INDEX

Explanations

technical terms related to computer science and software

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ulary

-0.16

McA

-0.15

endale

-0.14

lag

-0.14

ungal

-0.14

LAG

-0.14

atown

-0.13

ìĬ¤íģ¬

-0.13

 æ¹

-0.13

uder

-0.13

POSITIVE LOGITS

agi

0.15

xCD

0.15

_native

0.14

$MESS

0.14

riot

0.14

rou

0.13

æĳĺ

0.13

elder

0.13

 cÃ©

0.13

oir

0.13

Activations Density 0.092%

technical terms related to computer science and software

No Comments

No Known Activations