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

fast

-0.69

 Sapphire

-0.68

swick

-0.68

anguard

-0.68

DNA

-0.67

imb

-0.66

Pear

-0.66

 Commodore

-0.66

vic

-0.65

 Pearl

-0.64

POSITIVE LOGITS

COLOR

0.68

 Âł Âł Âł Âł Âł Âł Âł Âł

0.67

rils

0.65

 Âł Âł Âł Âł

0.63

olor

0.63

 professions

0.62

=-=-

0.62

 altogether

0.60

luaj

0.60

 JUSTICE

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.