INDEX

Explanations

phrases indicating a request for additional information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

aklÄ±

-0.16

uir

-0.15

ntag

-0.15

į¼

-0.14

æīįèĥ½

-0.14

uzzi

-0.13

uce

-0.13

ãģŁãĤģãģ®

-0.13

 knack

-0.13

POSITIVE LOGITS

 information

0.56

 info

0.48

information

0.45

ä¿¡æģ¯

0.41

 Information

0.41

 informatie

0.40

 informaciÃ³n

0.38

-information

0.38

Information

0.37

 details

0.36

Activations Density 0.028%

phrases indicating a request for additional information

No Comments

No Known Activations