INDEX

Explanations

terms related to the effect or impact on health or performance metrics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 AssemblyTitle

-0.50

 soumettre

-0.48

 tunn

-0.44

artifactId

-0.43

 Lengkap

-0.40

Bakgrunnsstoff

-0.39

chenkt

-0.38

 HomeComponent

-0.38

"]

-0.37

具备

-0.37

POSITIVE LOGITS

 affects

0.81

 affected

0.80

 Affect

0.78

 affecting

0.78

 affect

0.77

 Affecting

0.77

affect

0.77

 afect

0.76

affected

0.76

 influencing

0.76

Activations Density 0.396%

terms related to the effect or impact on health or performance metrics

No Comments

No Known Activations