INDEX

Explanations

references to drinking vessels and measurements related to beverages

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

cups or glasses

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

shooting

-0.42

 injecting

-0.41

Proveedor

-0.41

lossing

-0.40

licar

-0.40

 رأ

-0.39

 обе

-0.38

hooting

-0.38

processing

-0.38

threw

-0.38

POSITIVE LOGITS

cup

0.77

 cups

0.75

 Cups

0.69

 autorytatywna

0.63

AsUp

0.60

 bardak

0.59

 tumbler

0.57

cup

0.57

Cup

0.56

writeFieldEnd

0.56

Activations Density 0.063%

references to drinking vessels and measurements related to beverages

cups or glasses

No Comments

No Known Activations