INDEX

Explanations

key terms related to artistic and professional concepts, often in a context of recognition and influence

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

 fair

-0.14

 èī

-0.14

 ìŀĦ

-0.14

 watch

-0.13

illet

-0.13

preg

-0.13

idential

-0.13

 Pink

-0.13

 basket

-0.13

 olmam

-0.13

POSITIVE LOGITS

ichte

0.18

orz

0.15

erva

0.15

eni

0.14

ublic

0.14

igh

0.14

ICAL

0.13

ÑĥÑĤÑĮ

0.13

OOT

0.13

osc

0.13

Activations Density 0.038%

key terms related to artistic and professional concepts, often in a context of recognition and influence

No Comments

No Known Activations