INDEX

Explanations

programming-related syntax or constructs, particularly function declarations and parameters

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

user turn indicator

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

 invokingState

-0.65

 CreateTagHelper

-0.65

fromnode

-0.65

AndEndTag

-0.65

SourceChecksum

-0.63

parsedMessage

-0.62

 betweenstory

-0.59

Datuak

-0.56

WebServlet

-0.55

transQ

-0.54

POSITIVE LOGITS

astify

0.41

 Advance

0.36

MigrationBuilder

0.33

queryInterface

0.31

 ADVANCE

0.29

 advance

0.29

Glej

0.28

 confront

0.28

ndash

0.28

 gặp

0.28

Activations Density 0.004%

programming-related syntax or constructs, particularly function declarations and parameters

user turn indicator

No Comments

No Known Activations