INDEX

Explanations

technical specifications and comparisons in scientific contexts

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

different types and conditions

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

InputTagHelper

-0.43

 referrerpolicy

-0.40



-0.37

 thereunder

-0.36

 piso

-0.35

UserScript

-0.35

 Riviera

-0.35

dk

-0.34

reszcie

-0.34

رز

-0.34

POSITIVE LOGITS

 autorytatywna

0.60

parsedMessage

0.54

oredCriteria

0.50

ConstraintMaker

0.48

 незавершена

0.47

HasForeignKey

0.46

Erstellt

0.45

 égale

0.43

 intercepts

0.43

liceerd

0.43

Activations Density 0.271%

technical specifications and comparisons in scientific contexts

different types and conditions

No Comments

No Known Activations