Downloads · 30 days
0
lucky-initialization-atlas-100m-v2-checkpoints
lilywchen/lucky-initialization-atlas-100m-v2-checkpoints
lucky-initialization-atlas-100m-v2-checkpoints is a machine learning model from lilywchen. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for transformers. The card lists the license as apache-2.0.
Private, continuously synchronized model-only checkpoints for the prospectively frozen OLMo-3-ladder 100M initialization atlas. Every visible checkpoint was converted with the pinned olmocore.nn.hf.savehfmodel, serial…
Access
Public
Updated Aug 28, 2026
Repo size
507 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors340 GB · 98%
At a glance
- Library
- transformers
- License
- apache-2.0
- Access
- Public
- Created
- Aug 15, 2026
- Updated
- Aug 28, 2026
- SHA
- 26b7307c
Model card
From the Hugging Face model README
Lucky initialization atlas v2 checkpoints
Private, continuously synchronized model-only checkpoints for the prospectively
frozen OLMo-3-ladder 100M initialization atlas. Every visible checkpoint was
converted with the pinned olmo_core.nn.hf.save_hf_model, serialized with its
pinned tokenizer, and SHA-256 manifested before upload.
Load a specific model with standard Transformers APIs:
from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "lilywchen/lucky-initialization-atlas-100m-v2-checkpoints"
subfolder = "models/<run-id>/step-006210"
model = AutoModelForCausalLM.from_pretrained(repo, subfolder=subfolder)
tokenizer = AutoTokenizer.from_pretrained(repo, subfolder=subfolder)
These are model-only checkpoints. Exact in-progress recovery uses separate local full-state checkpoints; optimizer and RNG states are intentionally not published. The study's primary claims are conditional on its frozen FineWeb corpus/order, OLMo-core recipe, evaluation sets, and B200 runtime.
- Library
- transformers
- License
- apache-2.0
- Created
- Aug 15, 2026
- Updated
- Aug 28, 2026