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

agonist

-0.73

earcher

-0.64

ribes

-0.61

 Cato

-0.61

ournal

-0.60

Rai

-0.60

ional

-0.60

duo

-0.58

 Authors

-0.57

uru

-0.57

POSITIVE LOGITS

llan

0.79

rican

0.72

govtrack

0.71

ongyang

0.70

adelphia

0.67

ippi

0.66

 goose

0.66

rolet

0.64

kiss

0.64

river

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.