Finetuning QwQ - Validation Loss NaN

[DaddyCodesAlot]

Hi there, when finetuning Qwen QwQ on a train and validation set, the validation loss is reported as NaN.

I installed unsloth using:
!pip install unsloth

And here's my trainer:
from trl import SFTTrainer
from transformers import TrainingArguments, DataCollatorForSeq2Seq
from unsloth import is_bfloat16_supported

trainer = SFTTrainer(
model = model,
tokenizer = tokenizer,
train_dataset = dataset,
eval_dataset = validate_dataset,
dataset_text_field = "text",
max_seq_length = max_seq_length,
data_collator = DataCollatorForSeq2Seq(tokenizer = tokenizer),
dataset_num_proc = 2,
packing = False, # Can make training 5x faster for short sequences.
args = TrainingArguments(
per_device_train_batch_size = 2,
per_device_eval_batch_size = 2,
gradient_accumulation_steps = 4,
warmup_steps = 5,
num_train_epochs=num_epochs,
learning_rate = 2e-4,
# Do from unsloth import is_bfloat16_supported
# Do from unsloth import is_bfloat16_supported
fp16 = not is_bfloat16_supported(),
bf16 = is_bfloat16_supported(),
logging_steps = 1,
optim = "adamw_8bit",
weight_decay = 0.01,
lr_scheduler_type = "linear",
seed = 3407,
output_dir = "outputs",
save_strategy = "steps",
evaluation_strategy="steps",
eval_steps=checkpoint_count,
save_steps=checkpoint_count, # Saving checkpoint every 100 steps
),
)

[shimmyshimmer]

Do you have an example of the error? Like a screenshot/code? Thanks