INDEX

Explanations

network-related commands and protocols in code snippets

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

HTTP request 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

 invokingState

-0.69

 autorytatywna

-0.55

 للمعارف

-0.53

 disponibilités

-0.49

awtextra

-0.48

èdia

-0.48

 Eura

-0.47

Hod

-0.46

 PEND

-0.46

SWIG

-0.45

POSITIVE LOGITS

 AssemblyCompany

0.43

 anledning

0.39

expandindo

0.39

yntaxException

0.39

="@+

0.37

henswürdigkeiten

0.35

itanie

0.35

 sạn

0.35

 vagyok

0.35



0.34

Activations Density 0.042%

network-related commands and protocols in code snippets

HTTP request headers

No Comments

No Known Activations