INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

;;;;

-0.81

qv

-0.73

bda

-0.72

mington

-0.72

indal

-0.72

ibaba

-0.71

iership

-0.70

Downloadha

-0.70

antam

-0.69

POSITIVE LOGITS

 chancellor

0.71

Blocks

0.68

LOCK

0.65

enne

0.63

Operation

0.62

 Observer

0.62

clud

0.62

Mods

0.62

 Yellowstone

0.61

ysis

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.