INDEX

Explanations

function calls and method chaining in code

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

technical writing and code

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

Negative Logits

 invokingState

-0.52

onAttach

-0.49

 ComVisible

-0.48

 informée

-0.47



-0.47

 Wikimedijinoj

-0.46

 autorytatywna

-0.46

ValueGenerated

-0.44

➟

-0.43

🚔

-0.42

POSITIVE LOGITS

cstdio

0.46

 <<<<<<<<<<<<<<

0.43

 versch

0.38

 einzu

0.38

achal

0.36

tritts

0.36

 significativo

0.35

apunov

0.35

 Friedens

0.35

mspace

0.35

Activations Density 0.083%

function calls and method chaining in code

technical writing and code

No Known Activations