INDEX

Explanations

phrases related to user guidance and suggestions on how to engage with content and information on a website

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 proc

-0.17

 paralle

-0.17

ught

-0.16

nud

-0.16

mim

-0.16

æ¢

-0.16

Äįel

-0.15

 downs

-0.15

piv

-0.15

dw

-0.15

POSITIVE LOGITS

esses

0.15

ieux

0.14

ember

0.14

ÑĮÐµ

0.14

BI

0.14

ocos

0.14

oca

0.13

/generated

0.13

 narrow

0.13

Activations Density 0.397%

phrases related to user guidance and suggestions on how to engage with content and information on a website

No Comments

No Known Activations