INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

agogue

-0.67

 cook

-0.61

 risking

-0.60

eering

-0.60

going

-0.60

 accusing

-0.60

agog

-0.60

 Scholarship

-0.60

 BaseType

-0.59

eating

-0.58

POSITIVE LOGITS

Reviewer

0.77

iosyn

0.71

olor

0.70

 thous

0.67

units

0.67

ITS

0.65

owship

0.63

OME

0.63

ENS

0.63

Denver

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.