Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
pwython
29 days ago
|
parent
|
context
|
favorite
| on:
LLM from scratch, part 28 – training a base model ...
For those that have homebrewed a base model, does your output have the same AI-isms like overusing em dashes? If so/not, what dataset did you use?
itissid
29 days ago
|
next
[–]
Does yours also use the oxford comma and generally more commas?
miki123211
29 days ago
|
prev
|
next
[–]
AFAIK, those are mostly a consequence of posttraining.
whimsicalism
29 days ago
|
prev
[–]
that is a post-training artifact
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: