INDEX

Explanations

various identifiers or codes related to organizations, locations, or features in a document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ãĥ¼ãĥł

-0.15

itech

-0.15

abad

-0.15

ubat

-0.15

 FÃ©

-0.14

/I

-0.14

isy

-0.14

à¹Īà¸²à¸ĩà¸ģ

-0.14

ual

-0.14

iasi

-0.14

POSITIVE LOGITS

Autor

0.20

_EDEFAULT

0.17

ancel

0.16

AND

0.15

/AFP

0.15

'S

0.15

.TXT

0.15

raham

0.15

OF

0.15

’S

0.15

Activations Density 0.130%

various identifiers or codes related to organizations, locations, or features in a document

No Comments

No Known Activations