INDEX

Explanations

elements related to programming interfaces and records in code

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

code includes or headers

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

 AssemblyCulture

-0.64

IntoConstraints

-0.56

webElementXpaths

-0.56

 oprot

-0.56

ьажоргаш

-0.50

SharedDtor

-0.49

EDEFAULT

-0.48

preduce

-0.48

ValueGenerated

-0.48

achios

-0.47

POSITIVE LOGITS

 <<<<<<<<<<<<<<

0.38

heim

0.38

HEIM

0.33

Emer

0.32

 CanadaChoose

0.32

Tr

0.32

 חיצוניים

0.31

ThroughAttribute

0.30

ModelAdmin

0.30

最快更新

0.29

Activations Density 0.001%

elements related to programming interfaces and records in code

code includes or headers

No Comments

No Known Activations