INDEX

Explanations

sequences of numbers and symbols typically used in technical or structured data formats

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

acon

-0.15

inish

-0.15

ez

-0.15

ãĥ¼ãĤ¹

-0.14

credit

-0.14

OLON

-0.14

ilig

-0.14

.wr

-0.14

cion

-0.13

POSITIVE LOGITS

endor

0.15

lein

0.14

ritz

0.14

Ø±Ø§ÙĨÙĩ

0.14

 Real

0.14

awei

0.14

0.13

elling

0.13

unn

0.13

ront

0.13

Activations Density 0.015%

sequences of numbers and symbols typically used in technical or structured data formats

No Comments

No Known Activations