INDEX

Explanations

references to tree structures in programming or data organization contexts

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

tree structures and nodes

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

 ujednoznacz

-0.55

 utafitiHapana

-0.53

windowFixed

-0.52

UserScript

-0.45

 ویکی‌پدی

-0.45

期刊论文

-0.45

पया

-0.44

 disambiguazione

-0.42

 didst

-0.41

 Photocase

-0.41

POSITIVE LOGITS

SearchTree

0.74

 TreeNode

0.65

 tree

0.59

 Tree

0.58

 trees

0.57

 Trees

0.56

BinaryTree

0.53

树

0.52

 forest

0.51

tree

0.50

Activations Density 0.004%

references to tree structures in programming or data organization contexts

tree structures and nodes

No Comments

No Known Activations