INDEX

Explanations

references to updates or edits in content or discussions

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

Text after "edit" or "update"

np_acts-logits-general · gemini-2.0-flash

updates or resolutions

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

Not in Any Lists

No Comments

Negative Logits

 Wiktionnaire

-0.59

当時は

-0.58

 '\\;'

-0.56

 alltid

-0.52

 собой

-0.52

apimachinery

-0.52

esistenza

-0.51

際は

-0.51

才会

-0.50

 getItemCount

-0.50

POSITIVE LOGITS

 finally

1.27

finally

1.20

 FINALLY

1.12

Finally

1.04

 finalmente

1.02

 Finally

1.01

 Finalmente

0.92

 finalement

0.92

Finalmente

0.92

middels

0.89

Activations Density 0.156%

references to updates or edits in content or discussions

Text after "edit" or "update"

updates or resolutions

No Comments

No Known Activations