rirv938-grpo-to-compare_51856

developer_uid: rirv938

submission_id: rirv938-grpo-to-compare_51856_v1

model_name: rirv938-grpo-to-compare_51856_v1

model_group: rirv938/grpo_to_compare_

status: torndown

timestamp: 2025-04-25T20:33:03+00:00

num_battles: 11418

num_wins: 5649

celo_rating: 1285.68

family_friendly_score: 0.5564

family_friendly_standard_error: 0.007025938229161996

submission_type: basic

model_repo: rirv938/grpo_to_compare_to_dpo_10k_cp312_lr3_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': 0.5265051391370975, 'latency_mean': 1.8992621624469757, 'latency_p50': 1.9055958986282349, 'latency_p90': 2.1151338815689087}, {'batch_size': 3, 'throughput': 1.0766569700762716, 'latency_mean': 2.7807121539115904, 'latency_p50': 2.7921565771102905, 'latency_p90': 3.0512099266052246}, {'batch_size': 5, 'throughput': 1.3857161534023157, 'latency_mean': 3.5872720992565155, 'latency_p50': 3.597100019454956, 'latency_p90': 4.037728381156922}, {'batch_size': 6, 'throughput': 1.4988440047544145, 'latency_mean': 3.9774306297302244, 'latency_p50': 3.9783018827438354, 'latency_p90': 4.475658392906189}, {'batch_size': 8, 'throughput': 1.6383988928007482, 'latency_mean': 4.845288091897965, 'latency_p50': 4.83158004283905, 'latency_p90': 5.43770363330841}, {'batch_size': 10, 'throughput': 1.7404527741476596, 'latency_mean': 5.692353544235229, 'latency_p50': 5.730578780174255, 'latency_p90': 6.371566104888916}]

