INDEX

Explanations

elements related to layers and structure within programming or graphical contexts

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

layer or track relationships

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

WriteTagHelper

-0.56

BufferException

-0.48

المشاركات

-0.46

rungsseite

-0.45

awaiter

-0.45

SequentialGroup

-0.44

principalColumn

-0.44

contentLoaded

-0.43

ANTLR

-0.43

enumi

-0.42

POSITIVE LOGITS

 esternos

0.46

Hymen

0.46

 Layer

0.45

 layer

0.44

 Ligne

0.44

 layers

0.43

bpy

0.43

 Layers

0.42

 Roskov

0.42

GtkWidget

0.40

Activations Density 0.013%

elements related to layers and structure within programming or graphical contexts

layer or track relationships

No Comments

No Known Activations