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

rer

-0.72

Registered

-0.70

yright

-0.68

é¾įåĸļå£«

-0.68

Filename

-0.67

ccording

-0.66

];

-0.66

 Somers

-0.65

arer

-0.64

avement

-0.63

POSITIVE LOGITS

rig

0.73

pire

0.70

pour

0.68

ocr

0.66

ymm

0.66

iann

0.64

 Colony

0.63

ju

0.62

EV

0.62

 soil

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.