INDEX

Explanations

numeric ranges and comparative terms related to situations or conditions

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

is or does

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

دانشنامهٔ

-0.44

BagLayout

-0.44

nextLine

-0.40

 relegated

-0.39

 Vogt

-0.38

 amicable

-0.38

 seamlessly

-0.38

Искәрмәләр

-0.38

منابع

-0.38

 Motta

-0.37

POSITIVE LOGITS

 supérieurs

0.51

 plufieurs

0.49

StructEnd

0.48

 pleaſure

0.48

 queſta

0.48

 ainfi

0.48

 dezelve

0.47

 femininos

0.47

 opérés

0.46

thâu

0.46

Activations Density 0.001%

numeric ranges and comparative terms related to situations or conditions

is or does

No Comments

No Known Activations