INDEX

Explanations

intensifiers that enhance the degree of an adjective or adverb

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 absc

-0.55

γων

-0.54

zec

-0.53

 insegna

-0.53

richTextPanel

-0.52

jercicios

-0.51

 JPanel

-0.51

 vera

-0.51

EClass

-0.49

appBar

-0.49

POSITIVE LOGITS

 highly

3.59

highly

3.39

 Highly

3.27

Highly

3.27

 altamente

2.40

 heavily

1.86

 strongly

1.71

 hugely

1.58

 fortement

1.58

 greatly

1.45

Activations Density 0.091%

intensifiers that enhance the degree of an adjective or adverb

No Comments

No Known Activations