INDEX

Explanations

pasta dishes or related terms

oai_token-act-pair · gpt-3.5-turbo

mentions of "pa," particularly in various contexts and forms

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

wives

-0.74

RY

-0.72

WARE

-0.70

AGE

-0.70

rail

-0.69

MET

-0.68

Guest

-0.66

witch

-0.66

é¾įå¥ĳå£«

-0.65

roller

-0.65

POSITIVE LOGITS

pa

0.95

ican

0.90

Osw

0.78

xon

0.77

 Bras

0.76

issance

0.76

edia

0.73

bably

0.73

ornings

0.73

astic

0.72

Activations Density 0.017%

pasta dishes or related terms

mentions of "pa," particularly in various contexts and forms

No Comments

No Known Activations