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

gins

-0.74

asus

-0.74

minus

-0.69

atis

-0.68

BuyableInstoreAndOnline

-0.68

ators

-0.67

orns

-0.67

renheit

-0.66

ioxide

-0.64

ainment

-0.64

POSITIVE LOGITS

ãĥ³ãĤ¸

0.75

:{

0.65

lihood

0.62

juven

0.62

Merit

0.62

 PROT

0.61

{"

0.61

}}

0.61

prot

0.61

Zup

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.