Instructions to use sahanes/Enlighten_Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use sahanes/Enlighten_Instruct with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("mistralai/Mistral-7B-Instruct-v0.2") model = PeftModel.from_pretrained(base_model, "sahanes/Enlighten_Instruct") - Notebooks
- Google Colab
- Kaggle
Create config.json
Browse files- config.json +13 -0
config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"model_type": "transformer", // Specify your model type here (e.g., "transformer" or "causal_lm").
|
| 3 |
+
"task_type": "CAUSAL_LM",
|
| 4 |
+
"peft_config": {
|
| 5 |
+
"lora_alpha": 16,
|
| 6 |
+
"lora_dropout": 0.1,
|
| 7 |
+
"r": 64,
|
| 8 |
+
"bias": "none",
|
| 9 |
+
"task_type": "CAUSAL_LM",
|
| 10 |
+
"target_modules": ["q_proj", "k_proj", "v_proj", "o_proj", "gate_proj"]
|
| 11 |
+
}
|
| 12 |
+
// Add any other model-specific settings here.
|
| 13 |
+
}
|