INDEX

Explanations

references to company practices that involve unethical behavior or misconduct

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

.infinity

-0.14

ovnÃŃ

-0.14

 MAIL

-0.14

 mailbox

-0.14

macen

-0.14

endas

-0.14

 addCriterion

-0.14

 Telecom

-0.14

nn

-0.14

 mail

-0.14

POSITIVE LOGITS

 riders

0.37

 driver

0.37

 rider

0.36

 Uber

0.36

 rides

0.35

 ride

0.34

 drivers

0.33

 Driver

0.32

 Lyft

0.32

 fare

0.31

Activations Density 0.029%

references to company practices that involve unethical behavior or misconduct

No Comments

No Known Activations