INDEX

Explanations

terms related to government entities or organizations

oai_token-act-pair · gpt-3.5-turbo

references to various authorities or organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

-0.89

boxes

-0.83

lasses

-0.81

hematic

-0.79

alien

-0.72

irling

-0.72

lace

-0.71

yle

-0.70

mx

-0.69

box

-0.69

POSITIVE LOGITS

 Authority

1.12

uthor

0.85

 delegated

0.81

eous

0.79

oun

0.76

 authority

0.74

 Board

0.72

 derog

0.72

 nomine

0.71

 exercised

0.69

Activations Density 0.011%

terms related to government entities or organizations

references to various authorities or organizations

No Comments

No Known Activations