INDEX

Explanations

phrases related to expectations and predictions

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_10/width_16k/average_l0_113

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 Wikimedijinoj

-0.46

InitVars

-0.44

 Meinung

-0.43

ImageContext

-0.42

styleable

-0.42

visející

-0.42

 nothwendig

-0.41

UseVisualStyle

-0.41

 Wahrnehmung

-0.40

WebVitals

-0.40

POSITIVE LOGITS

 Expect

0.61

Expect

0.59

0.55

 expect

0.52

 beware

0.50

 Beware

0.48

andExpect

0.47

Beware

0.47

hoeddwyd

0.45

bootstrapcdn

0.42

Activations Density 0.272%

phrases related to expectations and predictions

No Known Activations