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

vana

-0.84

mens

-0.83

kefeller

-0.82

ahime

-0.80

anus

-0.76

 Cosponsors

-0.74

enko

-0.71

chwitz

-0.71

ovsky

-0.71

lies

-0.69

POSITIVE LOGITS

)]

0.67

Tes

0.65

XY

0.65

REL

0.64

GMT

0.63

OPLE

0.60

English

0.60

Param

0.60

 Mutant

0.59

ogue

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.