INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

schild

-0.71

ament

-0.71

stretched

-0.69

epad

-0.68

eret

-0.67

miah

-0.66

iland

-0.66

Fs

-0.66

ocado

-0.66

ourses

-0.66

POSITIVE LOGITS

âĸ¬

0.71

è£ħ

0.68

HIP

0.66

KEN

0.62

 Chandra

0.62

 Emin

0.62

Rai

0.61

Riy

0.61

cipled

0.61

EAR

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.