INDEX

Explanations

conditional statements and instructions or questions in a logical context

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

following question

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

 CreateTagHelper

-0.50

IntoConstraints

-0.47

webElementXpaths

-0.47

 betweenstory

-0.46

ngrx

-0.46

kaido

-0.46

principalColumn

-0.44

 kysy

-0.43

lines

-0.42

 neler

-0.42

POSITIVE LOGITS

enumi

0.53

 kasarigan

0.49

extAlignment

0.41

$_[

0.41

FormatException

0.40

Javadoc

0.37

 Organisateur

0.36

 tarifi

0.35

 getRule

0.35

 sepak

0.34

Activations Density 0.112%

conditional statements and instructions or questions in a logical context

following question

No Comments

No Known Activations