INDEX

Explanations

patterns in structured data and code formatting

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

code snippets

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 étrangers

-0.36

Australie

-0.34

 civilización

-0.33

gerät

-0.32

 Económica

-0.32

esterno

-0.32

 desayuno

-0.31

appartamento

-0.30

↵

-0.30

<eos>

-0.30

POSITIVE LOGITS

 ویکی‌پدی

0.80

 propOrder

0.76

RegressionTest

0.74

SharedDtor

0.74

GEBURTSDATUM

0.71

findpost

0.69

 noDo

0.69

 Signalez

0.68

parsedMessage

0.67

 informée

0.66

Activations Density 0.049%

patterns in structured data and code formatting

code snippets

No Known Activations