INDEX

Explanations

words related to disguise and deception

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "guise"

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hower

-0.87

ŃĶ

-0.80

 Spectrum

-0.74

 spect

-0.71

cling

-0.70

 eleph

-0.67

 shutter

-0.65

è¦ļéĨĴ

-0.63

 cort

-0.63

 footprint

-0.63

POSITIVE LOGITS

arant

1.14

idelines

1.11

arding

1.06

ilty

1.06

inea

1.05

ests

1.02

cci

0.99

errilla

0.99

ilar

0.96

vernment

0.95

Activations Density 0.009%

words related to disguise and deception

occurrences of the term "guise"

No Comments

No Known Activations