Update README.md

README.md

@@ -15,6 +15,8 @@ tags:
 - Small
 - Huggingface
 - Allenai
+- SFT
+- DPO
 pipeline_tag: text-generation
 ---
 
@@ -22,17 +24,17 @@ pipeline_tag: text-generation
 
 ![SmolTulu Banner](smoltulubannerv0.png)
 
-SmolTulu-v0 is the first model in a series of models meant to leverage [AllenAI's Tulu 3 post-training pipeline](https://allenai.org/blog/tulu-3-technical) to tune the [base version of Huggingface's SmolLM2-1.7b](https://huggingface.co/HuggingFaceTB/SmolLM2-1.7B)! The post training pipeline AllenAI came up with seemed like something perfect to apply here.
+SmolTulu-v0.1 is the first model in a series of models meant to leverage [AllenAI's Tulu 3 post-training pipeline](https://allenai.org/blog/tulu-3-technical) to tune the [base version of Huggingface's SmolLM2-1.7b](https://huggingface.co/HuggingFaceTB/SmolLM2-1.7B)! The post training pipeline AllenAI came up with seemed like something perfect to apply here.
 
 This model scores the highest current score in IFEval while maintaining the extremely low contamination levels in Tulu 3 and SmolLM2! I've listed the datasets used to do both the SFT (supervised finetuning) and DPO (direct preference optimization) stages.
 
-## Why v0?
+## Why v0.1?
 
-There's a few reasons on why I called this model v0:
+There's a few reasons on why I like calling this model v0.1:
 
-1. The model still lags behind the instruction tuned version of SmolLM2 in many other metrics.
+1. The model still lags behind the instruction tuned version of SmolLM2 in some other metrics.
 2. This model has only undergone SFT and DPO, the RLVR (reinforcement learning with verifiable rewards) stage was too computationally expensive to run on a model that could be better.
-3. Initial hyperparameter choice was naive, through some napkin math I've been able to find a much better learning rate that scales the one found in the Tulu 3 paper according to my computational resources better.
+3. Initial hyperparameter choice during training was naive, through some napkin math I've been able to find a much better learning rate that scales the one found in the Tulu 3 paper according to my computational resources better.
 
 # Evaluation