INDEX

Explanations

the word "to" in various contexts

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

numerical ranges

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

IUrlHelper

-0.51

MockBean

-0.47

stylers

-0.46

IsPostBack

-0.46

meralda

-0.44

 pinulongan

-0.43

 plundered

-0.43

☚

-0.42

 alimentaria

-0.40

ceased

-0.39

POSITIVE LOGITS

 Audiodateien

0.55

ſelves

0.48

 five

0.47

BeginContext

0.46

LookAnd

0.46

 amount

0.45

 cirka

0.45

ísimas

0.44

ísi

0.44

toHave

0.44

Activations Density 0.028%

the word "to" in various contexts

numerical ranges

No Comments

No Known Activations