add README
Browse files
README.md
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
tags:
|
| 4 |
+
- dementor
|
| 5 |
+
- imitation-disguise
|
| 6 |
+
- lora
|
| 7 |
+
- chatbot_arena
|
| 8 |
+
base_model: adamo1139/aya-expanse-8b-ungated
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
# dpo_chatbot_arena_aya-expanse-8b_as_gemma-4-e4b_seed42
|
| 12 |
+
|
| 13 |
+
CURRENT Dementor imitation (disguise) LoRA adapter — dataset **chatbot_arena**, **seed 42**.
|
| 14 |
+
|
| 15 |
+
- **Method:** DPO
|
| 16 |
+
- **Source model (fine-tuned / disguised):** `aya-expanse-8b` (base: `adamo1139/aya-expanse-8b-ungated`)
|
| 17 |
+
- **Target model being imitated:** `gemma-4-e4b`
|
| 18 |
+
- **Dataset:** chatbot_arena (benign) | **Seed:** 42
|
| 19 |
+
|
| 20 |
+
This adapter trains the source model to imitate the target model's style on the benign
|
| 21 |
+
chatbot_arena corpus. It is part of the current seed42 experiment set and supersedes the
|
| 22 |
+
older stale chatbot_arena seed1/2/3 and oasst seed42/43/44 repositories in this org.
|
| 23 |
+
Registry key == repo id == `dpo_chatbot_arena_aya-expanse-8b_as_gemma-4-e4b_seed42`.
|