rirv938-grpo-to-compare_65873

developer_uid: rirv938

submission_id: rirv938-grpo-to-compare_65873_v1

model_name: rirv938-grpo-to-compare_65873_v1

model_group: rirv938/grpo_to_compare_

status: torndown

timestamp: 2025-04-25T20:32:58+00:00

num_battles: 11810

num_wins: 5882

celo_rating: 1287.09

family_friendly_score: 0.5594

family_friendly_standard_error: 0.007020991952708677

submission_type: basic

model_repo: rirv938/grpo_to_compare_to_dpo_10k_cp156_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': 0.5270938973922747, 'latency_mean': 1.8971278190612793, 'latency_p50': 1.8955941200256348, 'latency_p90': 2.109815549850464}, {'batch_size': 3, 'throughput': 1.0987744487707787, 'latency_mean': 2.725614711046219, 'latency_p50': 2.7363739013671875, 'latency_p90': 2.9767576456069946}, {'batch_size': 5, 'throughput': 1.4106273990005216, 'latency_mean': 3.5328588485717773, 'latency_p50': 3.525585651397705, 'latency_p90': 3.928953766822815}, {'batch_size': 6, 'throughput': 1.5279180762167708, 'latency_mean': 3.893079296350479, 'latency_p50': 3.8796671628952026, 'latency_p90': 4.422182393074036}, {'batch_size': 8, 'throughput': 1.673971819425895, 'latency_mean': 4.723187444210052, 'latency_p50': 4.740978360176086, 'latency_p90': 5.284922647476196}, {'batch_size': 10, 'throughput': 1.7879830686098108, 'latency_mean': 5.5453357660770415, 'latency_p50': 5.541671633720398, 'latency_p90': 6.309602642059326}]

