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

 Sunshine

-0.74

 Investigative

-0.70

 Geek

-0.68

 Sandwich

-0.68

eries

-0.66

ffen

-0.65

icio

-0.62

ery

-0.61

ibrary

-0.60

cean

-0.60

POSITIVE LOGITS

ãĤ®

0.71

 sidx

0.70

ãĤ¸

0.70

 exting

0.69

oleon

0.67

ãĤ»

0.66

cx

0.66

abwe

0.65

 therap

0.65

urate

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.