INDEX

Explanations

terms related to data processing functions and programming constructs

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

pipe, filters, conversion

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

setContentView

-0.37

Välislingid

-0.36

 constructeur

-0.34

 securing

-0.33

atasaray

-0.33

 policiales

-0.32

btnNew

-0.32

 destined

-0.31

 secured

-0.31

kript

-0.30

POSITIVE LOGITS

Pipe

0.82

 Pipe

0.71

pipe

0.70

 wireType

0.69

pipes

0.67

Pipes

0.66

 Pipes

0.66

 pipe

0.66

 հղումներ

0.63

 PIPE

0.62

Activations Density 0.002%

terms related to data processing functions and programming constructs

pipe, filters, conversion

No Comments

No Known Activations