INDEX

Explanations

words related to directions or location, particularly indicating movement or positioning

oai_token-act-pair · gpt-3.5-turbo

references to directional movements or locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

MAT

-0.94

PROV

-0.76

soDeliveryDate

-0.74

MER

-0.74

ETHOD

-0.74

OTHER

-0.70

ICH

-0.69

MIC

-0.69

ASON

-0.66

ptive

-0.66

POSITIVE LOGITS

bound

1.20

pread

0.87

naire

0.85

 lanes

0.85

ridges

0.81

duty

0.81

uction

0.78

folios

0.77

loads

0.76

essel

0.74

Activations Density 0.014%

words related to directions or location, particularly indicating movement or positioning

references to directional movements or locations

No Comments

No Known Activations