INDEX

Explanations

references to specific brands and products, particularly in the automotive industry

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

apiro

-0.14

akov

-0.14

ãĥ¼ãĥĳãĥ¼

-0.13

ÑĦÐ¾ÑĢ

-0.13

Ã±ana

-0.13

VID

-0.12

opies

-0.12

ocaly

-0.12

ViewState

-0.12

errer

-0.12

POSITIVE LOGITS

 dÃ¡n

0.16

ukan

0.15

Hao

0.14

undle

0.14

ÅĻez

0.13

shoot

0.13

amm

0.13

ç½²

0.13

pcf

0.13

xo

0.13

Activations Density 0.011%

references to specific brands and products, particularly in the automotive industry

No Comments

No Known Activations