INDEX

Explanations

mathematical symbols and expressions associated with matrices

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

matrix dimensions and indices

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

AddTagHelper

-0.45

NextToken

-0.39

MethodManager

-0.38

WebVitals

-0.38

__))

-0.37

NSURL

-0.34

InstanceId

-0.34

 HasFactory

-0.34

Lugares

-0.33

dotenv

-0.32

POSITIVE LOGITS

 defaultstate

0.66

InjectAttribute

0.64

BagLayout

0.63

WindowConstants

0.59

setVerticalGroup

0.58

[:,

0.54

RTLI

0.53

parsedMessage

0.51

 gynhyrchwyd

0.51

 jspb

0.49

Activations Density 0.017%

mathematical symbols and expressions associated with matrices

matrix dimensions and indices

No Comments

No Known Activations