INDEX

Explanations

phrases that reference significant events or statements about individuals, particularly in the context of serious situations

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

conversational starters

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

 ProtoMessage

-0.71

 tartalomajánló

-0.67

 ujednoznacz

-0.62

 estekak

-0.61

+#+#

-0.60

 препратки

-0.56

URLException

-0.55

tagext

-0.55

 queſta

-0.54

Datuak

-0.53

POSITIVE LOGITS

 bomb

0.40

 load

0.36

 sulf

0.35

sal

0.34

bom

0.34

nestjs

0.34

bug

0.34

$_['

0.34

nav

0.33

az

0.33

Activations Density 0.111%

phrases that reference significant events or statements about individuals, particularly in the context of serious situations

conversational starters

No Comments

No Known Activations