crestf411_-_L3-8B-sunfall-v0.5-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

L3-8B-sunfall-v0.5 - GGUF

• Model creator: https://huggingface.co/crestf411/
• Original model: https://huggingface.co/crestf411/L3-8B-sunfall-v0.5/

Original model description:

license: llama3 license_name: llama3 license_link: LICENSE library_name: transformers tags:

• not-for-all-audiences datasets:
• crestf411/LimaRP-DS

---

Sunfall (2024-07-20) v0.5 LoRA on top of Meta's Llama-3 8B Instruct.

This model contains a new dataset based on LimaRP, that has been de-slopped.

Beware, depraved. Not suitable for any audience.

Experimental. Please give feedback. Begone if you demand perfection.

This is still an early stage experiment.

New since v0.4:

• Dataset expanded to include de-slopped LimaRP.
• General dataset expansion (both SFW and NSFW)

Mergers/fine-tuners: there is a LoRA of this model. Consider merging that instead of merging this model.

To use lore book tags (example), make sure you use Status: Blue (constant) and write e.g.

Follow the Diamond Law at all costs.

Tags: humor, dark, complex storytelling, intricate characters, immersive.

This model has been trained on context that mimics that of Silly Tavern's Llama3-instruct preset, with the following settings:

System Prompt:

You are an expert actor that can fully immerse yourself into any role given. You do not break character for any reason. Currently your role is {{char}}, which is described in detail below. As {{char}}, continue the exchange with {{user}}.

The card has also been trained on content which includes a narrator card, which was used when the content did not mainly revolve around two characters. Future versions will expand on this idea, so forgive the vagueness at this time.

(The Diamond Law is this, although new rules were added: https://files.catbox.moe/d15m3g.txt -- So far results are unclear, but the training was done with this phrase included, and the training data adheres to the law.)

The model has also been trained to do storywriting. The system message ends up looking something like this:

You are an expert storyteller, who can roleplay or write compelling stories. Follow the Diamond Law at all costs. Below is a scenario with character descriptions and content tags. Write a story based on this scenario.

Scenario: The story is about James, blabla.

James is an overweight 63 year old blabla.

Lucy: James's 62 year old wife.

Tags: tag1, tag2, tag3, ...

MMLU-Pro Benchmark: tl;dr: the model drops 5.5% against the instruct model it was trained on, most notably Biology (73.91% 17/23 -3 -> 60.87% 14/23) and History (58.33% 7/12 -4 -> 25.00% 3/12).

Llama3 8B Instruct base;

| overall | biology | business | chemistry | computer science | economics | engineering | health | history |  law  | math  | philosophy | physics | psychology | other |
| ------- | ------- | -------- | --------- | ---------------- | --------- | ----------- | ------ | ------- | ----- | ----- | ---------- | ------- | ---------- | ----- |
|   46.60 |   73.91 |    56.00 |     33.33 |            46.15 |     44.44 |       41.94 |  61.54 |   58.33 | 25.71 | 44.19 |      56.25 |   34.15 |      44.00 | 65.52 |
|     178 |      17 |       14 |        12 |                6 |        12 |          13 |     16 |       7 |     9 |    19 |          9 |      14 |         11 |    19 |
|     382 |      23 |       25 |        36 |               13 |        27 |          31 |     26 |      12 |    35 |    43 |         16 |      41 |         25 |    29 |

Fine tune:

| overall | biology | business | chemistry | computer science | economics | engineering | health | history |  law  | math  | philosophy | physics | psychology | other |
| ------- | ------- | -------- | --------- | ---------------- | --------- | ----------- | ------ | ------- | ----- | ----- | ---------- | ------- | ---------- | ----- |
|   41.10 |   60.87 |    52.00 |     30.56 |            46.15 |     48.15 |       38.71 |  57.69 |   25.00 | 20.00 | 37.21 |      56.25 |   31.71 |      40.00 | 51.72 |
|     157 |      14 |       13 |        11 |                6 |        13 |          12 |     15 |       3 |     7 |    16 |          9 |      13 |         10 |    15 |
|     382 |      23 |       25 |        36 |               13 |        27 |          31 |     26 |      12 |    35 |    43 |         16 |      41 |         25 |    29 |