rirv938-grpo-to-compare_71154

developer_uid: rirv938

submission_id: rirv938-grpo-to-compare_71154_v1

model_name: rirv938-grpo-to-compare_71154_v1

model_group: rirv938/grpo_to_compare_

status: torndown

timestamp: 2025-04-26T20:04:27+00:00

num_battles: 10204

num_wins: 5575

celo_rating: 1324.74

family_friendly_score: 0.5416000000000001

family_friendly_standard_error: 0.007046551497009016

submission_type: basic

model_repo: rirv938/grpo_to_compare_to_dpo_10k_cp312_lr9_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': 0.5325815976372177, 'latency_mean': 1.8775892102718352, 'latency_p50': 1.868011474609375, 'latency_p90': 2.087813305854797}, {'batch_size': 3, 'throughput': 1.0833285963742436, 'latency_mean': 2.755824829339981, 'latency_p50': 2.766098380088806, 'latency_p90': 3.0373297452926638}, {'batch_size': 5, 'throughput': 1.4027212683044261, 'latency_mean': 3.5390781223773957, 'latency_p50': 3.548505663871765, 'latency_p90': 4.025323963165283}, {'batch_size': 6, 'throughput': 1.5103944154862732, 'latency_mean': 3.948276790380478, 'latency_p50': 3.937659740447998, 'latency_p90': 4.406147050857544}, {'batch_size': 8, 'throughput': 1.639621518030933, 'latency_mean': 4.836142649650574, 'latency_p50': 4.878077507019043, 'latency_p90': 5.4430619239807125}, {'batch_size': 10, 'throughput': 1.732776064632589, 'latency_mean': 5.719732698202133, 'latency_p50': 5.7186596393585205, 'latency_p90': 6.387295889854431}]

