INDEX

Explanations

function definitions in programming code

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

<start_of_turn>user parsing

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

 betweenstory

-0.49

 juſ

-0.36

يكب

-0.36

 getSelected

-0.35

 DevComponents

-0.35

principalTable

-0.35

 eléctricas

-0.34

 electrico

-0.32

 iprot

-0.32

exp

-0.32

POSITIVE LOGITS

Hochspringen

0.69

Hozzáférés

0.63

endphp

0.62

Дереккөздер

0.61

="@+

0.60

ooker

0.58

meris

0.58

synthesize

0.58

Skocz

0.57

로운

0.56

Activations Density 0.000%

function definitions in programming code

<start_of_turn>user parsing

No Comments

No Known Activations