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

acet

-0.79

heric

-0.75

ope

-0.74

ascript

-0.74

yle

-0.72

ilateral

-0.72

adian

-0.70

emort

-0.70

uffs

-0.69

 expend

-0.68

POSITIVE LOGITS

WER

0.75

é¾

0.66

Cul

0.65

 Drift

0.65

 Come

0.64

 Dodd

0.63

](

0.63

OU

0.62

UAL

0.62

 Fitz

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.