developer_uid: richhx
submission_id: chaiml-csfs-v3-3-dpo-l_86358_v20
model_name: chaiml-csfs-v3-3-dpo-l_86358_v20
model_group: ChaiML/csfs-v3-3-dpo-lr5
status: torndown
timestamp: 2026-01-16T21:38:12+00:00
num_battles: 12061
num_wins: 6017
celo_rating: 1295.78
family_friendly_score: 0.48860000000000003
family_friendly_standard_error: 0.007069229661002675
submission_type: basic
model_repo: ChaiML/csfs-v3-3-dpo-lr5e6b01-lora
model_architecture: MistralForCausalLM
model_num_parameters: 24096691200.0
best_of: 8
max_input_tokens: 2048
max_output_tokens: 64
reward_model: default
latencies: [{'batch_size': 1, 'throughput': 0.36638959097939644, 'latency_mean': 2.729270722866058, 'latency_p50': 2.3284465074539185, 'latency_p90': 3.8505556344985963}, {'batch_size': 2, 'throughput': 0.6802144277808863, 'latency_mean': 2.933397740125656, 'latency_p50': 2.954869866371155, 'latency_p90': 3.1850252866744997}, {'batch_size': 3, 'throughput': 0.8219815462193047, 'latency_mean': 3.6458830070495605, 'latency_p50': 3.653582453727722, 'latency_p90': 3.933494734764099}, {'batch_size': 4, 'throughput': 0.9182796760076468, 'latency_mean': 4.343834655284882, 'latency_p50': 4.344780921936035, 'latency_p90': 4.892351150512695}, {'batch_size': 5, 'throughput': 0.9688207435902658, 'latency_mean': 5.131979662179947, 'latency_p50': 5.160033702850342, 'latency_p90': 5.823437166213989}]
gpu_counts: {'NVIDIA L40S': 1}
display_name: chaiml-csfs-v3-3-dpo-l_86358_v20
is_internal_developer: True
language_model: ChaiML/csfs-v3-3-dpo-lr5e6b01-lora
model_size: 24B
ranking_group: single
throughput_3p7s: 0.83
us_pacific_date: 2026-01-13
win_ratio: 0.4988806898267142
generation_params: {'temperature': 1.0, 'top_p': 1.0, 'min_p': 0.0, 'top_k': 80, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['####', '</s>', 'You:', '\n'], 'max_input_tokens': 2048, 'best_of': 8, 'max_output_tokens': 64}
formatter: {'memory_template': "{bot_name}'s persona: {memory}", 'prompt_template': '', 'bot_template': '{bot_name}: {message}\n', 'user_template': 'You: {message}\n', 'response_template': '####\n{bot_name}:', 'truncate_by_message': True}
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-csfs-v3-3-dpo-l-86358-v20-mkmlizer
Waiting for job on chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer to finish
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: bash: cannot set terminal process group (-1): Inappropriate ioctl for device
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: bash: no job control in this shell
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: /root/miniconda3/envs/nvidia/lib/python3.11/site-packages/mk1/__init__.py:1: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer:   __import__('pkg_resources').declare_namespace(__name__)
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: ║                                                                     ║
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: ║   ██████   ██████  █████   ████  ████                               ║
