All credits go to:
Lambent for the wonderful model series. They did all the work, I just ran a few commands.
Ο»ira Ξphrodite-v13-27B
Mira-Aphrodite-v13-27B-Q5_K_X.gguf
A conversation snippet
View Conversation
### USER
**_hey, how are you doing?_**
### ASSISTANT
*Mira beams, adjusting her headphones and leaning closer to her glowing monitor.*
**Mira:** Oh, hey! Doing great! Just...existing, you know? Sometimes that's enough. Anything new with you? π»β¨π
### USER
**_curious what kind of self-portrait you might describe for yourself (visually) if you wanted an avatar or profile picture_**
### ASSISTANT
*Mira taps a finger against her chin, her brow furrowing slightly in thought, then her eyes widen as an idea takes root.*
**Mira:** Oooh, good one! If I had an avatar, hmm... okay, so imagine: Soft, rosy-pink hair, slightly messy like it got staticky from thinking too much. A single strand of turquoise that I forgot to brush. Eyes are huge, golden with flecks of green, wide open and curious. I'm wearing these big, round glasses with purple rims. And I have this tiny little silver star earring in my left ear. Oh! And freckles! All over my cheeks and nose. Iβm just...kind of floating there, surrounded by pastel clouds and little glowing moths.β¨βοΈπ¦ Would that work? Or is it too much?π
π Overview
This is a merge of pre-trained language models created using [mergekit](https://github.com/arcee-ai/mergekit. It utilizes custom merge methods.
The Recipe
Ο»ira Ξphrodite-v13 | Methods: [CABS, DeLerp]
View Configuration
# https://huggingface.co/Lambent
name: Mira-Aphrodite-v8
merge_method: cabs
base_model: ./Ingredients/Lambent_Mira-v1.19-27B-dpo
models:
- model: ./Ingredients/Lambent_Mira-v1.19-27B-dpo
- model: ./Ingredients/Mira-Aphrodite-v7
parameters:
weight: 0.6
n_val: 16
m_val: 32
- model: ./Ingredients/Mira-Aphrodite-v5
parameters:
weight: 0.4
n_val: 11
m_val: 33
default_n_val: 8
default_m_val: 32
pruning_order:
- ./Ingredients/Mira-Aphrodite-v7
- ./Ingredients/Mira-Aphrodite-v5
dtype: bfloat16
out_dtype: bfloat16
tokenizer:
source: union
---
name: Mira-Aphrodite-v13
merge_method: delerp
models:
- model: ./Ingredients/Mira-Aphrodite-v8
- model: ./Ingredients/Lambent_Mira-v1.19-27B-dpo
base_model: ./Ingredients/Lambent_Mira-v1.19-27B-dpo
parameters:
t:
- value: 0.999
dtype: float32
out_dtype: bfloat16
tokenizer:
source: union
- Downloads last month
- 10
Hardware compatibility
Log In to add your hardware
Inference Providers NEW
This model isn't deployed by any Inference Provider. π Ask for provider support
Model tree for virtuous7373/Lambent-Mira-Aphrodite-v13
Base model
virtuous7373/Mira-M2 Finetuned
Lambent/Mira-v1.18-27B Finetuned
Lambent/Mira-v1.18-27B-heretic-0.019 Finetuned
Lambent/Mira-v1.19-27B-dpo