INDEX

Explanations

terms related to rotation and its variations

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

rotation and orientation

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

<h5>

-0.50

Imp

-0.43

chance

-0.42

ljiv

-0.41

 Buch

-0.41

Imp

-0.40

Gep

-0.40

ophyll

-0.40

 Thousand

-0.39

Buch

-0.39

POSITIVE LOGITS

 rotate

1.02

rotate

0.99

 rotates

0.93

 Rotate

0.92

Rotate

0.89

 Rotation

0.89

 rotation

0.84

 rotación

0.83

Rotation

0.81

 Rotating

0.81

Activations Density 0.013%

terms related to rotation and its variations

rotation and orientation

No Comments

No Known Activations