INDEX

Explanations

expressions related to assumptions and expectations about experiences or qualities being taken for granted

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

assumptions and granted

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Durchmesser

-0.45

 beliau

-0.38

 íntegra

-0.38

 jurí

-0.37

 betreffenden

-0.36

 internationalen

-0.36

 jurídico

-0.36

Jîn

-0.36

 Temperaturen

-0.36

帖最后由

-0.36

POSITIVE LOGITS

 disambiguazione

0.59

 مرئيه

0.53

 Assumptions

0.53

zzleHttp

0.52

RTLD

0.51

 assumed

0.51

 assumptions

0.51

 المعيارى

0.50

assumed

0.49

 assumption

0.48

Activations Density 0.031%

expressions related to assumptions and expectations about experiences or qualities being taken for granted

assumptions and granted

No Comments

No Known Activations