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

DragonMagazine

-0.90

 framing

-0.68

umbai

-0.66

asca

-0.64

 lightsaber

-0.64

 Lans

-0.64

 Riders

-0.63

 Curve

-0.62

 splits

-0.61

 Reading

-0.61

POSITIVE LOGITS

amn

0.79

aque

0.77

oster

0.76

iji

0.71

OO

0.70

proc

0.69

Pg

0.67

chy

0.66

mobi

0.66

urally

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.