INDEX

Explanations

elements related to precision, technical specifications, and programming concepts

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

technical documentation and programming terms

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

 rossa

-0.40

 trud

-0.35

sí

-0.33

RTGC

-0.33

 ListNode

-0.32

 merah

-0.31

 <>",

-0.30

 NSObject

-0.30

外部リンク

-0.29

Red

-0.29

POSITIVE LOGITS

vician

0.73

帖最后由

0.67

OGND

0.59

 betweenstory

0.59

PostInfinity

0.59

richTextPanel

0.57

+#+

0.56

 ब्रेकडाउन

0.51

⅘

0.51

PerformLayout

0.50

Activations Density 0.224%

elements related to precision, technical specifications, and programming concepts

technical documentation and programming terms

No Comments

No Known Activations