INDEX

Explanations

instances of embarrassment and the syllable "Tir"

oai_token-act-pair · gpt-4-turbo

words related to embarrassment or awkward situations

oai_token-act-pair · gpt-3.5-turbo

words associated with embarrassment and geographical locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

enz

-0.72

olves

-0.68

sheet

-0.66

entials

-0.65

iable

-0.65

iability

-0.65

ugg

-0.64

arettes

-0.64

vec

-0.64

CN

-0.63

POSITIVE LOGITS

Tir

0.89

ificial

0.88

isans

0.85

ertodd

0.82

alty

0.79

xon

0.79

rison

0.76

nered

0.76

nect

0.74

Downloadha

0.73

Activations Density 0.026%

instances of embarrassment and the syllable "Tir"

words related to embarrassment or awkward situations

words associated with embarrassment and geographical locations

No Comments

No Known Activations