INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

itness

-0.70

rahim

-0.68

mind

-0.67

 Templ

-0.67

SPONSORED

-0.66

minster

-0.66

omers

-0.65

eka

-0.65

 Javascript

-0.65

eln

-0.64

POSITIVE LOGITS

 sqor

0.92

Fuk

0.61

 gown

0.60

 quota

0.59

Ã¡n

0.58

ãĤ¦ãĤ¹

0.58

 anth

0.58

 speaker

0.58

 goodness

0.58

 hump

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.