INDEX

Explanations

phrases indicating time or periods of occurrence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

NEXT

-0.16

emailer

-0.15

aks

-0.15

onec

-0.14

bir

-0.14

orie

-0.14

chez

-0.14

WER

-0.14

 Ø¨ÙĪØ§Ø¨Ø©

-0.14

zim

-0.14

POSITIVE LOGITS

the

0.20

 years

0.20

rana

0.17

abant

0.16

InThe

0.15

endor

0.15

 times

0.15

å°Ĩ

0.14

å°ĩ

0.14

 days

0.14

Activations Density 0.060%

phrases indicating time or periods of occurrence

No Comments

No Known Activations