INDEX

Explanations

phrases associated with numerical values or statistics

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L8R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.8.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

ucch

-0.15

imd

-0.14

anko

-0.13

fds

-0.13

rike

-0.13

 domest

-0.13

Rub

-0.13

osa

-0.12

igram

-0.12

tridge

-0.12

POSITIVE LOGITS

ãģıãģł

0.15

NameValuePair

0.15

 Exit

0.15

ÑĩÐµÑģÐºÐ¸

0.15

 fast

0.14

(Unknown

0.13

 Shank

0.13

 exit

0.13

lrt

0.13

iddi

0.13

Activations Density 0.162%

phrases associated with numerical values or statistics

No Comments

No Known Activations