INDEX

Explanations

data and concepts associated with health and wellness

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

timestamps and citations

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

 Cast

-0.45

fromnode

-0.39

leyebilirsiniz

-0.37

uscita

-0.35

Bec

-0.35

ậm

-0.34

 pelan

-0.34

ագրություն

-0.34

zonder

-0.34

utilisons

-0.34

POSITIVE LOGITS

AddTagHelper

0.85

bootstrapcdn

0.62

 invokingState

0.61

 nahilalakip

0.60

 kaarangay

0.56

migrationBuilder

0.54

 tartalomajánló

0.54

 vPvB

0.52

AnimationsModule

0.51

TagMode

0.50

Activations Density 0.023%

data and concepts associated with health and wellness

timestamps and citations

No Comments

No Known Activations