INDEX

Explanations

instructional phrases and guidance for processes

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

instructions or procedures

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

yet

-0.45

Yet

-0.42

Yet

-0.40

yet

-0.37

pas

-0.37

Dur

-0.35

erol

-0.35

 pare

-0.35

 Accordingly

-0.35

dra

-0.35

POSITIVE LOGITS

ConstraintMaker

0.64

 كومونز

0.58

arangay

0.56

sizeCache

0.56

دانشنامهٔ

0.55

 EconPapers

0.54

FunctionFlags

0.52

ьаж

0.52

majánló

0.52

Personendaten

0.51

Activations Density 0.032%

instructional phrases and guidance for processes

instructions or procedures

No Comments

No Known Activations