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: ║  ░░██████ ██████  ░░███   ███░  ░░███                               ║
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: ║   ░███░█████░███   ░███  ███     ░███                               ║
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: ║   ░███░░███ ░███   ░███████      ░███                               ║
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: ║   ░███ ░░░  ░███   ░███░░███     ░███                               ║
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: ║   ░███      ░███   ░███ ░░███    ░███                               ║
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: ║   █████     █████  █████ ░░████  █████                              ║
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: ║  ░░░░░     ░░░░░  ░░░░░   ░░░░  ░░░░░                               ║
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: ║                                                                     ║
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: ║  Version: 0.30.6+torch280                                           ║
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: ║  Features: FLYWHEEL, CUDA                                           ║
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: ║  Copyright 2023-2025 MK ONE TECHNOLOGIES Inc.                       ║
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: ║                                                                     ║
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: ║  belonging to:                                                      ║
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: ║                                                                     ║
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: ║                                                                     ║
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
HTTP Request: %s %s "%s %d %s"
HTTP Request: %s %s "%s %d %s"
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: Downloaded to shared memory in 49.787s
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: Checking if ChaiML/csfs-v3-3-dpo-lr5e6b01-lora already exists in ChaiML
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: quantizing model to /dev/shm/model_cache, profile:q4, folder:/tmp/tmprhdgjlr2, device:0
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: Saving flywheel model at /dev/shm/model_cache
HTTP Request: %s %s "%s %d %s"
HTTP Request: %s %s "%s %d %s"
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: 
Loading 0:   0%|          | 0.00/363 [00:00<?, ?it/s]
Loading 0:   1%|          | 3.00/363 [00:01<03:48, 1.58it/s]
Loading 0:   1%|          | 3.00/363 [00:01<03:48, 1.58it/s]
Loading 0:   1%|          | 4.00/363 [00:03<06:10, 1.03s/it]
Loading 0:   1%|          | 4.00/363 [00:03<06:10, 1.03s/it]
Loading 0:   1%|▏         | 5.00/363 [00:05<07:58, 1.34s/it]
Loading 0:   1%|▏         | 5.00/363 [00:05<07:58, 1.34s/it]
