INDEX

Explanations

terms and phrases that refer to unintended consequences or byproducts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_16k/average_l0_84

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.13.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

ulkan

-0.57

 للمعارف

-0.54

quelize

-0.53

replacement

-0.52

noin

-0.51

AGEM

-0.51

Precise

-0.50

Сюжет

-0.49

marginVertical

-0.49

hach

-0.48

POSITIVE LOGITS

 incidentally

0.91

 incidental

0.87

 unintentionally

0.75

 inadvertently

0.71

 indirectly

0.71

 accidentally

0.68

idental

0.66

ついでに

0.64

Accidental

0.62

ついで

0.61

Activations Density 0.459%

terms and phrases that refer to unintended consequences or byproducts

No Comments

No Known Activations