INDEX

Explanations

the word "for" in various contexts and repetitions

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

plans for future

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

 bendera

-0.46

 Partagez

-0.46

 senhora

-0.45

 vapa

-0.45

ฟัง

-0.45

 Deixe

-0.43

 Cliquez

-0.43

 leyenda

-0.42

antaranya

-0.42

 Universitario

-0.42

POSITIVE LOGITS

Plans

0.63

 Plans

0.58

 plan

0.52

 plans

0.50

plans

0.48

 Plan

0.47

 PLAN

0.47

ValueGeneration

0.46

 PLANS

0.46

PLAN

0.45

Activations Density 0.039%

the word "for" in various contexts and repetitions

plans for future

No Comments

No Known Activations