INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

oop

-0.72

OTS

-0.72

pling

-0.71

 parach

-0.70

Ge

-0.66

Job

-0.66

 seams

-0.65

 scra

-0.64

arrass

-0.64

iry

-0.63

POSITIVE LOGITS

prep

0.69

 Norn

0.69

kson

0.68

ault

0.64

dn

0.63

channel

0.63

eport

0.63

sworth

0.62

nesota

0.61

said

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.