INDEX

Explanations

decision-making processes related to interface implementations

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

Code snippets or symbols

np_acts-logits-general · gemini-2.0-flash

selected cell

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

 Италијани

-1.07

 betweenstory

-1.02

RegressionTest

-0.84

ArrowToggle

-0.84

expandindo

-0.82

 gradova

-0.81

Autoritní

-0.74

 otomatig

-0.74

ьаж

-0.72

 Italijanski

-0.69

POSITIVE LOGITS

 BoxDecoration

0.92

\{\\

0.89

NOPQRST

0.76

UpInside

0.72

InputBorder

0.68

CloseOperation

0.67

󠁢

0.64

RectangleBorder

0.63

囗

0.63

 BoxFit

0.62

Activations Density 14.164%

decision-making processes related to interface implementations

Code snippets or symbols

selected cell

No Comments

No Known Activations