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

 Nest

-0.65

amy

-0.64

pict

-0.63

 dime

-0.63

vich

-0.63

aho

-0.62

 stripes

-0.61

ija

-0.59

zac

-0.59

icka

-0.58

POSITIVE LOGITS

 lawy

0.74

ĸļ

0.70

olkien

0.70

Gs

0.69

hyde

0.69

«

0.68

ertodd

0.68

theless

0.67

 destro

0.65

dayName

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.