chaiml-dpo-debug-ep1

developer_uid: bogoconic1

submission_id: chaiml-dpo-debug-ep1_v1

model_name: chaiml-dpo-debug-ep1_v1

model_group: ChaiML/dpo-debug-ep1

status: torndown

timestamp: 2025-07-02T12:12:56+00:00

num_battles: 8217

num_wins: 3551

celo_rating: 1239.16

family_friendly_score: 0.6126

family_friendly_standard_error: 0.006889430165115254

submission_type: basic

model_repo: ChaiML/dpo-debug-ep1

model_architecture: MistralForCausalLM

model_num_parameters: 12772070400.0

best_of: 8

max_input_tokens: 1024

max_output_tokens: 64

latencies: [{'batch_size': 1, 'throughput': 0.593805116860376, 'latency_mean': 1.6839180779457092, 'latency_p50': 1.6877539157867432, 'latency_p90': 1.83993661403656}, {'batch_size': 3, 'throughput': 1.060530296172797, 'latency_mean': 2.8181470859050752, 'latency_p50': 2.831488013267517, 'latency_p90': 3.073858642578125}, {'batch_size': 5, 'throughput': 1.277289118729925, 'latency_mean': 3.902527189254761, 'latency_p50': 3.9211615324020386, 'latency_p90': 4.383196759223938}, {'batch_size': 6, 'throughput': 1.3437370405070976, 'latency_mean': 4.4273417127132415, 'latency_p50': 4.434096693992615, 'latency_p90': 4.981665086746216}, {'batch_size': 8, 'throughput': 1.3969378238198737, 'latency_mean': 5.683202899694443, 'latency_p50': 5.710419535636902, 'latency_p90': 6.417232036590576}, {'batch_size': 10, 'throughput': 1.4230544797921298, 'latency_mean': 6.984424057006836, 'latency_p50': 6.925239443778992, 'latency_p90': 7.879940557479858}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: chaiml-dpo-debug-ep1_v1

is_internal_developer: True

language_model: ChaiML/dpo-debug-ep1

model_size: 13B

ranking_group: single

throughput_3p7s: 1.25

us_pacific_date: 2025-07-02

win_ratio: 0.43215285383960084

generation_params: {'temperature': 0.6, 'top_p': 0.95, 'min_p': 0.025, 'top_k': 60, 'presence_penalty': 0.4, 'frequency_penalty': 0.4, 'stopping_words': ['<|im_start|>', '<|im_end|>', '\n'], 'max_input_tokens': 1024, 'best_of': 8, 'max_output_tokens': 64}

