INDEX

Explanations

expressions of opinion or preference related to products or experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

 elsewhere

-0.17

 someone

-0.15

åĪ¥

-0.15

 occasionally

-0.14

 periodically

-0.14

 sometimes

-0.14

åı¦

-0.13

 another

-0.13

à¸ļà¸²à¸ĩ

-0.13

 Sometimes

-0.13

POSITIVE LOGITS

æīĢæľī

0.72

all

0.70

 every

0.68

 wszyst

0.64

ãģĻãģ¹ãģ¦

0.63

 semua

0.61

 ëª¨ëĵł

0.60

 everything

0.60

 Ð²ÑģÐµÑħ

0.57

 todas

0.55

Activations Density 1.118%

expressions of opinion or preference related to products or experiences

No Comments

No Known Activations