INDEX

Explanations

No Explanations Found

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

enburg

-0.79

 regrets

-0.70

IDENT

-0.66

ãĤ£

-0.63

inges

-0.62

IAL

-0.62

cture

-0.60

USER

-0.60

ENT

-0.60

 adop

-0.60

POSITIVE LOGITS

oppable

0.89

ochemistry

0.76

itaire

0.71

asm

0.70

phis

0.70

seek

0.70

ioxide

0.67

ifter

0.67

lace

0.66

offensive

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.