INDEX

Explanations

technical terms and concepts related to programming and session management

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

questions about it

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

 flesta

-0.63

 barnen

-0.63

 varandra

-0.59

baliknya

-0.56

 fleste

-0.54

 său

-0.51

 himself

-0.51

zkiem

-0.50

 stället

-0.50

 spesa

-0.49

POSITIVE LOGITS

حياتها

0.65

 kasarigan

0.62

awtextra

0.50

 która

0.49

 transfieras

0.48

 kterou

0.47

 ktorá

0.47

 která

0.47

 betweenstory

0.46

->___

0.46

Activations Density 0.092%

technical terms and concepts related to programming and session management

questions about it

No Comments

No Known Activations