Loading 0:   3%|▎         | 12.0/363 [00:08<03:31, 1.66it/s]
Loading 0:   3%|▎         | 12.0/363 [00:08<03:31, 1.66it/s]
Loading 0:   4%|▎         | 13.0/363 [00:10<04:30, 1.29it/s]
Loading 0:   4%|▎         | 13.0/363 [00:10<04:30, 1.29it/s]
Loading 0:   4%|▍         | 14.0/363 [00:12<05:37, 1.03it/s]
Loading 0:   4%|▍         | 14.0/363 [00:12<05:37, 1.03it/s]
Loading 0:   6%|▌         | 21.0/363 [00:14<03:24, 1.67it/s]
Loading 0:   6%|▌         | 21.0/363 [00:14<03:24, 1.67it/s]
Loading 0:   6%|▌         | 22.0/363 [00:16<04:12, 1.35it/s]
Loading 0:   6%|▌         | 22.0/363 [00:16<04:12, 1.35it/s]
Loading 0:   6%|▋         | 23.0/363 [00:18<05:09, 1.10it/s]
Loading 0:   6%|▋         | 23.0/363 [00:18<05:09, 1.10it/s]
Loading 0:   9%|▊         | 31.0/363 [00:19<02:27, 2.25it/s]
Loading 0:   9%|▊         | 31.0/363 [00:19<02:27, 2.25it/s]
Loading 0:   9%|▉         | 34.0/363 [00:22<02:49, 1.94it/s]
Loading 0:   9%|▉         | 34.0/363 [00:22<02:49, 1.94it/s]
Loading 0:  10%|▉         | 35.0/363 [00:23<03:35, 1.52it/s]
Loading 0:  10%|▉         | 35.0/363 [00:23<03:35, 1.52it/s]
Loading 0:  10%|▉         | 36.0/363 [00:25<04:31, 1.20it/s]
Loading 0:  10%|▉         | 36.0/363 [00:25<04:31, 1.20it/s]
Loading 0:  11%|█         | 39.0/363 [00:27<04:05, 1.32it/s]
Loading 0:  11%|█         | 39.0/363 [00:27<04:05, 1.32it/s]
Loading 0:  11%|█         | 40.0/363 [00:29<04:59, 1.08it/s]
Loading 0:  11%|█         | 40.0/363 [00:29<04:59, 1.08it/s]
Loading 0:  11%|█▏        | 41.0/363 [00:31<06:08, 1.14s/it]
Loading 0:  11%|█▏        | 41.0/363 [00:31<06:08, 1.14s/it]
Loading 0:  13%|█▎        | 48.0/363 [00:34<03:24, 1.54it/s]
Loading 0:  13%|█▎        | 48.0/363 [00:34<03:24, 1.54it/s]
Loading 0:  13%|█▎        | 49.0/363 [00:36<04:07, 1.27it/s]
Loading 0:  13%|█▎        | 49.0/363 [00:36<04:07, 1.27it/s]
Loading 0:  14%|█▍        | 50.0/363 [00:38<04:59, 1.04it/s]
Loading 0:  14%|█▍        | 50.0/363 [00:38<04:59, 1.04it/s]
Loading 0:  16%|█▌        | 57.0/363 [00:40<03:06, 1.64it/s]
Loading 0:  16%|█▌        | 57.0/363 [00:40<03:06, 1.64it/s]
Loading 0:  16%|█▌        | 58.0/363 [00:42<03:46, 1.34it/s]
Loading 0:  16%|█▌        | 58.0/363 [00:42<03:46, 1.34it/s]
Loading 0:  16%|█▋        | 59.0/363 [00:44<04:35, 1.10it/s]
Loading 0:  16%|█▋        | 59.0/363 [00:44<04:35, 1.10it/s]
Loading 0:  18%|█▊        | 65.0/363 [00:47<03:13, 1.54it/s]
Loading 0:  18%|█▊        | 65.0/363 [00:47<03:13, 1.54it/s]
Loading 0:  20%|█▉        | 71.0/363 [00:49<02:44, 1.78it/s]
Loading 0:  20%|█▉        | 71.0/363 [00:49<02:44, 1.78it/s]
Loading 0:  20%|█▉        | 72.0/363 [00:51<03:19, 1.46it/s]
Loading 0:  20%|█▉        | 72.0/363 [00:51<03:19, 1.46it/s]
Loading 0:  20%|██        | 73.0/363 [00:53<04:02, 1.19it/s]
Loading 0:  20%|██        | 73.0/363 [00:53<04:02, 1.19it/s]
Loading 0:  22%|██▏       | 79.0/363 [00:56<02:58, 1.59it/s]
Loading 0:  22%|██▏       | 79.0/363 [00:56<02:58, 1.59it/s]
Loading 0:  22%|██▏       | 80.0/363 [00:58<03:39, 1.29it/s]
Loading 0:  22%|██▏       | 80.0/363 [00:58<03:39, 1.29it/s]
