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

atre

-0.71

Rat

-0.70

ECTION

-0.69

ware

-0.68

jay

-0.68

yi

-0.68

ername

-0.68

chio

-0.67

thing

-0.67

phrine

-0.67

POSITIVE LOGITS

ingen

0.72

inn

0.68

ammy

0.62

ãĥ©ãĥ³

0.62

 anomal

0.62

paren

0.60

 bount

0.58

 iceberg

0.58

 Syracuse

0.58

 decl

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.