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

hea

-0.71

idia

-0.66

ivan

-0.64

identally

-0.63

arbon

-0.62

seeing

-0.62

 sweats

-0.61

Kids

-0.61

opian

-0.60

oping

-0.60

POSITIVE LOGITS

¶ħ

0.75

genre

0.73

mania

0.69

èĢħ

0.68

ãĤ¢ãĥ«

0.67

%%

0.67

ouver

0.67

rez

0.66

 Viking

0.65

 Stras

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.