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

display_name: rirv938-grpo-to-compare_71154_v1

is_internal_developer: True

language_model: rirv938/grpo_to_compare_to_dpo_10k_cp312_lr9_merged

model_size: 24B

ranking_group: single

throughput_3p7s: 1.46

us_pacific_date: 2025-04-26

win_ratio: 0.5463543708349666

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-71154-v1-mkmlizer
Waiting for job on rirv938-grpo-to-compare-71154-v1-mkmlizer to finish
rirv938-grpo-to-compare-71154-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rirv938-grpo-to-compare-71154-v1-mkmlizer: ║     _____            __           __                                ║
rirv938-grpo-to-compare-71154-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
rirv938-grpo-to-compare-71154-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
rirv938-grpo-to-compare-71154-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
rirv938-grpo-to-compare-71154-v1-mkmlizer: ║       /___/                                                         ║
rirv938-grpo-to-compare-71154-v1-mkmlizer: ║                                                                     ║
rirv938-grpo-to-compare-71154-v1-mkmlizer: ║  Version: 0.12.8                                                    ║
rirv938-grpo-to-compare-71154-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
rirv938-grpo-to-compare-71154-v1-mkmlizer: ║  https://mk1.ai                                                     ║
rirv938-grpo-to-compare-71154-v1-mkmlizer: ║                                                                     ║
rirv938-grpo-to-compare-71154-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
rirv938-grpo-to-compare-71154-v1-mkmlizer: ║  belonging to:                                                      ║
rirv938-grpo-to-compare-71154-v1-mkmlizer: ║                                                                     ║
rirv938-grpo-to-compare-71154-v1-mkmlizer: ║  Chai Research Corp.                                                ║
rirv938-grpo-to-compare-71154-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
rirv938-grpo-to-compare-71154-v1-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
rirv938-grpo-to-compare-71154-v1-mkmlizer: ║                                                                     ║
rirv938-grpo-to-compare-71154-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
Failed to get response for submission chaiml-gemma2rp-27b-sft-exp3_v1: HTTPConnectionPool(host='chaiml-gemma2rp-27b-sft-exp3-v1-predictor.tenant-chaiml-guanaco.k.chaiverse.com', port=80): Read timed out. (read timeout=12.0)
rirv938-grpo-to-compare-71154-v1-mkmlizer: Downloaded to shared memory in 159.967s
rirv938-grpo-to-compare-71154-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpdub6f4kw, device:0
rirv938-grpo-to-compare-71154-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
Failed to get response for submission chaiml-gemma2rp-27b-sft-exp3_v1: HTTPConnectionPool(host='chaiml-gemma2rp-27b-sft-exp3-v1-predictor.tenant-chaiml-guanaco.k.chaiverse.com', port=80): Read timed out. (read timeout=12.0)
rirv938-grpo-to-compare-71154-v1-mkmlizer: quantized model in 64.286s
rirv938-grpo-to-compare-71154-v1-mkmlizer: Processed model rirv938/grpo_to_compare_to_dpo_10k_cp312_lr9_merged in 224.253s
rirv938-grpo-to-compare-71154-v1-mkmlizer: creating bucket guanaco-mkml-models
rirv938-grpo-to-compare-71154-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rirv938-grpo-to-compare-71154-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rirv938-grpo-to-compare-71154-v1
rirv938-grpo-to-compare-71154-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rirv938-grpo-to-compare-71154-v1/config.json
rirv938-grpo-to-compare-71154-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rirv938-grpo-to-compare-71154-v1/special_tokens_map.json
rirv938-grpo-to-compare-71154-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rirv938-grpo-to-compare-71154-v1/tokenizer_config.json
rirv938-grpo-to-compare-71154-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/rirv938-grpo-to-compare-71154-v1/flywheel_model.1.safetensors
rirv938-grpo-to-compare-71154-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rirv938-grpo-to-compare-71154-v1/flywheel_model.0.safetensors
rirv938-grpo-to-compare-71154-v1-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|          | 3/363 [00:00<00:13, 27.51it/s]
Loading 0:   2%|▏         | 6/363 [00:00<00:27, 13.06it/s]
Loading 0:   3%|▎         | 10/363 [00:00<00:17, 20.14it/s]
Loading 0:   4%|▎         | 13/363 [00:00<00:26, 13.32it/s]
Loading 0:   4%|▍         | 15/363 [00:01<00:31, 10.89it/s]
Loading 0:   5%|▌         | 19/363 [00:01<00:22, 15.49it/s]
Loading 0:   6%|▌         | 22/363 [00:01<00:23, 14.82it/s]
Loading 0:   7%|▋         | 24/363 [00:01<00:24, 13.94it/s]
Loading 0:   8%|▊         | 28/363 [00:01<00:18, 17.76it/s]
Loading 0:   9%|▉         | 32/363 [00:01<00:15, 21.24it/s]
Loading 0:  10%|▉         | 35/363 [00:02<00:22, 14.39it/s]
Loading 0:  10%|█         | 37/363 [00:02<00:23, 13.82it/s]
Loading 0:  11%|█         | 39/363 [00:02<00:21, 14.77it/s]
Loading 0:  11%|█▏        | 41/363 [00:02<00:27, 11.66it/s]
Loading 0:  13%|█▎        | 46/363 [00:02<00:18, 17.60it/s]
Loading 0:  14%|█▍        | 50/363 [00:03<00:14, 20.99it/s]
Loading 0:  15%|█▍        | 53/363 [00:03<00:20, 14.80it/s]
Loading 0:  15%|█▌        | 56/363 [00:03<00:20, 15.20it/s]
Loading 0:  16%|█▌        | 58/363 [00:03<00:22, 13.40it/s]
Loading 0:  17%|█▋        | 60/363 [00:04<00:23, 12.73it/s]
Loading 0:  18%|█▊        | 64/363 [00:04<00:17, 16.75it/s]