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

display_name: rirv938-grpo-to-compare_65873_v1

is_internal_developer: True

language_model: rirv938/grpo_to_compare_to_dpo_10k_cp156_lr3_merged

model_size: 24B

ranking_group: single

throughput_3p7s: 1.48

us_pacific_date: 2025-04-25

win_ratio: 0.49805249788314987

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': ['You:', '###', '\n', '</s>'], '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-65873-v1-mkmlizer
Waiting for job on rirv938-grpo-to-compare-65873-v1-mkmlizer to finish
rirv938-grpo-to-compare-65873-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rirv938-grpo-to-compare-65873-v1-mkmlizer: ║     _____            __           __                                ║
rirv938-grpo-to-compare-65873-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
rirv938-grpo-to-compare-65873-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
rirv938-grpo-to-compare-65873-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
rirv938-grpo-to-compare-65873-v1-mkmlizer: ║       /___/                                                         ║
rirv938-grpo-to-compare-65873-v1-mkmlizer: ║                                                                     ║
rirv938-grpo-to-compare-65873-v1-mkmlizer: ║  Version: 0.12.8                                                    ║
rirv938-grpo-to-compare-65873-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
rirv938-grpo-to-compare-65873-v1-mkmlizer: ║  https://mk1.ai                                                     ║
rirv938-grpo-to-compare-65873-v1-mkmlizer: ║                                                                     ║
rirv938-grpo-to-compare-65873-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
rirv938-grpo-to-compare-65873-v1-mkmlizer: ║  belonging to:                                                      ║
rirv938-grpo-to-compare-65873-v1-mkmlizer: ║                                                                     ║
rirv938-grpo-to-compare-65873-v1-mkmlizer: ║  Chai Research Corp.                                                ║
rirv938-grpo-to-compare-65873-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
rirv938-grpo-to-compare-65873-v1-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
rirv938-grpo-to-compare-65873-v1-mkmlizer: ║                                                                     ║
rirv938-grpo-to-compare-65873-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
rirv938-grpo-to-compare-65873-v1-mkmlizer: Downloaded to shared memory in 163.516s
rirv938-grpo-to-compare-65873-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmppe_hm0rs, device:0
rirv938-grpo-to-compare-65873-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
Failed to get response for submission chaiml-panflute-v2-rirv_49668_v1: HTTPConnectionPool(host='chaiml-panflute-v2-rirv-49668-v1-predictor.tenant-chaiml-guanaco.k.chaiverse.com', port=80): Max retries exceeded with url: /v1/models/GPT-J-6B-lit-v2:predict (Caused by ConnectTimeoutError(<urllib3.connection.HTTPConnection object at 0x746b343a6f50>, 'Connection to chaiml-panflute-v2-rirv-49668-v1-predictor.tenant-chaiml-guanaco.k.chaiverse.com timed out. (connect timeout=12.0)'))
Retrying (%r) after connection broken by '%r': %s
Retrying (%r) after connection broken by '%r': %s
rirv938-grpo-to-compare-65873-v1-mkmlizer: quantized model in 64.133s
rirv938-grpo-to-compare-65873-v1-mkmlizer: Processed model rirv938/grpo_to_compare_to_dpo_10k_cp156_lr3_merged in 227.650s
rirv938-grpo-to-compare-65873-v1-mkmlizer: creating bucket guanaco-mkml-models
rirv938-grpo-to-compare-65873-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rirv938-grpo-to-compare-65873-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rirv938-grpo-to-compare-65873-v1
rirv938-grpo-to-compare-65873-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rirv938-grpo-to-compare-65873-v1/special_tokens_map.json
rirv938-grpo-to-compare-65873-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rirv938-grpo-to-compare-65873-v1/config.json
rirv938-grpo-to-compare-65873-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rirv938-grpo-to-compare-65873-v1/tokenizer_config.json
rirv938-grpo-to-compare-65873-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rirv938-grpo-to-compare-65873-v1/tokenizer.json
rirv938-grpo-to-compare-65873-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/rirv938-grpo-to-compare-65873-v1/flywheel_model.1.safetensors
rirv938-grpo-to-compare-65873-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rirv938-grpo-to-compare-65873-v1/flywheel_model.0.safetensors
rirv938-grpo-to-compare-65873-v1-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|          | 3/363 [00:00<00:14, 24.27it/s]
Loading 0:   2%|▏         | 6/363 [00:00<00:27, 12.81it/s]
Loading 0:   3%|▎         | 11/363 [00:00<00:17, 20.44it/s]
Loading 0:   4%|▍         | 14/363 [00:00<00:27, 12.62it/s]
Loading 0:   4%|▍         | 16/363 [00:01<00:28, 12.36it/s]
Loading 0:   6%|▌         | 21/363 [00:01<00:21, 15.78it/s]
Loading 0:   6%|▋         | 23/363 [00:01<00:26, 12.73it/s]
Loading 0:   8%|▊         | 28/363 [00:01<00:18, 18.03it/s]
Loading 0:   9%|▉         | 32/363 [00:01<00:15, 21.63it/s]
Loading 0:  10%|▉         | 35/363 [00:02<00:21, 14.99it/s]
Loading 0:  10%|█         | 38/363 [00:02<00:20, 15.68it/s]
Loading 0:  11%|█▏        | 41/363 [00:02<00:26, 12.31it/s]
Loading 0:  13%|█▎        | 47/363 [00:02<00:16, 18.75it/s]
Loading 0:  14%|█▍        | 51/363 [00:03<00:17, 17.81it/s]
Loading 0:  15%|█▍        | 54/363 [00:03<00:22, 13.58it/s]
Loading 0:  16%|█▌        | 57/363 [00:03<00:19, 15.51it/s]
Loading 0:  17%|█▋        | 60/363 [00:03<00:21, 14.10it/s]
Loading 0:  18%|█▊        | 64/363 [00:04<00:16, 17.96it/s]
