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

ORTS

-0.69

 clos

-0.66

imental

-0.65

iments

-0.65

eworld

-0.64

hw

-0.63

BuyableInstoreAndOnline

-0.63

 relics

-0.63

Reviewer

-0.62

ths

-0.61

POSITIVE LOGITS

Äĩ

0.70

©¶æ

0.68

ple

0.67

 Conversation

0.64

charged

0.61

 Speech

0.61

rie

0.61

dro

0.59

 Required

0.59

Required

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.