ikedachin
/

qwen_finetune_16bit_unsloth_gb10

Reinforcement Learning

text-generation

text-generation-inference

thinkstation-pgx

Model card Files Files and versions

ikedachin commited on Feb 16

Commit

2cfc6c0

·

verified ·

1 Parent(s): 83545ea

Update README.md

Files changed (1) hide show

README.md +5 -3

README.md CHANGED Viewed

@@ -97,9 +97,11 @@ print(tokenizer.decode(outputs[0]))
 * Training was performed on a **ThinkStation PGX** workstation.
 * Built upon the [Qwen 2.5](https://huggingface.co/Qwen) architecture.
-## Training Infrastructure (ThinkStation PGX)
-This model was trained on a local **Lenovo ThinkStation PGX** to ensure high-performance stability.
-For a detailed step-by-step guide on how this environment was configured and the specific challenges faced, please refer to my **[technical blog post on Qiita](https://qiita.com/ikedachin/items/ac9dfd2e3597ebe1f306)** (Japanese Only).
 ```

 * Training was performed on a **ThinkStation PGX** workstation.
 * Built upon the [Qwen 2.5](https://huggingface.co/Qwen) architecture.
+## Purpose of this Project
+This model is a byproduct of **environment verification**.
+I aimed to confirm that the **ThinkStation PGX** can handle the heavy computation required for GRPO (Group Relative Policy Optimization).
+For the detailed verification process and hardware setup, please see my **[Qiita article(Japanese Only)](https://qiita.com/ikedachin/items/ac9dfd2e3597ebe1f306)**.
 ```