INDEX

Explanations

phrases indicating uncertainty or evaluation of quality

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

windowFixed

-0.50

 AppCompatTheme

-0.48

 defaultstate

-0.47

èdia

-0.47

 Wikimedijinoj

-0.46

 auroit

-0.44

 utafitiHapana

-0.44

 TextAppearance

-0.43

ItemBackground

-0.43

 juridiques

-0.43

POSITIVE LOGITS

nodoc

0.45

Ƚ

0.42

imea

0.40

continue

0.40

bigcup

0.40

 undersigned

0.39

continues

0.39

 continue

0.39

mael

0.39

 iota

0.39

Activations Density 0.475%

phrases indicating uncertainty or evaluation of quality

No Comments

No Known Activations