INDEX

Explanations

the word 'naive'

oai_token-act-pair · gpt-3.5-turbo

terms describing naivety or gullibility

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

Downloadha

-0.83

ŃĶ

-0.79

foreseen

-0.75

ittee

-0.75

interrupted

-0.74

ngth

-0.73

alach

-0.72

hops

-0.71

avez

-0.69

orset

-0.68

POSITIVE LOGITS

 naive

1.07

 naÃ¯ve

0.96

Ã¯ve

0.92

sters

0.86

glers

0.79

lings

0.76

ly

0.74

 innocence

0.73

wd

0.71

ster

0.71

Activations Density 0.012%

the word 'naive'

terms describing naivety or gullibility

No Comments

No Known Activations