INDEX

Explanations

code snippets and data structures used in programming

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

lambda and programming constructs

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

MigrationBuilder

-0.57

 cellspacing

-0.55

 مشين

-0.49

EndProject

-0.47

SBATCH

-0.47

AddTagHelper

-0.46

tagHelperRunner

-0.44

лтамалар

-0.44

 <>",

-0.43

…~

-0.43

POSITIVE LOGITS

ariConfig

0.54

parsedMessage

0.49

prepareStatement

0.48

errHandler

0.45

RTGC

0.40

Бахар

0.38

 vivía

0.38

 vivió

0.37

Filmographie

0.37

 sientes

0.37

Activations Density 0.001%

code snippets and data structures used in programming

lambda and programming constructs

No Comments

No Known Activations