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

ãģį

-0.62

hay

-0.62

 ivory

-0.61

 knots

-0.58

about

-0.58

mys

-0.57

 discord

-0.56

arium

-0.56

 sweets

-0.56

 revel

-0.55

POSITIVE LOGITS

nesday

0.76

quickShipAvailable

0.76

arsity

0.74

Army

0.74

sbm

0.71

 Army

0.71

 Username

0.71

avier

0.70

 Plane

0.69

ORGE

0.69

Activations Density 0.000%

No Known Activations

This feature has no known activations.