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

strings

-0.73

 Christy

-0.69

rules

-0.66

sburg

-0.65

 Romans

-0.65

](

-0.62

 Olsen

-0.62

enegger

-0.61

istics

-0.60

 Greene

-0.60

POSITIVE LOGITS

£ı

1.05

 hemor

0.87

senal

0.81

channelAvailability

0.77

Reviewer

0.75

 newsp

0.72

 distingu

0.72

VK

0.68

ibo

0.67

keley

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.