INDEX

Explanations

references to governmental or administrative bodies with the word "Authority"

oai_token-act-pair · gpt-3.5-turbo

mentions of different authorities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

arget

-0.91

lasses

-0.85

alien

-0.82

boxes

-0.82

hematic

-0.77

irling

-0.77

ilde

-0.76

yle

-0.75

pocket

-0.75

bart

-0.74

POSITIVE LOGITS

 Authority

1.09

oun

0.79

uthor

0.74

 confir

0.73

 Board

0.72

BILITY

0.72

 delegated

0.71

 nomine

0.70

eous

0.68

 authority

0.67

Activations Density 0.020%

references to governmental or administrative bodies with the word "Authority"

mentions of different authorities

No Comments

No Known Activations