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

display_name: rirv938-grpo-to-compare_51856_v1

is_internal_developer: True

language_model: rirv938/grpo_to_compare_to_dpo_10k_cp312_lr3_merged

model_size: 24B

ranking_group: single

throughput_3p7s: 1.43

us_pacific_date: 2025-04-25

win_ratio: 0.49474513925380975

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>', '###', 'You:', '\n'], '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-51856-v1-mkmlizer
Waiting for job on rirv938-grpo-to-compare-51856-v1-mkmlizer to finish
rirv938-grpo-to-compare-51856-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rirv938-grpo-to-compare-51856-v1-mkmlizer: ║     _____            __           __                                ║
rirv938-grpo-to-compare-51856-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
rirv938-grpo-to-compare-51856-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
rirv938-grpo-to-compare-51856-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
rirv938-grpo-to-compare-51856-v1-mkmlizer: ║       /___/                                                         ║
rirv938-grpo-to-compare-51856-v1-mkmlizer: ║                                                                     ║
rirv938-grpo-to-compare-51856-v1-mkmlizer: ║  Version: 0.12.8                                                    ║
rirv938-grpo-to-compare-51856-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
rirv938-grpo-to-compare-51856-v1-mkmlizer: ║  https://mk1.ai                                                     ║
rirv938-grpo-to-compare-51856-v1-mkmlizer: ║                                                                     ║
rirv938-grpo-to-compare-51856-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
rirv938-grpo-to-compare-51856-v1-mkmlizer: ║  belonging to:                                                      ║
rirv938-grpo-to-compare-51856-v1-mkmlizer: ║                                                                     ║
rirv938-grpo-to-compare-51856-v1-mkmlizer: ║  Chai Research Corp.                                                ║
rirv938-grpo-to-compare-51856-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
rirv938-grpo-to-compare-51856-v1-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
rirv938-grpo-to-compare-51856-v1-mkmlizer: ║                                                                     ║
rirv938-grpo-to-compare-51856-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
rirv938-grpo-to-compare-51856-v1-mkmlizer: Downloaded to shared memory in 166.434s
rirv938-grpo-to-compare-51856-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpgaph2zyg, device:0
rirv938-grpo-to-compare-51856-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
rirv938-grpo-to-compare-51856-v1-mkmlizer: quantized model in 63.636s
rirv938-grpo-to-compare-51856-v1-mkmlizer: Processed model rirv938/grpo_to_compare_to_dpo_10k_cp312_lr3_merged in 230.071s
rirv938-grpo-to-compare-51856-v1-mkmlizer: creating bucket guanaco-mkml-models
rirv938-grpo-to-compare-51856-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rirv938-grpo-to-compare-51856-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rirv938-grpo-to-compare-51856-v1
rirv938-grpo-to-compare-51856-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rirv938-grpo-to-compare-51856-v1/config.json
rirv938-grpo-to-compare-51856-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rirv938-grpo-to-compare-51856-v1/special_tokens_map.json
rirv938-grpo-to-compare-51856-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rirv938-grpo-to-compare-51856-v1/tokenizer_config.json
rirv938-grpo-to-compare-51856-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rirv938-grpo-to-compare-51856-v1/tokenizer.json
rirv938-grpo-to-compare-51856-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/rirv938-grpo-to-compare-51856-v1/flywheel_model.1.safetensors
rirv938-grpo-to-compare-51856-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rirv938-grpo-to-compare-51856-v1/flywheel_model.0.safetensors
rirv938-grpo-to-compare-51856-v1-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|          | 3/363 [00:00<00:12, 29.13it/s]
Loading 0:   2%|▏         | 6/363 [00:00<00:25, 13.74it/s]
Loading 0:   3%|▎         | 11/363 [00:00<00:16, 21.25it/s]
Loading 0:   4%|▍         | 14/363 [00:00<00:27, 12.82it/s]
Loading 0:   4%|▍         | 16/363 [00:01<00:27, 12.48it/s]
Loading 0:   5%|▌         | 19/363 [00:01<00:22, 15.36it/s]
Loading 0:   6%|▌         | 21/363 [00:01<00:21, 15.83it/s]
Loading 0:   6%|▋         | 23/363 [00:01<00:27, 12.30it/s]
Loading 0:   8%|▊         | 28/363 [00:01<00:18, 18.18it/s]
Loading 0:   9%|▉         | 32/363 [00:01<00:15, 21.95it/s]
Loading 0:  10%|▉         | 35/363 [00:02<00:22, 14.79it/s]
Loading 0:  10%|█         | 38/363 [00:02<00:20, 15.52it/s]
Loading 0:  11%|█         | 40/363 [00:02<00:23, 13.88it/s]
Loading 0:  12%|█▏        | 42/363 [00:02<00:24, 13.36it/s]
Loading 0:  13%|█▎        | 46/363 [00:02<00:18, 17.56it/s]
Loading 0:  14%|█▍        | 50/363 [00:03<00:14, 20.99it/s]
Loading 0:  15%|█▍        | 53/363 [00:03<00:21, 14.17it/s]
Loading 0:  15%|█▌        | 55/363 [00:03<00:23, 13.26it/s]
Loading 0:  16%|█▌        | 57/363 [00:03<00:22, 13.83it/s]
Loading 0:  16%|█▋        | 59/363 [00:04<00:26, 11.43it/s]
Loading 0:  18%|█▊        | 64/363 [00:04<00:17, 17.28it/s]
Loading 0:  19%|█▊        | 68/363 [00:04<00:14, 20.90it/s]
Loading 0:  20%|█▉        | 71/363 [00:04<00:19, 14.71it/s]
Loading 0:  20%|██        | 74/363 [00:04<00:18, 15.30it/s]
