vlm_policy_init_7b_lora / train_results.json
dcahn12
Add policy init lora checkpoint
72f46ef
raw
history blame contribute delete
167 Bytes
{
"epoch": 1.0,
"train_loss": 1.4022162783462389,
"train_runtime": 13253.0183,
"train_samples_per_second": 2.186,
"train_steps_per_second": 0.017
}