INDEX

Explanations

references to the luxury sports car brand "Ferrari."

oai_token-act-pair · gpt-3.5-turbo

references to the luxury sports car brand Ferrari

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

umbn

-0.89

ablishment

-0.88

orative

-0.82

surv

-0.71

olver

-0.71

itcher

-0.70

cript

-0.70

alid

-0.70

retched

-0.68

upt

-0.66

POSITIVE LOGITS

 Ferrari

1.05

illo

0.81

 dealership

0.78

 parked

0.77

ettel

0.75

0.72

ucci

0.71

 capacitor

0.71

 driver

0.70

Fan

0.70

Activations Density 0.007%

references to the luxury sports car brand "Ferrari."

references to the luxury sports car brand Ferrari

No Comments

No Known Activations