INDEX

Explanations

terms related to morning and evening activities or news broadcasts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

HasKey

-0.90

.*")]

-0.87

styrelsen

-0.81

ajuato

-0.81

AndEndTag

-0.80

 виправивши

-0.80

 politiet

-0.78

UVWXYZ

-0.78

 considérons

-0.77

estries

-0.75

POSITIVE LOGITS

 morning

1.77

 Morning

1.67

 mornings

1.56

morning

1.52

 MORNING

1.51

Morning

1.47

 Morn

1.32

 Morgen

1.20

 morn

1.16

 afternoon

1.10

Activations Density 0.042%

terms related to morning and evening activities or news broadcasts

No Comments

No Known Activations