Loading 0:  21%|██        | 76/363 [00:04<00:20, 13.69it/s]
Loading 0:  21%|██▏       | 78/363 [00:05<00:21, 13.20it/s]
Loading 0:  23%|██▎       | 82/363 [00:05<00:16, 17.41it/s]
Loading 0:  24%|██▎       | 86/363 [00:05<00:13, 21.00it/s]
Loading 0:  25%|██▍       | 89/363 [00:05<00:18, 14.62it/s]
Loading 0:  25%|██▌       | 91/363 [00:05<00:19, 13.96it/s]
Loading 0:  26%|██▌       | 95/363 [00:06<00:14, 18.12it/s]
Loading 0:  27%|██▋       | 99/363 [00:06<00:12, 21.58it/s]
Loading 0:  28%|██▊       | 102/363 [00:06<00:13, 19.52it/s]
Loading 0:  29%|██▉       | 105/363 [00:06<00:17, 14.34it/s]
Loading 0:  29%|██▉       | 107/363 [00:06<00:19, 13.27it/s]
Loading 0:  30%|███       | 109/363 [00:07<00:19, 12.76it/s]
Loading 0:  31%|███       | 111/363 [00:07<00:18, 13.69it/s]
Loading 0:  31%|███       | 113/363 [00:07<00:22, 11.21it/s]
Loading 0:  33%|███▎      | 118/363 [00:07<00:14, 17.08it/s]
Loading 0:  34%|███▎      | 122/363 [00:07<00:11, 20.97it/s]
Loading 0:  34%|███▍      | 125/363 [00:08<00:16, 14.78it/s]
Loading 0:  35%|███▌      | 128/363 [00:08<00:15, 15.63it/s]
Loading 0:  36%|███▌      | 130/363 [00:08<00:16, 13.79it/s]
Loading 0:  36%|███▋      | 132/363 [00:08<00:17, 13.49it/s]
Loading 0:  37%|███▋      | 136/363 [00:08<00:12, 18.22it/s]
Loading 0:  39%|███▊      | 140/363 [00:08<00:09, 22.46it/s]
Loading 0:  39%|███▉      | 143/363 [00:09<00:14, 15.56it/s]
Loading 0:  40%|████      | 146/363 [00:09<00:13, 16.05it/s]
Loading 0:  41%|████      | 149/363 [00:09<00:17, 12.39it/s]
Loading 0:  42%|████▏     | 154/363 [00:09<00:11, 17.77it/s]
Loading 0:  44%|████▎     | 158/363 [00:09<00:09, 21.57it/s]
Loading 0:  45%|████▍     | 162/363 [00:10<00:14, 14.00it/s]
Loading 0:  45%|████▌     | 165/363 [00:10<00:12, 15.92it/s]
Loading 0:  46%|████▋     | 168/363 [00:10<00:13, 14.36it/s]
Loading 0:  47%|████▋     | 172/363 [00:10<00:10, 18.01it/s]
Loading 0:  48%|████▊     | 176/363 [00:10<00:08, 21.82it/s]
Loading 0:  49%|████▉     | 179/363 [00:11<00:11, 15.67it/s]
Loading 0:  50%|█████     | 182/363 [00:11<00:11, 16.04it/s]
Loading 0:  51%|█████     | 185/363 [00:11<00:14, 12.56it/s]
Loading 0:  52%|█████▏    | 190/363 [00:11<00:09, 17.80it/s]
Loading 0:  53%|█████▎    | 194/363 [00:12<00:07, 21.42it/s]
Loading 0:  55%|█████▍    | 198/363 [00:12<00:11, 14.02it/s]
Loading 0:  55%|█████▌    | 201/363 [00:27<03:24,  1.26s/it]
Loading 0:  56%|█████▌    | 203/363 [00:27<02:47,  1.05s/it]
Loading 0:  57%|█████▋    | 207/363 [00:27<01:46,  1.46it/s]
Loading 0:  58%|█████▊    | 210/363 [00:27<01:18,  1.96it/s]
Loading 0:  59%|█████▊    | 213/363 [00:27<00:59,  2.53it/s]
Loading 0:  59%|█████▉    | 215/363 [00:28<00:49,  3.02it/s]
Loading 0:  60%|█████▉    | 217/363 [00:28<00:40,  3.64it/s]
Loading 0:  60%|██████    | 219/363 [00:28<00:31,  4.54it/s]
Loading 0:  61%|██████    | 221/363 [00:28<00:28,  5.07it/s]
Loading 0:  62%|██████▏   | 226/363 [00:28<00:15,  8.82it/s]
Loading 0:  63%|██████▎   | 230/363 [00:28<00:11, 12.06it/s]
Loading 0:  64%|██████▍   | 233/363 [00:29<00:12, 10.78it/s]
Loading 0:  65%|██████▌   | 236/363 [00:29<00:10, 11.81it/s]
Loading 0:  66%|██████▌   | 238/363 [00:29<00:11, 11.29it/s]
Loading 0:  66%|██████▌   | 240/363 [00:29<00:10, 11.43it/s]
Loading 0:  67%|██████▋   | 244/363 [00:30<00:07, 15.50it/s]
Loading 0:  68%|██████▊   | 248/363 [00:30<00:05, 19.55it/s]
Loading 0:  69%|██████▉   | 251/363 [00:30<00:07, 14.38it/s]
Loading 0:  70%|██████▉   | 254/363 [00:30<00:07, 15.24it/s]
Loading 0:  71%|███████   | 256/363 [00:30<00:07, 13.59it/s]
Loading 0:  71%|███████   | 258/363 [00:31<00:08, 13.04it/s]
Loading 0:  72%|███████▏  | 262/363 [00:31<00:05, 17.30it/s]
Loading 0:  73%|███████▎  | 266/363 [00:31<00:04, 21.30it/s]
Loading 0:  74%|███████▍  | 269/363 [00:31<00:06, 15.06it/s]
Loading 0:  75%|███████▍  | 272/363 [00:31<00:05, 15.76it/s]
Loading 0:  75%|███████▌  | 274/363 [00:31<00:06, 13.76it/s]
Loading 0:  76%|███████▌  | 276/363 [00:32<00:06, 13.28it/s]
Loading 0:  77%|███████▋  | 280/363 [00:32<00:04, 17.92it/s]
Loading 0:  78%|███████▊  | 284/363 [00:32<00:03, 22.13it/s]
Loading 0:  79%|███████▉  | 287/363 [00:32<00:04, 15.68it/s]
Loading 0:  80%|███████▉  | 290/363 [00:32<00:04, 16.09it/s]
Loading 0:  81%|████████  | 293/363 [00:33<00:05, 12.15it/s]
Loading 0:  82%|████████▏ | 298/363 [00:33<00:03, 17.20it/s]
Loading 0:  83%|████████▎ | 302/363 [00:33<00:02, 20.62it/s]
Loading 0:  84%|████████▍ | 305/363 [00:33<00:03, 15.09it/s]
Loading 0:  85%|████████▍ | 308/363 [00:34<00:03, 15.88it/s]
Loading 0:  86%|████████▌ | 311/363 [00:34<00:04, 12.62it/s]
Loading 0:  87%|████████▋ | 317/363 [00:34<00:02, 19.07it/s]
Loading 0:  88%|████████▊ | 321/363 [00:34<00:02, 18.42it/s]
Loading 0:  89%|████████▉ | 324/363 [00:35<00:02, 14.23it/s]
Loading 0:  90%|█████████ | 327/363 [00:35<00:02, 16.38it/s]
Loading 0:  91%|█████████ | 330/363 [00:35<00:02, 14.78it/s]
Loading 0:  92%|█████████▏| 334/363 [00:35<00:01, 18.41it/s]
Loading 0:  93%|█████████▎| 338/363 [00:35<00:01, 22.21it/s]
Loading 0:  94%|█████████▍| 341/363 [00:36<00:01, 15.90it/s]
Loading 0:  95%|█████████▍| 344/363 [00:36<00:01, 16.50it/s]
Loading 0:  96%|█████████▌| 347/363 [00:36<00:01, 12.87it/s]
Loading 0:  97%|█████████▋| 352/363 [00:36<00:00, 18.03it/s]
Loading 0:  98%|█████████▊| 356/363 [00:36<00:00, 21.48it/s]
Loading 0:  99%|█████████▉| 359/363 [00:43<00:02,  1.60it/s]
Loading 0: 100%|█████████▉| 362/363 [00:44<00:00,  2.08it/s]
                                                            
