INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

place

-0.67

olo

-0.66

frog

-0.66

pring

-0.63

IQ

-0.62

 neglect

-0.61

 smoking

-0.61

mith

-0.61

ATOR

-0.61

 outright

-0.61

POSITIVE LOGITS

ĸļ

0.94

 Seraph

0.75

isine

0.73

 Sailor

0.67

Ô

0.66

Untitled

0.66

ukong

0.64

 satell

0.63

foreseen

0.63

[/

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.