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

 Eyes

-0.68

 Reck

-0.65

 Cere

-0.65

 Tact

-0.65

hyd

-0.64

yle

-0.64

 Recap

-0.62

 Frost

-0.62

 Prem

-0.61

arthy

-0.61

POSITIVE LOGITS

stanbul

0.84

lication

0.81

cies

0.81

ynski

0.81

ById

0.75

ULE

0.72

Îº

0.72

OPA

0.69

PsyNetMessage

0.68

emonium

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.