INDEX

Explanations

commands and phrases that suggest examination and analysis

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

"look" and its variations

np_acts-logits-general · gemini-2.0-flash

look at

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

djangoproject

-0.75

الحياه

-0.75

CPtr

-0.71

 noqa

-0.69

findpost

-0.69

bootstrapcdn

-0.69

RTEE

-0.68

CodedInputStream

-0.68

RegistryLite

-0.67

yntaxException

-0.65

POSITIVE LOGITS

 look

2.35

look

2.10

 Look

2.05

Look

2.01

 LOOK

1.82

LOOK

1.72

 glance

1.43

 looked

1.41

 looks

1.38

 looking

1.33

Activations Density 0.420%

commands and phrases that suggest examination and analysis

"look" and its variations

look at

No Comments

No Known Activations