INDEX

Explanations

phrases related to routine or regularity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

abis

-0.20

eba

-0.17

llib

-0.16

ABA

-0.16

 tehdy

-0.15

intl

-0.15

ç¥Ŀ

-0.14

clist

-0.14

(åľŁ

-0.14

ãĥ¼ãĥĩ

-0.14

POSITIVE LOGITS

0.18

reck

0.17

ÅĽcie

0.15

 normally

0.15

ocking

0.15

ETO

0.15

 companion

0.14

 Chip

0.14

Activations Density 0.155%

phrases related to routine or regularity

No Comments

No Known Activations