INDEX

Explanations

counts or numerical values

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

code snippets with line numbers and colons

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

 تضيفلها

-0.50

 terecht

-0.48

UrlResolution

-0.48

DECREF

-0.46

GOTREF

-0.46

BASEPATH

-0.44

 awakeFromNib

-0.44

StoryboardSegue

-0.44

 OnInit

-0.44

OCCURRED

-0.44

POSITIVE LOGITS

 betweenstory

0.66

 simplifié

0.46

Boe

0.44

 autorytatywna

0.41

 Breakthrough

0.41

LR

0.40

transQ

0.40

expandindo

0.39



0.39

sau

0.39

Activations Density 0.002%

counts or numerical values

code snippets with line numbers and colons

No Comments

No Known Activations