Dynosaur_-_llama3-8b-math-sft-mix-8-0-gguf – AI Model by RichardErkhov | AlphaNeural AI

You can deploy this model and start earning money today!

RichardErkhov/
Dynosaur_-_llama3-8b-math-sft-mix-8-0-gguf

like

0

endpoints_compatible

Views

No views yet

Quantization made by Richard Erkhov.

Request more models

llama3-8b-math-sft-mix-8-0 - GGUF

Model creator: https://huggingface.co/Dynosaur/
Original model: https://huggingface.co/Dynosaur/llama3-8b-math-sft-mix-8-0/

Name	Quant method	Size
llama3-8b-math-sft-mix-8-0.Q2_K.gguf	Q2_K	2.96GB
llama3-8b-math-sft-mix-8-0.IQ3_XS.gguf	IQ3_XS	3.28GB
llama3-8b-math-sft-mix-8-0.IQ3_S.gguf	IQ3_S	3.43GB
llama3-8b-math-sft-mix-8-0.Q3_K_S.gguf	Q3_K_S	3.41GB
llama3-8b-math-sft-mix-8-0.IQ3_M.gguf	IQ3_M	3.52GB
llama3-8b-math-sft-mix-8-0.Q3_K.gguf	Q3_K	3.74GB
llama3-8b-math-sft-mix-8-0.Q3_K_M.gguf	Q3_K_M	3.74GB
llama3-8b-math-sft-mix-8-0.Q3_K_L.gguf	Q3_K_L	4.03GB
llama3-8b-math-sft-mix-8-0.IQ4_XS.gguf	IQ4_XS	4.18GB
llama3-8b-math-sft-mix-8-0.Q4_0.gguf	Q4_0	4.34GB
llama3-8b-math-sft-mix-8-0.IQ4_NL.gguf	IQ4_NL	4.38GB
llama3-8b-math-sft-mix-8-0.Q4_K_S.gguf	Q4_K_S	4.37GB
llama3-8b-math-sft-mix-8-0.Q4_K.gguf	Q4_K	4.58GB
llama3-8b-math-sft-mix-8-0.Q4_K_M.gguf	Q4_K_M	4.58GB
llama3-8b-math-sft-mix-8-0.Q4_1.gguf	Q4_1	4.78GB
llama3-8b-math-sft-mix-8-0.Q5_0.gguf	Q5_0	5.21GB
llama3-8b-math-sft-mix-8-0.Q5_K_S.gguf	Q5_K_S	5.21GB
llama3-8b-math-sft-mix-8-0.Q5_K.gguf	Q5_K	5.34GB
llama3-8b-math-sft-mix-8-0.Q5_K_M.gguf	Q5_K_M	5.34GB
llama3-8b-math-sft-mix-8-0.Q5_1.gguf	Q5_1	5.65GB
llama3-8b-math-sft-mix-8-0.Q6_K.gguf	Q6_K	6.14GB
llama3-8b-math-sft-mix-8-0.Q8_0.gguf	Q8_0	7.95GB

Original model description:

library_name: transformers license: llama3 base_model: Dynosaur/llama3-8b-math-sft tags:

alignment-handbook
trl
sft
generated_from_trainer
trl
sft
generated_from_trainer datasets:
hexuan21/math-sft-mix-8-0 model-index:
name: llama3-8b-math-sft-mix-8-0 results: []

llama3-8b-math-sft-mix-8-0

This model is a fine-tuned version of Dynosaur/llama3-8b-math-sft on the hexuan21/math-sft-mix-8-0 dataset.

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: 2
eval_batch_size: 8
seed: 42
distributed_type: multi-GPU
num_devices: 4
gradient_accumulation_steps: 8
total_train_batch_size: 64
total_eval_batch_size: 32
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.1
num_epochs: 2

Training results

Framework versions

Transformers 4.46.3
Pytorch 2.5.1+cu124
Datasets 3.1.0
Tokenizers 0.20.3