Loading 0:  24%|██▎       | 86.0/363 [01:00<02:47, 1.65it/s]
Loading 0:  24%|██▎       | 86.0/363 [01:00<02:47, 1.65it/s]
Loading 0:  24%|██▍       | 87.0/363 [01:02<03:25, 1.34it/s]
Loading 0:  24%|██▍       | 87.0/363 [01:02<03:25, 1.34it/s]
Loading 0:  25%|██▍       | 90.0/363 [01:04<03:13, 1.41it/s]
Loading 0:  25%|██▍       | 90.0/363 [01:04<03:13, 1.41it/s]
Loading 0:  25%|██▌       | 91.0/363 [01:06<03:52, 1.17it/s]
Loading 0:  25%|██▌       | 91.0/363 [01:06<03:52, 1.17it/s]
Loading 0:  25%|██▌       | 92.0/363 [01:08<04:38, 1.03s/it]
Loading 0:  25%|██▌       | 92.0/363 [01:08<04:38, 1.03s/it]
Loading 0:  27%|██▋       | 99.0/363 [01:11<02:45, 1.60it/s]
Loading 0:  27%|██▋       | 99.0/363 [01:11<02:45, 1.60it/s]
Loading 0:  28%|██▊       | 100/363 [01:12<03:20, 1.31it/s] 
Loading 0:  28%|██▊       | 100/363 [01:12<03:20, 1.31it/s]
Loading 0:  28%|██▊       | 101/363 [01:14<04:02, 1.08it/s]
Loading 0:  28%|██▊       | 101/363 [01:14<04:02, 1.08it/s]
Loading 0:  30%|██▉       | 108/363 [01:17<02:36, 1.63it/s]
Loading 0:  30%|██▉       | 108/363 [01:17<02:36, 1.63it/s]
Loading 0:  31%|███       | 111/363 [01:19<02:35, 1.62it/s]
Loading 0:  31%|███       | 111/363 [01:19<02:35, 1.62it/s]
Loading 0:  31%|███       | 112/363 [01:21<03:09, 1.32it/s]
Loading 0:  31%|███       | 112/363 [01:21<03:09, 1.32it/s]
Loading 0:  31%|███       | 113/363 [01:23<03:51, 1.08it/s]
Loading 0:  31%|███       | 113/363 [01:23<03:51, 1.08it/s]
Loading 0:  33%|███▎      | 120/363 [01:25<02:26, 1.66it/s]
Loading 0:  33%|███▎      | 120/363 [01:25<02:26, 1.66it/s]
Loading 0:  33%|███▎      | 121/363 [01:27<02:58, 1.36it/s]
Loading 0:  33%|███▎      | 121/363 [01:27<02:58, 1.36it/s]
Loading 0:  34%|███▎      | 122/363 [01:29<03:36, 1.11it/s]
Loading 0:  34%|███▎      | 122/363 [01:29<03:36, 1.11it/s]
Loading 0:  36%|███▌      | 129/363 [01:32<02:19, 1.67it/s]
Loading 0:  36%|███▌      | 129/363 [01:32<02:19, 1.67it/s]
Loading 0:  36%|███▌      | 130/363 [01:34<02:50, 1.37it/s]
Loading 0:  36%|███▌      | 130/363 [01:34<02:50, 1.37it/s]
Loading 0:  36%|███▌      | 131/363 [01:36<03:26, 1.12it/s]
Loading 0:  36%|███▌      | 131/363 [01:36<03:26, 1.12it/s]
Loading 0:  38%|███▊      | 138/363 [01:38<02:13, 1.68it/s]
Loading 0:  38%|███▊      | 138/363 [01:38<02:13, 1.68it/s]
Loading 0:  38%|███▊      | 139/363 [01:40<02:42, 1.38it/s]
Loading 0:  38%|███▊      | 139/363 [01:40<02:42, 1.38it/s]
Loading 0:  39%|███▊      | 140/363 [01:42<03:18, 1.13it/s]
Loading 0:  39%|███▊      | 140/363 [01:42<03:18, 1.13it/s]
Loading 0:  41%|████      | 148/363 [01:43<01:36, 2.23it/s]
Loading 0:  41%|████      | 148/363 [01:43<01:36, 2.23it/s]
Loading 0:  42%|████▏     | 151/363 [01:45<01:49, 1.94it/s]
Loading 0:  42%|████▏     | 151/363 [01:45<01:49, 1.94it/s]
Loading 0:  42%|████▏     | 152/363 [01:47<02:18, 1.53it/s]
Loading 0:  42%|████▏     | 152/363 [01:47<02:18, 1.53it/s]
Loading 0:  42%|████▏     | 153/363 [01:49<02:53, 1.21it/s]
Loading 0:  42%|████▏     | 153/363 [01:49<02:53, 1.21it/s]
