INDEX

Explanations

phrases related to police investigations and legal matters

oai_token-act-pair · gpt-3.5-turbo

references to individuals with the surname "Piers" or variations of it

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

shaw

-0.80

Lot

-0.74

drops

-0.71

Patch

-0.69

Crash

-0.67

Shield

-0.65

Luck

-0.64

knit

-0.64

QB

-0.64

MER

-0.62

POSITIVE LOGITS

hip

1.00

iers

0.98

pread

0.95

aurus

0.92

hips

0.86

poons

0.86

*/(

0.85

cript

0.85

ystem

0.84

mith

0.80

Activations Density 0.013%

phrases related to police investigations and legal matters

references to individuals with the surname "Piers" or variations of it

No Comments

No Known Activations