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

©¶æ

-0.84

acts

-0.77

cius

-0.73

otin

-0.70

Children

-0.69

alted

-0.68

rouse

-0.67

eatures

-0.66

aneous

-0.66

olds

-0.66

POSITIVE LOGITS

 passer

0.67

 Haley

0.67

gib

0.67

nah

0.66

 otherwise

0.66

Ã±a

0.65

 Danielle

0.61

kowski

0.61

mary

0.61

 reck

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.