INDEX

Explanations

terms related to competence and competency

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

competence

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

-0.40

 Navidad

-0.38

as

-0.38

-0.35

“

-0.35

image

-0.34

 Otra

-0.34

//

-0.34

 gift

-0.33

POSITIVE LOGITS

 Competence

1.16

 competence

1.10

competent

1.09

 competencies

1.07

compet

1.07

 Competency

1.02

Compet

1.02

 Compet

1.02

 competency

1.01

 competent

0.95

Activations Density 0.006%

terms related to competence and competency

competence

No Comments

No Known Activations