INDEX

Explanations

special characters and formatting elements used in code

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

foreign language 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 @ 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

期刊论文

-0.52

delwed

-0.50

DockStyle

-0.47

findpost

-0.46

WriteBarrier

-0.45

styleable

-0.44

 multer

-0.41

san

-0.40

❋

-0.40

TabIndex

-0.38

POSITIVE LOGITS

CppMethod

0.54

 tirada

0.42

 mergeFrom

0.41

 étrangère

0.41

 Infórmanos

0.41

 SqlCommand

0.41

Datuak

0.40

 страх

0.40

 correcte

0.39

 compét

0.39

Activations Density 0.001%

special characters and formatting elements used in code

foreign language code snippets

No Comments

No Known Activations