INDEX

Explanations

references to specific groups or categories in a detailed context

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

endphp

-0.38

rawQuery

-0.34

 year

-0.32

 course

-0.32

 money

-0.30

 resulting

-0.30

req

-0.29

 sour

-0.29

Fazit

-0.29

yes

-0.29

POSITIVE LOGITS

others

0.85

 others

0.81

 anderen

0.80

 otros

0.76

 other

0.75

Others

0.74

 diğer

0.73

autres

0.71

 innych

0.69

 autres

0.68

Activations Density 0.537%

references to specific groups or categories in a detailed context

No Comments

No Known Activations