INDEX

Explanations

words related to advice or recommendations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

 baÅŁÄ±nda

-0.16

Ã¼b

-0.16

oÅ¡

-0.15

ÐºÑĥÐ»Ñı

-0.14

oons

-0.14

urre

-0.14

vere

-0.14

uppe

-0.14

UFFIX

-0.14

eper

-0.14

POSITIVE LOGITS

antages

0.45

ancement

0.45

ancing

0.42

antage

0.40

ancements

0.40

ocate

0.38

ances

0.37

ise

0.35

antaged

0.32

oc

0.31

Activations Density 0.005%

words related to advice or recommendations

No Comments

No Known Activations

words related to advice or recommendations

No Comments

No Known Activations