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

Elm

-0.73

 Wealth

-0.64

rations

-0.62

kat

-0.62

 somew

-0.61

NPC

-0.60

ME

-0.59

{"

-0.58

rons

-0.57

 NPCs

-0.57

POSITIVE LOGITS

ertodd

0.71

 overe

0.65

âĢ²

0.65

ument

0.62

thro

0.62

 grazing

0.62

Ly

0.60

eatured

0.60

orthern

0.59

©¶æ

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.