developer_uid: rirv938
submission_id: rirv938-grpo-to-compare-_3561_v1
model_name: rirv938-grpo-to-compare-_3561_v1
model_group: rirv938/grpo_to_compare_
status: torndown
timestamp: 2025-04-25T23:31:57+00:00
num_battles: 7719
num_wins: 3834
celo_rating: 1284.76
family_friendly_score: 0.5651999999999999
family_friendly_standard_error: 0.007010691264062339
submission_type: basic
model_repo: rirv938/grpo_to_compare_to_dpo_10k_cp312_g16_merged
model_architecture: MistralForCausalLM
model_num_parameters: 24096691200.0
best_of: 8
max_input_tokens: 1024
max_output_tokens: 64
reward_model: default
latencies: [{'batch_size': 1, 'throughput': 0.525195480675599, 'latency_mean': 1.9039873957633973, 'latency_p50': 1.9268687963485718, 'latency_p90': 2.0979568004608153}, {'batch_size': 3, 'throughput': 1.0825892044213539, 'latency_mean': 2.762799277305603, 'latency_p50': 2.7815574407577515, 'latency_p90': 3.0564582347869873}, {'batch_size': 5, 'throughput': 1.3703439724265087, 'latency_mean': 3.62459326505661, 'latency_p50': 3.6244051456451416, 'latency_p90': 4.0540040016174315}, {'batch_size': 6, 'throughput': 1.4952667071470507, 'latency_mean': 3.980293152332306, 'latency_p50': 3.9884084463119507, 'latency_p90': 4.491108250617981}, {'batch_size': 8, 'throughput': 1.6380291495606842, 'latency_mean': 4.842117869853974, 'latency_p50': 4.843689322471619, 'latency_p90': 5.461726093292237}, {'batch_size': 10, 'throughput': 1.7105294452867148, 'latency_mean': 5.796834483146667, 'latency_p50': 5.915446162223816, 'latency_p90': 6.492350602149964}]
gpu_counts: {'NVIDIA A100-SXM4-80GB': 1}
display_name: rirv938-grpo-to-compare-_3561_v1
is_internal_developer: True
language_model: rirv938/grpo_to_compare_to_dpo_10k_cp312_g16_merged
model_size: 24B
ranking_group: single
throughput_3p7s: 1.41
us_pacific_date: 2025-04-25
win_ratio: 0.4966964632724446
generation_params: {'temperature': 0.9, 'top_p': 0.9, 'min_p': 0.2, 'top_k': 80, 'presence_penalty': 0.5, 'frequency_penalty': 0.5, 'stopping_words': ['###', '</s>', '\n', 'You:'], 'max_input_tokens': 1024, 'best_of': 8, 'max_output_tokens': 64}
formatter: {'memory_template': '', 'prompt_template': '', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '{bot_name}:', 'truncate_by_message': False}
Resubmit model
Shutdown handler not registered because Python interpreter is not running in the main thread
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLizer
Starting job with name rirv938-grpo-to-compare-3561-v1-mkmlizer
Waiting for job on rirv938-grpo-to-compare-3561-v1-mkmlizer to finish
rirv938-grpo-to-compare-3561-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rirv938-grpo-to-compare-3561-v1-mkmlizer: ║ _____ __ __ ║
rirv938-grpo-to-compare-3561-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
rirv938-grpo-to-compare-3561-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
rirv938-grpo-to-compare-3561-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
