INDEX

Explanations

terms and roles related to news journalism and broadcasting

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

 setFrame

-0.15

Ð¾Ð»ÑĮÑĪ

-0.15

ï¼¬

-0.14

angl

-0.14

.TestTools

-0.14

PERT

-0.14

ihan

-0.14

ãĥ³ãĥĶ

-0.14

ully

-0.14

chluss

-0.14

POSITIVE LOGITS

 Giles

0.17

 station

0.16

uhn

0.14

 swim

0.14

 Swim

0.14

emm

0.14

ette

0.13

 property

0.13

 properties

0.13

Activations Density 0.024%

terms and roles related to news journalism and broadcasting

No Comments

No Known Activations