formatter: {'memory_template': '<|system|>Family Friendly{memory}\n', 'prompt_template': '', 'bot_template': '<|im_start|>assistant\n{message}<|im_end|>\n', 'user_template': '<|im_start|>user\nYou:{message}<|im_end|>\n', 'response_template': '<|im_start|>assistant\n', '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-dpo-debug-ep1-v1-mkmlizer
Waiting for job on chaiml-dpo-debug-ep1-v1-mkmlizer to finish
chaiml-dpo-debug-ep1-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-dpo-debug-ep1-v1-mkmlizer: ║                                                                     ║
chaiml-dpo-debug-ep1-v1-mkmlizer: ║   ██████   ██████  █████   ████  ████                               ║
chaiml-dpo-debug-ep1-v1-mkmlizer: ║  ░░██████ ██████  ░░███   ███░  ░░███                               ║
chaiml-dpo-debug-ep1-v1-mkmlizer: ║   ░███░█████░███   ░███  ███     ░███                               ║
chaiml-dpo-debug-ep1-v1-mkmlizer: ║   ░███░░███ ░███   ░███████      ░███                               ║
chaiml-dpo-debug-ep1-v1-mkmlizer: ║   ░███ ░░░  ░███   ░███░░███     ░███                               ║
chaiml-dpo-debug-ep1-v1-mkmlizer: ║   ░███      ░███   ░███ ░░███    ░███                               ║
chaiml-dpo-debug-ep1-v1-mkmlizer: ║   █████     █████  █████ ░░████  █████                              ║
chaiml-dpo-debug-ep1-v1-mkmlizer: ║  ░░░░░     ░░░░░  ░░░░░   ░░░░  ░░░░░                               ║
chaiml-dpo-debug-ep1-v1-mkmlizer: ║                                                                     ║
chaiml-dpo-debug-ep1-v1-mkmlizer: ║  Version: 0.29.8                                                    ║
chaiml-dpo-debug-ep1-v1-mkmlizer: ║  Features: FLYWHEEL, CUDA                                           ║
chaiml-dpo-debug-ep1-v1-mkmlizer: ║  Copyright 2023-2025 MK ONE TECHNOLOGIES Inc.                       ║
chaiml-dpo-debug-ep1-v1-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-dpo-debug-ep1-v1-mkmlizer: ║                                                                     ║
chaiml-dpo-debug-ep1-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-dpo-debug-ep1-v1-mkmlizer: ║  belonging to:                                                      ║
chaiml-dpo-debug-ep1-v1-mkmlizer: ║                                                                     ║
chaiml-dpo-debug-ep1-v1-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-dpo-debug-ep1-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-dpo-debug-ep1-v1-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
chaiml-dpo-debug-ep1-v1-mkmlizer: ║                                                                     ║
chaiml-dpo-debug-ep1-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
Unable to record family friendly update due to error: HTTPConnectionPool(host='chaiml-nemo-guard-merged-v2-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 0x7f05cc14cc50>, 'Connection to chaiml-nemo-guard-merged-v2-predictor.tenant-chaiml-guanaco.k.chaiverse.com timed out. (connect timeout=12.0)'))
chaiml-dpo-debug-ep1-v1-mkmlizer: Downloaded to shared memory in 39.596s
chaiml-dpo-debug-ep1-v1-mkmlizer: Checking if ChaiML/dpo-debug-ep1 already exists in ChaiML
chaiml-dpo-debug-ep1-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpe63cdjf2, device:0
chaiml-dpo-debug-ep1-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-dpo-debug-ep1-v1-mkmlizer: quantized model in 30.595s
chaiml-dpo-debug-ep1-v1-mkmlizer: Processed model ChaiML/dpo-debug-ep1 in 70.192s
chaiml-dpo-debug-ep1-v1-mkmlizer: creating bucket guanaco-mkml-models
chaiml-dpo-debug-ep1-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-dpo-debug-ep1-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-dpo-debug-ep1-v1/nvidia
chaiml-dpo-debug-ep1-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-dpo-debug-ep1-v1/nvidia/config.json
chaiml-dpo-debug-ep1-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-dpo-debug-ep1-v1/nvidia/special_tokens_map.json
chaiml-dpo-debug-ep1-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-dpo-debug-ep1-v1/nvidia/tokenizer_config.json
chaiml-dpo-debug-ep1-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-dpo-debug-ep1-v1/nvidia/flywheel_model.0.safetensors
