INDEX

Explanations

references to years and dates

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

year and month designations

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

 gynhyrchwyd

-0.46

 חיצוניים

-0.45

__(/*!

-0.44

BASEPATH

-0.44

 Winaray

-0.43

balleur

-0.43

 propOrder

-0.41

archiviato

-0.40

ThroughAttribute

-0.40

artney

-0.39

POSITIVE LOGITS

Year

1.11

 Year

1.05

 year

1.00

 YEAR

0.93

YEAR

0.91

getYear

0.89

year

0.85

 año

0.81

 Tahun

0.77

 année

0.75

Activations Density 0.012%

references to years and dates

year and month designations

No Comments

No Known Activations