INDEX

Explanations

phrases expressing preference or affection

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

Autoritní

-0.45

出版年

-0.45

TemporalType

-0.44

modelBuilder

-0.42

 sanitarias

-0.42

newBuilder

-0.41

 vPvB

-0.40

tschafts

-0.39

thenes

-0.38

 Ucrania

-0.38

POSITIVE LOGITS

enjoy

0.67

 enjoyed

0.60

Enjoy

0.57

 enjoyment

0.56

 Enjoy

0.56

 enjoys

0.55

 enjoy

0.54

 ENJOY

0.54

 appreciation

0.51

 enjoying

0.49

Activations Density 0.008%

phrases expressing preference or affection

No Comments

No Known Activations