chaiml-exp-grpo-cp312-_36146

developer_uid: rirv938
submission_id: chaiml-exp-grpo-cp312-_36146_v56
model_name: chaiml-exp-grpo-cp312-_36146_v56
model_group: ChaiML/exp_grpo_cp312_96
status: torndown
timestamp: 2025-10-21T18:24:47+00:00
num_battles: 5422
num_wins: 2987
celo_rating: 1309.1
family_friendly_score: 0.536
family_friendly_standard_error: 0.007052715788970942
submission_type: basic
model_repo: ChaiML/exp_grpo_cp312_96ff_b6_r2_less_ff_merged
model_architecture: MistralForCausalLM
model_num_parameters: 24096691200.0
best_of: 7
max_input_tokens: 1280
max_output_tokens: 64
reward_model: rirv938-iterative-rm-pa_44095_v1
latencies: [{'batch_size': 1, 'throughput': 0.5216379442692939, 'latency_mean': 1.9169148480892182, 'latency_p50': 1.9217958450317383, 'latency_p90': 2.106755757331848}, {'batch_size': 3, 'throughput': 1.0250112664559554, 'latency_mean': 2.9204288041591644, 'latency_p50': 2.9280399084091187, 'latency_p90': 3.167011260986328}, {'batch_size': 5, 'throughput': 1.3065421090857454, 'latency_mean': 3.8082323598861696, 'latency_p50': 3.807185411453247, 'latency_p90': 4.389211750030517}, {'batch_size': 6, 'throughput': 1.3841684300265813, 'latency_mean': 4.307720385789871, 'latency_p50': 4.306558847427368, 'latency_p90': 4.895227146148682}, {'batch_size': 8, 'throughput': 1.5043693336782706, 'latency_mean': 5.281377695798874, 'latency_p50': 5.297265410423279, 'latency_p90': 5.9166309356689455}, {'batch_size': 10, 'throughput': 1.5783145009889405, 'latency_mean': 6.299374957084655, 'latency_p50': 6.333407044410706, 'latency_p90': 7.240745615959168}]
gpu_counts: {'NVIDIA L40S': 1}
display_name: chaiml-exp-grpo-cp312-_36146_v56
ineligible_reason: num_battles<10000
is_internal_developer: True
language_model: ChaiML/exp_grpo_cp312_96ff_b6_r2_less_ff_merged
model_size: 24B
ranking_group: single
throughput_3p7s: 1.29
us_pacific_date: 2025-10-21
win_ratio: 0.5509037255625231
generation_params: {'temperature': 0.9, 'top_p': 0.95, 'min_p': 0.05, 'top_k': 80, 'presence_penalty': 0.45, 'frequency_penalty': 0.45, 'stopping_words': ['\n'], 'max_input_tokens': 1280, 'best_of': 7, '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 chaiml-exp-grpo-cp312-36146-v56-mkmlizer
Waiting for job on chaiml-exp-grpo-cp312-36146-v56-mkmlizer to finish
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: ║                                                                     ║
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: ║   ██████   ██████  █████   ████  ████                               ║
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: ║  ░░██████ ██████  ░░███   ███░  ░░███                               ║
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: ║   ░███░█████░███   ░███  ███     ░███                               ║
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: ║   ░███░░███ ░███   ░███████      ░███                               ║
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: ║   ░███ ░░░  ░███   ░███░░███     ░███                               ║
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: ║   ░███      ░███   ░███ ░░███    ░███                               ║
