INDEX

Explanations

phrases that indicate results or outcomes from analyses and studies

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

results of phrase

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ſelf

-0.76

 itſelf

-0.75

 pleaſure

-0.73

 snippetHide

-0.70

parsedMessage

-0.69

 propOrder

-0.68

 himſelf

-0.68

 myſelf

-0.68

 ſtre

-0.66

 AttributeSet

-0.64

POSITIVE LOGITS

getResult

0.54

 outcome

0.42

result

0.42

 result

0.42

 samping

0.41

 hasil

0.39

 cuộc

0.39

 results

0.38

การ

0.38

Result

0.38

Activations Density 0.055%

phrases that indicate results or outcomes from analyses and studies

results of phrase

No Comments

No Known Activations