rirv938-grpo-to-compare-3561-v1-mkmlizer: ║ /___/ ║
rirv938-grpo-to-compare-3561-v1-mkmlizer: ║ ║
rirv938-grpo-to-compare-3561-v1-mkmlizer: ║ Version: 0.12.8 ║
rirv938-grpo-to-compare-3561-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
rirv938-grpo-to-compare-3561-v1-mkmlizer: ║ https://mk1.ai ║
rirv938-grpo-to-compare-3561-v1-mkmlizer: ║ ║
rirv938-grpo-to-compare-3561-v1-mkmlizer: ║ The license key for the current software has been verified as ║
rirv938-grpo-to-compare-3561-v1-mkmlizer: ║ belonging to: ║
rirv938-grpo-to-compare-3561-v1-mkmlizer: ║ ║
rirv938-grpo-to-compare-3561-v1-mkmlizer: ║ Chai Research Corp. ║
rirv938-grpo-to-compare-3561-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
rirv938-grpo-to-compare-3561-v1-mkmlizer: ║ Expiration: 2028-03-31 23:59:59 ║
rirv938-grpo-to-compare-3561-v1-mkmlizer: ║ ║
rirv938-grpo-to-compare-3561-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
rirv938-grpo-to-compare-3561-v1-mkmlizer: Downloaded to shared memory in 165.109s
rirv938-grpo-to-compare-3561-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpu1v9gc8z, device:0
rirv938-grpo-to-compare-3561-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
rirv938-grpo-to-compare-3561-v1-mkmlizer: quantized model in 84.826s
rirv938-grpo-to-compare-3561-v1-mkmlizer: Processed model rirv938/grpo_to_compare_to_dpo_10k_cp312_g16_merged in 249.936s
rirv938-grpo-to-compare-3561-v1-mkmlizer: creating bucket guanaco-mkml-models
rirv938-grpo-to-compare-3561-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rirv938-grpo-to-compare-3561-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rirv938-grpo-to-compare-3561-v1
rirv938-grpo-to-compare-3561-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rirv938-grpo-to-compare-3561-v1/config.json
rirv938-grpo-to-compare-3561-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rirv938-grpo-to-compare-3561-v1/special_tokens_map.json
rirv938-grpo-to-compare-3561-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rirv938-grpo-to-compare-3561-v1/tokenizer_config.json
rirv938-grpo-to-compare-3561-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rirv938-grpo-to-compare-3561-v1/tokenizer.json
rirv938-grpo-to-compare-3561-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/rirv938-grpo-to-compare-3561-v1/flywheel_model.1.safetensors
rirv938-grpo-to-compare-3561-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rirv938-grpo-to-compare-3561-v1/flywheel_model.0.safetensors
rirv938-grpo-to-compare-3561-v1-mkmlizer: Loading 0: 0%| | 0/363 [00:00<?, ?it/s] Loading 0: 1%| | 3/363 [00:00<00:24, 14.60it/s] Loading 0: 1%|▏ | 5/363 [00:00<01:04, 5.54it/s] Loading 0: 3%|▎ | 10/363 [00:00<00:29, 11.97it/s] Loading 0: 4%|▎ | 13/363 [00:01<00:35, 9.90it/s] Loading 0: 4%|▍ | 15/363 [00:02<01:13, 4.75it/s] Loading 0: 5%|▌ | 19/363 [00:02<00:47, 7.19it/s] Loading 0: 6%|▌ | 21/363 [00:03<00:55, 6.16it/s] Loading 0: 6%|▋ | 23/363 [00:04<01:34, 3.61it/s] Loading 0: 7%|▋ | 27/363 [00:04<01:00, 5.59it/s] Loading 0: 8%|▊ | 29/363 [00:04<00:50, 6.66it/s] Loading 0: 9%|▊ | 31/363 [00:04<00:43, 7.63it/s] Loading 0: 9%|▉ | 33/363 [00:04<00:41, 7.98it/s] Loading 0: 10%|▉ | 35/363 [00:05<01:17, 4.26it/s] Loading 0: 10%|▉ | 36/363 [00:06<01:37, 3.34it/s] Loading 0: 11%|█ | 39/363 [00:07<01:22, 3.91it/s] Loading 0: 11%|█ | 40/363 [00:07<01:39, 3.24it/s] Loading 0: 11%|█▏ | 41/363 [00:08<01:57, 2.75it/s] Loading 0: 12%|█▏ | 45/363 [00:08<01:02, 5.09it/s] Loading 0: 13%|█▎ | 47/363 [00:08<00:50, 6.32it/s] Loading 0: 13%|█▎ | 49/363 [00:08<00:42, 7.43it/s] Loading 0: 14%|█▍ | 51/363 [00:09<00:39, 7.87it/s] Loading 0: 15%|█▍ | 53/363 [00:09<01:07, 4.57it/s] Loading 0: 15%|█▍ | 54/363 [00:10<01:11, 4.30it/s] Loading 0: 16%|█▌ | 57/363 [00:10<00:51, 5.92it/s] Loading 0: 16%|█▌ | 58/363 [00:10<00:55, 5.52it/s] Loading 0: 16%|█▋ | 59/363 [00:11<01:02, 4.86it/s] Loading 0: 18%|█▊ | 64/363 [00:11<00:31, 9.50it/s] Loading 0: 19%|█▊ | 68/363 [00:11<00:23, 12.62it/s] Loading 0: 19%|█▉ | 70/363 [00:11<00:29, 9.78it/s] Loading 0: 20%|█▉ | 72/363 [00:12<00:38, 7.51it/s] Loading 0: 21%|██ | 75/363 [00:12<00:31, 9.04it/s] Loading 0: 21%|██ | 77/363 [00:12<00:35, 8.13it/s] Loading 0: 23%|██▎ | 82/363 [00:12<00:21, 12.94it/s] Loading 0: 24%|██▎ | 86/363 [00:12<00:16, 16.48it/s] Loading 0: 25%|██▍ | 89/363 [00:13<00:20, 13.33it/s] Loading 0: 25%|██▌ | 91/363 [00:13<00:20, 13.03it/s] Loading 0: 26%|██▌ | 95/363 [00:13<00:15, 16.94it/s] Loading 0: 27%|██▋ | 99/363 [00:13<00:12, 20.38it/s] Loading 0: 28%|██▊ | 102/363 [00:13<00:13, 19.19it/s] Loading 0: 29%|██▉ | 105/363 [00:14<00:17, 14.52it/s] Loading 0: 29%|██▉ | 107/363 [00:14<00:18, 13.63it/s] Loading 0: 30%|███ | 109/363 [00:14<00:19, 13.29it/s] Loading 0: 31%|███ | 111/363 [00:14<00:17, 14.29it/s] Loading 0: 31%|███ | 113/363 [00:14<00:21, 11.80it/s] Loading 0: 33%|███▎ | 118/363 [00:15<00:13, 17.95it/s] Loading 0: 34%|███▎ | 122/363 [00:15<00:11, 21.43it/s] Loading 0: 34%|███▍ | 125/363 [00:15<00:15, 15.09it/s] Loading 0: 35%|███▌ | 128/363 [00:15<00:14, 15.93it/s] Loading 0: 36%|███▌ | 130/363 [00:15<00:16, 14.34it/s] Loading 0: 36%|███▋ | 132/363 [00:16<00:16, 13.87it/s] Loading 0: 37%|███▋ | 136/363 [00:16<00:12, 17.95it/s] Loading 0: 39%|███▊ | 140/363 [00:16<00:10, 21.55it/s] Loading 0: 39%|███▉ | 143/363 [00:16<00:14, 15.12it/s] Loading 0: 40%|███▉ | 145/363 [00:16<00:15, 14.34it/s] Loading 0: 40%|████ | 147/363 [00:16<00:14, 15.24it/s] Loading 0: 41%|████ | 149/363 [00:17<00:17, 12.49it/s] Loading 0: 42%|████▏ | 154/363 [00:17<00:11, 18.60it/s] Loading 0: 44%|████▎ | 158/363 [00:17<00:09, 21.96it/s] Loading 0: 44%|████▍ | 161/363 [00:17<00:13, 15.37it/s] Loading 0: 45%|████▌ | 164/363 [00:17<00:12, 16.15it/s] Loading 0: 46%|████▌ | 167/363 [00:18<00:15, 12.80it/s] Loading 0: 47%|████▋ | 172/363 [00:18<00:10, 17.82it/s] Loading 0: 48%|████▊ | 176/363 [00:18<00:08, 21.01it/s] Loading 0: 49%|████▉ | 179/363 [00:18<00:12, 15.16it/s] Loading 0: 50%|█████ | 