INDEX

Explanations

references to graphical representations or tables in the document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

_dll

-0.14

ardi

-0.14

abor

-0.14

 ÑĢÐ°ÑģÐº

-0.14

$__

-0.14

ederland

-0.13

modifiable

-0.13

ully

-0.13

 forCell

-0.13

-pointer

-0.12

POSITIVE LOGITS

}[

0.28

}↵

0.24

0.20

)[

0.19

 Vern

0.19

*↵

0.18

adora

0.18

↵

0.17

"[

0.17

Activations Density 0.021%

references to graphical representations or tables in the document

No Comments

No Known Activations