INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

lynn

-0.68

 borrow

-0.67

 entit

-0.66

 Representatives

-0.63

 authorized

-0.63

 guiActiveUn

-0.61

 Sail

-0.60

 Frenchman

-0.59

 passage

-0.58

 stip

-0.58

POSITIVE LOGITS

BILITY

0.78

iband

0.75

isky

0.70

CLASSIFIED

0.70

batch

0.68

essen

0.67

images

0.65

eared

0.65

endum

0.64

nces

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.