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

;;;;;;;;

-0.65

ben

-0.65

 bishop

-0.65

amm

-0.62

GP

-0.62

;;;;

-0.61

 disg

-0.61

 rake

-0.60

misc

-0.59

woods

-0.59

POSITIVE LOGITS

itivity

0.81

icut

0.79

oÄŁ

0.77

Faw

0.65

atform

0.63

vacc

0.63

othing

0.63

Fight

0.63

uga

0.62

 Templ

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.