INDEX

Explanations

sequences of command line arguments or flags used in shell commands

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

Followed by a lowercase letter

np_acts-logits-general · gemini-2.0-flash

commands and scripts

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

AndEndTag

-0.87

WriteBarrier

-0.81

 onCreateView

-0.70

onViewCreated

-0.68

principalTable

-0.66

^(@)

-0.66

AnimationFrame

-0.63

 useAppContext

-0.62

addCriterion

-0.61

AutoField

-0.60

POSITIVE LOGITS

aarrggbb

0.60

 ajudá

0.55

 Référence

0.52

force

0.51

 linkovi

0.48

ftagPool

0.47

helves

0.47

gebir

0.46

Activations Density 0.110%

sequences of command line arguments or flags used in shell commands

Followed by a lowercase letter

commands and scripts

No Comments

No Known Activations