INDEX

Explanations

references to mathematical or programming notation

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

EDEFAULT

-0.83



-0.68

 onBind

-0.65

 endforeach

-0.62

 Paglinawan

-0.55

וויק

-0.55

+#+

-0.53

 OFDb

-0.52

 exceptions

-0.52

getMinutes

-0.52

POSITIVE LOGITS

どちら

0.74

 whichever

0.72

 peindre

0.71

 either

0.67

 mourut

0.65

どっち

0.65

 Either

0.64

either

0.62

 avoient

0.62

Either

0.61

Activations Density 0.330%

references to mathematical or programming notation

No Comments

No Known Activations