INDEX

Explanations

references to times and dates

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

velle

-0.16

hower

-0.16

readcrumbs

-0.15

ÑģÑĤÑİ

-0.15

reach

-0.14

apo

-0.14

shima

-0.14

velt

-0.14

inerary

-0.14

hev

-0.14

POSITIVE LOGITS

 same

0.18

same

0.16

 mismo

0.15

 Beispiel

0.15

£

0.15

íĮĮ

0.15

ÏģÎ¯

0.15

ars

0.15

Ã¡bb

0.15

izon

0.14

Activations Density 0.005%

references to times and dates

No Comments

No Known Activations