INDEX

Explanations

references to the company IBM

oai_token-act-pair · gpt-3.5-turbo

mentions of the company IBM

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

igun

-0.79

alog

-0.75

adow

-0.72

alogue

-0.72

adows

-0.71

fulness

-0.70

 Niet

-0.70

icles

-0.70

icans

-0.67

fully

-0.66

POSITIVE LOGITS

 Watson

0.95

ITED

0.82

 Corporation

0.78

ERN

0.77

ERS

0.72

 Advantage

0.72

BM

0.71

Cola

0.71

ERT

0.71

BL

0.71

Activations Density 0.014%

references to the company IBM

mentions of the company IBM

No Comments

No Known Activations