INDEX

Explanations

terms related to data structures and their representation

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

Followed by words such as "like", "is", "has"

np_acts-logits-general · gemini-2.0-flash

how something looks

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

 betweenstory

-0.73

rawDesc

-0.63

 pymongo

-0.61

 kasarigan

-0.60

SBATCH

-0.59

ंदीखरीदारी

-0.57

σθαι

-0.56

égard

-0.56

脚注の使い方

-0.55

mahami

-0.54

POSITIVE LOGITS

 look

3.14

 looks

2.91

look

2.73

 looked

2.63

 Look

2.59

Look

2.50

 looking

2.48

 LOOK

2.40

looks

2.37

 Looks

2.30

Activations Density 0.329%

terms related to data structures and their representation

Followed by words such as "like", "is", "has"

how something looks

No Comments

No Known Activations