INDEX

Explanations

words that are proper nouns

oai_token-act-pair · gpt-3.5-turbo

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 @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 lash

-0.68

 contrace

-0.66

æ©Ł

-0.62

 divid

-0.61

 GOODMAN

-0.61

ãĤ´ãĥ³

-0.61

 compe

-0.61

 unequ

-0.60

cmp

-0.59

 coord

-0.59

POSITIVE LOGITS

olitan

1.26

ulse

1.18

isode

1.15

ocalypse

1.13

heric

1.06

ilot

1.04

udding

1.02

ersion

1.02

ossible

1.01

resents

1.01

Activations Density 0.043%

words that are proper nouns

the letter 'p'

No Comments

No Known Activations