182/363 [00:18<00:11, 15.98it/s] Loading 0: 51%|█████ | 185/363 [00:19<00:13, 13.02it/s] Loading 0: 52%|█████▏ | 190/363 [00:19<00:09, 17.96it/s] Loading 0: 53%|█████▎ | 194/363 [00:19<00:08, 21.07it/s] Loading 0: 54%|█████▍ | 197/363 [00:19<00:10, 15.85it/s] Loading 0: 55%|█████▌ | 200/363 [00:20<00:09, 16.42it/s] Loading 0: 55%|█████▌ | 200/363 [00:30<00:09, 16.42it/s] Loading 0: 55%|█████▌ | 201/363 [00:40<06:12, 2.30s/it] Loading 0: 56%|█████▌ | 203/363 [00:40<04:44, 1.78s/it] Loading 0: 57%|█████▋ | 207/363 [00:40<02:45, 1.06s/it] Loading 0: 58%|█████▊ | 210/363 [00:40<01:55, 1.33it/s] Loading 0: 59%|█████▊ | 213/363 [00:41<01:23, 1.81it/s] Loading 0: 59%|█████▉ | 215/363 [00:41<01:06, 2.22it/s] Loading 0: 60%|█████▉ | 217/363 [00:41<00:52, 2.78it/s] Loading 0: 60%|██████ | 219/363 [00:41<00:40, 3.56it/s] Loading 0: 61%|██████ | 221/363 [00:41<00:34, 4.15it/s] Loading 0: 62%|██████▏ | 226/363 [00:41<00:18, 7.44it/s] Loading 0: 63%|██████▎ | 230/363 [00:41<00:12, 10.35it/s] Loading 0: 64%|██████▍ | 233/363 [00:42<00:13, 9.79it/s] Loading 0: 65%|██████▌ | 236/363 [00:42<00:11, 11.28it/s] Loading 0: 66%|██████▌ | 239/363 [00:42<00:12, 10.33it/s] Loading 0: 67%|██████▋ | 244/363 [00:42<00:07, 15.03it/s] Loading 0: 68%|██████▊ | 248/363 [00:43<00:06, 18.41it/s] Loading 0: 69%|██████▉ | 251/363 [00:43<00:07, 14.50it/s] Loading 0: 70%|██████▉ | 254/363 [00:43<00:07, 15.42it/s] Loading 0: 71%|███████ | 257/363 [00:43<00:08, 12.57it/s] Loading 0: 72%|███████▏ | 262/363 [00:44<00:05, 17.38it/s] Loading 0: 73%|███████▎ | 266/363 [00:44<00:04, 20.66it/s] Loading 0: 74%|███████▍ | 269/363 [00:44<00:06, 15.17it/s] Loading 0: 75%|███████▍ | 272/363 [00:44<00:05, 15.78it/s] Loading 0: 76%|███████▌ | 275/363 [00:45<00:07, 12.57it/s] Loading 0: 77%|███████▋ | 280/363 [00:45<00:04, 17.04it/s] Loading 0: 78%|███████▊ | 284/363 [00:45<00:03, 20.08it/s] Loading 0: 79%|███████▉ | 287/363 [00:45<00:05, 14.51it/s] Loading 0: 80%|███████▉ | 290/363 [00:45<00:04, 15.13it/s] Loading 0: 80%|████████ | 292/363 [00:46<00:05, 13.70it/s] Loading 0: 81%|████████ | 294/363 [00:46<00:05, 13.25it/s] Loading 0: 82%|████████▏ | 298/363 [00:46<00:03, 17.51it/s] Loading 0: 83%|████████▎ | 302/363 [00:46<00:02, 21.13it/s] Loading 0: 84%|████████▍ | 305/363 [00:46<00:04, 14.27it/s] Loading 0: 85%|████████▍ | 307/363 [00:46<00:04, 13.64it/s] Loading 0: 85%|████████▌ | 309/363 [00:47<00:03, 14.41it/s] Loading 0: 86%|████████▌ | 311/363 [00:47<00:04, 11.77it/s] Loading 0: 87%|████████▋ | 316/363 [00:47<00:02, 17.78it/s] Loading 0: 88%|████████▊ | 320/363 [00:47<00:02, 21.22it/s] Loading 0: 89%|████████▉ | 323/363 [00:47<00:02, 14.58it/s] Loading 0: 90%|████████▉ | 326/363 [00:48<00:02, 15.24it/s] Loading 0: 90%|█████████ | 328/363 [00:48<00:02, 13.70it/s] Loading 0: 91%|█████████ | 330/363 [00:48<00:02, 13.14it/s] Loading 0: 92%|█████████▏| 334/363 [00:48<00:01, 17.36it/s] Loading 0: 93%|█████████▎| 338/363 [00:48<00:01, 20.86it/s] Loading 0: 