INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

Ļ½

-0.79

_>

-0.74

ENDED

-0.72

hedon

-0.69

RNA

-0.68

Ĺ

-0.66

 dilig

-0.66

pec

-0.65

cffff

-0.65

qa

-0.64

POSITIVE LOGITS

 landfall

0.68

 haunt

0.68

fect

0.64

 Echoes

0.63

 takeover

0.62

APH

0.61

actor

0.60

 booth

0.60

 Bast

0.59

 aven

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.