INDEX

Explanations

phrases that indicate repetition or variation in narratives and dialogue

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

preceding time periods

np_acts-logits-general · gemini-2.0-flash

during or throughout a time period

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

AnchorTagHelper

-0.59

AppCompat

-0.51

aneamente

-0.50

 AppBundle

-0.49

ondissement

-0.47

 actualidad

-0.46

toMatchSnapshot

-0.45

typeorm

-0.45

 OMITBAD

-0.44

nocache

-0.43

POSITIVE LOGITS

 during

1.59

 throughout

1.54

during

1.52

throughout

1.51

 DURING

1.40

During

1.38

 During

1.36

 Throughout

1.35

 durante

1.34

Throughout

1.33

Activations Density 0.683%

phrases that indicate repetition or variation in narratives and dialogue

preceding time periods

during or throughout a time period

No Known Activations