INDEX

Explanations

programming and console logging statements in code

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

code segments

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

InjectAttribute

-0.73

 <<<<<<<<<<<<<<

-0.67

 AssemblyCulture

-0.65

WriteTagHelper

-0.65

Искәрмәләр

-0.63

ContentAsync

-0.63

 StatelessWidget

-0.62

BufferException

-0.61

دانشنامهٔ

-0.60

addPreferredGap

-0.60

POSITIVE LOGITS

 accuracy

0.35

 whip

0.33

vé

0.30

 random

0.29

 Buche

0.29

 cualquiera

0.29

 dura

0.28

 ACCUR

0.28

 orang

0.28

 comfort

0.28

Activations Density 0.082%

programming and console logging statements in code

code segments

No Comments

No Known Activations