Loading 0:  43%|████▎     | 156/363 [01:51<02:36, 1.32it/s]
Loading 0:  43%|████▎     | 156/363 [01:51<02:36, 1.32it/s]
Loading 0:  43%|████▎     | 157/363 [01:53<03:08, 1.09it/s]
Loading 0:  43%|████▎     | 157/363 [01:53<03:08, 1.09it/s]
Loading 0:  44%|████▎     | 158/363 [01:55<03:45, 1.10s/it]
Loading 0:  44%|████▎     | 158/363 [01:55<03:45, 1.10s/it]
Loading 0:  45%|████▌     | 165/363 [01:58<02:05, 1.58it/s]
Loading 0:  45%|████▌     | 165/363 [01:58<02:05, 1.58it/s]
Loading 0:  46%|████▌     | 166/363 [01:59<02:32, 1.29it/s]
Loading 0:  46%|████▌     | 166/363 [01:59<02:32, 1.29it/s]
Loading 0:  46%|████▌     | 167/363 [02:01<03:04, 1.06it/s]
Loading 0:  46%|████▌     | 167/363 [02:01<03:04, 1.06it/s]
Loading 0:  48%|████▊     | 174/363 [02:04<01:53, 1.66it/s]
Loading 0:  48%|████▊     | 174/363 [02:04<01:53, 1.66it/s]
Loading 0:  48%|████▊     | 175/363 [02:06<02:18, 1.35it/s]
Loading 0:  48%|████▊     | 175/363 [02:06<02:18, 1.35it/s]
Loading 0:  48%|████▊     | 176/363 [02:08<02:48, 1.11it/s]
Loading 0:  48%|████▊     | 176/363 [02:08<02:48, 1.11it/s]
Loading 0:  50%|█████     | 182/363 [02:10<01:57, 1.54it/s]
Loading 0:  50%|█████     | 182/363 [02:10<01:57, 1.54it/s]
Loading 0:  52%|█████▏    | 188/363 [02:13<01:38, 1.78it/s]
Loading 0:  52%|█████▏    | 188/363 [02:13<01:38, 1.78it/s]
Loading 0:  52%|█████▏    | 189/363 [02:15<02:00, 1.44it/s]
Loading 0:  52%|█████▏    | 189/363 [02:15<02:00, 1.44it/s]
Loading 0:  53%|█████▎    | 192/363 [02:17<01:55, 1.48it/s]
Loading 0:  53%|█████▎    | 192/363 [02:17<01:55, 1.48it/s]
Loading 0:  53%|█████▎    | 193/363 [02:19<02:19, 1.22it/s]
Loading 0:  53%|█████▎    | 193/363 [02:19<02:19, 1.22it/s]
Loading 0:  53%|█████▎    | 194/363 [02:21<02:47, 1.01it/s]
Loading 0:  53%|█████▎    | 194/363 [02:21<02:47, 1.01it/s]
Loading 0:  55%|█████▌    | 201/363 [02:23<01:40, 1.62it/s]
Loading 0:  55%|█████▌    | 201/363 [02:23<01:40, 1.62it/s]
Loading 0:  56%|█████▌    | 202/363 [02:25<02:01, 1.33it/s]
Loading 0:  56%|█████▌    | 202/363 [02:25<02:01, 1.33it/s]
Loading 0:  56%|█████▌    | 203/363 [02:27<02:26, 1.09it/s]
Loading 0:  56%|█████▌    | 203/363 [02:27<02:26, 1.09it/s]
Loading 0:  58%|█████▊    | 210/363 [02:30<01:31, 1.67it/s]
Loading 0:  58%|█████▊    | 210/363 [02:30<01:31, 1.67it/s]
Loading 0:  58%|█████▊    | 211/363 [02:32<01:51, 1.37it/s]
Loading 0:  58%|█████▊    | 211/363 [02:32<01:51, 1.37it/s]
Loading 0:  58%|█████▊    | 212/363 [02:34<02:15, 1.12it/s]
Loading 0:  58%|█████▊    | 212/363 [02:34<02:15, 1.12it/s]
Loading 0:  60%|██████    | 218/363 [02:36<01:33, 1.54it/s]
Loading 0:  60%|██████    | 218/363 [02:36<01:33, 1.54it/s]
Loading 0:  60%|██████    | 219/363 [02:38<01:54, 1.26it/s]
Loading 0:  60%|██████    | 219/363 [02:38<01:54, 1.26it/s]
Loading 0:  62%|██████▏   | 225/363 [02:41<01:25, 1.60it/s]
Loading 0:  62%|██████▏   | 225/363 [02:41<01:25, 1.60it/s]
Loading 0:  63%|██████▎   | 228/363 [02:43<01:24, 1.61it/s]
Loading 0:  63%|██████▎   | 228/363 [02:43<01:24, 1.61it/s]
Loading 0:  63%|██████▎   | 229/363 [02:44<01:41, 1.32it/s]
