INDEX

Explanations

words ending in 'u'

oai_token-act-pair · gpt-3.5-turbo

occurrences of the letter 'u'

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 horizont

-0.75

 initials

-0.73

 Attribution

-0.71

 Revelations

-0.71

milo

-0.70

*/(

-0.70

 SERV

-0.70

 Cosponsors

-0.70

=-=-

-0.69

 constants

-0.69

POSITIVE LOGITS

pport

1.25

seless

1.07

cci

1.04

cca

1.04

uru

1.02

igi

0.99

isine

0.97

endo

0.96

asa

0.96

awei

0.92

Activations Density 0.027%

words ending in 'u'

occurrences of the letter 'u'

No Comments

No Known Activations