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

enegger

-0.84

wid

-0.71

zanne

-0.69

 native

-0.67

 mansion

-0.66

 wink

-0.63

 alias

-0.61

 maple

-0.61

 West

-0.60

oise

-0.60

POSITIVE LOGITS

Ħ¢

0.80

 Chains

0.77

skirts

0.75

Ã±

0.72

eering

0.70

ships

0.67

 sauces

0.67

Ĥª

0.66

 corrid

0.65

uses

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.