INDEX

Explanations

words related to disguises or masquerades

oai_token-act-pair · gpt-3.5-turbo

terms related to disguise and deception

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

aldi

-0.74

area

-0.70

ansk

-0.66

itar

-0.65

enge

-0.65

olics

-0.62

elve

-0.62

vals

-0.62

itsch

-0.62

ode

-0.61

POSITIVE LOGITS

querade

1.05

 disgu

0.99

 camoufl

0.89

pas

0.84

 smokes

0.82

 pretext

0.77

 disguise

0.77

 disguised

0.77

arial

0.76

 deceive

0.75

Activations Density 0.090%

words related to disguises or masquerades

terms related to disguise and deception

No Comments

No Known Activations