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: ║   █████     █████  █████ ░░████  █████                              ║
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: ║  ░░░░░     ░░░░░  ░░░░░   ░░░░  ░░░░░                               ║
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: ║                                                                     ║
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: ║  Version: 0.30.2                                                    ║
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: ║  Features: FLYWHEEL, CUDA                                           ║
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: ║  Copyright 2023-2025 MK ONE TECHNOLOGIES Inc.                       ║
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: ║                                                                     ║
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: ║  belonging to:                                                      ║
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: ║                                                                     ║
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: ║                                                                     ║
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
Unable to record family friendly update due to error: Invalid JSON input: Expecting value: line 1 column 1 (char 0)
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: Downloaded to shared memory in 87.405s
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: Checking if ChaiML/exp_grpo_cp312_96ff_b6_r2_less_ff_merged already exists in ChaiML
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: quantizing model to /dev/shm/model_cache, profile:q4, folder:/tmp/tmp9v0w5exo, device:0
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: quantized model in 290.413s
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: Processed model ChaiML/exp_grpo_cp312_96ff_b6_r2_less_ff_merged in 377.818s
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: creating bucket guanaco-mkml-models
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-exp-grpo-cp312-36146-v56/nvidia
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-exp-grpo-cp312-36146-v56/nvidia/tokenizer.json
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/chaiml-exp-grpo-cp312-36146-v56/nvidia/flywheel_model.1.safetensors
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-exp-grpo-cp312-36146-v56/nvidia/flywheel_model.0.safetensors
chaiml-exp-grpo-cp312-36146-v56-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|          | 3/363 [00:02<04:30,  1.33it/s]
Loading 0:   1%|          | 4/363 [00:04<06:51,  1.15s/it]
Loading 0:   1%|▏         | 5/363 [00:06<09:07,  1.53s/it]
Loading 0:   2%|▏         | 8/363 [00:06<04:16,  1.38it/s]
Loading 0:   2%|▏         | 9/363 [00:07<03:42,  1.59it/s]
Loading 0:   3%|▎         | 10/363 [00:07<03:00,  1.96it/s]
Loading 0:   3%|▎         | 11/363 [00:07<02:37,  2.24it/s]
Loading 0:   3%|▎         | 12/363 [00:07<02:10,  2.70it/s]
Loading 0:   4%|▎         | 13/363 [00:09<04:43,  1.23it/s]
Loading 0:   4%|▍         | 14/363 [00:11<06:37,  1.14s/it]
Loading 0:   4%|▍         | 15/363 [00:13<08:05,  1.39s/it]
