DongfuJiang_-_prm_version3_subsample_hf-gguf

RichardErkhov/DongfuJiang_-_prm_version3_subsample_hf-gguf

DongfuJiang_-_prm_version3_subsample_hf-gguf is a machine learning model from RichardErkhov. Use it for the machine learning task on the model card, and read the license before you ship it in a product.

Downloads · 30 days

372

28% of all-time downloads

All-time downloads

1.3K

Public

Repo size

92.9 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.gguf108 GB · 100%

At a glance

Access: Public
Created: Jul 25, 2025
Updated: Jul 25, 2025
SHA: f54510e3

Model card

From the Hugging Face model README

Quantization made by Richard Erkhov.

Github

Discord

Request more models

prm_version3_subsample_hf - GGUF

Model creator: https://huggingface.co/DongfuJiang/
Original model: https://huggingface.co/DongfuJiang/prm_version3_subsample_hf/

Name	Quant method	Size
prm_version3_subsample_hf.Q2_K.gguf	Q2_K	2.96GB
prm_version3_subsample_hf.IQ3_XS.gguf	IQ3_XS	3.28GB
prm_version3_subsample_hf.IQ3_S.gguf	IQ3_S	3.43GB
prm_version3_subsample_hf.Q3_K_S.gguf	Q3_K_S	3.41GB
prm_version3_subsample_hf.IQ3_M.gguf	IQ3_M	3.52GB
prm_version3_subsample_hf.Q3_K.gguf	Q3_K	3.74GB
prm_version3_subsample_hf.Q3_K_M.gguf	Q3_K_M	3.74GB
prm_version3_subsample_hf.Q3_K_L.gguf	Q3_K_L	4.03GB
prm_version3_subsample_hf.IQ4_XS.gguf	IQ4_XS	4.18GB
prm_version3_subsample_hf.Q4_0.gguf	Q4_0	4.34GB
prm_version3_subsample_hf.IQ4_NL.gguf	IQ4_NL	4.38GB
prm_version3_subsample_hf.Q4_K_S.gguf	Q4_K_S	4.37GB
prm_version3_subsample_hf.Q4_K.gguf	Q4_K	4.58GB
prm_version3_subsample_hf.Q4_K_M.gguf	Q4_K_M	4.58GB
prm_version3_subsample_hf.Q4_1.gguf	Q4_1	4.78GB
prm_version3_subsample_hf.Q5_0.gguf	Q5_0	5.21GB
prm_version3_subsample_hf.Q5_K_S.gguf	Q5_K_S	5.21GB
prm_version3_subsample_hf.Q5_K.gguf	Q5_K	5.34GB
prm_version3_subsample_hf.Q5_K_M.gguf	Q5_K_M	5.34GB
prm_version3_subsample_hf.Q5_1.gguf	Q5_1	5.65GB
prm_version3_subsample_hf.Q6_K.gguf	Q6_K	6.14GB
prm_version3_subsample_hf.Q8_0.gguf	Q8_0	7.95GB

Original model description:

library_name: transformers license: llama3.1 base_model: meta-llama/Meta-Llama-3.1-8B-Instruct tags:

llama-factory
full
generated_from_trainer model-index:
name: prm_version3_hf results: []

prm_version3_hf

This model is a fine-tuned version of meta-llama/Meta-Llama-3.1-8B-Instruct on the prm_conversations_prm_version3_math+webinstructsub-mcq+webinstructsub-oe+apps+gsm_mix_ref_subsample_hf dataset. It achieves the following results on the evaluation set:

Loss: 0.1262

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: 5e-06
train_batch_size: 1
eval_batch_size: 1
seed: 42
distributed_type: multi-GPU
num_devices: 8
gradient_accumulation_steps: 8
total_train_batch_size: 64
total_eval_batch_size: 8
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.05
num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
0.2014	0.1127	500	0.1967
0.1775	0.2253	1000	0.1792
0.1532	0.3380	1500	0.1660
0.1492	0.4506	2000	0.1544
0.1318	0.5633	2500	0.1443
0.1324	0.6759	3000	0.1358
0.1347	0.7886	3500	0.1296
0.1128	0.9012	4000	0.1268

Framework versions

Transformers 4.45.0
Pytorch 2.5.1+cu124
Datasets 3.1.0
Tokenizers 0.20.3

Created: Jul 25, 2025
Updated: Jul 25, 2025