INDEX

Explanations

actions related to posting, creating, and making changes or updates

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 newfound

-0.17

ëŁ¬

-0.16

zos

-0.16

 newly

-0.15

ÐµÐºÑĤÐ¸

-0.15

ãĥ¼ãĤ¹ãĥĪ

-0.14

æĸ°çļĦ

-0.14

 Newly

-0.14

@n

-0.14

aira

-0.14

POSITIVE LOGITS

 brand

0.45

brand

0.38

 entirely

0.37

 BRAND

0.36

 completely

0.35

 Brand

0.34

Brand

0.32

 totally

0.30

-brand

0.29

 altogether

0.29

Activations Density 0.109%

actions related to posting, creating, and making changes or updates

No Comments

No Known Activations