INDEX

Explanations

occurrences of the word "act" or its variations in various contexts

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

Zu

-0.53

posables

-0.51

tizers

-0.51

 Marguerite

-0.50

rehensive

-0.49

hera

-0.49

herme

-0.48

Zub

-0.48

meren

-0.48

isNullOrEmpty

-0.48

POSITIVE LOGITS

Act

0.98

Acts

0.96

act

0.95

Act

0.94

 Acts

0.90

ACT

0.85

 acts

0.84

act

0.84

 acted

0.83

 acting

0.75

Activations Density 0.162%

occurrences of the word "act" or its variations in various contexts

No Comments

No Known Activations