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

ACP

-0.70

ullah

-0.65

jad

-0.64

imeter

-0.62

FUL

-0.62

micro

-0.62

ground

-0.62

Ground

-0.62

pmwiki

-0.61

Plan

-0.59

POSITIVE LOGITS

 here

1.20

 HERE

0.82

indal

0.78

 speaking

0.75

 Here

0.73

ratulations

0.71

etts

0.71

 papers

0.71

orem

0.68

ients

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.