INDEX

Explanations

possessive pronouns and words that suggest ownership or personalization

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

Geplaatst

-1.07

tvguidetime

-1.01

RTLR

-0.90

ValueStyle

-0.89

 Waray

-0.86

 Мексичка

-0.85

Personensuche

-0.82

InjectAttribute

-0.82

+#+

-0.81

Portale

-0.81

POSITIVE LOGITS

원

0.53

Old

0.43

 Stelle

0.43

 де

0.42

ur

0.42

Makes

0.42

比

0.42

0.41

SI

0.41

Activations Density 0.250%

possessive pronouns and words that suggest ownership or personalization

No Comments

No Known Activations