INDEX

Explanations

specific phrases related to community, education, and organizational names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

uda

-0.15

sea

-0.15

ÏĥÏī

-0.14

UDA

-0.14

'n

-0.14

omik

-0.14

Äı

-0.13

_IRQ

-0.13

 adam

-0.13

 ÑĥÑģÑĤÑĢÐ°

-0.13

POSITIVE LOGITS

vag

0.17

dop

0.15

 Alto

0.15

äº

0.15

SystemService

0.14

lopedia

0.14

HITE

0.14

unning

0.14

Hol

0.14

vf

0.14

Activations Density 0.035%

specific phrases related to community, education, and organizational names

No Comments

No Known Activations