INDEX

Explanations

phrases indicating strong acknowledgment or affirmation of quality

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

"Well" followed by a past participle

np_acts-logits-general · gemini-2.0-flash

well followed by known/recognized

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 28-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_28/width_16k/average_l0_119

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.28.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

AndEndTag

-0.72

JsonPropertyName

-0.50

 snippetHide

-0.47

期刊论文

-0.47

뀝

-0.46

ciatore

-0.46

 tatuagem

-0.45

 iconTwitter

-0.45

 NSCoder

-0.44

ViewImports

-0.44

POSITIVE LOGITS

Well

0.85

 Well

0.84

 well

0.75

well

0.74

 WELL

0.71

WELL

0.63

 Wells

0.55

 versed

0.55

known

0.53

Wells

0.53

Activations Density 0.085%

phrases indicating strong acknowledgment or affirmation of quality

"Well" followed by a past participle

well followed by known/recognized

No Known Activations