rirv938-grpo-to-compare_32686

developer_uid: rirv938

submission_id: rirv938-grpo-to-compare_32686_v1

model_name: rirv938-grpo-to-compare_32686_v1

model_group: rirv938/grpo_to_compare_

status: torndown

timestamp: 2025-04-25T23:59:47+00:00

num_battles: 7445

num_wins: 3753

celo_rating: 1287.95

family_friendly_score: 0.5642

family_friendly_standard_error: 0.0070125367735221185

submission_type: basic

model_repo: rirv938/grpo_to_compare_to_dpo_10k_cp156_lr5_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': 0.5284057437673507, 'latency_mean': 1.8924194991588592, 'latency_p50': 1.9001344442367554, 'latency_p90': 2.0956629037857057}, {'batch_size': 3, 'throughput': 1.0813692539004902, 'latency_mean': 2.765026478767395, 'latency_p50': 2.7992273569107056, 'latency_p90': 3.032990264892578}, {'batch_size': 5, 'throughput': 1.4040473310354564, 'latency_mean': 3.526637850999832, 'latency_p50': 3.4954546689987183, 'latency_p90': 3.9494093894958495}, {'batch_size': 6, 'throughput': 1.5241891906214076, 'latency_mean': 3.906057902574539, 'latency_p50': 3.8733338117599487, 'latency_p90': 4.339727902412415}, {'batch_size': 8, 'throughput': 1.6706933273106694, 'latency_mean': 4.7627856171131135, 'latency_p50': 4.743297338485718, 'latency_p90': 5.3126041173934935}, {'batch_size': 10, 'throughput': 1.7598519438219191, 'latency_mean': 5.641620134115219, 'latency_p50': 5.633159160614014, 'latency_p90': 6.330276894569397}]

