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

 Izan

-0.73

 Takeru

-0.64

enium

-0.64

 globe

-0.63

AST

-0.60

 Frames

-0.60

 genesis

-0.58

 Orig

-0.58

 UCHIJ

-0.58

 idols

-0.57

POSITIVE LOGITS

olkien

0.70

 concess

0.70

 marqu

0.69

ledge

0.68

uddy

0.65

lean

0.64

pocket

0.64

amph

0.63

aid

0.63

earch

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.