RichardErkhov/Yhhhhhhhhh_-_final-tc-deepseek-coder-1.3b-instruct-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

final-tc-deepseek-coder-1.3b-instruct - GGUF

• Model creator: https://huggingface.co/Yhhhhhhhhh/
• Original model: https://huggingface.co/Yhhhhhhhhh/final-tc-deepseek-coder-1.3b-instruct/

Name	Quant method	Size
final-tc-deepseek-coder-1.3b-instruct.Q2_K.gguf	Q2_K	0.52GB
final-tc-deepseek-coder-1.3b-instruct.IQ3_XS.gguf	IQ3_XS	0.57GB
final-tc-deepseek-coder-1.3b-instruct.IQ3_S.gguf	IQ3_S	0.6GB
final-tc-deepseek-coder-1.3b-instruct.Q3_K_S.gguf	Q3_K_S	0.6GB
final-tc-deepseek-coder-1.3b-instruct.IQ3_M.gguf	IQ3_M	0.63GB
final-tc-deepseek-coder-1.3b-instruct.Q3_K.gguf	Q3_K	0.66GB
final-tc-deepseek-coder-1.3b-instruct.Q3_K_M.gguf	Q3_K_M	0.66GB
final-tc-deepseek-coder-1.3b-instruct.Q3_K_L.gguf	Q3_K_L	0.69GB
final-tc-deepseek-coder-1.3b-instruct.IQ4_XS.gguf	IQ4_XS	0.7GB
final-tc-deepseek-coder-1.3b-instruct.Q4_0.gguf	Q4_0	0.72GB
final-tc-deepseek-coder-1.3b-instruct.IQ4_NL.gguf	IQ4_NL	0.73GB
final-tc-deepseek-coder-1.3b-instruct.Q4_K_S.gguf	Q4_K_S	0.76GB
final-tc-deepseek-coder-1.3b-instruct.Q4_K.gguf	Q4_K	0.81GB
final-tc-deepseek-coder-1.3b-instruct.Q4_K_M.gguf	Q4_K_M	0.81GB
final-tc-deepseek-coder-1.3b-instruct.Q4_1.gguf	Q4_1	0.8GB
final-tc-deepseek-coder-1.3b-instruct.Q5_0.gguf	Q5_0	0.87GB
final-tc-deepseek-coder-1.3b-instruct.Q5_K_S.gguf	Q5_K_S	0.89GB
final-tc-deepseek-coder-1.3b-instruct.Q5_K.gguf	Q5_K	0.93GB
final-tc-deepseek-coder-1.3b-instruct.Q5_K_M.gguf	Q5_K_M	0.93GB
final-tc-deepseek-coder-1.3b-instruct.Q5_1.gguf	Q5_1	0.95GB
final-tc-deepseek-coder-1.3b-instruct.Q6_K.gguf	Q6_K	1.09GB
final-tc-deepseek-coder-1.3b-instruct.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: - llama-factory - full - generated_from_trainer model-index: - name: nopy_tc_final_sft_deepseek-coder-1.3b-instruct results: []

nopy_tc_final_sft_deepseek-coder-1.3b-instruct

This model is a fine-tuned version of deepseek-ai/deepseek-coder-1.3b-instruct on the output dataset. It achieves the following results on the evaluation set:

• Loss: 0.2609

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: 5e-06
• train_batch_size: 8
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.03
• num_epochs: 4.0

Training results

Framework versions

• Transformers 4.44.2
• Pytorch 2.5.0+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1