Loading 0:   5%|▍         | 18/363 [00:14<03:55,  1.46it/s]
Loading 0:   5%|▌         | 19/363 [00:14<03:14,  1.77it/s]
Loading 0:   6%|▌         | 21/363 [00:16<04:07,  1.38it/s]
Loading 0:   6%|▌         | 22/363 [00:18<05:36,  1.01it/s]
Loading 0:   6%|▋         | 23/363 [00:20<06:58,  1.23s/it]
Loading 0:   7%|▋         | 26/363 [00:20<03:45,  1.49it/s]
Loading 0:   7%|▋         | 27/363 [00:20<03:17,  1.70it/s]
Loading 0:   8%|▊         | 28/363 [00:20<02:43,  2.05it/s]
Loading 0:   8%|▊         | 30/363 [00:21<01:58,  2.80it/s]
Loading 0:   9%|▊         | 31/363 [00:21<01:49,  3.03it/s]
Loading 0:   9%|▉         | 32/363 [00:21<01:32,  3.57it/s]
Loading 0:   9%|▉         | 33/363 [00:21<01:19,  4.17it/s]
Loading 0:   9%|▉         | 34/363 [00:23<03:49,  1.43it/s]
Loading 0:  10%|▉         | 35/363 [00:25<05:43,  1.05s/it]
Loading 0:  10%|▉         | 36/363 [00:27<07:14,  1.33s/it]
Loading 0:  11%|█         | 39/363 [00:29<05:13,  1.04it/s]
Loading 0:  11%|█         | 40/363 [00:31<06:19,  1.18s/it]
Loading 0:  11%|█▏        | 41/363 [00:33<07:24,  1.38s/it]
Loading 0:  12%|█▏        | 44/363 [00:33<03:57,  1.34it/s]
Loading 0:  12%|█▏        | 45/363 [00:34<03:26,  1.54it/s]
Loading 0:  13%|█▎        | 46/363 [00:34<02:49,  1.87it/s]
Loading 0:  13%|█▎        | 48/363 [00:34<02:03,  2.56it/s]
Loading 0:  13%|█▎        | 49/363 [00:34<01:53,  2.77it/s]
Loading 0:  14%|█▍        | 50/363 [00:35<01:35,  3.29it/s]
Loading 0:  14%|█▍        | 51/363 [00:35<01:20,  3.86it/s]
Loading 0:  14%|█▍        | 52/363 [00:37<03:41,  1.40it/s]
Loading 0:  15%|█▍        | 53/363 [00:39<05:28,  1.06s/it]
Loading 0:  15%|█▍        | 54/363 [00:41<06:54,  1.34s/it]
Loading 0:  16%|█▌        | 57/363 [00:43<04:57,  1.03it/s]
Loading 0:  16%|█▌        | 58/363 [00:45<06:00,  1.18s/it]
Loading 0:  16%|█▋        | 59/363 [00:47<07:00,  1.38s/it]
Loading 0:  17%|█▋        | 62/363 [00:47<03:45,  1.34it/s]
Loading 0:  17%|█▋        | 63/363 [00:47<03:15,  1.54it/s]
Loading 0:  18%|█▊        | 64/363 [00:47<02:40,  1.86it/s]
Loading 0:  18%|█▊        | 66/363 [00:48<01:56,  2.56it/s]
Loading 0:  18%|█▊        | 67/363 [00:48<01:46,  2.77it/s]
Loading 0:  19%|█▊        | 68/363 [00:48<01:29,  3.29it/s]
Loading 0:  19%|█▉        | 69/363 [00:48<01:15,  3.87it/s]
Loading 0:  19%|█▉        | 70/363 [00:50<03:28,  1.40it/s]
Loading 0:  20%|█▉        | 71/363 [00:52<05:09,  1.06s/it]
Loading 0:  20%|█▉        | 72/363 [00:54<06:30,  1.34s/it]
Loading 0:  21%|██        | 75/363 [00:56<04:39,  1.03it/s]
Loading 0:  21%|██        | 76/363 [00:58<05:38,  1.18s/it]
Loading 0:  21%|██        | 77/363 [01:00<06:33,  1.38s/it]
Loading 0:  22%|██▏       | 80/363 [01:01<03:31,  1.34it/s]
Loading 0:  22%|██▏       | 81/363 [01:01<03:04,  1.53it/s]
Loading 0:  23%|██▎       | 82/363 [01:01<02:31,  1.85it/s]
Loading 0:  23%|██▎       | 84/363 [01:01<01:49,  2.56it/s]
Loading 0:  23%|██▎       | 85/363 [01:01<01:40,  2.77it/s]
Loading 0:  24%|██▎       | 86/363 [01:02<01:24,  3.30it/s]
Loading 0:  24%|██▍       | 87/363 [01:02<01:11,  3.86it/s]
Loading 0:  24%|██▍       | 88/363 [01:04<03:15,  1.40it/s]
Loading 0:  25%|██▍       | 89/363 [01:06<04:49,  1.06s/it]
Loading 0:  25%|██▍       | 90/363 [01:08<06:06,  1.34s/it]
