INDEX

Explanations

instances of interaction and engagement with content, particularly related to reading and communication

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 BoxDecoration

-0.66

 Infórmanos

-0.65

AddTagHelper

-0.63

rsiniz

-0.59

 invokingState

-0.56

Datuak

-0.55

 dourada

-0.53

 Citadel

-0.52

 ProtoMessage

-0.52

各样的

-0.52

POSITIVE LOGITS

...

1.06

ldots

0.93

...

0.91

…

0.91

....

0.84

?...

0.84

!..

0.81

)...

0.81

!...

0.81

…

0.80

Activations Density 0.200%

instances of interaction and engagement with content, particularly related to reading and communication

No Comments

No Known Activations