INDEX

Explanations

names, likely focusing on last names

oai_token-act-pair · gpt-3.5-turbo

names and identifiers of people involved in incidents or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

GH

-0.95

GH

-0.95

 Hair

-0.85

ALT

-0.83

 Haku

-0.82

 ASIC

-0.80

 Andrews

-0.80

gc

-0.80

hex

-0.79

 Malk

-0.78

POSITIVE LOGITS

ner

1.11

kn

1.03

ider

1.02

NER

1.00

ners

1.00

Kn

0.95

KN

0.93

iers

0.86

Len

0.85

 Feinstein

0.85

Activations Density 0.609%

names, likely focusing on last names

names and identifiers of people involved in incidents or events

No Comments

No Known Activations