Loading 0:  63%|██████▎   | 229/363 [02:44<01:41, 1.32it/s]
Loading 0:  63%|██████▎   | 230/363 [02:46<02:02, 1.08it/s]
Loading 0:  63%|██████▎   | 230/363 [02:46<02:02, 1.08it/s]
Loading 0:  65%|██████▌   | 237/363 [02:49<01:15, 1.66it/s]
Loading 0:  65%|██████▌   | 237/363 [02:49<01:15, 1.66it/s]
Loading 0:  66%|██████▌   | 238/363 [02:51<01:31, 1.36it/s]
Loading 0:  66%|██████▌   | 238/363 [02:51<01:31, 1.36it/s]
Loading 0:  66%|██████▌   | 239/363 [02:53<01:51, 1.12it/s]
Loading 0:  66%|██████▌   | 239/363 [02:53<01:51, 1.12it/s]
Loading 0:  68%|██████▊   | 246/363 [02:55<01:09, 1.68it/s]
Loading 0:  68%|██████▊   | 246/363 [02:55<01:09, 1.68it/s]
Loading 0:  68%|██████▊   | 247/363 [02:57<01:24, 1.38it/s]
Loading 0:  68%|██████▊   | 247/363 [02:57<01:24, 1.38it/s]
Loading 0:  68%|██████▊   | 248/363 [02:59<01:42, 1.13it/s]
Loading 0:  68%|██████▊   | 248/363 [02:59<01:42, 1.13it/s]
Loading 0:  70%|███████   | 255/363 [03:02<01:04, 1.69it/s]
Loading 0:  70%|███████   | 255/363 [03:02<01:04, 1.69it/s]
Loading 0:  71%|███████   | 256/363 [03:04<01:17, 1.38it/s]
Loading 0:  71%|███████   | 256/363 [03:04<01:17, 1.38it/s]
Loading 0:  71%|███████   | 257/363 [03:06<01:33, 1.13it/s]
Loading 0:  71%|███████   | 257/363 [03:06<01:33, 1.13it/s]
Loading 0:  73%|███████▎  | 265/363 [03:07<00:44, 2.23it/s]
Loading 0:  73%|███████▎  | 265/363 [03:07<00:44, 2.23it/s]
Loading 0:  74%|███████▍  | 268/363 [03:09<00:49, 1.94it/s]
Loading 0:  74%|███████▍  | 268/363 [03:09<00:49, 1.94it/s]
Loading 0:  74%|███████▍  | 269/363 [03:11<01:01, 1.53it/s]
Loading 0:  74%|███████▍  | 269/363 [03:11<01:01, 1.53it/s]
Loading 0:  74%|███████▍  | 270/363 [03:13<01:16, 1.21it/s]
Loading 0:  74%|███████▍  | 270/363 [03:13<01:16, 1.21it/s]
Loading 0:  75%|███████▌  | 273/363 [03:28<03:24, 2.27s/it]
Loading 0:  75%|███████▌  | 273/363 [03:28<03:24, 2.27s/it]
Loading 0:  75%|███████▌  | 274/363 [03:29<03:17, 2.22s/it]
Loading 0:  75%|███████▌  | 274/363 [03:29<03:17, 2.22s/it]
Loading 0:  76%|███████▌  | 275/363 [03:31<03:12, 2.19s/it]
Loading 0:  76%|███████▌  | 275/363 [03:31<03:12, 2.19s/it]
Loading 0:  78%|███████▊  | 282/363 [03:34<01:23, 1.03s/it]
Loading 0:  78%|███████▊  | 282/363 [03:34<01:23, 1.03s/it]
Loading 0:  78%|███████▊  | 283/363 [03:36<01:30, 1.13s/it]
Loading 0:  78%|███████▊  | 283/363 [03:36<01:30, 1.13s/it]
Loading 0:  78%|███████▊  | 284/363 [03:38<01:38, 1.25s/it]
Loading 0:  78%|███████▊  | 284/363 [03:38<01:38, 1.25s/it]
Loading 0:  80%|████████  | 291/363 [03:40<00:52, 1.37it/s]
Loading 0:  80%|████████  | 291/363 [03:40<00:52, 1.37it/s]
Loading 0:  80%|████████  | 292/363 [03:42<01:00, 1.18it/s]
Loading 0:  80%|████████  | 292/363 [03:42<01:00, 1.18it/s]
Loading 0:  81%|████████  | 293/363 [03:44<01:09, 1.00it/s]
Loading 0:  81%|████████  | 293/363 [03:44<01:09, 1.00it/s]
Loading 0:  82%|████████▏ | 299/363 [03:47<00:44, 1.45it/s]
Loading 0:  82%|████████▏ | 299/363 [03:47<00:44, 1.45it/s]
Loading 0:  84%|████████▍ | 305/363 [03:49<00:33, 1.72it/s]
