INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ciation

-0.81

 mosqu

-0.78

ospons

-0.73

oute

-0.72

odcast

-0.71

visor

-0.70

ription

-0.69

Winged

-0.68

onduct

-0.68

ortality

-0.67

POSITIVE LOGITS

ãĥ¼ãĥĨ

0.72

soever

0.71

umers

0.71

ãĤ±

0.71

ãĥ¼ãĥĨãĤ£

0.69

igans

0.66

zer

0.65

Gamer

0.65

Times

0.62

 shooters

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.