INDEX

Explanations

mention of the word "mold" or variations of it in the text

oai_token-act-pair · gpt-3.5-turbo

mentions of mold and related terms

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

IGHTS

-0.74

externalActionCode

-0.74

WTO

-0.73

avez

-0.73

vernment

-0.72

×Ļ×

-0.69

ZE

-0.67

 derog

-0.66

ights

-0.66

--------------------------------------------------------

-0.65

POSITIVE LOGITS

 mold

1.06

 mould

1.03

ering

0.94

ovan

0.87

eus

0.86

 spores

0.84

 molded

0.80

eers

0.77

lings

0.77

eer

0.76

Activations Density 0.018%

mention of the word "mold" or variations of it in the text

mentions of mold and related terms

No Comments

No Known Activations