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

 Malf

-0.63

-0.61

 regards

-0.61

 airports

-0.61

 CTRL

-0.60

 hindsight

-0.59

 Wars

-0.57

 Spit

-0.57

ã

-0.56

MIL

-0.56

POSITIVE LOGITS

raq

0.77

 sender

0.69

ija

0.69

gra

0.67

phabet

0.67

berman

0.65

izont

0.65

nesty

0.64

ktop

0.64

abi

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.