INDEX

Explanations

references to data retrieval and manipulation methods in a programming context

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

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

 itſelf

-0.76

 CURIAM

-0.75

 resourceCulture

-0.74

 increí

-0.74

 leſs

-0.73

 ſei

-0.71

 verſ

-0.70

 ſein

-0.70

 ſont

-0.70

 Infórmanos

-0.70

POSITIVE LOGITS

initWithFrame

0.38

el

0.38

↵↵

0.38

and

0.36

mal

0.35

or

0.34

ax

0.34

 Fischer

0.34

&&

0.33

Activations Density 0.011%

references to data retrieval and manipulation methods in a programming context

No Comments

No Known Activations