L3.3-Electra-R1-70b / README.md
Steelskull's picture
Update README.md
10fb146 verified
|
raw
history blame
70.5 kB
metadata
base_model:
  - EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.0
  - Sao10K/L3.1-70B-Hanami-x1
  - Sao10K/L3.3-70B-Euryale-v2.3
  - LatitudeGames/Wayfarer-Large-70B-Llama-3.3
  - TheDrummer/Anubis-70B-v1
  - TheSkullery/L3.1x3.3-Hydroblated-R1-70B-v4.4
  - SicariusSicariiStuff/Negative_LLAMA_70B
  - Sao10K/70B-L3.3-Cirrus-x1
library_name: transformers

L3.3-Electra-R1-70b

Electra Model Mascot
⚡ Top Sponsors

Model Information

L3.3-Electra-R1-70b v0.5.OG

L3.3 = Llama 3.3 SCE Merge R1 = Deepseek R1 70b Parameters

Model Composition

Model Series Overview

L3.3-Electra-R1-70b represents the foundational release in a three-part model series, followed by L3.3-Cu-Mai-R1-70b (Version A) and L3.3-Mokume-Gane-R1-70b (Version C). The name "Electra" draws inspiration from the electric-powered aesthetic of the model's mascot, representing the powerful capabilities and lightning-fast responses that define this model's performance.

Technical Architecture

Built on a custom DeepSeek R1 Distill base (TheSkullery/L3.1x3.3-Hydroblated-R1-70B-v4.4), Electra-R1 integrates specialized components through the SCE merge method with a select_topk parameter of 0.16. The model uses float32 dtype during processing with a bfloat16 output dtype for optimized performance.

  • EVA and Wayfarer foundations for creative expression and scene comprehension
  • Euryale, Cirrus and Hanami elements for enhanced reasoning capabilities
  • Anubis components for detailed scene description
  • Negative_LLAMA integration for balanced perspective and response

Core Capabilities

As the OG model in the series, Electra-R1 serves as the gold standard and reliable baseline. User feedback consistently highlights its superior intelligence, coherence, and unique ability to provide deep character insights. Through proper prompting, the model demonstrates advanced reasoning capabilities and an "X-factor" that enables unprompted exploration of character inner thoughts and motivations.

Base Architecture

The model utilizes the custom Hydroblated-R1 base, engineered for stability and enhanced reasoning. The SCE merge method's settings are precisely tuned based on extensive community feedback, ensuring optimal component integration while maintaining model coherence and reliability. This foundation establishes Electra-R1 as the benchmark upon which its variant models build and expand.

Recommended Sampler Settings

⚡ By: @Geechan

Static Temperature: 1.0
Dynamic Temp (Alternative): 0.8 - 1.05
Min P: 0.025-0.03
DRY:
- Multiplier: 0.8
- Base: 1.74
- Length: 4-6

Good Starting Templates & Prompts

LeCeption v2 by @Steel > A revamped XML version of Llam@ception 1.5.2 with stepped thinking and Reasoning added

⚡ ST REASONING CONFIGURATION:

Start Reply With:

'<think> OK, as an objective, detached narrative analyst, let's think this through carefully:'

Reasoning Formatting (no spaces):

Prefix: '<think>'
Suffix: '</think>'

Support & Community:

Special Thanks

  • @Geechan for feedback and sampler settings
  • @Konnect for their feedback and templates
  • @Kistara for their feedback and help with the model mascot design on past model's
  • @Thana Alt for their feedback
  • @Lightning_missile for their feedback
  • The Arli community for feedback and testers
  • The BeaverAI communty for feedback and testers

I wish I could add everyone but im pretty sure it would be as long as the card!