INDEX

Explanations

numerical values, specifically related to contact information and business details

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

size after number

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

 stationnement

-0.51

 Gegenstand

-0.47

 Pacífico

-0.45

 libremente

-0.44

Життєпис

-0.42

 Parigi

-0.42

 Bibel

-0.42

 élevé

-0.42

 útiles

-0.41

 nedeniyle

-0.40

POSITIVE LOGITS

########.

0.61

tagHelperRunner

0.55

 betweenstory

0.52

IsContent

0.52

clusal

0.51

 rhestr

0.49

tinyos

0.48

ьаж

0.48

 CreateTagHelper

0.47

FieldBuilder

0.46

Activations Density 0.003%

numerical values, specifically related to contact information and business details

size after number

No Comments

No Known Activations