INDEX

Explanations

programming-related terms and syntax within coding instructions or algorithms

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

selection determination

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_131k/average_l0_97

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.24.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

 BufferedWriter

-0.53

 kasarigan

-0.53

″]

-0.51

FormTagHelper

-0.47

 tranſ

-0.46

بوابة

-0.45

 tartalomajánló

-0.45

WireFormatLite

-0.44

 ſta

-0.43

 ſur

-0.43

POSITIVE LOGITS

GEBURTSDATUM

0.59

 selection

0.52

 seleção

0.49

 selezion

0.48

 للمعارف

0.47

 sélection

0.47

 Selection

0.46

selection

0.45

 SELECTION

0.44

 détermination

0.43

Activations Density 0.093%

programming-related terms and syntax within coding instructions or algorithms

selection determination

No Comments

No Known Activations