INDEX

Explanations

specific references to ongoing research, opinions, or discussions about various topics

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

preceding verbs like "prefer" or "suggest"

np_acts-logits-general · gemini-2.0-flash

opted, chose, prefer

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

gameserver

-0.39

ibid

-0.38

halt

-0.37

fillText

-0.37

libft

-0.36

偃

-0.36

invokeLater

-0.36

Epil

-0.35

дю

-0.35

middot

-0.35

POSITIVE LOGITS

 préf

0.55

 prefer

0.53

 prefers

0.52

 preferring

0.50

 préféré

0.49

+#+

0.48

 chose

0.47

 opted

0.46

preferred

0.46

 preferred

0.45

Activations Density 0.594%

specific references to ongoing research, opinions, or discussions about various topics

preceding verbs like "prefer" or "suggest"

opted, chose, prefer

No Comments

No Known Activations