INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

xxx

-0.71

 Reloaded

-0.67

Developer

-0.67

zzo

-0.66

atis

-0.66

QB

-0.65

().

-0.64

Attribute

-0.63

=-=-

-0.62

 Socket

-0.62

POSITIVE LOGITS

 Folk

0.70

hran

0.69

mson

0.68

miah

0.65

 turnout

0.65

 Cumm

0.64

Synopsis

0.62

chet

0.62

duino

0.61

mobi

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.