INDEX

Explanations

date and time-related formats

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

timestamps and dates

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

Kub

-0.40

Lähteet

-0.39

 الأخ

-0.36

 оку

-0.35

 Denk

-0.35

 site

-0.35

 Betrieb

-0.34

 Segel

-0.34

目が

-0.34

ยุ

-0.33

POSITIVE LOGITS

 instagram

0.60



0.57

Instagram

0.57

 Instagram

0.54

instagram

0.53

0.51

 photoshoot

0.51

 fotografico

0.50

 oprot

0.50

 nahilalakip

0.49

Activations Density 0.001%

date and time-related formats

timestamps and dates

No Comments

No Known Activations