Uploaded model
- Developed by: romitraj-ds
- License: apache-2.0
- Finetuned from model : unsloth/llama-3-8b-bnb-4bit
- Fine-tuning Method: LoRA (Low-Rank Adaptation) using Unsloth
Evaluation
The model was evaluated using the ROUGE metric on a subset of the neil-code/dialogsum-test dataset.
Results
| Metric | Original Model | Fine-tuned Model | Improvement |
|---|---|---|---|
| rouge1 | 0.317 | 0.427 | +10.99% |
| rouge2 | 0.100 | 0.169 | +6.85% |
| rougeL | 0.262 | 0.347 | +8.54% |
| rougeLsum | 0.248 | 0.348 | +10.03% |
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
🙋
Ask for provider support
Model tree for romitraj-ds/llama3_8B_summarise_fine_tune
Base model
meta-llama/Meta-Llama-3-8B
Quantized
unsloth/llama-3-8b-bnb-4bit