INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 treasury

-0.72

 detrim

-0.72

 beginnings

-0.72

 cassette

-0.71

 millenn

-0.66

 confessions

-0.64

Zin

-0.62

 memos

-0.61

 syll

-0.61

 Telegram

-0.60

POSITIVE LOGITS

joy

0.77

ä¸ī

0.76

intend

0.74

distance

0.73

 isEnabled

0.72

tackle

0.72

bang

0.69

levard

0.68

lev

0.67

aire

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.