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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uli

-0.69

loo

-0.69

 Gloves

-0.68

 Logged

-0.64

pointers

-0.63

Ski

-0.62

nesota

-0.62

 Gale

-0.62

isoft

-0.62

 guides

-0.61

POSITIVE LOGITS

rant

0.85

ted

0.77

ting

0.70

auder

0.66

dam

0.64

 Juda

0.63

lain

0.62

asar

0.61

EXT

0.60

ively

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.