INDEX

Explanations

technical elements related to code structure and documentation

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

Code generated by client

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

fromnode

-0.60

帖最后由

-0.56

verifyException

-0.55

SourceChecksum

-0.54

 AssemblyCompany

-0.53

GraphicsUnit

-0.53

GenerationType

-0.52

+#+

-0.49

JspWriter

-0.48

 Meksiku

-0.48

POSITIVE LOGITS

 Lingkungan

0.45

 gyhoeddwyd

0.41

 mourut

0.41

 prawo

0.41

 travailleurs

0.40

 Italij

0.39

 fué

0.38

 almendras

0.38

 poète

0.37

 picioare

0.37

Activations Density 0.012%

technical elements related to code structure and documentation

Code generated by client

No Comments

No Known Activations