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

iltr

-0.68

ructose

-0.67

icion

-0.66

inder

-0.66

CONCLUS

-0.65

urned

-0.65

 Actions

-0.65

 await

-0.65

enna

-0.65

NK

-0.63

POSITIVE LOGITS

yip

0.68

 Remem

0.68

 warr

0.65

å¥³

0.65

ktop

0.65

edom

0.64

 athlet

0.63

 Presidents

0.62

 Sapp

0.61

 elig

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.