INDEX

Explanations

programming constructs and operations related to initialization and setup processes

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

Code-related terms: "update", "read", "load", "return"

np_acts-logits-general · gemini-2.0-flash

programmatic operations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 28-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_28/width_16k/average_l0_119

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.28.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

 totiž

-0.35

 dessutom

-0.35

 ciascuno

-0.33

 appunto

-0.33

 solcher

-0.32

 např

-0.32

 perciò

-0.31

 dziś

-0.31

 bijvoorbeeld

-0.30

 inilah

-0.30

POSITIVE LOGITS

 surla

0.66

ỡng

0.66

 appropriate

0.64

Jeografia

0.61

揄

0.60

BibitemShut

0.60

 blank

0.59

஢

0.59



0.59

 authorization

0.58

Activations Density 0.406%

programming constructs and operations related to initialization and setup processes

Code-related terms: "update", "read", "load", "return"

programmatic operations

No Comments

No Known Activations