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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

isions

-0.82

 Clarks

-0.78

uci

-0.71

rac

-0.69

 Dayton

-0.69

RET

-0.68

ND

-0.68

encer

-0.67

DEV

-0.67

uron

-0.67

POSITIVE LOGITS

 feast

0.70

gered

0.70

 advantage

0.65

rely

0.65

></

0.64

 amusement

0.63

 playful

0.62

 banquet

0.60

ews

0.60

 implanted

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.