INDEX

Explanations

instructions for preparing or processing food items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ujednoznacz

-0.72

dymyr

-0.67

enumi

-0.66

 Pourtant

-0.63

avyzd

-0.60

 للمعارف

-0.57

 ligiloj

-0.56

 فريبيس

-0.55

owal

-0.55

+:+

-0.54

POSITIVE LOGITS

 preferably

0.81

 Preferably

0.68

preferably

0.62

 предвари

0.59

EndInit

0.59

 gently

0.53

½

0.51

 previamente

0.51

 preferable

0.50

最好

0.50

Activations Density 0.062%

instructions for preparing or processing food items

No Comments

No Known Activations