INDEX

Explanations

phrases related to legal, criminal, and political topics

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 encomp

-0.92

 disreg

-0.81

 excru

-0.74

 inev

-0.73

 depic

-0.73

 unve

-0.72

 antem

-0.72

 macrop

-0.71

 suscep

-0.69

 accla

-0.68

POSITIVE LOGITS

 HasFactory

0.63

pertise

0.59

 TextAppearance

0.57

KELEY

0.55

 aspi

0.53

 experimente

0.53

 religione

0.53

 tille

0.52

 locu

0.52

 visse

0.52

Activations Density 0.583%

phrases related to legal, criminal, and political topics

No Comments

No Known Activations