Job rirv938-grpo-to-compare-51856-v1-mkmlizer completed after 258.17s with status: succeeded
Stopping job with name rirv938-grpo-to-compare-51856-v1-mkmlizer
Pipeline stage MKMLizer completed in 258.63s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.13s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service rirv938-grpo-to-compare-51856-v1
Waiting for inference service rirv938-grpo-to-compare-51856-v1 to be ready
Inference service rirv938-grpo-to-compare-51856-v1 ready after 140.64977955818176s
Pipeline stage MKMLDeployer completed in 141.11s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.414189100265503s
Received healthy response to inference request in 1.9145817756652832s
Received healthy response to inference request in 1.9346907138824463s
Received healthy response to inference request in 1.9208929538726807s
Received healthy response to inference request in 1.8118417263031006s
5 requests
0 failed requests
5th percentile: 1.832389736175537
10th percentile: 1.8529377460479737
20th percentile: 1.8940337657928468
30th percentile: 1.9158440113067627
40th percentile: 1.9183684825897216
50th percentile: 1.9208929538726807
60th percentile: 1.9264120578765869
70th percentile: 1.931931161880493
80th percentile: 2.0305903911590577
90th percentile: 2.22238974571228
95th percentile: 2.3182894229888915
99th percentile: 2.3950091648101806
mean time: 1.9992392539978028
Pipeline stage StressChecker completed in 11.09s
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 0.62s
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 0.58s
Shutdown handler de-registered
rirv938-grpo-to-compare_51856_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 0.15s
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-51856-v1-profiler
Waiting for inference service rirv938-grpo-to-compare-51856-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
%s, retrying in %s seconds...
Evaluating %s Family Friendly Score with %s threads
Pipeline stage OfflineFamilyFriendlyScorer completed in 4956.15s
Shutdown handler de-registered
rirv938-grpo-to-compare_51856_v1 status is now inactive due to auto deactivation removed underperforming models
rirv938-grpo-to-compare_51856_v1 status is now torndown due to DeploymentManager action