INDEX

Explanations

terms related to screen orientation and layout adjustments

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

GOTREF

-0.45

principalTable

-0.40

 expressão

-0.40

esgue

-0.40

 <=",

-0.39

 للمعارف

-0.39

 Signalez

-0.38

⤒

-0.36

liceerd

-0.35

 CURIAM

-0.34

POSITIVE LOGITS

 orientation

0.59

orientation

0.57

Orientation

0.54

ORIENTATION

0.51

 perpendicular

0.50

 Orientation

0.49

setVerticalGroup

0.49

 VERTICAL

0.48

 vertical

0.47

 Vertical

0.47

Activations Density 0.005%

terms related to screen orientation and layout adjustments

No Comments

No Known Activations