INDEX

Explanations

terms related to processes of understanding, development, and interaction

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

related to context or tool

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

 Numerade

-0.61

FlatList

-0.52

 duquel

-0.51

MLLoader

-0.50

 mahdol

-0.50

 kautta

-0.48

 jopa

-0.48

keyColumn

-0.48

 avulla

-0.48

 therefore

-0.47

POSITIVE LOGITS

脚注の使い方

0.47

 مشين

0.45

PyExc

0.42

Rujuakan

0.40

 sumpay

0.39

Portale

0.39

PerformLayout

0.39

 चीज़ों

0.37

xase

0.37

.*")]

0.35

Activations Density 0.037%

terms related to processes of understanding, development, and interaction

related to context or tool

No Comments

No Known Activations