INDEX

Explanations

data related to structured information, likely in a programming or markup context

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

following "the"

np_acts-logits-general · gemini-2.0-flash

redstone torch powering

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.31.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 الحره

-0.54

StoryboardSegue

-0.47

 anſ

-0.37

ijiet

-0.37

اریخ

-0.36

清明

-0.35

 ekor

-0.35

 neutralizing

-0.35

 nonatomic

-0.35

 Winnipeg

-0.35

POSITIVE LOGITS

mod

0.76

Mod

0.75

Mod

0.75

 modded

0.65

 mods

0.62

mod

0.60

 Mods

0.56

 modding

0.55

MOD

0.53

MOD

0.53

Activations Density 0.025%

data related to structured information, likely in a programming or markup context

following "the"

redstone torch powering

No Comments

No Known Activations