INDEX

Explanations

references to the letter "m" in various contexts

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

references to a specific variable or memory pointer in code-related contexts

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

m followed by specific characters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.20.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

MLLoader

-0.60

ent

-0.56

gameserver

-0.54

bootstrapcdn

-0.52

expo

-0.49

getitem

-0.49

 GenerationType

-0.49

enu

-0.48

ABOUT

-0.48

ाहरण

-0.48

POSITIVE LOGITS

 Majefty

0.75

Bm

0.75

 itſelf

0.74

 fhew

0.74

 neceff

0.73

 purpoſe

0.73

Efq

0.71

mmm

0.69

 ſtate

0.66

 Mover

0.66

Activations Density 0.237%

references to the letter "m" in various contexts

references to a specific variable or memory pointer in code-related contexts

m followed by specific characters

No Comments

No Known Activations