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

brew

-0.80

uum

-0.76

ohyd

-0.76

Thumbnail

-0.76

uscript

-0.76

nerg

-0.74

iche

-0.73

iframe

-0.73

 recomb

-0.73

rogens

-0.72

POSITIVE LOGITS

ãĤ¢ãĥ«

0.74

 Samson

0.71

 Tycoon

0.71

 Lion

0.66

Poc

0.65

 Siem

0.65

 indoors

0.60

 Siberia

0.60

 rings

0.60

 Mongolia

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.