INDEX

Explanations

terms related to technology and user interfaces

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

åĴ²

-0.14

.ncbi

-0.14

rophe

-0.14

.Dir

-0.14

outu

-0.14

chein

-0.13

atrix

-0.13

é£¾

-0.13

icha

-0.13

 ÐºÐ¾Ð½ÑĤÑĢÐ°

-0.13

POSITIVE LOGITS

pen

0.40

Pen

0.35

 pens

0.33

Pen

0.30

 drawing

0.30

 styl

0.30

pen

0.29

 pressure

0.29

 Pens

0.29

_pen

0.28

Activations Density 0.011%

terms related to technology and user interfaces

No Comments

No Known Activations