chaiml-dpo-debug-ep1-v1-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|▏         | 5/363 [00:00<00:11, 30.10it/s]
Loading 0:   4%|▎         | 13/363 [00:00<00:07, 49.52it/s]
Loading 0:   5%|▌         | 19/363 [00:00<00:07, 44.09it/s]
Loading 0:   7%|▋         | 24/363 [00:00<00:08, 42.31it/s]
Loading 0:   9%|▊         | 31/363 [00:00<00:06, 49.11it/s]
Loading 0:  10%|█         | 37/363 [00:00<00:07, 45.79it/s]
Loading 0:  12%|█▏        | 42/363 [00:00<00:07, 44.61it/s]
Loading 0:  13%|█▎        | 49/363 [00:01<00:06, 49.77it/s]
Loading 0:  15%|█▌        | 55/363 [00:01<00:07, 43.62it/s]
Loading 0:  17%|█▋        | 60/363 [00:01<00:06, 43.31it/s]
Loading 0:  18%|█▊        | 65/363 [00:01<00:10, 28.65it/s]
Loading 0:  20%|█▉        | 71/363 [00:01<00:08, 34.38it/s]
Loading 0:  21%|██        | 76/363 [00:01<00:08, 35.68it/s]
Loading 0:  22%|██▏       | 81/363 [00:02<00:07, 38.10it/s]
Loading 0:  24%|██▍       | 87/363 [00:02<00:07, 38.81it/s]
Loading 0:  25%|██▌       | 92/363 [00:02<00:06, 38.82it/s]
Loading 0:  27%|██▋       | 99/363 [00:02<00:05, 44.22it/s]
Loading 0:  29%|██▊       | 104/363 [00:02<00:05, 44.20it/s]
Loading 0:  30%|███       | 109/363 [00:02<00:05, 45.28it/s]
Loading 0:  31%|███▏      | 114/363 [00:02<00:06, 38.02it/s]
Loading 0:  33%|███▎      | 119/363 [00:02<00:06, 38.06it/s]
Loading 0:  35%|███▍      | 126/363 [00:03<00:05, 43.54it/s]
Loading 0:  36%|███▌      | 131/363 [00:03<00:05, 45.06it/s]
Loading 0:  37%|███▋      | 136/363 [00:03<00:06, 37.55it/s]
Loading 0:  39%|███▉      | 142/363 [00:03<00:07, 31.38it/s]
Loading 0:  40%|████      | 146/363 [00:03<00:06, 32.22it/s]
Loading 0:  41%|████▏     | 150/363 [00:03<00:06, 30.65it/s]
Loading 0:  43%|████▎     | 156/363 [00:03<00:05, 36.39it/s]
Loading 0:  44%|████▍     | 160/363 [00:04<00:05, 36.46it/s]
Loading 0:  46%|████▌     | 166/363 [00:04<00:04, 40.32it/s]
Loading 0:  47%|████▋     | 171/363 [00:04<00:04, 41.80it/s]
Loading 0:  48%|████▊     | 176/363 [00:04<00:05, 35.52it/s]
Loading 0:  50%|█████     | 183/363 [00:04<00:04, 43.26it/s]
Loading 0:  52%|█████▏    | 188/363 [00:04<00:04, 42.73it/s]
Loading 0:  53%|█████▎    | 193/363 [00:04<00:03, 42.52it/s]
Loading 0:  55%|█████▍    | 198/363 [00:04<00:03, 44.40it/s]
Loading 0:  56%|█████▌    | 203/363 [00:05<00:04, 37.19it/s]
Loading 0:  58%|█████▊    | 210/363 [00:05<00:03, 44.12it/s]
Loading 0:  59%|█████▉    | 215/363 [00:05<00:03, 42.62it/s]
Loading 0:  61%|██████    | 220/363 [00:05<00:03, 44.40it/s]
Loading 0:  62%|██████▏   | 225/363 [00:05<00:05, 27.01it/s]
Loading 0:  63%|██████▎   | 230/363 [00:05<00:04, 29.45it/s]
Loading 0:  65%|██████▌   | 237/363 [00:06<00:03, 36.72it/s]
Loading 0:  67%|██████▋   | 242/363 [00:06<00:03, 38.00it/s]
Loading 0:  68%|██████▊   | 247/363 [00:06<00:02, 39.34it/s]
Loading 0:  69%|██████▉   | 252/363 [00:06<00:02, 41.83it/s]
Loading 0:  71%|███████   | 257/363 [00:06<00:03, 34.44it/s]
Loading 0:  73%|███████▎  | 264/363 [00:06<00:02, 41.93it/s]
Loading 0:  74%|███████▍  | 269/363 [00:06<00:02, 40.82it/s]
Loading 0:  75%|███████▌  | 274/363 [00:06<00:02, 40.29it/s]
Loading 0:  77%|███████▋  | 279/363 [00:07<00:02, 40.61it/s]
Loading 0:  78%|███████▊  | 284/363 [00:07<00:02, 34.33it/s]
Loading 0:  80%|████████  | 292/363 [00:07<00:01, 42.46it/s]
Loading 0:  82%|████████▏ | 297/363 [00:07<00:01, 43.28it/s]
Loading 0:  83%|████████▎ | 302/363 [00:07<00:01, 43.25it/s]
Loading 0:  85%|████████▍ | 307/363 [00:08<00:02, 19.89it/s]
Loading 0:  86%|████████▌ | 312/363 [00:08<00:02, 21.76it/s]
Loading 0:  88%|████████▊ | 319/363 [00:08<00:01, 28.79it/s]
Loading 0:  89%|████████▉ | 324/363 [00:08<00:01, 31.68it/s]
Loading 0:  91%|█████████ | 329/363 [00:08<00:00, 34.30it/s]
Loading 0:  92%|█████████▏| 334/363 [00:08<00:00, 36.90it/s]
Loading 0:  93%|█████████▎| 339/363 [00:09<00:00, 32.22it/s]
Loading 0:  95%|█████████▌| 346/363 [00:09<00:00, 39.58it/s]
Loading 0:  97%|█████████▋| 351/363 [00:09<00:00, 39.40it/s]
Loading 0:  98%|█████████▊| 356/363 [00:09<00:00, 40.47it/s]
Loading 0:  99%|█████████▉| 361/363 [00:09<00:00, 40.98it/s]
                                                            
