INDEX

Explanations

prominent pronouns and article usage

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

possessive pronoun followed by noun

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

ValueStyle

-0.50

 beginnetje

-0.50

новниш

-0.42

expandindo

-0.39

))));

-0.35

السكان

-0.35

AnimationsModule

-0.35

RemoveField

-0.35

 utafitiHapana

-0.34

يكب

-0.33

POSITIVE LOGITS

thym

0.48

 Folio

0.47

wagon

0.47

GOB

0.47

iſen

0.46

culti

0.46

civi

0.46

 CULT

0.46

Wet

0.46

 Empfehlungen

0.46

Activations Density 0.206%

prominent pronouns and article usage

possessive pronoun followed by noun

No Comments

No Known Activations