INDEX

Explanations

references to specific identifiers or names in a technical context

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

java annotations and code snippets

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

 MainAxisSize

-0.47

IANGLES

-0.45

 setId

-0.43

ChromeDriver

-0.42

loger

-0.42

 queſta

-0.42

bronze

-0.42

Programmer

-0.41

Rid

-0.40

 EconPapers

-0.40

POSITIVE LOGITS

RequestMapping

0.78

 AssemblyVersion

0.59

WebServlet

0.53

PostMapping

0.49

RequestMethod

0.46

 ModelExpression

0.43

 Bagian

0.42

JsonProperty

0.42

principalColumn

0.41

GetMapping

0.41

Activations Density 0.000%

references to specific identifiers or names in a technical context

java annotations and code snippets

No Comments

No Known Activations