INDEX

Explanations

punctuation and syntax elements typical in programming code

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

function call boundaries

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

principalTable

-0.59

GOTREF

-0.52

 CURIAM

-0.47

migrationBuilder

-0.45

LabelTagHelper

-0.44

makeText

-0.43

 doInBackground

-0.42

ynb

-0.41

raulic

-0.40

 interfering

-0.40

POSITIVE LOGITS

GEBURTSDATUM

0.51

Personendaten

0.48

sizeCache

0.47

Lähteet

0.42

ställning

0.40

exitRule

0.40

TacToe

0.39

Filmo

0.39

省市镇

0.39

AttributeSet

0.38

Activations Density 0.008%

punctuation and syntax elements typical in programming code

function call boundaries

No Comments

No Known Activations