Loading 0:  26%|██▌       | 93/363 [01:08<02:59,  1.51it/s]
Loading 0:  26%|██▌       | 94/363 [01:08<02:35,  1.73it/s]
Loading 0:  26%|██▌       | 95/363 [01:08<02:06,  2.11it/s]
Loading 0:  27%|██▋       | 97/363 [01:09<01:32,  2.88it/s]
Loading 0:  27%|██▋       | 98/363 [01:09<01:26,  3.07it/s]
Loading 0:  27%|██▋       | 99/363 [01:09<01:12,  3.63it/s]
Loading 0:  28%|██▊       | 101/363 [01:11<02:28,  1.76it/s]
Loading 0:  28%|██▊       | 102/363 [01:13<03:50,  1.13it/s]
Loading 0:  28%|██▊       | 103/363 [01:15<05:04,  1.17s/it]
Loading 0:  29%|██▉       | 105/363 [01:15<03:05,  1.39it/s]
Loading 0:  29%|██▉       | 106/363 [01:17<04:17,  1.00s/it]
Loading 0:  29%|██▉       | 107/363 [01:19<05:17,  1.24s/it]
Loading 0:  30%|██▉       | 108/363 [01:21<06:10,  1.45s/it]
Loading 0:  31%|███       | 111/363 [01:23<04:21,  1.04s/it]
Loading 0:  31%|███       | 112/363 [01:25<05:07,  1.23s/it]
Loading 0:  31%|███       | 113/363 [01:27<05:52,  1.41s/it]
Loading 0:  32%|███▏      | 116/363 [01:28<03:10,  1.29it/s]
Loading 0:  32%|███▏      | 117/363 [01:28<02:45,  1.48it/s]
Loading 0:  33%|███▎      | 118/363 [01:28<02:16,  1.80it/s]
Loading 0:  33%|███▎      | 120/363 [01:28<01:38,  2.47it/s]
Loading 0:  33%|███▎      | 121/363 [01:28<01:29,  2.69it/s]
Loading 0:  34%|███▎      | 122/363 [01:29<01:16,  3.17it/s]
Loading 0:  34%|███▍      | 123/363 [01:29<01:04,  3.73it/s]
Loading 0:  34%|███▍      | 124/363 [01:31<02:52,  1.39it/s]
Loading 0:  34%|███▍      | 125/363 [01:33<04:13,  1.06s/it]
Loading 0:  35%|███▍      | 126/363 [01:35<05:17,  1.34s/it]
Loading 0:  36%|███▌      | 129/363 [01:37<03:47,  1.03it/s]
Loading 0:  36%|███▌      | 130/363 [01:39<04:35,  1.18s/it]
Loading 0:  36%|███▌      | 131/363 [01:41<05:20,  1.38s/it]
Loading 0:  37%|███▋      | 134/363 [01:41<02:51,  1.34it/s]
Loading 0:  37%|███▋      | 135/363 [01:41<02:28,  1.53it/s]
Loading 0:  37%|███▋      | 136/363 [01:41<02:02,  1.86it/s]
Loading 0:  38%|███▊      | 138/363 [01:42<01:28,  2.55it/s]
Loading 0:  38%|███▊      | 139/363 [01:42<01:20,  2.77it/s]
Loading 0:  39%|███▊      | 140/363 [01:42<01:08,  3.28it/s]
Loading 0:  39%|███▉      | 141/363 [01:42<00:57,  3.83it/s]
Loading 0:  39%|███▉      | 142/363 [01:44<02:38,  1.40it/s]
Loading 0:  39%|███▉      | 143/363 [01:46<03:53,  1.06s/it]
Loading 0:  40%|███▉      | 144/363 [01:48<04:53,  1.34s/it]
Loading 0:  40%|████      | 147/363 [01:50<03:29,  1.03it/s]
Loading 0:  41%|████      | 148/363 [01:52<04:13,  1.18s/it]
Loading 0:  41%|████      | 149/363 [01:54<04:55,  1.38s/it]
Loading 0:  42%|████▏     | 152/363 [01:55<02:38,  1.33it/s]
Loading 0:  42%|████▏     | 153/363 [01:55<02:17,  1.52it/s]
Loading 0:  42%|████▏     | 154/363 [01:55<01:52,  1.85it/s]
Loading 0:  43%|████▎     | 156/363 [01:55<01:20,  2.56it/s]
Loading 0:  43%|████▎     | 157/363 [01:56<01:14,  2.78it/s]
Loading 0:  44%|████▎     | 158/363 [01:56<01:02,  3.30it/s]
Loading 0:  44%|████▍     | 159/363 [01:56<00:52,  3.86it/s]
Loading 0:  44%|████▍     | 160/363 [01:58<02:24,  1.40it/s]
Loading 0:  44%|████▍     | 161/363 [02:00<03:33,  1.06s/it]
Loading 0:  45%|████▍     | 162/363 [02:02<04:28,  1.34s/it]
Loading 0:  45%|████▌     | 165/363 [02:04<03:11,  1.03it/s]
