INDEX

Explanations

terms related to work and productivity

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

 Lynch

-0.15

NOP

-0.15

à¥ģà¤µ

-0.15

urette

-0.15

ovic

-0.15

'&'

-0.15

à¸ŀ

-0.14

onic

-0.14

ifton

-0.14

stellen

-0.14

POSITIVE LOGITS

outs

0.60

-outs

0.52

 Outs

0.50

 outs

0.48

-Out

0.48

OUT

0.46

out

0.43

-out

0.43

Out

0.42

outed

0.37

Activations Density 0.044%

terms related to work and productivity

No Comments

No Known Activations