INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

onson

-0.74

eton

-0.71

*/(

-0.68

achel

-0.68

btn

-0.67

 Wells

-0.66

livious

-0.66

ably

-0.65

å§«

-0.64

ubb

-0.64

POSITIVE LOGITS

Plot

0.71

WithNo

0.69

 Imag

0.68

FTWARE

0.68

opsis

0.64

Origin

0.63

 Demonic

0.62

 Mehran

0.61

 Moons

0.58

 Leaves

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.