INDEX

Explanations

words related to security or safety

oai_token-act-pair · gpt-3.5-turbo

the conjunction "and" 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 @ 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

Deal

-0.86

Report

-0.81

Recomm

-0.74

LIST

-0.72

bo

-0.71

inational

-0.70

daq

-0.69

ĨĴ

-0.69

account

-0.69

Names

-0.69

POSITIVE LOGITS

romeda

0.89

rogens

0.86

rogen

0.82

 therefore

0.79

 consequently

0.77

 other

0.71

 assorted

0.69

hester

0.68

 related

0.65

 Other

0.64

Activations Density 0.583%

words related to security or safety

the conjunction "and" in various contexts

No Comments

No Known Activations