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

soever

-0.69

Ĥª

-0.68

 pity

-0.67

 dreaded

-0.65

 undone

-0.64

 inactive

-0.62

 Redux

-0.62

 pleasant

-0.61

 bene

-0.61

bable

-0.61

POSITIVE LOGITS

elf

0.77

ultan

0.76

erker

0.74

iggs

0.74

achus

0.73

ymph

0.73

rss

0.72

sylv

0.71

odcast

0.70

ewski

0.70

Activations Density 0.000%

No Known Activations

This feature has no known activations.