INDEX

Explanations

negative adjectives and phrases related to issues of fairness, obligation, and unresolved matters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

point

-0.17

 Rouge

-0.16

pack

-0.15

OrUpdate

-0.15

ä¸įå¥½

-0.15

usz

-0.15

 Karlov

-0.15

ä¸įè¶³

-0.15

izr

-0.14

AZE

-0.14

POSITIVE LOGITS

/un

0.32

(Un

0.20

ably

0.19

/il

0.19

vably

0.17

ly

0.16

ertainty

0.16

ables

0.16

ments

0.15

Ùĩ

0.15

Activations Density 0.151%

negative adjectives and phrases related to issues of fairness, obligation, and unresolved matters

No Comments

No Known Activations