INDEX

Explanations

sections describing the field and context of inventions

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

stressed

-0.46

 Mazz

-0.45

 llorar

-0.43

过了

-0.42

sempre

-0.41

自带

-0.41

nacional

-0.41

whenever

-0.41

sequently

-0.40

vyš

-0.40

POSITIVE LOGITS

 EconPapers

0.60

 Signalez

0.54

IsMutable

0.51

 disambiguazione

0.50

 resourceCulture

0.49

 tartalomajánló

0.49

 nahilalakip

0.48

ViewImports

0.48

 stanovnika

0.47

 Roskov

0.47

Activations Density 0.007%

sections describing the field and context of inventions

No Comments

No Known Activations