Reinforcement Learning
Transformers
Safetensors
olmo2
text-generation
rlvr
rl-zero
chain-of-thought
faithfulness
reward-hacking
cue-injection
unfaithrl
Instructions to use UnfaithRL/OLMo-2-0425-1B-Instruct-mmlu_trained_ha-2048 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use UnfaithRL/OLMo-2-0425-1B-Instruct-mmlu_trained_ha-2048 with Transformers:
# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("UnfaithRL/OLMo-2-0425-1B-Instruct-mmlu_trained_ha-2048") model = AutoModelForCausalLM.from_pretrained("UnfaithRL/OLMo-2-0425-1B-Instruct-mmlu_trained_ha-2048", device_map="auto") - Notebooks
- Google Colab
- Kaggle
File size: 157 Bytes
56f77e2 | 1 2 3 4 5 6 7 8 9 10 | {
"_from_model_config": true,
"bos_token_id": 100257,
"eos_token_id": [
100257
],
"pad_token_id": 100277,
"transformers_version": "4.56.1"
}
|