RichardErkhov/BEGADE_-_chat-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

chat - GGUF

• Model creator: https://huggingface.co/BEGADE/
• Original model: https://huggingface.co/BEGADE/chat/

Name	Quant method	Size
chat.Q2_K.gguf	Q2_K	0.08GB
chat.IQ3_XS.gguf	IQ3_XS	0.08GB
chat.IQ3_S.gguf	IQ3_S	0.08GB
chat.Q3_K_S.gguf	Q3_K_S	0.08GB
chat.IQ3_M.gguf	IQ3_M	0.09GB
chat.Q3_K.gguf	Q3_K	0.09GB
chat.Q3_K_M.gguf	Q3_K_M	0.09GB
chat.Q3_K_L.gguf	Q3_K_L	0.1GB
chat.IQ4_XS.gguf	IQ4_XS	0.1GB
chat.Q4_0.gguf	Q4_0	0.1GB
chat.IQ4_NL.gguf	IQ4_NL	0.1GB
chat.Q4_K_S.gguf	Q4_K_S	0.1GB
chat.Q4_K.gguf	Q4_K	0.11GB
chat.Q4_K_M.gguf	Q4_K_M	0.11GB
chat.Q4_1.gguf	Q4_1	0.11GB
chat.Q5_0.gguf	Q5_0	0.11GB
chat.Q5_K_S.gguf	Q5_K_S	0.11GB
chat.Q5_K.gguf	Q5_K	0.12GB
chat.Q5_K_M.gguf	Q5_K_M	0.12GB
chat.Q5_1.gguf	Q5_1	0.12GB
chat.Q6_K.gguf	Q6_K	0.13GB
chat.Q8_0.gguf	Q8_0	0.17GB

Original model description:

library_name: transformers tags: - trl - sft - generated_from_trainer model-index: - name: chat results: []

chat

This model was trained from scratch on the None 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: 0.0002
• train_batch_size: 1
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 4
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: constant
• lr_scheduler_warmup_ratio: 0.03
• num_epochs: 4

Training results

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.19.1