INDEX

Explanations

references to classes and functions in a programming context

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

Code, especially related to Google

np_acts-logits-general · gemini-2.0-flash

General Motors and Supplemental material

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

 localObject

-0.62

 ***!

-0.59

 незавершена

-0.54

 surla

-0.52

ductory

-0.50

TagMode

-0.49

:+:

-0.49

Externé

-0.49

出版年

-0.49

 BorderLayout

-0.48

POSITIVE LOGITS

google

0.67

 absl

0.59

 Google

0.59

 GOOGLE

0.59

 google

0.58

 gainera

0.57

Google

0.55

 goog

0.53

новниш

0.51

GOOGLE

0.49

Activations Density 0.563%

references to classes and functions in a programming context

Code, especially related to Google

General Motors and Supplemental material

No Comments

No Known Activations