94%|█████████▍| 341/363 [00:49<00:01, 14.32it/s] Loading 0: 94%|█████████▍| 343/363 [00:49<00:01, 13.72it/s] Loading 0: 95%|█████████▌| 345/363 [00:49<00:01, 14.62it/s] Loading 0: 96%|█████████▌| 347/363 [00:49<00:01, 11.84it/s] Loading 0: 97%|█████████▋| 352/363 [00:49<00:00, 17.91it/s] Loading 0: 98%|█████████▊| 356/363 [00:49<00:00, 21.38it/s] Loading 0: 99%|█████████▉| 359/363 [00:57<00:02, 1.34it/s] Loading 0: 99%|█████████▉| 361/363 [00:57<00:01, 1.62it/s]
Job rirv938-grpo-to-compare-3561-v1-mkmlizer completed after 279.09s with status: succeeded
Stopping job with name rirv938-grpo-to-compare-3561-v1-mkmlizer
Pipeline stage MKMLizer completed in 279.57s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.15s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service rirv938-grpo-to-compare-3561-v1
Waiting for inference service rirv938-grpo-to-compare-3561-v1 to be ready
Inference service rirv938-grpo-to-compare-3561-v1 ready after 150.55409598350525s
Pipeline stage MKMLDeployer completed in 151.23s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.4850144386291504s
Received healthy response to inference request in 1.4551365375518799s
Received healthy response to inference request in 2.06927490234375s
Received healthy response to inference request in 1.7857739925384521s
Received healthy response to inference request in 1.7717094421386719s
5 requests
0 failed requests
5th percentile: 1.5184511184692382
10th percentile: 1.5817656993865967
20th percentile: 1.7083948612213136
30th percentile: 1.774522352218628
40th percentile: 1.78014817237854
50th percentile: 1.7857739925384521
60th percentile: 1.8991743564605712
70th percentile: 2.0125747203826903
80th percentile: 2.15242280960083
90th percentile: 2.3187186241149904
95th percentile: 2.4018665313720704
99th percentile: 2.4683848571777345
mean time: 1.913381862640381
Pipeline stage StressChecker completed in 10.87s
run pipeline stage %s
Running pipeline stage OfflineFamilyFriendlyTriggerPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%s
triggered trigger_guanaco_pipeline args=%s
Pipeline stage OfflineFamilyFriendlyTriggerPipeline completed in 0.80s
run pipeline stage %s
Running pipeline stage TriggerMKMLProfilingPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%s
triggered trigger_guanaco_pipeline args=%s
Pipeline stage TriggerMKMLProfilingPipeline completed in 0.63s
Shutdown handler de-registered
rirv938-grpo-to-compare-_3561_v1 status is now deployed due to DeploymentManager action
Shutdown handler registered
run pipeline %s
run pipeline stage %s
Running pipeline stage OfflineFamilyFriendlyScorer
Evaluating %s Family Friendly Score with %s threads
Pipeline stage OfflineFamilyFriendlyScorer completed in 2744.80s
Shutdown handler de-registered
rirv938-grpo-to-compare-_3561_v1 status is now inactive due to auto deactivation removed underperforming models
rirv938-grpo-to-compare-_3561_v1 status is now torndown due to DeploymentManager action