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

ampton

-0.69

ÃŃn

-0.66

 unfocusedRange

-0.64

 Britann

-0.64

 horizont

-0.63

ottage

-0.62

ãģ®é

-0.61

 exits

-0.61

 Majesty

-0.61

ients

-0.59

POSITIVE LOGITS

advertising

0.88

interstitial

0.77

Attach

0.74

yip

0.70

article

0.67

glomer

0.66

rities

0.66

post

0.65

SPONSORED

0.65

dose

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.