INDEX

Explanations

phrases related to the beginning or introduction of various works

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

kop

-0.15

pur

-0.15

lew

-0.15

hz

-0.14

adier

-0.14

erness

-0.14

íĮĮ

-0.14

enes

-0.13

ÐµÐ½Ð¸

-0.13

 twice

-0.13

POSITIVE LOGITS

/start

0.19

/tutorial

0.16

 opening

0.15

 sequence

0.15

 Sequence

0.15

ductory

0.15

opening

0.15

azole

0.15

 Opening

0.15

 credits

0.15

Activations Density 0.020%

phrases related to the beginning or introduction of various works

No Comments

No Known Activations