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

locks

-0.73

locked

-0.71

 Zamb

-0.69

obe

-0.69

abor

-0.64

lock

-0.62

vati

-0.62

lyn

-0.61

 disob

-0.61

 corro

-0.60

POSITIVE LOGITS

":"/

0.93

ãĤ¤ãĥĪ

0.86

icago

0.75

://

0.73

ãĥīãĥ©

0.69

çīĪ

0.69

ãĥĳ

0.68

ãĤ¢ãĥ«

0.67

ãĤ½

0.67

eas

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.