INDEX

Explanations

references to doors and their mechanisms

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Palmas

-0.83

 Allociné

-0.76

"")

-0.74

liesslich

-0.73

")),

-0.72

 Macon

-0.72

'}),

-0.72

__))

-0.71

 vermogen

-0.71

 mijne

-0.70

POSITIVE LOGITS

 doors

1.68

 door

1.65

 Doors

1.53

 Door

1.53

door

1.51

 DOOR

1.49

Door

1.41

Doors

1.40

doors

1.33

DOOR

1.19

Activations Density 0.047%

references to doors and their mechanisms

No Comments

No Known Activations