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

display_name: rirv938-grpo-to-compare_32686_v1

is_internal_developer: True

language_model: rirv938/grpo_to_compare_to_dpo_10k_cp156_lr5_merged

model_size: 24B

ranking_group: single

throughput_3p7s: 1.47

us_pacific_date: 2025-04-25

win_ratio: 0.504096709200806

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': ['###', '\n', '</s>', '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-32686-v1-mkmlizer
Waiting for job on rirv938-grpo-to-compare-32686-v1-mkmlizer to finish
rirv938-grpo-to-compare-32686-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rirv938-grpo-to-compare-32686-v1-mkmlizer: ║     _____            __           __                                ║
rirv938-grpo-to-compare-32686-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
rirv938-grpo-to-compare-32686-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
rirv938-grpo-to-compare-32686-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
rirv938-grpo-to-compare-32686-v1-mkmlizer: ║       /___/                                                         ║
rirv938-grpo-to-compare-32686-v1-mkmlizer: ║                                                                     ║
rirv938-grpo-to-compare-32686-v1-mkmlizer: ║  Version: 0.12.8                                                    ║
rirv938-grpo-to-compare-32686-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
rirv938-grpo-to-compare-32686-v1-mkmlizer: ║  https://mk1.ai                                                     ║
rirv938-grpo-to-compare-32686-v1-mkmlizer: ║                                                                     ║
rirv938-grpo-to-compare-32686-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
rirv938-grpo-to-compare-32686-v1-mkmlizer: ║  belonging to:                                                      ║
rirv938-grpo-to-compare-32686-v1-mkmlizer: ║                                                                     ║
rirv938-grpo-to-compare-32686-v1-mkmlizer: ║  Chai Research Corp.                                                ║
rirv938-grpo-to-compare-32686-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
rirv938-grpo-to-compare-32686-v1-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
rirv938-grpo-to-compare-32686-v1-mkmlizer: ║                                                                     ║
rirv938-grpo-to-compare-32686-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
rirv938-grpo-to-compare-32686-v1-mkmlizer: Downloaded to shared memory in 167.330s
rirv938-grpo-to-compare-32686-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpg05tlzhe, device:0
rirv938-grpo-to-compare-32686-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
rirv938-grpo-to-compare-32686-v1-mkmlizer: quantized model in 62.891s
rirv938-grpo-to-compare-32686-v1-mkmlizer: Processed model rirv938/grpo_to_compare_to_dpo_10k_cp156_lr5_merged in 230.221s
rirv938-grpo-to-compare-32686-v1-mkmlizer: creating bucket guanaco-mkml-models
rirv938-grpo-to-compare-32686-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rirv938-grpo-to-compare-32686-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rirv938-grpo-to-compare-32686-v1
rirv938-grpo-to-compare-32686-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rirv938-grpo-to-compare-32686-v1/config.json
rirv938-grpo-to-compare-32686-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rirv938-grpo-to-compare-32686-v1/special_tokens_map.json
rirv938-grpo-to-compare-32686-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rirv938-grpo-to-compare-32686-v1/tokenizer_config.json
rirv938-grpo-to-compare-32686-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rirv938-grpo-to-compare-32686-v1/tokenizer.json
rirv938-grpo-to-compare-32686-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/rirv938-grpo-to-compare-32686-v1/flywheel_model.1.safetensors
rirv938-grpo-to-compare-32686-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rirv938-grpo-to-compare-32686-v1/flywheel_model.0.safetensors
rirv938-grpo-to-compare-32686-v1-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|          | 3/363 [00:00<00:12, 29.97it/s]
Loading 0:   2%|▏         | 6/363 [00:00<00:25, 14.23it/s]
Loading 0:   3%|▎         | 11/363 [00:00<00:16, 21.66it/s]
Loading 0:   4%|▍         | 14/363 [00:00<00:25, 13.54it/s]
Loading 0:   4%|▍         | 16/363 [00:01<00:26, 13.28it/s]
Loading 0:   6%|▌         | 21/363 [00:01<00:20, 16.74it/s]
Loading 0:   6%|▋         | 23/363 [00:01<00:25, 13.43it/s]
Loading 0:   8%|▊         | 28/363 [00:01<00:17, 19.42it/s]
Loading 0:   9%|▉         | 32/363 [00:01<00:14, 22.96it/s]
Loading 0:  10%|▉         | 35/363 [00:02<00:21, 15.33it/s]
Loading 0:  10%|█         | 38/363 [00:02<00:20, 16.01it/s]
Loading 0:  11%|█▏        | 41/363 [00:02<00:25, 12.52it/s]
Loading 0:  13%|█▎        | 46/363 [00:02<00:18, 17.60it/s]
Loading 0:  14%|█▍        | 50/363 [00:02<00:14, 20.95it/s]
Loading 0:  15%|█▍        | 53/363 [00:03<00:20, 15.17it/s]
Loading 0:  15%|█▌        | 56/363 [00:03<00:19, 15.90it/s]
Loading 0:  16%|█▋        | 59/363 [00:03<00:24, 12.54it/s]
Loading 0:  18%|█▊        | 64/363 [00:03<00:17, 17.58it/s]
Loading 0:  19%|█▊        | 68/363 [00:04<00:14, 20.93it/s]
