INDEX

Explanations

proper nouns and locations

oai_token-act-pair · gpt-3.5-turbo

mentions of locations or geographic entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

orer

-0.74

antha

-0.72

uds

-0.69

udeb

-0.69

accompanied

-0.69

ãĥł

-0.67

opsy

-0.66

fusc

-0.65

lda

-0.65

ãĥ¼ãĥĨãĤ£

-0.62

POSITIVE LOGITS

 CITY

1.16

MEN

1.03

POL

1.01

 COUNTY

0.96

MEN

0.96

 PRES

0.96

VILLE

0.95

 EDITION

0.94

 NEWS

0.93

 TODAY

0.90

Activations Density 0.055%

proper nouns and locations

mentions of locations or geographic entities

No Comments

No Known Activations