INDEX

Explanations

frequent personal pronouns and phrases that express opinions or actions

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

preceding "can" or "cannot"

np_acts-logits-general · gemini-2.0-flash

can or cannot

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

MigrationBuilder

-0.57

-0.53

 оригіналу

-0.53

setVerticalGroup

-0.53

Salut

-0.52

ThroughAttribute

-0.51

pyplot

-0.50

RegistryLite

-0.49

Dabei

-0.49

kháu

-0.49

POSITIVE LOGITS

 cannot

2.78

cannot

2.44

 Cannot

2.31

 couldn

2.29

Cannot

2.17

 cant

2.15

 CANNOT

2.01

 unable

1.96

couldn

1.93

 Couldn

1.87

Activations Density 0.628%

frequent personal pronouns and phrases that express opinions or actions

preceding "can" or "cannot"

can or cannot

No Comments

No Known Activations