rirv938-grpo-20250711-c_20147

developer_uid: rirv938

submission_id: rirv938-grpo-20250711-c_20147_v2

model_name: rirv938-grpo-20250711-c_20147_v2

model_group: rirv938/grpo_20250711_cp

status: torndown

timestamp: 2025-07-12T04:26:26+00:00

num_battles: 9882

num_wins: 4947

celo_rating: 1287.43

family_friendly_score: 0.4766

family_friendly_standard_error: 0.007063319899310805

submission_type: basic

model_repo: rirv938/grpo_20250711_cp624_sid_mistral_24b_dpo_40k__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': 0.5182795050408139, 'latency_mean': 1.9293631410598755, 'latency_p50': 1.9441053867340088, 'latency_p90': 2.1339847087860107}, {'batch_size': 3, 'throughput': 1.0485527824200154, 'latency_mean': 2.8505406093597414, 'latency_p50': 2.858670234680176, 'latency_p90': 3.137733554840088}, {'batch_size': 5, 'throughput': 1.3366925971546855, 'latency_mean': 3.726157379150391, 'latency_p50': 3.73051118850708, 'latency_p90': 4.138593339920044}, {'batch_size': 6, 'throughput': 1.4519128132569956, 'latency_mean': 4.100167000293732, 'latency_p50': 4.103496313095093, 'latency_p90': 4.566594576835632}, {'batch_size': 8, 'throughput': 1.5783918983636707, 'latency_mean': 5.021753886938095, 'latency_p50': 4.980723142623901, 'latency_p90': 5.6626249074935915}, {'batch_size': 10, 'throughput': 1.6669619474001376, 'latency_mean': 5.943215144872665, 'latency_p50': 5.97039270401001, 'latency_p90': 6.6939586162567135}]

gpu_counts: {'NVIDIA A100-SXM4-80GB': 1}

display_name: rirv938-grpo-20250711-c_20147_v2

is_internal_developer: True

language_model: rirv938/grpo_20250711_cp624_sid_mistral_24b_dpo_40k__merged

model_size: 24B

ranking_group: single

throughput_3p7s: 1.34

us_pacific_date: 2025-07-11

win_ratio: 0.5006071645415908

generation_params: {'temperature': 0.9, 'top_p': 0.9, 'min_p': 0.6, 'top_k': 80, 'presence_penalty': 0.5, 'frequency_penalty': 0.5, 'stopping_words': ['###', '\n', '</s>', 'You:'], 'max_input_tokens': 1024, 'best_of': 8, 'max_output_tokens': 64}