Loading 0:  46%|████▌     | 166/363 [02:06<03:51,  1.18s/it]
Loading 0:  46%|████▌     | 167/363 [02:08<04:29,  1.37s/it]
Loading 0:  47%|████▋     | 170/363 [02:08<02:23,  1.34it/s]
Loading 0:  47%|████▋     | 171/363 [02:08<02:04,  1.54it/s]
Loading 0:  47%|████▋     | 172/363 [02:08<01:42,  1.87it/s]
Loading 0:  48%|████▊     | 174/363 [02:09<01:13,  2.56it/s]
Loading 0:  48%|████▊     | 175/363 [02:09<01:07,  2.80it/s]
Loading 0:  48%|████▊     | 176/363 [02:09<00:56,  3.31it/s]
Loading 0:  49%|████▉     | 177/363 [02:09<00:48,  3.87it/s]
Loading 0:  49%|████▉     | 178/363 [02:11<02:11,  1.40it/s]
Loading 0:  49%|████▉     | 179/363 [02:13<03:14,  1.06s/it]
Loading 0:  50%|████▉     | 180/363 [02:15<04:04,  1.34s/it]
Loading 0:  50%|█████     | 183/363 [02:17<02:54,  1.03it/s]
Loading 0:  51%|█████     | 184/363 [02:19<03:30,  1.18s/it]
Loading 0:  51%|█████     | 185/363 [02:21<04:04,  1.37s/it]
Loading 0:  52%|█████▏    | 188/363 [02:22<02:10,  1.34it/s]
Loading 0:  52%|█████▏    | 189/363 [02:22<01:53,  1.53it/s]
Loading 0:  52%|█████▏    | 190/363 [02:22<01:33,  1.86it/s]
Loading 0:  53%|█████▎    | 192/363 [02:22<01:07,  2.55it/s]
Loading 0:  53%|█████▎    | 193/363 [02:22<01:01,  2.76it/s]
Loading 0:  53%|█████▎    | 194/363 [02:23<00:51,  3.27it/s]
Loading 0:  54%|█████▎    | 195/363 [02:23<00:43,  3.85it/s]
Loading 0:  54%|█████▍    | 196/363 [02:25<01:58,  1.41it/s]
Loading 0:  54%|█████▍    | 197/363 [02:27<02:55,  1.06s/it]
Loading 0:  55%|█████▍    | 198/363 [02:29<03:40,  1.34s/it]
Loading 0:  55%|█████▌    | 201/363 [02:31<02:36,  1.03it/s]
Loading 0:  56%|█████▌    | 202/363 [02:33<03:09,  1.18s/it]
Loading 0:  56%|█████▌    | 203/363 [02:35<03:39,  1.37s/it]
Loading 0:  57%|█████▋    | 206/363 [02:35<01:57,  1.34it/s]
Loading 0:  57%|█████▋    | 207/363 [02:35<01:41,  1.53it/s]
Loading 0:  57%|█████▋    | 208/363 [02:35<01:23,  1.86it/s]
Loading 0:  58%|█████▊    | 210/363 [02:36<00:59,  2.58it/s]
Loading 0:  58%|█████▊    | 211/363 [02:36<00:54,  2.79it/s]
Loading 0:  58%|█████▊    | 212/363 [02:36<00:45,  3.31it/s]
Loading 0:  59%|█████▊    | 213/363 [02:36<00:38,  3.89it/s]
Loading 0:  59%|█████▉    | 214/363 [02:38<01:45,  1.41it/s]
Loading 0:  59%|█████▉    | 215/363 [02:40<02:36,  1.05s/it]
Loading 0:  60%|█████▉    | 216/363 [02:42<03:15,  1.33s/it]
Loading 0:  60%|██████    | 219/363 [02:44<02:18,  1.04it/s]
Loading 0:  61%|██████    | 220/363 [02:46<02:47,  1.17s/it]
Loading 0:  61%|██████    | 221/363 [02:48<03:15,  1.37s/it]
Loading 0:  62%|██████▏   | 224/363 [02:49<01:43,  1.34it/s]
Loading 0:  62%|██████▏   | 225/363 [02:49<01:30,  1.53it/s]
Loading 0:  62%|██████▏   | 226/363 [02:49<01:13,  1.86it/s]
Loading 0:  63%|██████▎   | 228/363 [02:49<00:53,  2.54it/s]
Loading 0:  63%|██████▎   | 229/363 [02:49<00:48,  2.76it/s]
Loading 0:  63%|██████▎   | 230/363 [02:50<00:40,  3.25it/s]
Loading 0:  64%|██████▎   | 231/363 [02:50<00:34,  3.82it/s]
Loading 0:  64%|██████▍   | 232/363 [02:52<01:33,  1.40it/s]
Loading 0:  64%|██████▍   | 233/363 [02:54<02:17,  1.06s/it]
Loading 0:  64%|██████▍   | 234/363 [02:56<02:51,  1.33s/it]
Loading 0:  65%|██████▌   | 237/363 [02:58<02:01,  1.04it/s]
Loading 0:  66%|██████▌   | 238/363 [03:00<02:26,  1.17s/it]
