INDEX

Explanations

questions and rhetorical inquiries in dialogue

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.38.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

 همچنین

-0.50

 также

-0.49

 також

-0.47

 additionally

-0.47

 wikihow

-0.44

녁

-0.43

BorderFactory

-0.43

StoreMessageInfo

-0.42

 also

-0.42

 επίσης

-0.42

POSITIVE LOGITS

 replied

0.61

 Hearing

0.58

 hearing

0.58

Hearing

0.58

 laughed

0.56

 retorted

0.55

 Suddenly

0.53

 suddenly

0.52

hearing

0.52

 terdengar

0.52

Activations Density 0.071%

questions and rhetorical inquiries in dialogue

No Comments

No Known Activations