INDEX

Explanations

mentions of someone being foolish or acting in a silly manner

oai_token-act-pair · gpt-3.5-turbo

references to being foolish or deceived

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Parenthood

-0.69

ials

-0.67

orney

-0.67

foreseen

-0.62

lined

-0.61

accompan

-0.61

apers

-0.60

ŃĶ

-0.60

apa

-0.59

Delivery

-0.59

POSITIVE LOGITS

hard

0.86

sonian

0.85

ery

0.85

proof

0.84

ingly

0.77

 fooled

0.76

pas

0.76

ibility

0.74

 naive

0.74

hemer

0.74

Activations Density 0.053%

mentions of someone being foolish or acting in a silly manner

references to being foolish or deceived

No Comments

No Known Activations