Loading 0:  66%|██████▌   | 239/363 [03:02<02:50,  1.38s/it]
Loading 0:  67%|██████▋   | 242/363 [03:02<01:30,  1.33it/s]
Loading 0:  67%|██████▋   | 243/363 [03:02<01:18,  1.53it/s]
Loading 0:  67%|██████▋   | 244/363 [03:02<01:04,  1.85it/s]
Loading 0:  68%|██████▊   | 246/363 [03:03<00:46,  2.54it/s]
Loading 0:  68%|██████▊   | 247/363 [03:03<00:41,  2.77it/s]
Loading 0:  68%|██████▊   | 248/363 [03:03<00:35,  3.28it/s]
Loading 0:  69%|██████▊   | 249/363 [03:03<00:29,  3.85it/s]
Loading 0:  69%|██████▉   | 250/363 [03:05<01:20,  1.41it/s]
Loading 0:  69%|██████▉   | 251/363 [03:07<01:58,  1.06s/it]
Loading 0:  69%|██████▉   | 252/363 [03:09<02:27,  1.33s/it]
Loading 0:  70%|███████   | 255/363 [03:11<01:44,  1.03it/s]
Loading 0:  71%|███████   | 256/363 [03:13<02:05,  1.17s/it]
Loading 0:  71%|███████   | 257/363 [03:15<02:25,  1.38s/it]
Loading 0:  72%|███████▏  | 260/363 [03:16<01:17,  1.34it/s]
Loading 0:  72%|███████▏  | 261/363 [03:16<01:06,  1.53it/s]
Loading 0:  72%|███████▏  | 262/363 [03:16<00:54,  1.86it/s]
Loading 0:  73%|███████▎  | 264/363 [03:16<00:38,  2.55it/s]
Loading 0:  73%|███████▎  | 265/363 [03:16<00:35,  2.76it/s]
Loading 0:  73%|███████▎  | 266/363 [03:17<00:29,  3.28it/s]
Loading 0:  74%|███████▎  | 267/363 [03:17<00:25,  3.84it/s]
Loading 0:  74%|███████▍  | 268/363 [03:19<01:07,  1.40it/s]
Loading 0:  74%|███████▍  | 269/363 [03:21<01:39,  1.06s/it]
Loading 0:  74%|███████▍  | 270/363 [03:23<02:03,  1.33s/it]
Loading 0:  75%|███████▌  | 273/363 [03:38<05:04,  3.38s/it]
Loading 0:  75%|███████▌  | 274/363 [03:40<04:35,  3.09s/it]
Loading 0:  76%|███████▌  | 275/363 [03:42<04:12,  2.87s/it]
Loading 0:  77%|███████▋  | 278/363 [03:42<02:07,  1.51s/it]
Loading 0:  77%|███████▋  | 279/363 [03:43<01:45,  1.26s/it]
Loading 0:  77%|███████▋  | 280/363 [03:43<01:24,  1.02s/it]
Loading 0:  78%|███████▊  | 282/363 [03:43<00:55,  1.46it/s]
Loading 0:  78%|███████▊  | 283/363 [03:43<00:46,  1.71it/s]
Loading 0:  78%|███████▊  | 284/363 [03:43<00:37,  2.10it/s]
Loading 0:  79%|███████▊  | 285/363 [03:43<00:30,  2.57it/s]
Loading 0:  79%|███████▉  | 286/363 [03:45<01:01,  1.24it/s]
Loading 0:  79%|███████▉  | 287/363 [03:47<01:25,  1.12s/it]
Loading 0:  79%|███████▉  | 288/363 [03:49<01:43,  1.38s/it]
Loading 0:  80%|████████  | 291/363 [03:51<01:11,  1.01it/s]
Loading 0:  80%|████████  | 292/363 [03:53<01:24,  1.19s/it]
Loading 0:  81%|████████  | 293/363 [03:55<01:37,  1.39s/it]
Loading 0:  82%|████████▏ | 296/363 [03:56<00:50,  1.33it/s]
Loading 0:  82%|████████▏ | 297/363 [03:56<00:42,  1.54it/s]
Loading 0:  82%|████████▏ | 298/363 [03:56<00:34,  1.86it/s]
Loading 0:  83%|████████▎ | 300/363 [03:56<00:24,  2.58it/s]
Loading 0:  83%|████████▎ | 301/363 [03:57<00:21,  2.84it/s]
Loading 0:  83%|████████▎ | 302/363 [03:57<00:18,  3.36it/s]
Loading 0:  83%|████████▎ | 303/363 [03:57<00:15,  3.94it/s]
Loading 0:  84%|████████▎ | 304/363 [03:59<00:41,  1.41it/s]
Loading 0:  84%|████████▍ | 305/363 [04:01<01:00,  1.05s/it]
Loading 0:  84%|████████▍ | 306/363 [04:03<01:15,  1.33s/it]
Loading 0:  85%|████████▌ | 309/363 [04:05<00:52,  1.04it/s]
Loading 0:  85%|████████▌ | 310/363 [04:07<01:02,  1.17s/it]
Loading 0:  86%|████████▌ | 311/363 [04:09<01:11,  1.37s/it]
