INDEX

Explanations

mentions of a specific place name

oai_token-act-pair · gpt-3.5-turbo

references to the word "Glad."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Mutant

-0.71

 Moore

-0.70

 temper

-0.68

 session

-0.66

 Nero

-0.63

 Hawkins

-0.63

 practice

-0.63

 square

-0.62

 Apocalypse

-0.61

imp

-0.61

POSITIVE LOGITS

lad

4.64

Lad

1.54

 Glad

1.30

lav

1.17

adder

1.05

lass

1.02

laden

1.02

lol

1.02

lam

1.01

los

1.01

Activations Density 0.010%

mentions of a specific place name

references to the word "Glad."

No Comments

No Known Activations