INDEX

Explanations

references to visual documentation, particularly photographs and screenshots

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

Hochspringen

-0.66

Referințe

-0.65

 préfé

-0.58

 colic

-0.57

 Thessal

-0.56

bitat

-0.56

 Krakow

-0.55

 senses

-0.54

 Erben

-0.52

 نهر

-0.52

POSITIVE LOGITS

 photo

0.72

 screenshot

0.70

 screenshots

0.70

 selfies

0.68

AnchorStyles

0.65

 selfie

0.64

 photos

0.63

 foto

0.62

Screenshot

0.61

 Screenshot

0.60

Activations Density 0.235%

references to visual documentation, particularly photographs and screenshots

No Comments

No Known Activations