INDEX

Explanations

references to categories or classifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

}}\

-0.54

}\

-0.49

 ioutil

-0.47

-0.45

}\

-0.45

 doPost

-0.44

}}\

-0.43

())))

-0.42

<bos>

-0.41

]\

-0.41

POSITIVE LOGITS

 Category

1.09

 category

1.07

 getCategory

1.06

getCategory

1.02

 CATEGORY

1.02

Category

0.99

category

0.98

CATEGORY

0.91

ategory

0.90

 Categories

0.88

Activations Density 0.063%

references to categories or classifications

No Comments

No Known Activations