INDEX

Explanations

actions related to progression or advancement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ManagedBean

-0.66

 Roskov

-0.62

basicConfig

-0.61

glLoadIdentity

-0.61

 Вікі

-0.59

NSNotification

-0.59

czo

-0.57

calaureate

-0.55

ouncy

-0.55

 MainAxisSize

-0.55

POSITIVE LOGITS

things

0.62

 things

0.61

ฏิ

0.58

 began

0.56

 abusing

0.56

 empiezan

0.54

ticing

0.53

Things

0.53

ämään

0.52

hassee

0.52

Activations Density 0.168%

actions related to progression or advancement

No Comments

No Known Activations