INDEX

Explanations

terms related to scientific and technical concepts

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

æĮ¯ãĤĬ

-0.15

 Manual

-0.14

oug

-0.14

 Enum

-0.14

/manual

-0.13

Ð¸ÑĩÐ°

-0.13

preload

-0.13

resi

-0.13

avir

-0.13

estre

-0.13

POSITIVE LOGITS

 excit

0.29

 solar

0.28

 hole

0.28

 device

0.27

dye

0.27

 devices

0.26

 OLED

0.25

 organic

0.24

 sensit

0.24

 efficiencies

0.24

Activations Density 0.009%

terms related to scientific and technical concepts

No Comments

No Known Activations