INDEX

Explanations

references to specific individuals or names

oai_token-act-pair · gpt-3.5-turbo

names and titles, particularly related to individuals in politics and entertainment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

eering

-0.77

ARK

-0.71

NCT

-0.66

ingham

-0.65

eers

-0.63

 Johns

-0.63

 replication

-0.62

¥ŀ

-0.61

eer

-0.60

 payoff

-0.59

POSITIVE LOGITS

coni

1.18

lus

1.14

cano

1.03

cul

0.96

cious

0.96

colo

0.88

bane

0.87

ucha

0.83

cus

0.82

bian

0.82

Activations Density 0.011%

references to specific individuals or names

names and titles, particularly related to individuals in politics and entertainment

No Comments

No Known Activations