INDEX

Explanations

complex legal and technical language

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

neurons with acetylcholinesterase

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

StoryboardSegue

-0.77

fromnode

-0.61

 ThemeData

-0.57

angliski

-0.55

 jsPsych

-0.54

 AssemblyCulture

-0.49

parsedMessage

-0.47

principalColumn

-0.47

MemoryWarning

-0.46

 crossorigin

-0.46

POSITIVE LOGITS

 enfans

0.40

 occafion

0.39

 Belén

0.38

record

0.37

Facade

0.36

 infin

0.36

 ucapnya

0.36

NSObject

0.35

 pleaſure

0.35

 Salom

0.35

Activations Density 0.179%

complex legal and technical language

neurons with acetylcholinesterase

No Comments

No Known Activations