INDEX

Explanations

sequences that involve user interactions and system responses in an application context

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

Conditional logic like "when," "if," "then," etc

np_acts-logits-general · gemini-2.0-flash

when user action

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

SourceChecksum

-0.76

nestjs

-0.64

migrationBuilder

-0.61

LookAnd

-0.58

discuss

-0.58

TypedDataSet

-0.57

arResult

-0.57

 useDispatch

-0.55

 deleteById

-0.55

 pourra

-0.54

POSITIVE LOGITS

 someone

0.59

 somebody

0.55

Someone

0.54

Somebody

0.53

 nào

0.52

 <=",

0.52

someone

0.51

 user

0.49

 пользователем

0.49

 Someone

0.49

Activations Density 0.559%

sequences that involve user interactions and system responses in an application context

Conditional logic like "when," "if," "then," etc

when user action

No Comments

No Known Activations