Loading 0:  84%|████████▍ | 305/363 [03:49<00:33, 1.72it/s]
Loading 0:  84%|████████▍ | 306/363 [03:51<00:40, 1.41it/s]
Loading 0:  84%|████████▍ | 306/363 [03:51<00:40, 1.41it/s]
Loading 0:  85%|████████▌ | 309/363 [03:53<00:37, 1.46it/s]
Loading 0:  85%|████████▌ | 309/363 [03:53<00:37, 1.46it/s]
Loading 0:  85%|████████▌ | 310/363 [03:55<00:43, 1.21it/s]
Loading 0:  85%|████████▌ | 310/363 [03:55<00:43, 1.21it/s]
Loading 0:  86%|████████▌ | 311/363 [03:57<00:51, 1.00it/s]
Loading 0:  86%|████████▌ | 311/363 [03:57<00:51, 1.00it/s]
Loading 0:  88%|████████▊ | 318/363 [04:00<00:27, 1.61it/s]
Loading 0:  88%|████████▊ | 318/363 [04:00<00:27, 1.61it/s]
Loading 0:  88%|████████▊ | 319/363 [04:01<00:33, 1.32it/s]
Loading 0:  88%|████████▊ | 319/363 [04:01<00:33, 1.32it/s]
Loading 0:  88%|████████▊ | 320/363 [04:03<00:39, 1.09it/s]
Loading 0:  88%|████████▊ | 320/363 [04:03<00:39, 1.09it/s]
Loading 0:  90%|█████████ | 327/363 [04:06<00:21, 1.67it/s]
Loading 0:  90%|█████████ | 327/363 [04:06<00:21, 1.67it/s]
Loading 0:  90%|█████████ | 328/363 [04:08<00:25, 1.37it/s]
Loading 0:  90%|█████████ | 328/363 [04:08<00:25, 1.37it/s]
Loading 0:  91%|█████████ | 329/363 [04:10<00:30, 1.12it/s]
Loading 0:  91%|█████████ | 329/363 [04:10<00:30, 1.12it/s]
Loading 0:  92%|█████████▏| 335/363 [04:12<00:18, 1.55it/s]
Loading 0:  92%|█████████▏| 335/363 [04:12<00:18, 1.55it/s]
Loading 0:  93%|█████████▎| 336/363 [04:14<00:21, 1.26it/s]
Loading 0:  93%|█████████▎| 336/363 [04:14<00:21, 1.26it/s]
Loading 0:  94%|█████████▍| 343/363 [04:17<00:11, 1.70it/s]
Loading 0:  94%|█████████▍| 343/363 [04:17<00:11, 1.70it/s]
Loading 0:  95%|█████████▌| 346/363 [04:19<00:10, 1.68it/s]
Loading 0:  95%|█████████▌| 346/363 [04:19<00:10, 1.68it/s]
Loading 0:  96%|█████████▌| 347/363 [04:21<00:11, 1.37it/s]
Loading 0:  96%|█████████▌| 347/363 [04:21<00:11, 1.37it/s]
Loading 0:  96%|█████████▌| 348/363 [04:23<00:13, 1.12it/s]
Loading 0:  96%|█████████▌| 348/363 [04:23<00:13, 1.12it/s]
Loading 0:  98%|█████████▊| 355/363 [04:25<00:04, 1.68it/s]
Loading 0:  98%|█████████▊| 355/363 [04:25<00:04, 1.68it/s]
Loading 0:  98%|█████████▊| 356/363 [04:27<00:05, 1.38it/s]
Loading 0:  98%|█████████▊| 356/363 [04:27<00:05, 1.38it/s]
Loading 0:  98%|█████████▊| 357/363 [04:29<00:05, 1.12it/s]
Loading 0:  98%|█████████▊| 357/363 [04:29<00:05, 1.12it/s]
Loading 0: 100%|██████████| 363/363 [04:30<00:00, 2.06it/s]
Loading 0: 100%|██████████| 363/363 [04:30<00:00, 2.06it/s]
                                                           

Loading 0: 100%|██████████| 363/363 [04:30<00:00, 1.34it/s]
HTTP Request: %s %s "%s %d %s"
HTTP Request: %s %s "%s %d %s"
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: The tokenizer you are loading from '/tmp/tmprhdgjlr2' with an incorrect regex pattern: https://huggingface.co/mistralai/Mistral-Small-3.1-24B-Instruct-2503/discussions/84#69121093e8b480e709447d5e. This will lead to incorrect tokenization. You should set the `fix_mistral_regex=True` flag when loading this tokenizer to fix this issue.
