INDEX

Explanations

phrases related to the application of principles or guidelines in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ates

-0.15

 allem

-0.14

ceiver

-0.14

onium

-0.14

enser

-0.13

unan

-0.13

ules

-0.13

uario

-0.13

cape

-0.13

POSITIVE LOGITS

ynchronously

0.14

/vnd

0.14

Ãªn

0.14

 Bout

0.14

 larger

0.14

ettel

0.13

Î¯Îº

0.13

emmel

0.13

versible

0.13

Äįka

0.13

Activations Density 0.176%

phrases related to the application of principles or guidelines in various contexts

No Comments

No Known Activations