INDEX

Explanations

references to specific numbers or quantities

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

after "the", "both", "each", "latter"

np_acts-logits-general · gemini-2.0-flash

both and comparison

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 28-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_28/width_16k/average_l0_119

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.28.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

BagLayout

-0.67

 Chwiliwch

-0.65

DockStyle

-0.65

 propOrder

-0.63

 AspNetCore

-0.63

OGND

-0.59

BagConstraints

-0.59

zzleHttp

-0.58

Mutagenicity

-0.58

 ProtoMessage

-0.58

POSITIVE LOGITS

 project

0.40

 prosjek

0.40

 peça

0.37

作品

0.36

<?

0.36

 program

0.35

 azért

0.35

 tool

0.33

 piece

0.32

 اثر

0.32

Activations Density 1.332%

references to specific numbers or quantities

after "the", "both", "each", "latter"

both and comparison

No Comments

No Known Activations