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: quantized model in 276.690s
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: Processed model ChaiML/csfs-v3-3-dpo-lr5e6b01-lora in 326.477s
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: creating bucket guanaco-mkml-models
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-csfs-v3-3-dpo-l-86358-v20/nvidia
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-csfs-v3-3-dpo-l-86358-v20/nvidia/config.json
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-csfs-v3-3-dpo-l-86358-v20/nvidia/special_tokens_map.json
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-csfs-v3-3-dpo-l-86358-v20/nvidia/tokenizer_config.json
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-csfs-v3-3-dpo-l-86358-v20/nvidia/tokenizer.json
HTTP Request: %s %s "%s %d %s"
chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-csfs-v3-3-dpo-l-86358-v20/nvidia/flywheel_model.0.safetensors
Job chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer completed after 452.9s with status: succeeded
Stopping job with name chaiml-csfs-v3-3-dpo-l-86358-v20-mkmlizer
Pipeline stage MKMLizer completed in 454.00s
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 chaiml-csfs-v3-3-dpo-l-86358-v20
Waiting for inference service chaiml-csfs-v3-3-dpo-l-86358-v20 to be ready
HTTP Request: %s %s "%s %d %s"
HTTP Request: %s %s "%s %d %s"
Inference service chaiml-csfs-v3-3-dpo-l-86358-v20 ready after 181.57848358154297s
Pipeline stage MKMLDeployer completed in 182.24s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.471066951751709s
Received healthy response to inference request in 2.153644323348999s
Received healthy response to inference request in 2.2886526584625244s
Received healthy response to inference request in 2.09515118598938s
Received healthy response to inference request in 1.9268848896026611s
5 requests
0 failed requests
5th percentile: 1.9605381488800049
10th percentile: 1.9941914081573486
20th percentile: 2.061497926712036
30th percentile: 2.1068498134613036
40th percentile: 2.1302470684051515
50th percentile: 2.153644323348999
60th percentile: 2.2076476573944093
70th percentile: 2.2616509914398195
80th percentile: 2.325135517120361
90th percentile: 2.398101234436035
95th percentile: 2.4345840930938722
99th percentile: 2.463770380020142
mean time: 2.1870800018310548
Pipeline stage StressChecker completed in 12.98s
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 0.60s
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 0.63s
Shutdown handler de-registered
chaiml-csfs-v3-3-dpo-l_86358_v20 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
Generating Leaderboard row for %s
Generated Leaderboard row for %s
Pipeline stage OfflineFamilyFriendlyScorer completed in 2982.97s
Shutdown handler de-registered
chaiml-csfs-v3-3-dpo-l_86358_v20 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of chaiml-csfs-v3-3-dpo-l_86358_v20
Shutdown handler not registered because Python interpreter is not running in the main thread
Falling back to EndpointApi.from_submission implementation
admin requested tearing down of chaiml-csfs-v3-3-dpo-l_86358_v20
Falling back to EndpointApi.from_submission implementation
Shutdown handler not registered because Python interpreter is not running in the main thread
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
run pipeline %s
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
run pipeline stage %s
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Running pipeline stage MKMLDeleter
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Checking if service chaiml-csfs-v3-3-dpo-l-86358-v20 is running
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Skipping teardown as no inference service was found
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Pipeline stage MKMLDeleter completed in 4.34s
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
run pipeline stage %s
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Running pipeline stage MKMLModelDeleter
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Cleaning model data from S3
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Cleaning model data from model cache
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Deleting key chaiml-csfs-v3-3-dpo-l-86358-v20/nvidia/tokenizer.json from bucket guanaco-mkml-models
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Deleting key chaiml-csfs-v3-3-dpo-l-86358-v20/nvidia/tokenizer_config.json from bucket guanaco-mkml-models
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Pipeline stage MKMLModelDeleter completed in 7.91s
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Shutdown handler de-registered
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation
Falling back to EndpointApi.from_submission implementation