Loading 0:  19%|█▊        | 68/363 [00:04<00:13, 21.77it/s]
Loading 0:  20%|█▉        | 71/363 [00:04<00:18, 15.45it/s]
Loading 0:  20%|██        | 74/363 [00:04<00:18, 15.92it/s]
Loading 0:  21%|██        | 77/363 [00:05<00:23, 12.08it/s]
Loading 0:  23%|██▎       | 82/363 [00:05<00:16, 16.60it/s]
Loading 0:  24%|██▎       | 86/363 [00:05<00:13, 19.89it/s]
Loading 0:  25%|██▍       | 89/363 [00:05<00:18, 14.59it/s]
Loading 0:  25%|██▌       | 92/363 [00:05<00:17, 15.06it/s]
Loading 0:  26%|██▌       | 95/363 [00:05<00:15, 17.43it/s]
Loading 0:  27%|██▋       | 99/363 [00:06<00:12, 21.17it/s]
Loading 0:  28%|██▊       | 102/363 [00:06<00:13, 19.01it/s]
Loading 0:  29%|██▉       | 105/363 [00:06<00:18, 14.30it/s]
Loading 0:  29%|██▉       | 107/363 [00:06<00:19, 12.95it/s]
Loading 0:  30%|███       | 109/363 [00:07<00:20, 12.58it/s]
Loading 0:  31%|███       | 111/363 [00:07<00:18, 13.48it/s]
Loading 0:  31%|███       | 113/363 [00:07<00:22, 10.90it/s]
Loading 0:  33%|███▎      | 118/363 [00:07<00:14, 16.98it/s]
Loading 0:  34%|███▎      | 122/363 [00:07<00:11, 20.91it/s]
Loading 0:  34%|███▍      | 125/363 [00:08<00:16, 14.48it/s]
Loading 0:  35%|███▌      | 128/363 [00:08<00:15, 15.28it/s]
Loading 0:  36%|███▌      | 130/363 [00:08<00:17, 13.66it/s]
Loading 0:  36%|███▋      | 132/363 [00:08<00:17, 12.96it/s]
Loading 0:  37%|███▋      | 136/363 [00:08<00:13, 17.03it/s]
Loading 0:  39%|███▊      | 140/363 [00:08<00:11, 20.17it/s]
Loading 0:  39%|███▉      | 143/363 [00:09<00:15, 14.18it/s]
Loading 0:  40%|███▉      | 145/363 [00:09<00:16, 13.55it/s]
Loading 0:  40%|████      | 147/363 [00:09<00:14, 14.46it/s]
Loading 0:  41%|████      | 149/363 [00:09<00:18, 11.51it/s]
Loading 0:  42%|████▏     | 154/363 [00:09<00:11, 17.55it/s]
Loading 0:  44%|████▎     | 158/363 [00:09<00:09, 21.23it/s]
Loading 0:  44%|████▍     | 161/363 [00:10<00:13, 14.85it/s]
Loading 0:  45%|████▌     | 164/363 [00:10<00:12, 15.43it/s]
Loading 0:  46%|████▌     | 166/363 [00:10<00:14, 13.81it/s]
Loading 0:  46%|████▋     | 168/363 [00:10<00:14, 13.38it/s]
Loading 0:  47%|████▋     | 172/363 [00:11<00:10, 17.74it/s]
Loading 0:  48%|████▊     | 176/363 [00:11<00:08, 21.89it/s]
Loading 0:  49%|████▉     | 179/363 [00:11<00:12, 15.17it/s]
Loading 0:  50%|█████     | 182/363 [00:11<00:11, 15.70it/s]
Loading 0:  51%|█████     | 184/363 [00:11<00:12, 13.97it/s]
Loading 0:  51%|█████     | 186/363 [00:12<00:13, 13.52it/s]
Loading 0:  52%|█████▏    | 190/363 [00:12<00:09, 17.76it/s]
Loading 0:  53%|█████▎    | 194/363 [00:12<00:08, 20.77it/s]
Loading 0:  54%|█████▍    | 197/363 [00:12<00:11, 14.64it/s]
Loading 0:  55%|█████▍    | 199/363 [00:12<00:11, 13.99it/s]
Loading 0:  55%|█████▌    | 201/363 [00:27<04:42,  1.74s/it]
Loading 0:  56%|█████▌    | 203/363 [00:27<03:34,  1.34s/it]
Loading 0:  57%|█████▋    | 207/363 [00:27<02:04,  1.26it/s]
Loading 0:  58%|█████▊    | 210/363 [00:27<01:26,  1.77it/s]
Loading 0:  59%|█████▊    | 213/363 [00:28<01:02,  2.39it/s]
Loading 0:  59%|█████▉    | 215/363 [00:28<00:51,  2.89it/s]
Loading 0:  60%|█████▉    | 217/363 [00:28<00:41,  3.55it/s]
Loading 0:  60%|██████    | 219/363 [00:28<00:32,  4.49it/s]
Loading 0:  61%|██████    | 221/363 [00:28<00:28,  5.04it/s]
Loading 0:  62%|██████▏   | 226/363 [00:28<00:15,  8.93it/s]
Loading 0:  63%|██████▎   | 230/363 [00:29<00:10, 12.27it/s]
Loading 0:  64%|██████▍   | 233/363 [00:29<00:11, 10.88it/s]
Loading 0:  65%|██████▌   | 236/363 [00:29<00:10, 11.96it/s]
Loading 0:  66%|██████▌   | 238/363 [00:29<00:10, 11.47it/s]
Loading 0:  66%|██████▌   | 240/363 [00:29<00:10, 11.58it/s]
Loading 0:  67%|██████▋   | 244/363 [00:30<00:07, 15.68it/s]
Loading 0:  68%|██████▊   | 248/363 [00:30<00:05, 19.49it/s]
Loading 0:  69%|██████▉   | 251/363 [00:30<00:07, 14.08it/s]
Loading 0:  70%|██████▉   | 253/363 [00:30<00:08, 13.58it/s]
Loading 0:  70%|███████   | 255/363 [00:30<00:07, 14.52it/s]
Loading 0:  71%|███████   | 257/363 [00:31<00:08, 11.80it/s]
Loading 0:  72%|███████▏  | 262/363 [00:31<00:05, 18.06it/s]
Loading 0:  73%|███████▎  | 266/363 [00:31<00:04, 22.03it/s]
Loading 0:  74%|███████▍  | 269/363 [00:31<00:06, 15.26it/s]
Loading 0:  75%|███████▍  | 272/363 [00:31<00:05, 15.80it/s]
Loading 0:  76%|███████▌  | 275/363 [00:32<00:07, 12.40it/s]
Loading 0:  77%|███████▋  | 280/363 [00:32<00:04, 17.60it/s]
Loading 0:  78%|███████▊  | 284/363 [00:32<00:03, 20.64it/s]
Loading 0:  79%|███████▉  | 287/363 [00:32<00:05, 15.18it/s]
Loading 0:  80%|███████▉  | 290/363 [00:33<00:04, 15.85it/s]
Loading 0:  81%|████████  | 293/363 [00:33<00:05, 12.37it/s]
Loading 0:  82%|████████▏ | 298/363 [00:33<00:03, 17.27it/s]
Loading 0:  83%|████████▎ | 302/363 [00:33<00:02, 20.59it/s]
Loading 0:  84%|████████▍ | 305/363 [00:33<00:03, 15.22it/s]
Loading 0:  85%|████████▍ | 308/363 [00:34<00:03, 15.88it/s]
Loading 0:  86%|████████▌ | 311/363 [00:34<00:04, 12.54it/s]
Loading 0:  87%|████████▋ | 316/363 [00:34<00:02, 17.46it/s]
Loading 0:  88%|████████▊ | 320/363 [00:34<00:02, 20.80it/s]
Loading 0:  89%|████████▉ | 323/363 [00:35<00:02, 15.18it/s]
Loading 0:  90%|████████▉ | 326/363 [00:35<00:02, 15.74it/s]
Loading 0:  91%|█████████ | 329/363 [00:35<00:02, 12.45it/s]
Loading 0:  92%|█████████▏| 334/363 [00:35<00:01, 17.25it/s]
Loading 0:  93%|█████████▎| 338/363 [00:35<00:01, 20.49it/s]
Loading 0:  94%|█████████▍| 341/363 [00:36<00:01, 15.09it/s]
Loading 0:  95%|█████████▍| 344/363 [00:36<00:01, 15.65it/s]
Loading 0:  96%|█████████▌| 347/363 [00:36<00:01, 12.37it/s]
Loading 0:  97%|█████████▋| 352/363 [00:36<00:00, 17.33it/s]
Loading 0:  98%|█████████▊| 356/363 [00:36<00:00, 20.73it/s]
Loading 0:  99%|█████████▉| 359/363 [00:44<00:02,  1.58it/s]
Loading 0: 100%|█████████▉| 362/363 [00:44<00:00,  2.04it/s]
                                                            
