INDEX

Explanations

references to individuals in temporary or role-specific leadership positions

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "acting" and its variations in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

xual

-0.83

ilings

-0.69

 Blackwell

-0.66

joy

-0.65

nda

-0.64

front

-0.64

 Sovere

-0.64

ciating

-0.63

fore

-0.61

fac

-0.61

POSITIVE LOGITS

uary

0.97

uate

0.84

iott

0.80

inic

0.79

uated

0.78

uations

0.78

icter

0.75

ional

0.75

ives

0.72

uka

0.69

Activations Density 0.016%

references to individuals in temporary or role-specific leadership positions

mentions of the word "acting" and its variations in various contexts

No Comments

No Known Activations