INDEX

Explanations

intensifiers or modifiers that express completeness or totality in a statement

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

@[+][

-0.67

 braccia

-0.67

Datuak

-0.66

 navideño

-0.65

 Connectez

-0.64

 doulou

-0.64

جوايز

-0.64

 phosphory

-0.63

 Junge

-0.63

 ولد

-0.62

POSITIVE LOGITS

 completely

2.22

 Completely

2.14

 totally

2.08

Completely

2.06

 Totally

1.95

completely

1.94

 entirely

1.93

Totally

1.88

totally

1.80

 totalmente

1.58

Activations Density 0.139%

intensifiers or modifiers that express completeness or totality in a statement

No Comments

No Known Activations