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

argon

-0.71

Lewis

-0.70

 Matter

-0.68

Beck

-0.66

gow

-0.66

Ast

-0.65

Constructed

-0.65

Minor

-0.64

Sensor

-0.63

Rail

-0.62

POSITIVE LOGITS

TABLE

0.83

FTWARE

0.74

urses

0.73

 teasp

0.71

hower

0.69

 arrang

0.67

 confir

0.66

ards

0.65

theless

0.65

EStream

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.