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

 equival

-0.70

theless

-0.64

 integrity

-0.59

dry

-0.59

ãĤ¨ãĥ«

-0.58

 polymer

-0.58

-0.57

 pollut

-0.57

POSITIVE LOGITS

isson

0.85

romeda

0.81

agascar

0.80

bane

0.80

anon

0.79

azar

0.79

etary

0.78

galitarian

0.78

oslav

0.77

auri

0.77

Activations Density 0.000%

No Known Activations

This feature has no known activations.