INDEX

Explanations

mathematical and scientific notation formats, particularly those used in equations and expressions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

ombine

-0.16

ÑīÐ¸Ð½Ñĭ

-0.16

quia

-0.16

deen

-0.15

mony

-0.15

ä»ª

-0.15

echa

-0.14

obra

-0.14

 Tamb

-0.14

ifo

-0.14

POSITIVE LOGITS

}↵

0.16

aries

0.16

oret

0.15

anela

0.15

oris

0.15

ware

0.15

Å¡ov

0.14

ories

0.14

aling

0.14

SHARE

0.14

Activations Density 0.060%

mathematical and scientific notation formats, particularly those used in equations and expressions

No Comments

No Known Activations