INDEX

Explanations

phrases or terms related to knowledge and understanding

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

guards

-0.15

ubern

-0.15

-0.14

Ð¿Ð¸Ñģ

-0.14

mav

-0.14

/Sub

-0.14

urator

-0.14

POSITIVE LOGITS

agine

0.15

Vac

0.15

heim

0.15

eki

0.15

chein

0.14

woke

0.14

Nun

0.14

á»ģn

0.14

olist

0.14

ogne

0.14

Activations Density 0.048%

phrases or terms related to knowledge and understanding

No Comments

No Known Activations