INDEX

Explanations

complex technical structures and formatting in data or programming contexts

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

The letter "l" followed by various characters

np_acts-logits-general · gemini-2.0-flash

specific tokens followed by 'l' or 'L'

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 '\\;'

-0.43

 hini

-0.40

poj

-0.40

Personendaten

-0.36

OfWork

-0.35

 Dhoni

-0.35

 Copernicus

-0.33

 στε

-0.33

νος

-0.33

POSITIVE LOGITS

0.78

0.74

LP

0.71

 league

0.70

 liters

0.68

 lieutenant

0.68

LO

0.67

 linkovi

0.67

Lieutenant

0.67

Lee

0.67

Activations Density 0.776%

complex technical structures and formatting in data or programming contexts

The letter "l" followed by various characters

specific tokens followed by 'l' or 'L'

No Comments

No Known Activations