INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

jri

-0.82

Court

-0.75

proc

-0.74

shall

-0.74

 cooks

-0.73

 exceptions

-0.69

soType

-0.68

grain

-0.67

 Contrary

-0.67

encers

-0.66

POSITIVE LOGITS

 stake

0.75

Pac

0.70

roups

0.66

ascus

0.64

asus

0.64

 plank

0.64

Peg

0.63

urated

0.62

antis

0.62

foe

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.