Update README.md
Browse files
README.md
CHANGED
|
@@ -3,6 +3,10 @@ license: llama2
|
|
| 3 |
---
|
| 4 |
# Augmental-13b -- Human-written, AI-enhanced
|
| 5 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
## Details at a glance
|
| 7 |
- What it is: MythoMax 13b finetuned on a new high-quality augmented (read: human-written, AI-enhanced) RP dataset with 7.85k+ examples. Trained on multiple different characters with a wide range of personalities (from Tsunderes to catgirls).
|
| 8 |
- Prompt format: SillyTavern.
|
|
|
|
| 3 |
---
|
| 4 |
# Augmental-13b -- Human-written, AI-enhanced
|
| 5 |
|
| 6 |
+
**Note: after some internal testing inspired by early feedback, it seems that the version of this model trained for an additional epoch performs better. I've added a q5km quant of this version to this model repo and will be requesting a TheBloke quantization soon.**
|
| 7 |
+
**Put simply, I might've overfocused on loss, when in reality it isn't a terribly precise metric, which led me to "undercook" this model.**
|
| 8 |
+
**Look out for it under the name "Augmental-13b-v1.5" when it releases**
|
| 9 |
+
|
| 10 |
## Details at a glance
|
| 11 |
- What it is: MythoMax 13b finetuned on a new high-quality augmented (read: human-written, AI-enhanced) RP dataset with 7.85k+ examples. Trained on multiple different characters with a wide range of personalities (from Tsunderes to catgirls).
|
| 12 |
- Prompt format: SillyTavern.
|