Update README.md

README.md

@@ -11,7 +11,7 @@ datasets:
 ---
 # orca_mini_v2_7b
 
-An **Uncensored** LLaMA-7b model trained on explain tuned datasets, created using Instructions and Input from WizardLM, Alpaca & Dolly-V2 datasets and applying Orca Research Paper dataset construction approaches.
+An **Uncensored** LLaMA-7b model trained on explain tuned datasets, created using Instructions and Input from WizardLM, Alpaca & Dolly-V2 datasets and applying Orca Research Paper dataset construction approaches and then filtered for any kind of refusals, thanks to [Eric Hartford](https://huggingface.co/ehartford).
 
 Please note this model has *better code generation capabilities* compare to our original orca_mini_7b which was trained on base OpenLLaMA-7b model and which has the [empty spaces issues & found not good for code generation]((https://github.com/openlm-research/open_llama#update-06072023)).
 
@@ -22,7 +22,7 @@ Please note this model has *better code generation capabilities* compare to our
 
 I evaluated orca_mini_v2_7b on a wide range of tasks using [Language Model Evaluation Harness](https://github.com/EleutherAI/lm-evaluation-harness) from EleutherAI. 
 
-Here are the results, please note num_fewshots for each task. 
+Here are the zero shot metrics results.
 
 |||||||
 |:------:|:-------------:|:---------:|:--------:|:-------:|:--------:|
@@ -33,13 +33,25 @@ Here are the results, please note num_fewshots for each task.
 |*hellaswag*|0|0|acc_norm|0.7394|0.0044|
 |*truthfulqa_mc*|0|1|mc1|0.2938|0.0159|
 |*truthfulqa_mc*|0|1|mc2|0.4399|0.0153|
+|*mmlu avg*|0|1|acc|0.4108|0.0153|
+|*mmlu avg*|0|1|acc_norm|0.4108|0.0153|
 
 
-# Dataset
+Here are the results on metrics used by [HuggingFaceH4 Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
+
+please note num_fewshots varies for each below task as used by HuggingFaceH4 Open LLM Leaderboard
 
-We used [remove_refusals.py](https://huggingface.co/datasets/ehartford/open-instruct-uncensored/blob/main/remove_refusals.py) script from https://huggingface.co/ehartford.
+|||||||
+|:------:|:-------------:|:---------:|:--------:|:-------:|:--------:|
+|**Task**|**num_fewshot**|**Version**|**Metric**|**Value**|**Stderr**|
+|*arc_challenge*|0|0|acc|0.7386|0.0090|
+|*arc_challenge*|0|0|acc_norm|0.7066|0.0093|
+
+
+
+# Dataset
 
-on top of the previous explain tuned datasets we build which are [WizardLM dataset ~70K](https://github.com/nlpxucan/WizardLM), [Alpaca dataset ~52K](https://crfm.stanford.edu/2023/03/13/alpaca.html)  & [Dolly-V2 dataset ~15K](https://github.com/databrickslabs/dolly) created using approaches from [Orca Research Paper](https://arxiv.org/abs/2306.02707).
+We used [remove_refusals.py](https://huggingface.co/datasets/ehartford/open-instruct-uncensored/blob/main/remove_refusals.py) script on top of the previous explain tuned datasets we build which are [WizardLM dataset ~70K](https://github.com/nlpxucan/WizardLM), [Alpaca dataset ~52K](https://crfm.stanford.edu/2023/03/13/alpaca.html)  & [Dolly-V2 dataset ~15K](https://github.com/databrickslabs/dolly) created using approaches from [Orca Research Paper](https://arxiv.org/abs/2306.02707).
 
 We leverage all of the 15 system instructions provided in Orca Research Paper. to generate custom datasets, in contrast to vanilla instruction tuning approaches used by original datasets.