INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

anz

-0.69

ulations

-0.68

Wax

-0.67

uers

-0.66

antz

-0.64

ãĤĵ

-0.63

ulpt

-0.62

 Fell

-0.61

getic

-0.61

abl

-0.60

POSITIVE LOGITS

/-

0.76

inion

0.71

gencies

0.71

IME

0.70

 mishand

0.63

ammy

0.60

dom

0.60

aunder

0.59

ife

0.58

 therefore

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.