INDEX

Explanations

references to healthcare and medical practices

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

 Snape

-0.17

ï¼Īå¹³æĪĲ

-0.15

utsche

-0.14

metro

-0.14

usat

-0.13

SelectionMode

-0.13

 IHttpActionResult

-0.13

Ð·ÑĮ

-0.13

sect

-0.13

quential

-0.13

POSITIVE LOGITS

 COVID

0.36

COVID

0.34

 pandemic

0.30

 Covid

0.30

 coronavirus

0.29

 Virtual

0.28

 virtual

0.28

 ì½Ķë¡ľëĤĺ

0.27

 covid

0.27

 Coronavirus

0.25

Activations Density 0.818%

references to healthcare and medical practices

No Comments

No Known Activations