INDEX

Explanations

references to canned products or packaging

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

canned, cans, Canning

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

Predecesor

-0.43

Ful

-0.40

े

-0.40

istle

-0.40

 path

-0.39

 area

-0.38

ing

-0.37

itoriale

-0.37

AttributeSet

-0.37

XmlType

-0.37

POSITIVE LOGITS

 canned

1.78

 Canned

1.70

 cans

1.38

 Cans

1.21

 canning

1.20

 Canning

0.98

 latas

0.95

罐

0.86

anned

0.85

cans

0.84

Activations Density 0.006%

references to canned products or packaging

canned, cans, Canning

No Comments

No Known Activations