INDEX

Explanations

comparative phrases and connections related to control groups

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

un- words

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

 autorytatywna

-0.47

ChildScrollView

-0.42

 onBind

-0.39

Autoritní

-0.38

ötzlich

-0.37

 OnDestroy

-0.36

ultuous

-0.35

XmlAttribute

-0.34

withOpacity

-0.33

SequentialGroup

-0.33

POSITIVE LOGITS

un

2.11

未

1.72

uns

1.63

Un

1.52

 unre

1.49

unt

1.48

Un

1.45

未

1.35

 unas

1.34

 untreated

1.33

Activations Density 0.636%

comparative phrases and connections related to control groups

un- words

No Comments

No Known Activations

comparative phrases and connections related to control groups

un- words

No Comments

No Known Activations