Loading 0:  20%|█▉        | 71/363 [00:04<00:19, 15.20it/s]
Loading 0:  20%|██        | 74/363 [00:04<00:18, 15.85it/s]
Loading 0:  21%|██        | 77/363 [00:04<00:23, 12.40it/s]
Loading 0:  23%|██▎       | 82/363 [00:05<00:16, 17.38it/s]
Loading 0:  24%|██▎       | 86/363 [00:05<00:13, 20.70it/s]
Loading 0:  25%|██▍       | 89/363 [00:05<00:18, 15.03it/s]
Loading 0:  25%|██▌       | 92/363 [00:05<00:17, 15.52it/s]
Loading 0:  26%|██▋       | 96/363 [00:05<00:13, 19.35it/s]
Loading 0:  28%|██▊       | 101/363 [00:05<00:12, 21.26it/s]
Loading 0:  29%|██▊       | 104/363 [00:06<00:14, 17.68it/s]
Loading 0:  29%|██▉       | 107/363 [00:06<00:18, 13.80it/s]
Loading 0:  30%|███       | 109/363 [00:06<00:18, 13.47it/s]
Loading 0:  31%|███       | 111/363 [00:06<00:17, 14.34it/s]
Loading 0:  31%|███       | 113/363 [00:07<00:20, 12.04it/s]
Loading 0:  33%|███▎      | 118/363 [00:07<00:13, 17.96it/s]
Loading 0:  34%|███▎      | 122/363 [00:07<00:11, 21.76it/s]
Loading 0:  34%|███▍      | 125/363 [00:07<00:15, 15.55it/s]
Loading 0:  35%|███▌      | 128/363 [00:07<00:14, 16.38it/s]
Loading 0:  36%|███▌      | 131/363 [00:08<00:18, 12.70it/s]
Loading 0:  37%|███▋      | 136/363 [00:08<00:12, 17.83it/s]
Loading 0:  39%|███▊      | 140/363 [00:08<00:10, 21.24it/s]
Loading 0:  39%|███▉      | 143/363 [00:08<00:14, 15.56it/s]
Loading 0:  40%|████      | 146/363 [00:08<00:13, 16.25it/s]
Loading 0:  41%|████      | 149/363 [00:09<00:16, 12.80it/s]
Loading 0:  42%|████▏     | 154/363 [00:09<00:11, 17.74it/s]
Loading 0:  44%|████▎     | 158/363 [00:09<00:09, 21.01it/s]
Loading 0:  44%|████▍     | 161/363 [00:09<00:13, 15.34it/s]
Loading 0:  45%|████▌     | 164/363 [00:10<00:12, 16.07it/s]
Loading 0:  46%|████▌     | 167/363 [00:10<00:15, 12.71it/s]
Loading 0:  47%|████▋     | 172/363 [00:10<00:10, 17.82it/s]
Loading 0:  48%|████▊     | 176/363 [00:10<00:08, 21.07it/s]
Loading 0:  49%|████▉     | 179/363 [00:11<00:11, 15.35it/s]
Loading 0:  50%|█████     | 182/363 [00:11<00:11, 16.13it/s]
Loading 0:  51%|█████     | 185/363 [00:11<00:14, 12.62it/s]
Loading 0:  52%|█████▏    | 190/363 [00:11<00:09, 17.54it/s]
Loading 0:  53%|█████▎    | 194/363 [00:11<00:08, 20.85it/s]
Loading 0:  54%|█████▍    | 197/363 [00:12<00:10, 15.40it/s]
Loading 0:  55%|█████▌    | 200/363 [00:12<00:10, 15.78it/s]
Loading 0:  56%|█████▌    | 203/363 [00:27<03:43,  1.39s/it]
Loading 0:  57%|█████▋    | 207/363 [00:27<02:24,  1.08it/s]
Loading 0:  58%|█████▊    | 210/363 [00:27<01:45,  1.45it/s]
Loading 0:  59%|█████▊    | 213/363 [00:27<01:18,  1.91it/s]
Loading 0:  60%|█████▉    | 216/363 [00:27<01:00,  2.45it/s]
Loading 0:  60%|██████    | 219/363 [00:28<00:43,  3.32it/s]
Loading 0:  61%|██████    | 222/363 [00:28<00:33,  4.20it/s]
Loading 0:  62%|██████▏   | 226/363 [00:28<00:22,  6.15it/s]
Loading 0:  63%|██████▎   | 230/363 [00:28<00:15,  8.55it/s]
Loading 0:  64%|██████▍   | 233/363 [00:28<00:15,  8.54it/s]
Loading 0:  65%|██████▌   | 236/363 [00:29<00:12,  9.97it/s]
Loading 0:  66%|██████▌   | 239/363 [00:29<00:13,  9.38it/s]
Loading 0:  67%|██████▋   | 244/363 [00:29<00:08, 13.73it/s]
Loading 0:  68%|██████▊   | 248/363 [00:29<00:06, 17.04it/s]
Loading 0:  69%|██████▉   | 251/363 [00:29<00:08, 13.82it/s]
Loading 0:  70%|██████▉   | 254/363 [00:30<00:07, 14.80it/s]
Loading 0:  71%|███████   | 257/363 [00:30<00:08, 12.26it/s]
Loading 0:  72%|███████▏  | 262/363 [00:30<00:05, 17.20it/s]
Loading 0:  73%|███████▎  | 266/363 [00:30<00:04, 20.50it/s]
Loading 0:  74%|███████▍  | 269/363 [00:31<00:06, 15.44it/s]
Loading 0:  75%|███████▍  | 272/363 [00:31<00:05, 16.26it/s]
Loading 0:  76%|███████▌  | 275/363 [00:31<00:06, 12.94it/s]
Loading 0:  77%|███████▋  | 280/363 [00:31<00:04, 18.01it/s]
Loading 0:  78%|███████▊  | 284/363 [00:31<00:03, 21.36it/s]
Loading 0:  79%|███████▉  | 287/363 [00:32<00:04, 15.82it/s]
Loading 0:  80%|███████▉  | 290/363 [00:32<00:04, 16.57it/s]
Loading 0:  81%|████████  | 293/363 [00:32<00:05, 13.23it/s]
Loading 0:  82%|████████▏ | 298/363 [00:32<00:03, 18.16it/s]
Loading 0:  83%|████████▎ | 302/363 [00:32<00:02, 21.16it/s]
Loading 0:  84%|████████▍ | 305/363 [00:33<00:03, 15.48it/s]
Loading 0:  85%|████████▍ | 308/363 [00:33<00:03, 16.06it/s]
Loading 0:  86%|████████▌ | 311/363 [00:33<00:04, 12.58it/s]
Loading 0:  87%|████████▋ | 316/363 [00:33<00:02, 17.51it/s]
Loading 0:  88%|████████▊ | 320/363 [00:34<00:02, 20.83it/s]
Loading 0:  89%|████████▉ | 323/363 [00:34<00:02, 14.80it/s]
Loading 0:  90%|████████▉ | 326/363 [00:34<00:02, 15.40it/s]
Loading 0:  91%|█████████ | 329/363 [00:34<00:02, 12.57it/s]
Loading 0:  92%|█████████▏| 334/363 [00:35<00:01, 17.52it/s]
Loading 0:  93%|█████████▎| 338/363 [00:35<00:01, 20.76it/s]
Loading 0:  94%|█████████▍| 341/363 [00:35<00:01, 15.37it/s]
Loading 0:  95%|█████████▍| 344/363 [00:35<00:01, 16.18it/s]
Loading 0:  96%|█████████▌| 347/363 [00:36<00:01, 12.80it/s]
Loading 0:  97%|█████████▋| 352/363 [00:36<00:00, 17.91it/s]
Loading 0:  98%|█████████▊| 356/363 [00:36<00:00, 21.30it/s]
Loading 0:  99%|█████████▉| 359/363 [00:43<00:02,  1.62it/s]
Loading 0: 100%|█████████▉| 362/363 [00:43<00:00,  2.09it/s]
                                                            
