INDEX

Explanations

expressions of apology or regret

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

AntiForgeryToken

-0.72

]='\

-0.69

GHIJKLM

-0.69

trường

-0.68

 ThemeData

-0.64

Xna

-0.64

SerializedSize

-0.63

 FetchType

-0.62

NameInMap

-0.62

XmlAccessorType

-0.60

POSITIVE LOGITS

 offended

0.59

 inconvenience

0.58

 offend

0.57

 intrusion

0.55

Rüyada

0.52

lite

0.52

 late

0.51

 defStyle

0.51

maq

0.51

Activations Density 0.040%

expressions of apology or regret

No Comments

No Known Activations