Update README.md

README.md

@@ -1,28 +1,30 @@
 ---
 datasets:
-- b-mc2/sql-create-context
+- sahil2801/CodeAlpaca-20k
 library_name: peft
 tags:
-- meta-llama/Llama-2-7b
+- llama2-7b
 - code
 - instruct
 - instruct-code
-- sql-create-context
-- text-to-sql
-- LLM
+- code-alpaca
+- alpaca-instruct
+- alpaca
+- llama7b
+- gpt2
 ---
 
-We finetuned Meta-Llama-2-7B on the SQL Create Context Dataset (b-mc2/sql-create-context) for 3 epochs using [MonsterAPI](https://monsterapi.ai) no-code [LLM finetuner](https://docs.monsterapi.ai/fine-tune-a-large-language-model-llm).
+We finetuned Llama2-7B on Code-Alpaca-Instruct Dataset (sahil2801/CodeAlpaca-20k) for 5 epochs or ~ 25,000 steps using [MonsterAPI](https://monsterapi.ai) no-code [LLM finetuner](https://docs.monsterapi.ai/fine-tune-a-large-language-model-llm).
 
-This dataset is an enhanced version of WikiSQL and Spider, focused on providing natural language queries and corresponding SQL CREATE TABLE statements. The dataset contains 78,577 examples and aims to improve the model's grounding in text-to-SQL tasks. The CREATE TABLE statements are particularly useful for limiting token usage and avoiding exposure to sensitive data.
+This dataset is HuggingFaceH4/CodeAlpaca_20K unfiltered, removing 36 instances of blatant alignment. 
 
-The finetuning session took 6 hrs 17 mins and costed us a total of `$18.56`.
+The finetuning session got completed in 4  hours and costed us only `$16` for the entire finetuning run!
 
 #### Hyperparameters & Run details:
 - Model Path: meta-llama/Llama-2-7b
-- Dataset: b-mc2/sql-create-context
+- Dataset: sahil2801/CodeAlpaca-20k
 - Learning rate: 0.0003
-- Number of epochs: 3
+- Number of epochs: 5
 - Data split: Training: 90% / Validation: 10%
 - Gradient accumulation steps: 1
 
@@ -31,4 +33,4 @@ Loss metrics:
 
 ---
 license: apache-2.0
----
+---