INDEX

Explanations

institutional affiliations and locations related to academic or research entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 resourceCulture

-1.05

 intptr

-0.88

awtextra

-0.86

 jsPsych

-0.84

 propOrder

-0.83

AndEndTag

-0.79

 يتيمه

-0.79

PreferredItem

-0.77

TagMode

-0.75

دانشنامهٔ

-0.75

POSITIVE LOGITS

 apologise

0.47

 angesch

0.43

oherty

0.41

 chercheurs

0.39

 tearDown

0.38

Gruß

0.38

 apologised

0.38

 Housewives

0.38

یتی

0.37

AfterViewInit

0.37

Activations Density 0.532%

institutional affiliations and locations related to academic or research entities

No Comments

No Known Activations