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

MIT

-0.83

BIP

-0.77

GET

-0.75

JUST

-0.73

NH

-0.72

SEE

-0.70

TX

-0.70

LOAD

-0.70

 bask

-0.69

gd

-0.69

POSITIVE LOGITS

 Fever

0.74

 ozone

0.69

 Isles

0.69

 Gors

0.65

 Flavoring

0.64

 Death

0.64

ensable

0.62

 Bever

0.62

 Dough

0.61

iability

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.