INDEX

Explanations

code elements related to detecting and handling specific conditions or events

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

start of turn user

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

DockStyle

-0.55

 Италијани

-0.49

 createSlice

-0.48

 ivelany

-0.47

msgTypes

-0.46

)++;

-0.45

UnknownFieldSet

-0.44

arraycopy

-0.44

 autorytatywna

-0.44

期刊论文

-0.42

POSITIVE LOGITS

뀜

0.40

 Exactos

0.38

rigen

0.38

 doInBackground

0.37

Enllaces

0.37

tonode

0.36

ordeel

0.35

 shutters

0.35

 פע

0.35

profesor

0.34

Activations Density 0.010%

code elements related to detecting and handling specific conditions or events

start of turn user

No Comments

No Known Activations