INDEX

Explanations

terms related to advanced scientific or technical capabilities and analysis

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

éŁ

-0.15

 electron

-0.15

auth

-0.14

_misc

-0.14

jvu

-0.13

altet

-0.13

vÄĽÅĻ

-0.13

ecies

-0.13

 aras

-0.13

POSITIVE LOGITS

automation

0.21

 wells

0.19

 Wells

0.17

Automation

0.17

Array

0.17

Platform

0.17

zym

0.17

 micro

0.16

 Rapids

0.16

 chips

0.16

Activations Density 0.044%

terms related to advanced scientific or technical capabilities and analysis

No Comments

No Known Activations