Loading 0:  87%|████████▋ | 314/363 [04:09<00:36,  1.35it/s]
Loading 0:  87%|████████▋ | 315/363 [04:09<00:30,  1.55it/s]
Loading 0:  87%|████████▋ | 316/363 [04:09<00:25,  1.87it/s]
Loading 0:  88%|████████▊ | 318/363 [04:10<00:17,  2.59it/s]
Loading 0:  88%|████████▊ | 319/363 [04:10<00:15,  2.82it/s]
Loading 0:  88%|████████▊ | 320/363 [04:10<00:12,  3.34it/s]
Loading 0:  88%|████████▊ | 321/363 [04:10<00:10,  3.91it/s]
Loading 0:  89%|████████▊ | 322/363 [04:12<00:29,  1.41it/s]
Loading 0:  89%|████████▉ | 323/363 [04:14<00:42,  1.05s/it]
Loading 0:  89%|████████▉ | 324/363 [04:16<00:51,  1.33s/it]
Loading 0:  90%|█████████ | 327/363 [04:18<00:34,  1.03it/s]
Loading 0:  90%|█████████ | 328/363 [04:20<00:41,  1.17s/it]
Loading 0:  91%|█████████ | 329/363 [04:22<00:46,  1.38s/it]
Loading 0:  91%|█████████▏| 332/363 [04:23<00:23,  1.34it/s]
Loading 0:  92%|█████████▏| 333/363 [04:23<00:19,  1.54it/s]
Loading 0:  92%|█████████▏| 334/363 [04:23<00:15,  1.87it/s]
Loading 0:  93%|█████████▎| 336/363 [04:23<00:10,  2.58it/s]
Loading 0:  93%|█████████▎| 337/363 [04:23<00:09,  2.83it/s]
Loading 0:  93%|█████████▎| 338/363 [04:24<00:07,  3.36it/s]
Loading 0:  93%|█████████▎| 339/363 [04:24<00:06,  3.93it/s]
Loading 0:  94%|█████████▎| 340/363 [04:26<00:16,  1.41it/s]
Loading 0:  94%|█████████▍| 341/363 [04:28<00:23,  1.05s/it]
Loading 0:  94%|█████████▍| 342/363 [04:30<00:27,  1.33s/it]
Loading 0:  95%|█████████▌| 345/363 [04:32<00:17,  1.03it/s]
Loading 0:  95%|█████████▌| 346/363 [04:34<00:19,  1.17s/it]
Loading 0:  96%|█████████▌| 347/363 [04:36<00:22,  1.38s/it]
Loading 0:  96%|█████████▋| 350/363 [04:36<00:09,  1.34it/s]
Loading 0:  97%|█████████▋| 351/363 [04:36<00:07,  1.54it/s]
Loading 0:  97%|█████████▋| 352/363 [04:36<00:05,  1.87it/s]
Loading 0:  98%|█████████▊| 354/363 [04:37<00:03,  2.56it/s]
Loading 0:  98%|█████████▊| 355/363 [04:37<00:02,  2.79it/s]
Loading 0:  98%|█████████▊| 356/363 [04:37<00:02,  3.32it/s]
Loading 0:  98%|█████████▊| 357/363 [04:37<00:02,  2.97it/s]
Loading 0:  99%|█████████▉| 359/363 [04:39<00:02,  1.61it/s]
Loading 0:  99%|█████████▉| 360/363 [04:41<00:02,  1.07it/s]
Loading 0:  99%|█████████▉| 361/363 [04:43<00:02,  1.21s/it]
                                                            
Job chaiml-exp-grpo-cp312-36146-v56-mkmlizer completed after 490.76s with status: succeeded
Stopping job with name chaiml-exp-grpo-cp312-36146-v56-mkmlizer
Pipeline stage MKMLizer completed in 491.58s
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 chaiml-exp-grpo-cp312-36146-v56
Waiting for inference service chaiml-exp-grpo-cp312-36146-v56 to be ready
Inference service chaiml-exp-grpo-cp312-36146-v56 ready after 70.3958899974823s
Pipeline stage MKMLDeployer completed in 71.00s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.304175853729248s
Received healthy response to inference request in 2.1345760822296143s
Received healthy response to inference request in 1.7507219314575195s
Received healthy response to inference request in 1.7034344673156738s
Received healthy response to inference request in 1.6972432136535645s
5 requests
0 failed requests
5th percentile: 1.6984814643859862
10th percentile: 1.6997197151184082
20th percentile: 1.702196216583252
