INDEX

Explanations

words reflecting strong emotional reactions or experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

upa

-0.15

undo

-0.15

Ø³Ø§ÙĨ

-0.15

 mastur

-0.15

bre

-0.14

ëĿ¼ëıĦ

-0.14

-0.13

InParameter

-0.13

POSITIVE LOGITS

eki

0.15

Nob

0.15

oni

0.14

 topLeft

0.14

obel

0.14

yle

0.14

sworth

0.13

 Surprise

0.13

 instead

0.13

Calibri

0.13

Activations Density 0.163%

words reflecting strong emotional reactions or experiences

No Comments

No Known Activations