Inference Providers NEW

This model is not currently available via any of the supported third-party Inference Providers, and HF Inference API was unable to determine this model's library.

Model tree for iqwiki-kor/Llama3.2-3B-MP-RM

Base model

meta-llama/Llama-3.2-3B-Instruct

Finetuned

(227)

this model

Collection including iqwiki-kor/Llama3.2-3B-MP-RM

Cross-lingual Transfer of Reward Models

This is the collection of synthetic preference data and trained reward models in "Cross-lingual Transfer of Reward Models in Multilingual Alignment". • 5 items • Updated Oct 31, 2024

Evaluation results

Metadata error: specify a dataset to view leaderboard