INDEX

Explanations

words related to political events or actions

oai_token-act-pair · gpt-3.5-turbo

instances of the word "office" in relation to political positions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ostic

-0.74

alog

-0.72

oken

-0.70

essen

-0.69

asca

-0.66

oise

-0.66

Tang

-0.64

Side

-0.63

TERN

-0.62

osphere

-0.62

POSITIVE LOGITS

holders

0.88

clinton

0.78

yrim

0.72

holder

0.70

earable

0.67

prison

0.64

 pledging

0.64

 hopeful

0.63

long

0.61

matical

0.61

Activations Density 0.020%

words related to political events or actions

instances of the word "office" in relation to political positions

No Comments

No Known Activations