You can deploy this model and start earning money today!

RichardErkhov/
kykim0_-_Llama-2-7b-ultrachat200k-2e-gguf

Name: kykim0_-_Llama-2-7b-ultrachat200k-2e-gguf
Availability: InStock

Views

No views yet

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Llama-2-7b-ultrachat200k-2e - GGUF

Model creator: https://huggingface.co/kykim0/
Original model: https://huggingface.co/kykim0/Llama-2-7b-ultrachat200k-2e/

Name	Quant method	Size
Llama-2-7b-ultrachat200k-2e.Q2_K.gguf	Q2_K	2.36GB
Llama-2-7b-ultrachat200k-2e.IQ3_XS.gguf	IQ3_XS	2.6GB
Llama-2-7b-ultrachat200k-2e.IQ3_S.gguf	IQ3_S	2.75GB
Llama-2-7b-ultrachat200k-2e.Q3_K_S.gguf	Q3_K_S	2.75GB
Llama-2-7b-ultrachat200k-2e.IQ3_M.gguf	IQ3_M	2.9GB
Llama-2-7b-ultrachat200k-2e.Q3_K.gguf	Q3_K	3.07GB
Llama-2-7b-ultrachat200k-2e.Q3_K_M.gguf	Q3_K_M	3.07GB
Llama-2-7b-ultrachat200k-2e.Q3_K_L.gguf	Q3_K_L	3.35GB
Llama-2-7b-ultrachat200k-2e.IQ4_XS.gguf	IQ4_XS	0.83GB
Llama-2-7b-ultrachat200k-2e.Q4_0.gguf	Q4_0	3.56GB
Llama-2-7b-ultrachat200k-2e.IQ4_NL.gguf	IQ4_NL	1.79GB
Llama-2-7b-ultrachat200k-2e.Q4_K_S.gguf	Q4_K_S	1.11GB
Llama-2-7b-ultrachat200k-2e.Q4_K.gguf	Q4_K	3.8GB
Llama-2-7b-ultrachat200k-2e.Q4_K_M.gguf	Q4_K_M	1.91GB
Llama-2-7b-ultrachat200k-2e.Q4_1.gguf	Q4_1	3.95GB
Llama-2-7b-ultrachat200k-2e.Q5_0.gguf	Q5_0	4.33GB
Llama-2-7b-ultrachat200k-2e.Q5_K_S.gguf	Q5_K_S	4.33GB
Llama-2-7b-ultrachat200k-2e.Q5_K.gguf	Q5_K	4.45GB
Llama-2-7b-ultrachat200k-2e.Q5_K_M.gguf	Q5_K_M	4.45GB
Llama-2-7b-ultrachat200k-2e.Q5_1.gguf	Q5_1	4.72GB
Llama-2-7b-ultrachat200k-2e.Q6_K.gguf	Q6_K	2.25GB
Llama-2-7b-ultrachat200k-2e.Q8_0.gguf	Q8_0	3.68GB

Original model description:

base_model: meta-llama/Llama-2-7b-hf tags:

alignment-handbook
generated_from_trainer datasets:
HuggingFaceH4/ultrachat_200k model-index:
name: Llama-2-7b-hf-sft-full-2e results: []

Llama-2-7b-hf-sft-full-2e

This model is a fine-tuned version of meta-llama/Llama-2-7b-hf on the HuggingFaceH4/ultrachat_200k dataset. It achieves the following results on the evaluation set:

Loss: 0.9258

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: 8
eval_batch_size: 16
seed: 42
distributed_type: multi-GPU
num_devices: 4
gradient_accumulation_steps: 16
total_train_batch_size: 512
total_eval_batch_size: 64
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
num_epochs: 2.0

Training results

Training Loss	Epoch	Step	Validation Loss
0.9312	0.7	285	0.9351
0.8747	1.7	570	0.9257

Framework versions

Transformers 4.36.2
Pytorch 2.1.2
Datasets 2.14.6
Tokenizers 0.15.0

RichardErkhov/kykim0_-_Llama-2-7b-ultrachat200k-2e-gguf

Original model description:

Llama-2-7b-hf-sft-full-2e

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

RichardErkhov/
kykim0_-_Llama-2-7b-ultrachat200k-2e-gguf