INDEX

Explanations

references to physical injuries and medical conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

uet

-0.16

 textured

-0.15

èĥİ

-0.15

ueur

-0.14

à¤®à¤¨

-0.14

.squeeze

-0.14

 kicker

-0.14

orate

-0.14

 skirts

-0.13

ÏĦÎ¹ÎºÏĮ

-0.13

POSITIVE LOGITS

 injuries

0.60

 injury

0.55

 Injury

0.49

 injured

0.47

 wounds

0.41

juries

0.39

ä¼¤

0.38

 wound

0.36

 wounded

0.36

åĤ·

0.35

Activations Density 0.325%

references to physical injuries and medical conditions

No Comments

No Known Activations