INDEX

Explanations

references to names or specific nouns

oai_token-act-pair · gpt-3.5-turbo

mentions of individuals and episodes from certain series or events

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

oat

-0.93

bing

-0.85

ret

-0.77

oths

-0.75

reth

-0.72

 Telegram

-0.68

bor

-0.67

rosse

-0.67

gest

-0.66

bage

-0.64

POSITIVE LOGITS

elson

0.79

letal

0.79

umat

0.77

opoulos

0.77

ewski

0.75

cart

0.74

ansen

0.74

umatic

0.73

emic

0.73

aimon

0.73

Activations Density 0.045%

references to names or specific nouns

mentions of individuals and episodes from certain series or events

No Comments

No Known Activations