INDEX

Explanations

names containing the syllable "man"

oai_token-act-pair · gpt-3.5-turbo

the character "man" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

uesday

-0.78

avorite

-0.66

ÃįÃį

-0.66

urus

-0.66

CAST

-0.66

anked

-0.63

PsyNetMessage

-0.63

anyon

-0.62

¥ŀ

-0.62

ORK

-0.62

POSITIVE LOGITS

ifest

1.38

ufact

1.17

ning

1.14

uscript

1.08

iac

1.05

ned

1.01

hood

0.96

å£«

0.94

agements

0.91

iasis

0.91

Activations Density 0.038%

names containing the syllable "man"

the character "man" in various contexts

No Comments

No Known Activations