INDEX

Explanations

words and phrases related to casualties and injuries in emergency situations

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

zby

-0.15

emia

-0.14

ibern

-0.14

.Debugger

-0.14

incoming

-0.13

ASTE

-0.13

 à¹Ĩ

-0.13

typeid

-0.13

Incoming

-0.13

aste

-0.13

POSITIVE LOGITS

kar

0.17

 Toll

0.15

*)_

0.15

 Wade

0.15

 Firm

0.14

PING

0.14

errat

0.14

insky

0.14

ikon

0.14

utr

0.13

Activations Density 0.038%

words and phrases related to casualties and injuries in emergency situations

No Comments

No Known Activations