INDEX

Explanations

references to movies and related entertainment content

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

바

-0.47

žem

-0.46

 navigazione

-0.45

či

-0.45

AxisAlignment

-0.44

 instrumentos

-0.44

Msp

-0.42

forhold

-0.41

łąd

-0.41

 músicos

-0.41

POSITIVE LOGITS

 movies

1.68

 movie

1.67

 films

1.56

 cinema

1.53

 Movies

1.46

 film

1.43

 Movie

1.43

 cinemas

1.39

movie

1.38

 MOVIE

1.37

Activations Density 0.296%

references to movies and related entertainment content

No Comments

No Known Activations