Job rirv938-grpo-to-compare-32686-v1-mkmlizer completed after 258.65s with status: succeeded
Stopping job with name rirv938-grpo-to-compare-32686-v1-mkmlizer
Pipeline stage MKMLizer completed in 259.12s
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-to-compare-32686-v1
Waiting for inference service rirv938-grpo-to-compare-32686-v1 to be ready
Inference service rirv938-grpo-to-compare-32686-v1 ready after 130.5520076751709s
Pipeline stage MKMLDeployer completed in 130.99s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.2706894874572754s
Received healthy response to inference request in 1.8207414150238037s
Received healthy response to inference request in 1.8503892421722412s
Received healthy response to inference request in 2.009718656539917s
Received healthy response to inference request in 1.8317937850952148s
5 requests
0 failed requests
5th percentile: 1.822951889038086
10th percentile: 1.8251623630523681
20th percentile: 1.8295833110809325
30th percentile: 1.8355128765106201
40th percentile: 1.8429510593414307
50th percentile: 1.8503892421722412
60th percentile: 1.9141210079193116
70th percentile: 1.9778527736663818
80th percentile: 2.0619128227233885
90th percentile: 2.166301155090332
95th percentile: 2.218495321273804
99th percentile: 2.260250654220581
mean time: 1.9566665172576905
Pipeline stage StressChecker completed in 11.17s
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 0.70s
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 0.69s
Shutdown handler de-registered
rirv938-grpo-to-compare_32686_v1 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 0.13s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.11s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service rirv938-grpo-to-compare-32686-v1-profiler
Waiting for inference service rirv938-grpo-to-compare-32686-v1-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 2828.62s
Shutdown handler de-registered
rirv938-grpo-to-compare_32686_v1 status is now inactive due to auto deactivation removed underperforming models
rirv938-grpo-to-compare_32686_v1 status is now torndown due to DeploymentManager action