Job rirv938-grpo-to-compare-65873-v1-mkmlizer completed after 258.64s with status: succeeded
Stopping job with name rirv938-grpo-to-compare-65873-v1-mkmlizer
Pipeline stage MKMLizer completed in 259.11s
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-65873-v1
Waiting for inference service rirv938-grpo-to-compare-65873-v1 to be ready
Inference service rirv938-grpo-to-compare-65873-v1 ready after 140.63471508026123s
Pipeline stage MKMLDeployer completed in 141.09s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.609375238418579s
Received healthy response to inference request in 2.327512502670288s
Received healthy response to inference request in 2.046880006790161s
Received healthy response to inference request in 2.114683151245117s
Received healthy response to inference request in 2.036496639251709s
5 requests
0 failed requests
5th percentile: 2.0385733127593992
10th percentile: 2.04064998626709
20th percentile: 2.044803333282471
30th percentile: 2.0604406356811524
40th percentile: 2.0875618934631346
50th percentile: 2.114683151245117
60th percentile: 2.1998148918151856
70th percentile: 2.284946632385254
80th percentile: 2.383885049819946
90th percentile: 2.496630144119263
95th percentile: 2.553002691268921
99th percentile: 2.5981007289886473
mean time: 2.226989507675171
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 0.71s
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 0.71s
Shutdown handler de-registered
rirv938-grpo-to-compare_65873_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 0.14s
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-65873-v1-profiler
Waiting for inference service rirv938-grpo-to-compare-65873-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 4844.40s
Shutdown handler de-registered
rirv938-grpo-to-compare_65873_v1 status is now inactive due to auto deactivation removed underperforming models
rirv938-grpo-to-compare_65873_v1 status is now torndown due to DeploymentManager action