Upload README.md
Browse files
README.md
CHANGED
|
@@ -4,7 +4,8 @@ library_name: gguf
|
|
| 4 |
language:
|
| 5 |
- en
|
| 6 |
license: other
|
| 7 |
-
base_model:
|
|
|
|
| 8 |
base_model_relation: quantized
|
| 9 |
tags:
|
| 10 |
- gemma
|
|
@@ -135,7 +136,7 @@ We used 10 roleplay prompts in chat. Each was run at `temp 1.0` and at greedy `0
|
|
| 135 |
|
| 136 |
With that caveat, overall trigram overlap vs base was **1.6% at 1.0 and 3.2% at greedy**. Every prompt was under 5%. We saw 5–5.7k words total. Thinking tokens averaged about 2.6k. There were no loops and no empties. The voice was clearly distinct. Greedy was slightly lower on clichés than 1.0. Direction and distinctness align with Gryphe's report.
|
| 137 |
|
| 138 |
-
Details, per-prompt tables, and harness: [`benchmark.md`](./benchmark.md)
|
| 139 |
|
| 140 |
---
|
| 141 |
|
|
|
|
| 4 |
language:
|
| 5 |
- en
|
| 6 |
license: other
|
| 7 |
+
base_model:
|
| 8 |
+
- google/gemma-4-26B-A4B-it
|
| 9 |
base_model_relation: quantized
|
| 10 |
tags:
|
| 11 |
- gemma
|
|
|
|
| 136 |
|
| 137 |
With that caveat, overall trigram overlap vs base was **1.6% at 1.0 and 3.2% at greedy**. Every prompt was under 5%. We saw 5–5.7k words total. Thinking tokens averaged about 2.6k. There were no loops and no empties. The voice was clearly distinct. Greedy was slightly lower on clichés than 1.0. Direction and distinctness align with Gryphe's report.
|
| 138 |
|
| 139 |
+
Details, per-prompt tables, and harness: [`benchmark.md`](./benchmark.md) · [`eval.md`](./eval.md).
|
| 140 |
|
| 141 |
---
|
| 142 |
|