INDEX

Explanations

emotional expressions and references to personal experiences

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

feeling negative states

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

RectangleBorder

-0.63

enumii

-0.58

ArgumentParser

-0.53

 Proble

-0.52

 PROBLE

-0.52

PyExc

-0.52

NameInMap

-0.51

 problematic

-0.51

 समीक्षक

-0.51

argout

-0.50

POSITIVE LOGITS

 realizing

0.66

 inability

0.66

 unable

0.62

 losing

0.58

 realize

0.57

 realising

0.56

 realizes

0.54

 realization

0.54

 seeing

0.54

 kehilangan

0.54

Activations Density 0.087%

emotional expressions and references to personal experiences

feeling negative states

No Comments

No Known Activations