30th percentile: 1.7128919601440429
40th percentile: 1.7318069458007812
50th percentile: 1.7507219314575195
60th percentile: 1.9042635917663575
70th percentile: 2.057805252075195
80th percentile: 2.168496036529541
90th percentile: 2.2363359451293947
95th percentile: 2.2702558994293214
99th percentile: 2.297391862869263
mean time: 1.918030309677124
Pipeline stage StressChecker completed in 11.04s
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 0.73s
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 0.88s
Shutdown handler de-registered
chaiml-exp-grpo-cp312-_36146_v56 status is now deployed due to DeploymentManager action
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
%s, retrying in %s seconds...
Evaluating %s Family Friendly Score with %s threads
%s, retrying in %s seconds...
Shutdown handler registered
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service chaiml-exp-grpo-cp312-36146-v56-profiler is running
Skipping teardown as no inference service was found
Pipeline stage MKMLProfilerDeleter completed in 0.77s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.09s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service chaiml-exp-grpo-cp312-36146-v56-profiler
Waiting for inference service chaiml-exp-grpo-cp312-36146-v56-profiler to be ready
Inference service chaiml-exp-grpo-cp312-36146-v56-profiler ready after 121.07024145126343s
Pipeline stage MKMLProfilerDeployer completed in 121.51s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl --kubeconfig /code/guanaco/guanaco_services/resources/kchai_coreweave_us_east_04a.yaml cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/chaiml-exp-grpo-cp31d13b92b983e259109fcb683206c0ffc5-deploshcf5:/code/chaiverse_profiler_1761072118 --namespace tenant-chaiml-guanaco
kubectl --kubeconfig /code/guanaco/guanaco_services/resources/kchai_coreweave_us_east_04a.yaml exec -it chaiml-exp-grpo-cp31d13b92b983e259109fcb683206c0ffc5-deploshcf5 --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1761072118 && python profiles.py profile --best_of_n 7 --auto_batch 5 --batches 1,5,10,15,20,25,30,35,40,45,50,55,60,65,70,75,80,85,90,95,100,105,110,115,120,125,130,135,140,145,150,155,160,165,170,175,180,185,190,195 --samples 200 --input_tokens 1280 --output_tokens 64 --summary /code/chaiverse_profiler_1761072118/summary.json'
kubectl --kubeconfig /code/guanaco/guanaco_services/resources/kchai_coreweave_us_east_04a.yaml exec -it chaiml-exp-grpo-cp31d13b92b983e259109fcb683206c0ffc5-deploshcf5 --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1761072118/summary.json'
Pipeline stage MKMLProfilerRunner completed in 1143.74s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service chaiml-exp-grpo-cp312-36146-v56-profiler is running
Tearing down inference service chaiml-exp-grpo-cp312-36146-v56-profiler
Service chaiml-exp-grpo-cp312-36146-v56-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 0.75s
Shutdown handler de-registered
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 2837.33s
Shutdown handler de-registered
chaiml-exp-grpo-cp312-_36146_v56 status is now inactive due to auto deactivation removed underperforming models
chaiml-exp-grpo-cp312-_36146_v56 status is now torndown due to DeploymentManager action