INDEX

Explanations

phrases indicating time periods or actions occurring within timeframes

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

Rolf

-0.69

 Rolf

-0.62

BrowserModule

-0.61

énario

-0.58

']?>

-0.58

 Balboa

-0.58

 Beatrix

-0.57

 Paxton

-0.55

 creatinine

-0.55

 propOrder

-0.54

POSITIVE LOGITS

during

1.43

 During

1.40

During

1.33

 DURING

1.28

durante

1.28

 during

1.26

 periods

1.10

 Durante

1.09

Durante

1.07

 durant

1.06

Activations Density 0.117%

phrases indicating time periods or actions occurring within timeframes

No Comments

No Known Activations