INDEX

Explanations

references to specific political figures

oai_token-act-pair · gpt-3.5-turbo

repeated mentions of senators

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

Pwr

-0.66

 manual

-0.66

 Siberian

-0.65

 factor

-0.63

 factors

-0.63

 oper

-0.60

 towed

-0.60

 Mania

-0.60

LIA

-0.60

 ponies

-0.59

POSITIVE LOGITS

iors

1.40

pai

1.07

eca

1.04

escent

1.01

egal

1.00

esse

0.92

ileaks

0.91

esis

0.91

IOR

0.91

olt

0.90

Activations Density 0.011%

references to specific political figures

repeated mentions of senators

No Comments

No Known Activations