INDEX

Explanations

words related to criticism or disapproval

oai_token-act-pair · gpt-3.5-turbo

variations of the prefix "unc," indicating negation or absence

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

å§«

-0.89

uyomi

-0.87

Ò

-0.85

tery

-0.80

geist

-0.70

=-=-=-=-

-0.69

=-=-=-=-=-=-=-=-

-0.69

Fo

-0.68

 Dynamics

-0.67

MENT

-0.66

POSITIVE LOGITS

redited

1.17

ooked

1.16

ursed

1.16

reated

1.08

aleb

1.05

umbered

1.04

ount

1.03

ivil

1.00

onduct

0.97

irc

0.97

Activations Density 0.010%

words related to criticism or disapproval

variations of the prefix "unc," indicating negation or absence

No Comments

No Known Activations