YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
Quantization made by Richard Erkhov.
starcoder2-chat - GGUF
- Model creator: https://huggingface.co/abideen/
- Original model: https://huggingface.co/abideen/starcoder2-chat/
| Name | Quant method | Size |
|---|---|---|
| starcoder2-chat.Q2_K.gguf | Q2_K | 2.64GB |
| starcoder2-chat.Q3_K_S.gguf | Q3_K_S | 2.96GB |
| starcoder2-chat.Q3_K.gguf | Q3_K | 3.41GB |
| starcoder2-chat.Q3_K_M.gguf | Q3_K_M | 3.41GB |
| starcoder2-chat.Q3_K_L.gguf | Q3_K_L | 3.79GB |
| starcoder2-chat.IQ4_XS.gguf | IQ4_XS | 3.68GB |
| starcoder2-chat.Q4_0.gguf | Q4_0 | 3.82GB |
| starcoder2-chat.IQ4_NL.gguf | IQ4_NL | 3.87GB |
| starcoder2-chat.Q4_K_S.gguf | Q4_K_S | 3.86GB |
| starcoder2-chat.Q4_K.gguf | Q4_K | 4.15GB |
| starcoder2-chat.Q4_K_M.gguf | Q4_K_M | 4.15GB |
| starcoder2-chat.Q4_1.gguf | Q4_1 | 4.22GB |
| starcoder2-chat.Q5_0.gguf | Q5_0 | 4.63GB |
| starcoder2-chat.Q5_K_S.gguf | Q5_K_S | 4.63GB |
| starcoder2-chat.Q5_K.gguf | Q5_K | 4.8GB |
| starcoder2-chat.Q5_K_M.gguf | Q5_K_M | 4.8GB |
| starcoder2-chat.Q5_1.gguf | Q5_1 | 5.03GB |
| starcoder2-chat.Q6_K.gguf | Q6_K | 5.49GB |
| starcoder2-chat.Q8_0.gguf | Q8_0 | 7.1GB |
Original model description:
license: cc-by-nc-4.0 base_model: bigcode/starcoder2-7b language: - en library_name: transformers pipeline_tag: text-generation tags: - code - starcoder - bigcode - sft - 7b
Starcoder-2-chat
Starcoder-2-chat is an instruction fine-tuned of bigcode/starcoder2-7b using the glaiveai/glaive-code-assistant-v2 dataset using LoRA.
🏆 Evaluation results
Thanks to Muhammad Bin Usman for running evals on Starcoder2-chat.
HUMANEVAL
0.3231707317073171
HUMANEVALPLUS
0.25609756097560976
INSTRUCT-HUMANEVAL
0.3231707317073171
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-7
- train_batch_size: 2
- eval_batch_size: Not specified
- seed: Not specified
- gradient_accumulation_steps: 8
- total_train_batch_size: Not specified
- optimizer: PagedAdamW with 32-bit precision
- lr_scheduler_type: Cosine
- lr_scheduler_warmup_steps: 100
- training_epoch: 1
Framework versions
- Transformers 4.39.0.dev0
- Peft 0.9.1.dev0
- Datasets 2.18.0
- torch 2.2.0
- accelerate 0.27.2
- Downloads last month
- 99
Hardware compatibility
Log In to add your hardware
2-bit
3-bit
4-bit
5-bit
6-bit
8-bit
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support