INDEX

Explanations

references to teams and teamwork

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

New Auto-Interp

Configuration

fnlp/Llama3_1-8B-Base-LXM-8x/Llama3_1-8B-Base-L22M-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.hook_mlp_out

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

 team

-1.16

team

-0.98

 Team

-0.94

-team

-0.91

 teams

-0.90

Team

-0.87

_team

-0.79

 TEAM

-0.79

teams

-0.73

(team

-0.73

POSITIVE LOGITS

iá»ĩn

0.15

ëĶ°

0.15

Ø¨ÙĨ

0.15

/Internal

0.14

inder

0.14

Ã³st

0.14

gil

0.14

aho

0.14

,vector

0.14

.IsSuccess

0.13

Activations Density 0.280%

references to teams and teamwork

No Comments

No Known Activations