INDEX

Explanations

proper nouns and names

oai_token-act-pair · gpt-3.5-turbo

occurrences of the letter 'P'

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ãĤ¼ãĤ¦ãĤ¹

-0.90

 GOODMAN

-0.81

 Aval

-0.79

é¾įå¥ĳå£«

-0.79

æ©

-0.79

æ©Ł

-0.77

åŃĲ

-0.76

adm

-0.76

ãĥ³ãĤ¸

-0.76

é¾

-0.75

POSITIVE LOGITS

aired

1.14

ossession

1.13

redict

1.10

ricia

1.08

ardon

1.08

hew

1.06

ulse

1.03

ossible

1.03

etition

1.02

ivot

1.01

Activations Density 0.035%

proper nouns and names

occurrences of the letter 'P'

No Comments

No Known Activations