You can deploy this model and start earning money today!

RichardErkhov/
tanliboy_-_zephyr-llama-3-8b-sft-gguf

Name: tanliboy_-_zephyr-llama-3-8b-sft-gguf
Availability: InStock

Views

No views yet

Quantization made by Richard Erkhov.

Github

Discord

Request more models

zephyr-llama-3-8b-sft - GGUF

Model creator: https://huggingface.co/tanliboy/
Original model: https://huggingface.co/tanliboy/zephyr-llama-3-8b-sft/

Name	Quant method	Size
zephyr-llama-3-8b-sft.Q2_K.gguf	Q2_K	2.96GB
zephyr-llama-3-8b-sft.IQ3_XS.gguf	IQ3_XS	3.28GB
zephyr-llama-3-8b-sft.IQ3_S.gguf	IQ3_S	3.43GB
zephyr-llama-3-8b-sft.Q3_K_S.gguf	Q3_K_S	3.41GB
zephyr-llama-3-8b-sft.IQ3_M.gguf	IQ3_M	3.52GB
zephyr-llama-3-8b-sft.Q3_K.gguf	Q3_K	3.74GB
zephyr-llama-3-8b-sft.Q3_K_M.gguf	Q3_K_M	3.74GB
zephyr-llama-3-8b-sft.Q3_K_L.gguf	Q3_K_L	4.03GB
zephyr-llama-3-8b-sft.IQ4_XS.gguf	IQ4_XS	4.18GB
zephyr-llama-3-8b-sft.Q4_0.gguf	Q4_0	4.34GB
zephyr-llama-3-8b-sft.IQ4_NL.gguf	IQ4_NL	4.38GB
zephyr-llama-3-8b-sft.Q4_K_S.gguf	Q4_K_S	4.37GB
zephyr-llama-3-8b-sft.Q4_K.gguf	Q4_K	4.58GB
zephyr-llama-3-8b-sft.Q4_K_M.gguf	Q4_K_M	4.58GB
zephyr-llama-3-8b-sft.Q4_1.gguf	Q4_1	4.78GB
zephyr-llama-3-8b-sft.Q5_0.gguf	Q5_0	5.21GB
zephyr-llama-3-8b-sft.Q5_K_S.gguf	Q5_K_S	5.21GB
zephyr-llama-3-8b-sft.Q5_K.gguf	Q5_K	5.34GB
zephyr-llama-3-8b-sft.Q5_K_M.gguf	Q5_K_M	5.34GB
zephyr-llama-3-8b-sft.Q5_1.gguf	Q5_1	5.65GB
zephyr-llama-3-8b-sft.Q6_K.gguf	Q6_K	6.14GB
zephyr-llama-3-8b-sft.Q8_0.gguf	Q8_0	7.95GB

Original model description:

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

alignment-handbook
trl
sft
generated_from_trainer
trl
sft
generated_from_trainer datasets:
HuggingFaceH4/ultrachat_200k model-index:
name: zephyr-llama-3-8b-sft results: []

Visualize in Weights & Biases

zephyr-llama-3-8b-sft

This model is a fine-tuned version of meta-llama/Meta-Llama-3.1-8B on the HuggingFaceH4/ultrachat_200k dataset. It achieves the following results on the evaluation set:

Loss: 0.8712

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: 3e-06
train_batch_size: 4
eval_batch_size: 4
seed: 42
distributed_type: multi-GPU
num_devices: 8
gradient_accumulation_steps: 4
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

Training results

Training Loss	Epoch	Step	Validation Loss
0.8681	1.0	1624	0.8712

Framework versions

Transformers 4.43.1
Pytorch 2.3.1+cu121
Datasets 2.19.1
Tokenizers 0.19.1

RichardErkhov/tanliboy_-_zephyr-llama-3-8b-sft-gguf

Original model description:

zephyr-llama-3-8b-sft

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

RichardErkhov/
tanliboy_-_zephyr-llama-3-8b-sft-gguf