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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

oglu

-0.79

utterstock

-0.76

munition

-0.74

irling

-0.74

oppable

-0.73

hemy

-0.73

orce

-0.71

cipline

-0.70

xual

-0.70

inos

-0.70

POSITIVE LOGITS

 unden

0.87

 Cosponsors

0.79

 toile

0.76

CLE

0.72

CLASS

0.70

Prev

0.69

 aloud

0.68

TAG

0.67

 Falls

0.65

Beg

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.