INDEX

Explanations

phrases related to news stories or competitions with rankings

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

standard

Context Size

1,024

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Minang

-0.82

 Compañ

-0.82

 Pekan

-0.75

 Embaj

-0.74

 prek

-0.72

 policia

-0.70

 Karang

-0.70

 impon

-0.69

 Galería

-0.68

 teka

-0.66

POSITIVE LOGITS

 friable

0.80

 cryst

0.75

 inext

0.70

 mutagen

0.68

 tetrach

0.68

 unden

0.68

0.67

 polymorph

0.66

 ampoule

0.66

 tupperware

0.65

Activations Density 0.184%

phrases related to news stories or competitions with rankings

No Comments

No Known Activations