INDEX

Explanations

phrases emphasizing magnitude and intensity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

olib

-0.15

agen

-0.15

ntag

-0.15

ÄĻ

-0.15

-lfs

-0.15

rg

-0.14

vro

-0.14

cha

-0.14

INET

-0.14

ieber

-0.14

POSITIVE LOGITS

0.16

ars

0.16

isbury

0.16

0.15

asso

0.14

iman

0.14

stakes

0.14

Įĵ

0.14

kas

0.14

Activations Density 0.200%

phrases emphasizing magnitude and intensity

No Comments

No Known Activations