INDEX

Explanations

references to significant cultural and social topics, particularly those related to media, notable figures, and historical events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ander

-0.15

å®ŀåľ¨

-0.13

dG

-0.13

nat

-0.13

bern

-0.13

åĪļæīį

-0.13

 ÙģØ±ÙĪ

-0.12

/latest

-0.12

vrier

-0.12

.chapter

-0.12

POSITIVE LOGITS

 titular

0.14

fragistics

0.14

orry

0.14

.mdl

0.12

vido

0.12

 cual

0.12

LOC

0.12

ä¹ĭä¸Ģ

0.12

pew

0.12

 famously

0.12

Activations Density 0.777%

references to significant cultural and social topics, particularly those related to media, notable figures, and historical events

No Comments

No Known Activations