INDEX

Explanations

references to orientations and positions, particularly "upside down" and related terminologies

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

Upside down or inverted orientation

np_acts-logits-general · gemini-2.0-flash

upside down and inverted

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.31.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rrggbb

-0.48

uxxxx

-0.30

rensa

-0.29

romyal

-0.27

วง

-0.26

琐

-0.25

 Council

-0.25

 council

-0.25

詰

-0.24

 remp

-0.24

POSITIVE LOGITS

 upside

1.80

 flipped

1.55

 Upside

1.54

 inverted

1.52

 flip

1.32

 flips

1.27

 inversion

1.27

 flipping

1.26

 Flip

1.21

 invert

1.20

Activations Density 0.197%

references to orientations and positions, particularly "upside down" and related terminologies

Upside down or inverted orientation

upside down and inverted

No Comments

No Known Activations