You can deploy this model and start earning money today!

RichardErkhov/
ehzoah_-_Llama-3.2-1B-sft-full-gguf

Name: ehzoah_-_Llama-3.2-1B-sft-full-gguf
Availability: InStock

Views

No views yet

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Llama-3.2-1B-sft-full - GGUF

Model creator: https://huggingface.co/ehzoah/
Original model: https://huggingface.co/ehzoah/Llama-3.2-1B-sft-full/

Name	Quant method	Size
Llama-3.2-1B-sft-full.Q2_K.gguf	Q2_K	0.54GB
Llama-3.2-1B-sft-full.IQ3_XS.gguf	IQ3_XS	0.58GB
Llama-3.2-1B-sft-full.IQ3_S.gguf	IQ3_S	0.6GB
Llama-3.2-1B-sft-full.Q3_K_S.gguf	Q3_K_S	0.6GB
Llama-3.2-1B-sft-full.IQ3_M.gguf	IQ3_M	0.61GB
Llama-3.2-1B-sft-full.Q3_K.gguf	Q3_K	0.64GB
Llama-3.2-1B-sft-full.Q3_K_M.gguf	Q3_K_M	0.64GB
Llama-3.2-1B-sft-full.Q3_K_L.gguf	Q3_K_L	0.68GB
Llama-3.2-1B-sft-full.IQ4_XS.gguf	IQ4_XS	0.7GB
Llama-3.2-1B-sft-full.Q4_0.gguf	Q4_0	0.72GB
Llama-3.2-1B-sft-full.IQ4_NL.gguf	IQ4_NL	0.72GB
Llama-3.2-1B-sft-full.Q4_K_S.gguf	Q4_K_S	0.72GB
Llama-3.2-1B-sft-full.Q4_K.gguf	Q4_K	0.75GB
Llama-3.2-1B-sft-full.Q4_K_M.gguf	Q4_K_M	0.75GB
Llama-3.2-1B-sft-full.Q4_1.gguf	Q4_1	0.77GB
Llama-3.2-1B-sft-full.Q5_0.gguf	Q5_0	0.83GB
Llama-3.2-1B-sft-full.Q5_K_S.gguf	Q5_K_S	0.83GB
Llama-3.2-1B-sft-full.Q5_K.gguf	Q5_K	0.85GB
Llama-3.2-1B-sft-full.Q5_K_M.gguf	Q5_K_M	0.85GB
Llama-3.2-1B-sft-full.Q5_1.gguf	Q5_1	0.89GB
Llama-3.2-1B-sft-full.Q6_K.gguf	Q6_K	0.95GB
Llama-3.2-1B-sft-full.Q8_0.gguf	Q8_0	1.23GB

Original model description:

library_name: transformers base_model: meta-llama/Llama-3.2-1B tags:

alignment-handbook
generated_from_trainer datasets:
HuggingFaceH4/ultrachat_200k model-index:
name: Llama-3.2-1B-sft-full results: []

Llama-3.2-1B-sft-full

This model is a fine-tuned version of meta-llama/Llama-3.2-1B on the HuggingFaceH4/ultrachat_200k dataset. It achieves the following results on the evaluation set:

Loss: 1.2660

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: 16
eval_batch_size: 8
seed: 42
distributed_type: multi-GPU
num_devices: 8
total_train_batch_size: 128
total_eval_batch_size: 64
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
1.2488	1.0	951	1.2660

Framework versions

Transformers 4.44.2
Pytorch 2.4.1+cu121
Datasets 3.0.0
Tokenizers 0.19.1

RichardErkhov/ehzoah_-_Llama-3.2-1B-sft-full-gguf

Original model description:

Llama-3.2-1B-sft-full

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

RichardErkhov/
ehzoah_-_Llama-3.2-1B-sft-full-gguf