INDEX

Explanations

instances of the word "embrace" and its variations, reflecting themes of acceptance and affection

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 2-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 betweenstory

-0.46

featureID

-0.40

renderItem

-0.38

FieldError

-0.38

Werdegang

-0.38

 ()=>

-0.37

AutoScale

-0.36

Outstanding

-0.36

 Turi

-0.35

reprodu

-0.35

POSITIVE LOGITS

ita

0.75

 embrace

0.73

 embraced

0.69

hug

0.68

 pants

0.67

ITA

0.67

AnchorTagHelper

0.64

 embracing

0.64

 Embrace

0.63

 tribe

0.63

Activations Density 0.375%

instances of the word "embrace" and its variations, reflecting themes of acceptance and affection

No Comments

No Known Activations