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

 Clicker

-0.82

 agre

-0.71

 assum

-0.71

 livest

-0.70

osate

-0.68

avascript

-0.67

 undermin

-0.66

 motions

-0.66

 dilig

-0.66

 scrut

-0.65

POSITIVE LOGITS

Von

0.78

ynthesis

0.77

ogenesis

0.76

 Rider

0.74

ãĥ¼ãĥ³

0.74

Å«

0.73

Ã

0.72

Zar

0.71

Â¯

0.68

anova

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.