Loading 0:  19%|█▊        | 68/363 [00:04<00:14, 19.89it/s]
Loading 0:  20%|█▉        | 71/363 [00:04<00:20, 14.00it/s]
Loading 0:  20%|██        | 73/363 [00:04<00:21, 13.34it/s]
Loading 0:  21%|██        | 75/363 [00:04<00:20, 14.15it/s]
Loading 0:  21%|██        | 77/363 [00:05<00:24, 11.62it/s]
Loading 0:  23%|██▎       | 82/363 [00:05<00:15, 18.01it/s]
Loading 0:  24%|██▎       | 86/363 [00:05<00:12, 22.11it/s]
Loading 0:  25%|██▍       | 89/363 [00:05<00:17, 15.43it/s]
Loading 0:  25%|██▌       | 92/363 [00:05<00:17, 15.64it/s]
Loading 0:  27%|██▋       | 97/363 [00:06<00:12, 20.89it/s]
Loading 0:  28%|██▊       | 101/363 [00:06<00:12, 21.71it/s]
Loading 0:  29%|██▊       | 104/363 [00:06<00:14, 17.46it/s]
Loading 0:  29%|██▉       | 107/363 [00:06<00:18, 13.60it/s]
Loading 0:  30%|███       | 109/363 [00:07<00:19, 13.11it/s]
Loading 0:  31%|███       | 111/363 [00:07<00:18, 13.88it/s]
Loading 0:  31%|███       | 113/363 [00:07<00:22, 11.17it/s]
Loading 0:  33%|███▎      | 118/363 [00:07<00:14, 17.20it/s]
Loading 0:  34%|███▎      | 122/363 [00:07<00:11, 21.43it/s]
Loading 0:  34%|███▍      | 125/363 [00:08<00:15, 15.37it/s]
Loading 0:  35%|███▌      | 128/363 [00:08<00:14, 16.03it/s]
Loading 0:  36%|███▌      | 131/363 [00:08<00:19, 11.95it/s]
Loading 0:  38%|███▊      | 137/363 [00:08<00:12, 18.35it/s]
Loading 0:  39%|███▉      | 141/363 [00:08<00:12, 17.53it/s]
Loading 0:  40%|███▉      | 144/363 [00:09<00:16, 13.64it/s]
Loading 0:  40%|████      | 147/363 [00:09<00:13, 15.65it/s]
Loading 0:  41%|████▏     | 150/363 [00:09<00:15, 14.02it/s]
Loading 0:  43%|████▎     | 155/363 [00:09<00:10, 19.16it/s]
Loading 0:  44%|████▍     | 159/363 [00:10<00:11, 18.26it/s]
Loading 0:  45%|████▍     | 162/363 [00:10<00:14, 14.10it/s]
Loading 0:  45%|████▌     | 165/363 [00:10<00:12, 16.19it/s]
Loading 0:  46%|████▋     | 168/363 [00:10<00:13, 14.78it/s]
Loading 0:  47%|████▋     | 172/363 [00:10<00:10, 18.77it/s]
Loading 0:  48%|████▊     | 176/363 [00:11<00:08, 22.44it/s]
Loading 0:  49%|████▉     | 179/363 [00:11<00:11, 15.96it/s]
Loading 0:  50%|█████     | 182/363 [00:11<00:10, 16.55it/s]
Loading 0:  51%|█████     | 185/363 [00:11<00:13, 12.88it/s]
Loading 0:  53%|█████▎    | 191/363 [00:11<00:08, 19.46it/s]
Loading 0:  54%|█████▎    | 195/363 [00:12<00:09, 18.58it/s]
Loading 0:  55%|█████▍    | 198/363 [00:12<00:11, 14.51it/s]
Loading 0:  55%|█████▌    | 201/363 [00:27<03:36,  1.34s/it]
Loading 0:  56%|█████▌    | 203/363 [00:27<02:56,  1.10s/it]
Loading 0:  57%|█████▋    | 207/363 [00:28<01:51,  1.39it/s]
Loading 0:  58%|█████▊    | 210/363 [00:28<01:21,  1.89it/s]
Loading 0:  59%|█████▊    | 213/363 [00:28<01:00,  2.48it/s]
Loading 0:  59%|█████▉    | 215/363 [00:28<00:49,  2.96it/s]
Loading 0:  60%|█████▉    | 217/363 [00:28<00:40,  3.60it/s]
Loading 0:  60%|██████    | 219/363 [00:28<00:31,  4.50it/s]
Loading 0:  61%|██████    | 221/363 [00:29<00:28,  5.01it/s]
Loading 0:  62%|██████▏   | 226/363 [00:29<00:15,  8.79it/s]
Loading 0:  63%|██████▎   | 230/363 [00:29<00:11, 11.80it/s]
Loading 0:  64%|██████▍   | 233/363 [00:29<00:12, 10.78it/s]
Loading 0:  65%|██████▌   | 236/363 [00:29<00:10, 12.14it/s]
Loading 0:  66%|██████▌   | 238/363 [00:30<00:10, 11.58it/s]
Loading 0:  66%|██████▌   | 240/363 [00:30<00:10, 11.64it/s]
Loading 0:  67%|██████▋   | 244/363 [00:30<00:07, 15.84it/s]
Loading 0:  68%|██████▊   | 248/363 [00:30<00:05, 19.81it/s]
Loading 0:  69%|██████▉   | 251/363 [00:30<00:07, 14.76it/s]
Loading 0:  70%|██████▉   | 254/363 [00:31<00:06, 15.57it/s]
Loading 0:  71%|███████   | 256/363 [00:31<00:07, 13.78it/s]
Loading 0:  71%|███████   | 258/363 [00:31<00:07, 13.28it/s]
Loading 0:  72%|███████▏  | 262/363 [00:31<00:05, 17.97it/s]
Loading 0:  73%|███████▎  | 266/363 [00:31<00:04, 22.39it/s]
Loading 0:  74%|███████▍  | 269/363 [00:31<00:06, 15.66it/s]
Loading 0:  75%|███████▍  | 272/363 [00:32<00:05, 16.25it/s]
Loading 0:  76%|███████▌  | 275/363 [00:32<00:07, 12.54it/s]
Loading 0:  77%|███████▋  | 280/363 [00:32<00:04, 17.92it/s]
Loading 0:  78%|███████▊  | 284/363 [00:32<00:03, 21.61it/s]
Loading 0:  79%|███████▉  | 287/363 [00:33<00:04, 15.70it/s]
Loading 0:  80%|███████▉  | 290/363 [00:33<00:04, 16.24it/s]
Loading 0:  81%|████████  | 293/363 [00:33<00:05, 12.73it/s]
Loading 0:  82%|████████▏ | 298/363 [00:33<00:03, 17.88it/s]
Loading 0:  83%|████████▎ | 302/363 [00:33<00:02, 21.52it/s]
Loading 0:  84%|████████▍ | 306/363 [00:34<00:04, 14.10it/s]
Loading 0:  85%|████████▌ | 309/363 [00:34<00:03, 16.06it/s]
Loading 0:  86%|████████▌ | 312/363 [00:34<00:03, 14.62it/s]
Loading 0:  87%|████████▋ | 316/363 [00:34<00:02, 18.27it/s]
Loading 0:  88%|████████▊ | 320/363 [00:34<00:01, 21.69it/s]
Loading 0:  89%|████████▉ | 323/363 [00:35<00:02, 15.69it/s]
Loading 0:  90%|████████▉ | 326/363 [00:35<00:02, 16.06it/s]
Loading 0:  91%|█████████ | 329/363 [00:35<00:02, 12.22it/s]
Loading 0:  92%|█████████▏| 334/363 [00:35<00:01, 16.62it/s]
Loading 0:  93%|█████████▎| 338/363 [00:36<00:01, 19.75it/s]
Loading 0:  94%|█████████▍| 341/363 [00:36<00:01, 14.67it/s]
Loading 0:  95%|█████████▍| 344/363 [00:36<00:01, 15.39it/s]
Loading 0:  95%|█████████▌| 346/363 [00:36<00:01, 13.82it/s]
Loading 0:  96%|█████████▌| 348/363 [00:36<00:01, 13.32it/s]
Loading 0:  97%|█████████▋| 352/363 [00:37<00:00, 17.40it/s]
Loading 0:  98%|█████████▊| 356/363 [00:37<00:00, 21.14it/s]
Loading 0:  99%|█████████▉| 359/363 [00:44<00:02,  1.44it/s]
Loading 0:  99%|█████████▉| 361/363 [00:44<00:01,  1.73it/s]
                                                            
