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

ibrary

-0.63

 Greenberg

-0.62

 shorts

-0.60

 coron

-0.60

IDA

-0.59

 obsc

-0.59

 foreskin

-0.58

 compartment

-0.58

 sweats

-0.57

 Neuroscience

-0.56

POSITIVE LOGITS

REC

0.78

terness

0.78

sem

0.76

ience

0.71

rate

0.71

hire

0.69

ffee

0.68

beat

0.68

rov

0.68

stead

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.