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

ĸļ

-0.81

heid

-0.72

sworth

-0.67

HAM

-0.65

mans

-0.65

ahu

-0.63

kid

-0.63

<<

-0.63

ciating

-0.61

glers

-0.60

POSITIVE LOGITS

TPPStreamerBot

0.77

ossibility

0.74

OGR

0.71

 helicop

0.70

avascript

0.68

ople

0.66

 confir

0.63

PF

0.62

ntil

0.62

Orderable

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.