formatter: {'memory_template': '[INST]', 'prompt_template': '', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '[/INST]{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-20250711-c-20147-v2-mkmlizer
Waiting for job on rirv938-grpo-20250711-c-20147-v2-mkmlizer to finish
rirv938-grpo-20250711-c-20147-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rirv938-grpo-20250711-c-20147-v2-mkmlizer: ║                                                                     ║
rirv938-grpo-20250711-c-20147-v2-mkmlizer: ║   ██████   ██████  █████   ████  ████                               ║
rirv938-grpo-20250711-c-20147-v2-mkmlizer: ║  ░░██████ ██████  ░░███   ███░  ░░███                               ║
rirv938-grpo-20250711-c-20147-v2-mkmlizer: ║   ░███░█████░███   ░███  ███     ░███                               ║
rirv938-grpo-20250711-c-20147-v2-mkmlizer: ║   ░███░░███ ░███   ░███████      ░███                               ║
rirv938-grpo-20250711-c-20147-v2-mkmlizer: ║   ░███ ░░░  ░███   ░███░░███     ░███                               ║
rirv938-grpo-20250711-c-20147-v2-mkmlizer: ║   ░███      ░███   ░███ ░░███    ░███                               ║
rirv938-grpo-20250711-c-20147-v2-mkmlizer: ║   █████     █████  █████ ░░████  █████                              ║
rirv938-grpo-20250711-c-20147-v2-mkmlizer: ║  ░░░░░     ░░░░░  ░░░░░   ░░░░  ░░░░░                               ║
rirv938-grpo-20250711-c-20147-v2-mkmlizer: ║                                                                     ║
rirv938-grpo-20250711-c-20147-v2-mkmlizer: ║  Version: 0.29.15                                                   ║
rirv938-grpo-20250711-c-20147-v2-mkmlizer: ║  Features: FLYWHEEL, CUDA                                           ║
rirv938-grpo-20250711-c-20147-v2-mkmlizer: ║  Copyright 2023-2025 MK ONE TECHNOLOGIES Inc.                       ║
rirv938-grpo-20250711-c-20147-v2-mkmlizer: ║  https://mk1.ai                                                     ║
rirv938-grpo-20250711-c-20147-v2-mkmlizer: ║                                                                     ║
rirv938-grpo-20250711-c-20147-v2-mkmlizer: ║  The license key for the current software has been verified as      ║
rirv938-grpo-20250711-c-20147-v2-mkmlizer: ║  belonging to:                                                      ║
rirv938-grpo-20250711-c-20147-v2-mkmlizer: ║                                                                     ║
rirv938-grpo-20250711-c-20147-v2-mkmlizer: ║  Chai Research Corp.                                                ║
rirv938-grpo-20250711-c-20147-v2-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
rirv938-grpo-20250711-c-20147-v2-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
rirv938-grpo-20250711-c-20147-v2-mkmlizer: ║                                                                     ║
rirv938-grpo-20250711-c-20147-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
rirv938-grpo-20250711-c-20147-v2-mkmlizer: Downloaded to shared memory in 120.337s
rirv938-grpo-20250711-c-20147-v2-mkmlizer: Checking if rirv938/grpo_20250711_cp624_sid_mistral_24b_dpo_40k__merged already exists in ChaiML
rirv938-grpo-20250711-c-20147-v2-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpumb0mlaq, device:0
rirv938-grpo-20250711-c-20147-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
rirv938-grpo-20250711-c-20147-v2-mkmlizer: quantized model in 57.964s
rirv938-grpo-20250711-c-20147-v2-mkmlizer: Processed model rirv938/grpo_20250711_cp624_sid_mistral_24b_dpo_40k__merged in 178.407s
rirv938-grpo-20250711-c-20147-v2-mkmlizer: creating bucket guanaco-mkml-models
rirv938-grpo-20250711-c-20147-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rirv938-grpo-20250711-c-20147-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rirv938-grpo-20250711-c-20147-v2/nvidia
