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

maxwell

-0.91

align

-0.82

notation

-0.78

grain

-0.78

reference

-0.77

Course

-0.76

afety

-0.74

minist

-0.73

plain

-0.72

scl

-0.72

POSITIVE LOGITS

 Stur

0.84

Bey

0.66

 Stoke

0.66

 Weeks

0.63

 polling

0.63

 Watergate

0.62

Ud

0.61

Dro

0.61

kne

0.60

 Nixon

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.