INDEX

Explanations

requests for information or assistance

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

 Roskov

-0.51

Diweddarwch

-0.49

titleMargin

-0.49

 pulumi

-0.48

OGND

-0.48

таратура

-0.47

 للاسماء

-0.46

 հղումներ

-0.46

insee

-0.44

PropertyChanging

-0.44

POSITIVE LOGITS

 himself

0.42

 bisous

0.42

himself

0.39

who

0.39

انتهای

0.38

новништво

0.38

Val

0.37

<!--[

0.36

AccessorTable

0.36

Ingredienti

0.35

Activations Density 0.014%

requests for information or assistance

No Comments

No Known Activations