INDEX

Explanations

sections of text with relevant quantitative data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

Autoritní

-0.93

HasForeignKey

-0.90

<bos>

-0.83

XmlAccessType

-0.81

 '\\;'

-0.81

 kaarangay

-0.80

GEBURTSDATUM

-0.80

protoc

-0.80

WithMany

-0.79

elemField

-0.78

POSITIVE LOGITS

 picioare

0.58

 Vergine

0.58

 înal

0.58

 tevens

0.56

äiv

0.56

↵↵

0.55

 emel

0.55

[toxicity=0]

0.55

 sienta

0.55

 dumne

0.53

Activations Density 0.003%

sections of text with relevant quantitative data

No Comments

No Known Activations