INDEX

Explanations

phrases related to time-sensitive announcements or updates

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ãĤĴãģĭ

-0.15

edral

-0.15

iquÃ©

-0.14

Ã¸y

-0.14

our

-0.14

emiz

-0.14

ÄŁan

-0.14

 (“

-0.14

ÑĩÐµ

-0.13

zia

-0.13

POSITIVE LOGITS

ãĥ¼ãĥ©

0.17

 unspecified

0.17

"[

0.16

lang

0.16

uxe

0.16

"(

0.15

 meanwhile

0.15

 unnamed

0.14

Activations Density 0.482%

phrases related to time-sensitive announcements or updates

No Comments

No Known Activations

phrases related to time-sensitive announcements or updates

No Comments

No Known Activations