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

adan

-0.79

 pand

-0.75

humans

-0.74

immune

-0.70

lik

-0.68

emo

-0.68

ican

-0.68

ividual

-0.64

omal

-0.63

 è£ıè

-0.63

POSITIVE LOGITS

 Carroll

0.69

 ---------

0.67

 Verse

0.66

 Nept

0.64

 Gilmore

0.63

REDACTED

0.62

 Rollins

0.62

Urs

0.62

MpServer

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.