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

sonian

-0.69

ication

-0.64

reditary

-0.63

gee

-0.61

Plex

-0.60

çīĪ

-0.60

oria

-0.60

Äĩ

-0.59

Viz

-0.59

uria

-0.59

POSITIVE LOGITS

ron

0.85

Anonymous

0.82

RON

0.80

yrics

0.78

 Transcript

0.74

 Anonymous

0.73

Quote

0.72

rons

0.71

anon

0.70

 robbers

0.69

Activations Density 0.000%

No Known Activations

This feature has no known activations.