INDEX

Explanations

references to technical or programming-related concepts and components

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

negotiate from strength

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

newInstance

-0.39

 émotion

-0.39

 réve

-0.38

 Ahli

-0.37

 Polisi

-0.37

 women

-0.36

 souverain

-0.35

 panoramique

-0.34

 sensibilité

-0.34

 prisoners

-0.34

POSITIVE LOGITS

GHIJKLM

1.15

expandindo

0.68

parsedMessage

0.65

ffine

0.65

يكب

0.59

 հղումներ

0.59

featureID

0.57

AddTagHelper

0.57

postMessage

0.57

发表于

0.56

Activations Density 0.001%

references to technical or programming-related concepts and components

negotiate from strength

No Comments

No Known Activations