INDEX

Explanations

phrases indicating the significance or importance of various subjects or ideas

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

anzi

-0.14

eref

-0.13

áº¯t

-0.13

¶Į

-0.13

 mÃª

-0.13

áº¿

-0.13

Ã¤tz

-0.12

abwe

-0.12

alg

-0.12

Ð°Ð½Ð½Ñı

-0.12

POSITIVE LOGITS

 importance

0.48

 need

0.41

 necessity

0.35

 significance

0.33

 value

0.33

 Importance

0.33

 dangers

0.31

need

0.31

import

0.31

 centr

0.30

Activations Density 0.187%

phrases indicating the significance or importance of various subjects or ideas

No Comments

No Known Activations