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

cast

-0.77

 Mysteries

-0.75

Translation

-0.70

ribution

-0.69

âĶĢâĶĢ

-0.66

Dispatch

-0.66

uzzle

-0.66

faced

-0.64

ipolar

-0.63

iscovery

-0.62

POSITIVE LOGITS

too

0.99

too

0.75

 sake

0.66

Too

0.66

HW

0.65

 stops

0.63

iae

0.62

 pedal

0.61

BF

0.61

enza

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.