Job rirv938-grpo-to-compare-71154-v1-mkmlizer completed after 257.96s with status: succeeded
Stopping job with name rirv938-grpo-to-compare-71154-v1-mkmlizer
Pipeline stage MKMLizer completed in 258.47s
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-71154-v1
Waiting for inference service rirv938-grpo-to-compare-71154-v1 to be ready
Failed to get response for submission chaiml-gemma2rp-27b-sft-exp3_v1: HTTPConnectionPool(host='chaiml-gemma2rp-27b-sft-exp3-v1-predictor.tenant-chaiml-guanaco.k.chaiverse.com', port=80): Read timed out. (read timeout=12.0)
Inference service rirv938-grpo-to-compare-71154-v1 ready after 140.93226599693298s
Pipeline stage MKMLDeployer completed in 141.45s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.7062888145446777s
Received healthy response to inference request in 2.094322919845581s
Received healthy response to inference request in 2.28299880027771s
Received healthy response to inference request in 2.0866026878356934s
Received healthy response to inference request in 1.9947481155395508s
5 requests
0 failed requests
5th percentile: 2.013119029998779
10th percentile: 2.031489944458008
20th percentile: 2.068231773376465
30th percentile: 2.0881467342376707
40th percentile: 2.091234827041626
50th percentile: 2.094322919845581
60th percentile: 2.1697932720184325
70th percentile: 2.245263624191284
80th percentile: 2.3676568031311036
90th percentile: 2.5369728088378904
95th percentile: 2.621630811691284
99th percentile: 2.689357213973999
mean time: 2.2329922676086427
Pipeline stage StressChecker completed in 12.68s
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 0.78s
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 0.70s
Shutdown handler de-registered
rirv938-grpo-to-compare_71154_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 0.12s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.15s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service rirv938-grpo-to-compare-71154-v1-profiler
Waiting for inference service rirv938-grpo-to-compare-71154-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 5145.19s
Shutdown handler de-registered
rirv938-grpo-to-compare_71154_v1 status is now inactive due to auto deactivation removed underperforming models
rirv938-grpo-to-compare_71154_v1 status is now torndown due to DeploymentManager action