INDEX

Explanations

variations and forms of the word "trick."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 17-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L17R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.17.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

į°

-0.16

hma

-0.15

æŀ¶

-0.15

ÎµÎ¹Î±

-0.15

 hÆ°á»Łng

-0.15

emoc

-0.14

leck

-0.14

ÑıÐ¶

-0.14

izmet

-0.14

 derece

-0.14

POSITIVE LOGITS

ster

0.24

sters

0.22

ery

0.22

ERY

0.20

 tricks

0.19

 trick

0.16

 isolated

0.16

icular

0.16

 Learned

0.16

 learned

0.16

Activations Density 0.029%

variations and forms of the word "trick."

No Comments

No Known Activations