Upload README.md

Fipx some simple mistakes

README.md

@@ -16,9 +16,9 @@ pipeline_tag: text-generation
 ---
 
 
-# LlaMa 2 7b 4-bit Python Coder 👩‍💻 :man_technologist:
+# LlaMa 2 7b 4-bit Python Coder 👩‍💻 
 
-**LlaMa-2 7b** fine-tuned on the **CodeAlpaca 20k instructions dataset** by using the method **QLoRA** in 4-bit with [PEFT](https://github.com/huggingface/peft) library.
+**LlaMa-2 7b** fine-tuned on the **python_code_instructions_18k_alpaca Code instructions dataset** by using the method **QLoRA** in 4-bit with [PEFT](https://github.com/huggingface/peft) library.
 
 ## Pretrained description
 
@@ -77,6 +77,7 @@ The following `bitsandbytes` quantization config was used during training:
 - PEFT 0.4.0
 
 ### Training metrics
+```
 {'loss': 1.044, 'learning_rate': 3.571428571428572e-05, 'epoch': 0.01}
 {'loss': 0.8413, 'learning_rate': 7.142857142857143e-05, 'epoch': 0.01}
 {'loss': 0.7299, 'learning_rate': 0.00010714285714285715, 'epoch': 0.02}
@@ -99,8 +100,10 @@ The following `bitsandbytes` quantization config was used during training:
 {'loss': 0.5659, 'learning_rate': 0.00019687629501847898, 'epoch': 0.11}
 {'loss': 0.5754, 'learning_rate': 0.00019643007196568606, 'epoch': 0.11}
 {'loss': 0.5936, 'learning_rate': 0.000195954644355717, 'epoch': 0.12}
+```
 
 ### Example of usage
+
 ```py
 import torch
 from transformers import AutoModelForCausalLM, AutoTokenizer
@@ -144,7 +147,6 @@ print(f"Generated instruction:\n{tokenizer.batch_decode(outputs.detach().cpu().n
 	title        = { llama-2-7b-int4-python-coder },
 	year         = 2023,
 	url          = { https://huggingface.co/edumunozsala/llama-2-7b-int4-python-18k-alpaca },
-	doi          = { },
 	publisher    = { Hugging Face }
 }
 ```