INDEX

Explanations

terms related to giving or providing help or resources

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ledon

-0.15

adera

-0.15

è¤

-0.15

ÐµÐ¸

-0.15

agon

-0.15

 past

-0.15

rouch

-0.14

à¹Ģà¸Ķ

-0.14

 Mush

-0.14

/Area

-0.14

POSITIVE LOGITS

.EventQueue

0.16

emer

0.15

mass

0.15

 Toledo

0.15

STM

0.14

 cita

0.14

ires

0.14

æĻ´

0.14

/sec

0.14

brief

0.14

Activations Density 0.009%

terms related to giving or providing help or resources

No Comments

No Known Activations