rirv938-grpo-20250711-c-20147-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rirv938-grpo-20250711-c-20147-v2/nvidia/config.json
rirv938-grpo-20250711-c-20147-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rirv938-grpo-20250711-c-20147-v2/nvidia/special_tokens_map.json
rirv938-grpo-20250711-c-20147-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rirv938-grpo-20250711-c-20147-v2/nvidia/tokenizer_config.json
rirv938-grpo-20250711-c-20147-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rirv938-grpo-20250711-c-20147-v2/nvidia/tokenizer.json
rirv938-grpo-20250711-c-20147-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/rirv938-grpo-20250711-c-20147-v2/nvidia/flywheel_model.1.safetensors
rirv938-grpo-20250711-c-20147-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rirv938-grpo-20250711-c-20147-v2/nvidia/flywheel_model.0.safetensors
rirv938-grpo-20250711-c-20147-v2-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|          | 3/363 [00:00<00:12, 28.24it/s]
Loading 0:   2%|▏         | 6/363 [00:00<00:25, 13.87it/s]
Loading 0:   3%|▎         | 11/363 [00:00<00:16, 21.49it/s]
Loading 0:   4%|▍         | 14/363 [00:00<00:26, 12.97it/s]
Loading 0:   4%|▍         | 16/363 [00:01<00:27, 12.71it/s]
Loading 0:   6%|▌         | 21/363 [00:01<00:21, 16.05it/s]
Loading 0:   6%|▋         | 23/363 [00:01<00:26, 13.08it/s]
Loading 0:   8%|▊         | 28/363 [00:01<00:17, 18.92it/s]
Loading 0:   9%|▉         | 32/363 [00:01<00:14, 22.25it/s]
Loading 0:  10%|▉         | 35/363 [00:02<00:21, 15.11it/s]
Loading 0:  10%|█         | 38/363 [00:02<00:20, 15.75it/s]
Loading 0:  11%|█▏        | 41/363 [00:02<00:26, 12.38it/s]
Loading 0:  13%|█▎        | 46/363 [00:02<00:18, 17.45it/s]
Loading 0:  14%|█▍        | 50/363 [00:02<00:15, 20.47it/s]
Loading 0:  15%|█▍        | 53/363 [00:03<00:21, 14.60it/s]
Loading 0:  15%|█▌        | 56/363 [00:03<00:19, 15.37it/s]
Loading 0:  16%|█▋        | 59/363 [00:03<00:24, 12.25it/s]
Loading 0:  18%|█▊        | 64/363 [00:04<00:17, 16.91it/s]
Loading 0:  19%|█▊        | 68/363 [00:04<00:14, 19.97it/s]
Loading 0:  20%|█▉        | 71/363 [00:04<00:20, 14.32it/s]
Loading 0:  20%|██        | 74/363 [00:04<00:19, 14.98it/s]
Loading 0:  21%|██        | 76/363 [00:04<00:21, 13.43it/s]
Loading 0:  21%|██▏       | 78/363 [00:05<00:21, 13.00it/s]
Loading 0:  23%|██▎       | 82/363 [00:05<00:16, 17.22it/s]
Loading 0:  24%|██▎       | 86/363 [00:05<00:13, 20.79it/s]
Loading 0:  25%|██▍       | 89/363 [00:05<00:19, 14.18it/s]
Loading 0:  25%|██▌       | 91/363 [00:05<00:20, 13.51it/s]
Loading 0:  26%|██▌       | 95/363 [00:05<00:15, 17.69it/s]
Loading 0:  27%|██▋       | 99/363 [00:06<00:12, 21.38it/s]
Loading 0:  28%|██▊       | 102/363 [00:06<00:13, 19.23it/s]
Loading 0:  29%|██▉       | 105/363 [00:06<00:17, 14.46it/s]
Loading 0:  29%|██▉       | 107/363 [00:06<00:19, 13.23it/s]
Loading 0:  30%|███       | 109/363 [00:07<00:19, 12.94it/s]
Loading 0:  31%|███       | 111/363 [00:07<00:18, 13.86it/s]
Loading 0:  31%|███       | 113/363 [00:07<00:21, 11.46it/s]
Loading 0:  33%|███▎      | 118/363 [00:07<00:13, 17.93it/s]
Loading 0:  34%|███▎      | 122/363 [00:07<00:11, 21.81it/s]
Loading 0:  34%|███▍      | 125/363 [00:07<00:15, 14.99it/s]
Loading 0:  35%|███▌      | 128/363 [00:08<00:15, 15.60it/s]
Loading 0:  36%|███▌      | 131/363 [00:08<00:18, 12.33it/s]
Loading 0:  37%|███▋      | 136/363 [00:08<00:13, 17.34it/s]
Loading 0:  39%|███▊      | 140/363 [00:08<00:10, 20.69it/s]
Loading 0:  39%|███▉      | 143/363 [00:09<00:14, 14.76it/s]
Loading 0:  40%|████      | 146/363 [00:09<00:13, 15.58it/s]
Loading 0:  41%|████      | 149/363 [00:09<00:17, 12.37it/s]
