INDEX

Explanations

abbreviations or acronyms ending in "DI"

oai_token-act-pair · gpt-3.5-turbo

references to Direct Investment

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

storm

-0.78

icago

-0.75

ifications

-0.72

sburg

-0.71

itude

-0.71

days

-0.69

rooms

-0.69

storms

-0.69

sburgh

-0.67

night

-0.67

POSITIVE LOGITS

RECT

1.37

DI

1.22

ANE

1.02

BC

0.95

ANCE

0.92

ARY

0.88

ENTS

0.88

VICE

0.88

LE

0.86

OTAL

0.86

Activations Density 0.022%

abbreviations or acronyms ending in "DI"

references to Direct Investment

No Comments

No Known Activations