INDEX

Explanations

references to writing instruments and tools

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

è¾

-0.16

bild

-0.15

ATIO

-0.15

 mÃ¡lo

-0.14

.pow

-0.14

unda

-0.13

orent

-0.13

POSITIVE LOGITS

 pens

0.32

nib

0.31

pen

0.31

ink

0.28

 fountain

0.27

Pen

0.26

 Fountain

0.25

Ink

0.24

Pen

0.24

 Pens

0.24

Activations Density 0.007%

references to writing instruments and tools

No Comments

No Known Activations