INDEX

Explanations

assertive statements and references to documents or reports

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

cav

-0.16

ieu

-0.15

 exposition

-0.14

CAF

-0.14

 Ferguson

-0.14

 Tooth

-0.14

UIL

-0.14

uis

-0.13

ys

-0.13

and

-0.13

POSITIVE LOGITS

veau

0.18

ivol

0.17

ystate

0.16

RefCount

0.14

uga

0.14

zdy

0.14

onian

0.14

agina

0.14

villa

0.14

-NLS

0.14

Activations Density 0.059%

assertive statements and references to documents or reports

No Comments

No Known Activations