INDEX

Explanations

references to source paths and library imports in code

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

path components

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.16.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

*>(

-0.49

ophers

-0.47

(&:

-0.47

 adipis

-0.46

<bos>

-0.45

 ModelExpression

-0.44

*>(

-0.44

PhysRev

-0.43

COOH

-0.43

Espèce

-0.42

POSITIVE LOGITS

reportWebVitals

0.89

 Wikimedijinoj

0.73

Glej

0.71

argout

0.69

adaptiveStyles

0.66

 consultato

0.65

SuspendLayout

0.64

TagHelper

0.64

 NDEBUG

0.61

Externí

0.60

Activations Density 0.008%

references to source paths and library imports in code

path components

No Comments

No Known Activations