INDEX

Explanations

frequent words and pronouns that emphasize connection and personal reflection

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

from the

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

存于互联网档案馆

-0.67

 мәкал

-0.61

tagHelperRunner

-0.60

AsUp

-0.58

VIAF

-0.58

sizeCache

-0.57

 pinulongan

-0.56

__(/*!

-0.55

 WaitForSeconds

-0.55

 AssemblyTitle

-0.55

POSITIVE LOGITS

 olmayan

0.42

地

0.40

<strong>

0.40

eners

0.39

 royal

0.38

shit

0.37

 revs

0.37

地の

0.36

 Bancroft

0.36

The

0.36

Activations Density 0.126%

frequent words and pronouns that emphasize connection and personal reflection

from the

No Comments

No Known Activations