INDEX

Explanations

references to positions, roles, or classifications within organizational structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

icer

-0.15

 itself

-0.14

illac

-0.14

á½·

-0.14

icens

-0.14

olean

-0.13

 ÑĤÑĢÐ¾

-0.13

ãĢģäºĮ

-0.13

Pey

-0.13

 settings

-0.13

POSITIVE LOGITS

 respectively

0.19

åĪĨåĪ«

0.17

ati

0.17

/features

0.15

 ÑģÐ¾Ð¾ÑĤÐ²ÐµÑĤ

0.15

fait

0.15

esser

0.15

ISO

0.14

akov

0.14

anta

0.13

Activations Density 0.200%

references to positions, roles, or classifications within organizational structures

No Comments

No Known Activations