INDEX

Explanations

phrases that imply manipulation or diversion of attention

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

ntag

-0.16

enco

-0.16

icio

-0.16

 upstream

-0.15

dater

-0.15

InterfaceOrientation

-0.15

odega

-0.15

bish

-0.15

uts

-0.14

 HOLDERS

-0.14

POSITIVE LOGITS

 away

0.36

 attention

0.33

 toward

0.31

 Away

0.30

 towards

0.29

attention

0.29

 Attention

0.28

 divert

0.27

 diverted

0.27

 onto

0.25

Activations Density 0.060%

phrases that imply manipulation or diversion of attention

No Comments

No Known Activations