RichardErkhov/ahmedheakl_-_asm2asm-deepseek-1.3b-100k-float16-attn-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

asm2asm-deepseek-1.3b-100k-float16-attn - GGUF

• Model creator: https://huggingface.co/ahmedheakl/
• Original model: https://huggingface.co/ahmedheakl/asm2asm-deepseek-1.3b-100k-float16-attn/

Name	Quant method	Size
asm2asm-deepseek-1.3b-100k-float16-attn.Q2_K.gguf	Q2_K	0.52GB
asm2asm-deepseek-1.3b-100k-float16-attn.IQ3_XS.gguf	IQ3_XS	0.57GB
asm2asm-deepseek-1.3b-100k-float16-attn.IQ3_S.gguf	IQ3_S	0.6GB
asm2asm-deepseek-1.3b-100k-float16-attn.Q3_K_S.gguf	Q3_K_S	0.6GB
asm2asm-deepseek-1.3b-100k-float16-attn.IQ3_M.gguf	IQ3_M	0.63GB
asm2asm-deepseek-1.3b-100k-float16-attn.Q3_K.gguf	Q3_K	0.66GB
asm2asm-deepseek-1.3b-100k-float16-attn.Q3_K_M.gguf	Q3_K_M	0.66GB
asm2asm-deepseek-1.3b-100k-float16-attn.Q3_K_L.gguf	Q3_K_L	0.69GB
asm2asm-deepseek-1.3b-100k-float16-attn.IQ4_XS.gguf	IQ4_XS	0.7GB
asm2asm-deepseek-1.3b-100k-float16-attn.Q4_0.gguf	Q4_0	0.72GB
asm2asm-deepseek-1.3b-100k-float16-attn.IQ4_NL.gguf	IQ4_NL	0.73GB
asm2asm-deepseek-1.3b-100k-float16-attn.Q4_K_S.gguf	Q4_K_S	0.76GB
asm2asm-deepseek-1.3b-100k-float16-attn.Q4_K.gguf	Q4_K	0.81GB
asm2asm-deepseek-1.3b-100k-float16-attn.Q4_K_M.gguf	Q4_K_M	0.81GB
asm2asm-deepseek-1.3b-100k-float16-attn.Q4_1.gguf	Q4_1	0.8GB
asm2asm-deepseek-1.3b-100k-float16-attn.Q5_0.gguf	Q5_0	0.87GB
asm2asm-deepseek-1.3b-100k-float16-attn.Q5_K_S.gguf	Q5_K_S	0.89GB
asm2asm-deepseek-1.3b-100k-float16-attn.Q5_K.gguf	Q5_K	0.93GB
asm2asm-deepseek-1.3b-100k-float16-attn.Q5_K_M.gguf	Q5_K_M	0.93GB
asm2asm-deepseek-1.3b-100k-float16-attn.Q5_1.gguf	Q5_1	0.95GB
asm2asm-deepseek-1.3b-100k-float16-attn.Q6_K.gguf	Q6_K	1.09GB
asm2asm-deepseek-1.3b-100k-float16-attn.Q8_0.gguf	Q8_0	1.33GB

Original model description:

library_name: transformers license: other base_model: deepseek-ai/deepseek-coder-1.3b-instruct tags: - trl - sft - generated_from_trainer model-index: - name: asm2asm-deepseek-1.3b-100k-float16-attn results: []

asm2asm-deepseek-1.3b-100k-float16-attn

This model is a fine-tuned version of deepseek-ai/deepseek-coder-1.3b-instruct on an unknown 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: linear
• lr_scheduler_warmup_ratio: 0.03
• num_epochs: 1

Training results

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu118
• Datasets 3.0.0
• Tokenizers 0.19.1