INDEX

Explanations

words related to object-oriented programming and file manipulation

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

file and point operations

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

featureID

-0.67

 propOrder

-0.60

WriteTagHelper

-0.58

 CanadaChoose

-0.57

AnimationsModule

-0.55

CppCodeGen

-0.54

webElementXpaths

-0.53

jsxFileName

-0.52

AndEndTag

-0.52

Clik

-0.51

POSITIVE LOGITS

Bond

0.38

printList

0.38

 saites

0.37

 onResponse

0.35

bond

0.34

ByExample

0.34

Ủ

0.33

 صوتيه

0.33

package

0.33

div

0.32

Activations Density 0.044%

words related to object-oriented programming and file manipulation

file and point operations

No Comments

No Known Activations