INDEX

Explanations

sequences of numbers and codes, possibly related to identifiers or versions

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

BECTS or code structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.16.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

textra

-0.47

 CGPoint

-0.47

 goal

-0.46

ñora

-0.45

deaux

-0.45

 colgantes

-0.45

Petsc

-0.45

extrême

-0.44

 extreme

-0.44

 kiểm

-0.43

POSITIVE LOGITS

 ExecuteAsync

0.79

 ComVisible

0.75

Хьажоргаш

0.71

 <>",

0.69

MigrationBuilder

0.66

 незавершена

0.65

)':

0.63

SerializedName

0.63

)».

0.63

 يتيمه

0.60

Activations Density 0.007%

sequences of numbers and codes, possibly related to identifiers or versions

BECTS or code structures

No Comments

No Known Activations