INDEX

Explanations

references to specific individuals, locations, and cultural products

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 ÐĲÐ»ÐµÐºÑģÐ°Ð½Ð´ÑĢ

-0.20

ller

-0.19

inÃ¡ÅĻ

-0.18

Ð¿Ð¸Ð¾Ð½

-0.17

 Ð´Ð¾ÐºÑĤÐ¾ÑĢ

-0.17

ÏĦÎ®ÏĤ

-0.17

ajan

-0.17

 Ð¿Ð¾Ð»ÑĮÐ·Ð¾Ð²Ð°ÑĤ

-0.17

Î¹ÏĥÎ¼ÏĮÏĤ

-0.17

orer

-0.16

POSITIVE LOGITS

Ð¸ÑĤÐµÐ»Ñı

0.23

Ð²Ð°Ð½Ð°

0.22

Ð°ÑĤÐµÐ»Ñı

0.22

iego

0.22

ÐµÐºÑĤÐ¾ÑĢÐ°

0.21

mana

0.20

onga

0.20

 Paula

0.20

Ð°Ð½Ð°

0.20

 Arth

0.19

Activations Density 0.107%

references to specific individuals, locations, and cultural products

No Comments

No Known Activations