INDEX

Explanations

references to organizations and their associated activities or announcements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

Ð»Ð¾Ð´

-0.15

redo

-0.14

èĩ

-0.14

ItemAt

-0.14

 Ð½Ð¸

-0.14

lops

-0.14

uesta

-0.13

à¸£à¸Ķ

-0.13

 crÃ©d

-0.13

itto

-0.13

POSITIVE LOGITS

istry

0.15

ÑĢÐ¾Ðº

0.15

igi

0.15

Tiny

0.15

board

0.14

vida

0.14

 lateral

0.14

quito

0.14

 resid

0.14

asca

0.13

Activations Density 0.008%

references to organizations and their associated activities or announcements

No Comments

No Known Activations