Loading 0:  42%|████▏     | 154/363 [00:09<00:12, 17.34it/s]
Loading 0:  44%|████▎     | 158/363 [00:09<00:09, 20.68it/s]
Loading 0:  44%|████▍     | 161/363 [00:10<00:13, 15.02it/s]
Loading 0:  45%|████▌     | 164/363 [00:10<00:12, 15.79it/s]
Loading 0:  46%|████▌     | 167/363 [00:10<00:15, 12.56it/s]
Loading 0:  47%|████▋     | 172/363 [00:10<00:10, 17.43it/s]
Loading 0:  48%|████▊     | 176/363 [00:10<00:09, 20.68it/s]
Loading 0:  49%|████▉     | 179/363 [00:11<00:12, 15.01it/s]
Loading 0:  50%|█████     | 182/363 [00:11<00:11, 15.80it/s]
Loading 0:  51%|█████     | 185/363 [00:11<00:14, 12.47it/s]
Loading 0:  52%|█████▏    | 190/363 [00:12<00:09, 17.40it/s]
Loading 0:  53%|█████▎    | 194/363 [00:12<00:08, 20.67it/s]
Loading 0:  54%|█████▍    | 197/363 [00:12<00:11, 14.90it/s]
Loading 0:  55%|█████▌    | 200/363 [00:12<00:10, 15.68it/s]
Loading 0:  56%|█████▌    | 203/363 [00:27<03:47,  1.42s/it]
Loading 0:  57%|█████▋    | 207/363 [00:27<02:27,  1.06it/s]
Loading 0:  58%|█████▊    | 210/363 [00:28<01:47,  1.42it/s]
Loading 0:  59%|█████▊    | 213/363 [00:28<01:20,  1.87it/s]
Loading 0:  60%|█████▉    | 216/363 [00:28<01:01,  2.39it/s]
Loading 0:  60%|██████    | 219/363 [00:28<00:44,  3.24it/s]
Loading 0:  61%|██████    | 222/363 [00:29<00:34,  4.11it/s]
Loading 0:  62%|██████▏   | 226/363 [00:29<00:22,  6.01it/s]
Loading 0:  63%|██████▎   | 230/363 [00:29<00:15,  8.34it/s]
Loading 0:  64%|██████▍   | 233/363 [00:29<00:15,  8.38it/s]
Loading 0:  65%|██████▌   | 236/363 [00:29<00:12,  9.89it/s]
Loading 0:  66%|██████▌   | 239/363 [00:30<00:13,  9.30it/s]
Loading 0:  67%|██████▋   | 244/363 [00:30<00:08, 13.64it/s]
Loading 0:  68%|██████▊   | 248/363 [00:30<00:06, 16.94it/s]
Loading 0:  69%|██████▉   | 251/363 [00:30<00:08, 13.34it/s]
Loading 0:  70%|██████▉   | 254/363 [00:30<00:07, 14.38it/s]
Loading 0:  71%|███████   | 257/363 [00:31<00:08, 11.90it/s]
Loading 0:  72%|███████▏  | 262/363 [00:31<00:06, 16.66it/s]
Loading 0:  73%|███████▎  | 266/363 [00:31<00:04, 19.97it/s]
Loading 0:  74%|███████▍  | 269/363 [00:31<00:06, 14.79it/s]
Loading 0:  75%|███████▍  | 272/363 [00:32<00:05, 15.46it/s]
Loading 0:  76%|███████▌  | 275/363 [00:32<00:07, 12.40it/s]
Loading 0:  77%|███████▋  | 280/363 [00:32<00:04, 17.17it/s]
Loading 0:  78%|███████▊  | 284/363 [00:32<00:03, 20.38it/s]
Loading 0:  79%|███████▉  | 287/363 [00:33<00:05, 14.93it/s]
Loading 0:  80%|███████▉  | 290/363 [00:33<00:04, 15.83it/s]
Loading 0:  81%|████████  | 293/363 [00:33<00:05, 12.59it/s]
Loading 0:  82%|████████▏ | 298/363 [00:33<00:03, 17.54it/s]
Loading 0:  83%|████████▎ | 302/363 [00:33<00:02, 20.60it/s]
Loading 0:  84%|████████▍ | 305/363 [00:34<00:03, 15.32it/s]
Loading 0:  85%|████████▍ | 308/363 [00:34<00:03, 16.10it/s]
Loading 0:  86%|████████▌ | 311/363 [00:34<00:04, 12.91it/s]
Loading 0:  87%|████████▋ | 316/363 [00:34<00:02, 17.70it/s]
Loading 0:  88%|████████▊ | 320/363 [00:34<00:02, 20.65it/s]
Loading 0:  89%|████████▉ | 323/363 [00:35<00:02, 15.08it/s]
Loading 0:  90%|████████▉ | 326/363 [00:35<00:02, 15.53it/s]
Loading 0:  91%|█████████ | 329/363 [00:35<00:02, 12.26it/s]
Loading 0:  92%|█████████▏| 334/363 [00:35<00:01, 17.05it/s]
Loading 0:  93%|█████████▎| 338/363 [00:36<00:01, 20.00it/s]
Loading 0:  94%|█████████▍| 341/363 [00:36<00:01, 14.58it/s]
Loading 0:  95%|█████████▍| 344/363 [00:36<00:01, 15.24it/s]
Loading 0:  96%|█████████▌| 347/363 [00:36<00:01, 12.16it/s]
Loading 0:  97%|█████████▋| 352/363 [00:37<00:00, 17.09it/s]
Loading 0:  98%|█████████▊| 356/363 [00:37<00:00, 20.17it/s]
Loading 0:  99%|█████████▉| 359/363 [00:37<00:00, 10.84it/s]
Loading 0: 100%|█████████▉| 362/363 [00:38<00:00, 11.02it/s]
                                                            
