INDEX

Explanations

references to personal achievements or milestones

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

ancias

-0.15

pite

-0.15

resse

-0.14

ÏĢÎ¿

-0.14

iliz

-0.14

irsch

-0.13

eneg

-0.13

 even

-0.13

à¹īà¸²à¸ķ

-0.13

istra

-0.13

POSITIVE LOGITS

 definitely

0.24

 probably

0.18

 Definitely

0.18

 certainly

0.17

depends

0.16

 depends

0.16

 Depends

0.16

urrent

0.15

 definite

0.15

 creo

0.15

Activations Density 0.255%

references to personal achievements or milestones

No Comments

No Known Activations