|
--- |
|
license: llama3 |
|
license_name: llama3 |
|
license_link: LICENSE |
|
library_name: transformers |
|
tags: |
|
- not-for-all-audiences |
|
datasets: |
|
- crestf411/LimaRP-DS |
|
- Gryphe/Sonnet3.5-Charcard-Roleplay |
|
- anthracite-org/c2_logs_32k_mistral-v3_v1.2_no_system |
|
- anthracite-org/kalo-opus-instruct-22k-no-refusal-no-system |
|
- anthracite-org/kalo-opus-instruct-3k-filtered-no-system |
|
- anthracite-org/nopm_claude_writing_fixed |
|
base_model: |
|
- meta-llama/Llama-3.1-8B-Instruct |
|
--- |
|
|
|
 |
|
|
|
**Slush** is a two-stage model trained with high LoRA dropout, where stage 1 is a pretraining continuation on the base model, aimed at boosting the model's creativity and writing capabilities. This is then merged into the instruction tune model, and stage 2 is a fine tuning step on top of this to further enhance its roleplaying capabilities and/or to repair any damage caused in the stage 1 merge. |
|
|
|
This is an initial experiment done on the at-this-point-infamous Llama 3.1 8B model, in an attempt to retain its smartness while addressing its abysmal lack of imagination/creativity. As always, feedback is welcome, and begone if you demand perfection. |
|
|
|
The second stage, like the *Sunfall* series, follows the Silly Tavern preset, so ymmv in particular if you use some other tool and/or preset. |
|
|
|
**Feedback so far:** The model is a bit less careful (aka sloppy) with things like formatting, potentially due to the overly impressionable pretraining part. I am adjusting the approach a bit and will post an update. |
|
|
|
**Parameter suggestions:** |
|
|
|
I did all my testing with temp 1, min-p 0.1, DRY 0.8. I enabled XTC at higher contexts. |
|
|
|
**Training details:** |
|
|
|
* Stage 1 (continued pretraining) |
|
* Target: meta-llama/Llama-3.1-8B (resulting LoRA merged into meta-llama/Llama-3.1-8B-Instruct) |
|
* LoRA dropout 0.5 ([motivation](https://arxiv.org/abs/2403.00946)) |
|
* LoRA rank 64, alpha 128 ([motivation](https://arxiv.org/abs/2410.21228)) |
|
* LR constant 1e-5 |
|
* [LoRA+](https://arxiv.org/abs/2402.12354) with LR Ratio: 5 |
|
* Context size: 16384 |
|
* Gradient accumulation steps: 4 |
|
* Epochs: 1 |
|
* Stage 2 (fine tune) |
|
* Target: Stage 1 model |
|
* LoRA dropout 0.5 |
|
* LoRA rank 32, alpha 64 |
|
* LR cosine 1.5e-5 (min 1.5e-6) |
|
* [LoRA+](https://arxiv.org/abs/2402.12354) with LR Ratio: 5 |
|
* Context size: 16384 |
|
* Gradient accumulation steps: 4 |
|
* Epochs: 1 |
|
|
|
Initially I tried to keep stage 2 ultra thin/light to only minimally repair the merge-damage and trying to retain the base model's IQ, but this did not work out well, so I ultimately ramped up the dataset to include external sources (see dataset metadata) to boost the RP capacity of the model. |
|
|
