INDEX

Explanations

locations or places, particularly cities and countries

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "in."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

SourceFile

-0.77

ascript

-0.75

lly

-0.72

gio

-0.70

selves

-0.69

rie

-0.67

estic

-0.66

miah

-0.65

itte

-0.65

estial

-0.64

POSITIVE LOGITS

 captivity

1.24

 limbo

1.04

activity

1.01

 office

0.95

 hiber

0.85

 Wonderland

0.83

 obscurity

0.81

effect

0.80

prison

0.80

 prison

0.79

Activations Density 0.148%

locations or places, particularly cities and countries

occurrences of the word "in."

No Comments

No Known Activations