INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

URI

-0.78

é¾

-0.77

DX

-0.74

é¾įå

-0.69

 motto

-0.67

 Maze

-0.66

 Achieve

-0.66

 Donation

-0.62

 Hardcore

-0.61

DIT

-0.61

POSITIVE LOGITS

pt

0.65

rie

0.65

 cons

0.64

cedes

0.63

etch

0.62

eus

0.61

ovy

0.60

erek

0.59

 Available

0.58

yrs

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.