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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ement

-0.74

 recogn

-0.70

rification

-0.70

 counting

-0.67

adm

-0.66

 Totem

-0.65

 desp

-0.65

isation

-0.64

uscript

-0.63

raf

-0.63

POSITIVE LOGITS

ãĥĩãĤ£

0.75

0.73

DVD

0.73

ãĤ·ãĥ£

0.73

.)

0.73

Ops

0.72

DX

0.71

bc

0.71

0.70

Activations Density 0.000%

No Known Activations

This feature has no known activations.