Job rirv938-grpo-20250711-c-20147-v2-mkmlizer completed after 217.65s with status: succeeded
Stopping job with name rirv938-grpo-20250711-c-20147-v2-mkmlizer
Pipeline stage MKMLizer completed in 218.17s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.16s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service rirv938-grpo-20250711-c-20147-v2
Waiting for inference service rirv938-grpo-20250711-c-20147-v2 to be ready
Inference service rirv938-grpo-20250711-c-20147-v2 ready after 191.22810173034668s
Pipeline stage MKMLDeployer completed in 191.85s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.431605577468872s
Received healthy response to inference request in 1.9993906021118164s
Received healthy response to inference request in 2.208544969558716s
Received healthy response to inference request in 1.8435852527618408s
Received healthy response to inference request in 2.0261950492858887s
5 requests
0 failed requests
5th percentile: 1.874746322631836
10th percentile: 1.905907392501831
20th percentile: 1.9682295322418213
30th percentile: 2.0047514915466307
40th percentile: 2.0154732704162597
50th percentile: 2.0261950492858887
60th percentile: 2.0991350173950196
70th percentile: 2.1720749855041506
80th percentile: 2.2531570911407472
90th percentile: 2.3423813343048097
95th percentile: 2.3869934558868406
99th percentile: 2.422683153152466
mean time: 2.101864290237427
Pipeline stage StressChecker completed in 11.91s
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 0.67s
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 0.72s
Shutdown handler de-registered
rirv938-grpo-20250711-c_20147_v2 status is now deployed due to DeploymentManager action
Shutdown handler registered
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Skipping teardown as no inference service was successfully deployed
Pipeline stage MKMLProfilerDeleter completed in 0.11s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.10s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service rirv938-grpo-20250711-c-20147-v2-profiler
Waiting for inference service rirv938-grpo-20250711-c-20147-v2-profiler to be ready
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 3057.42s
Shutdown handler de-registered
rirv938-grpo-20250711-c_20147_v2 status is now inactive due to auto deactivation removed underperforming models
rirv938-grpo-20250711-c_20147_v2 status is now torndown due to DeploymentManager action