INDEX

Explanations

references to carpets or rugs

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

carpet

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

 Buchanan

-0.36

 chua

-0.35

<<"

-0.34

 Zeus

-0.33

"])

-0.33

'],'

-0.32

'])

-0.32

 Nova

-0.32

})

-0.32

 Chua

-0.31

POSITIVE LOGITS

 carpet

2.08

Carpet

2.06

 Carpet

2.06

carpet

1.97

 carpets

1.70

 alfombra

1.47

 tapete

1.00

peted

0.95

rug

0.94

 tapis

0.94

Activations Density 0.002%

references to carpets or rugs

carpet

No Comments

No Known Activations