Job chaiml-dpo-debug-ep1-v1-mkmlizer completed after 94.29s with status: succeeded
Stopping job with name chaiml-dpo-debug-ep1-v1-mkmlizer
Pipeline stage MKMLizer completed in 94.93s
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-dpo-debug-ep1-v1
Waiting for inference service chaiml-dpo-debug-ep1-v1 to be ready
Failed to get response for submission chaiml-mistral31-24b-s_69496_v16: HTTPConnectionPool(host='chaiml-mistral31-24b-s-69496-v16-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 0x7f064c64d390>, 'Connection to chaiml-mistral31-24b-s-69496-v16-predictor.tenant-chaiml-guanaco.k.chaiverse.com timed out. (connect timeout=12.0)'))
Inference service chaiml-dpo-debug-ep1-v1 ready after 220.73511838912964s
Pipeline stage MKMLDeployer completed in 221.26s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.2504961490631104s
Received healthy response to inference request in 1.5028245449066162s
Received healthy response to inference request in 1.7699849605560303s
Received healthy response to inference request in 1.4177541732788086s
{"detail":"HTTPConnectionPool(host='chaiml-dpo-debug-ep1-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 0x7d0a5026c410>, 'Connection to chaiml-dpo-debug-ep1-v1-predictor.tenant-chaiml-guanaco.k.chaiverse.com timed out. (connect timeout=12.0)'))"}
Received unhealthy response to inference request!
5 requests
1 failed requests
5th percentile: 1.4347682476043702
10th percentile: 1.4517823219299317
20th percentile: 1.4858104705810546
30th percentile: 1.5562566280364991
40th percentile: 1.6631207942962647
50th percentile: 1.7699849605560303
60th percentile: 1.9621894359588623
70th percentile: 2.1543939113616943
80th percentile: 4.237921714782717
90th percentile: 8.212772846221924
95th percentile: 10.200198411941527
99th percentile: 11.790138864517212
mean time: 3.8257367610931396
%s, retrying in %s seconds...
Received healthy response to inference request in 1.7090976238250732s
Received healthy response to inference request in 1.7739675045013428s
Received healthy response to inference request in 1.7213985919952393s
Received healthy response to inference request in 1.6517882347106934s
Received healthy response to inference request in 1.5060474872589111s
5 requests
0 failed requests
5th percentile: 1.5351956367492676
10th percentile: 1.5643437862396241
20th percentile: 1.6226400852203369
30th percentile: 1.6632501125335692
40th percentile: 1.6861738681793212
50th percentile: 1.7090976238250732
60th percentile: 1.7140180110931396
70th percentile: 1.718938398361206
80th percentile: 1.73191237449646
90th percentile: 1.7529399394989014
95th percentile: 1.763453722000122
99th percentile: 1.7718647480010987
mean time: 1.672459888458252
Pipeline stage StressChecker completed in 29.85s
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 0.95s
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 0.81s
Shutdown handler de-registered
chaiml-dpo-debug-ep1_v1 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
Pipeline stage OfflineFamilyFriendlyScorer completed in 3216.71s
Shutdown handler de-registered
chaiml-dpo-debug-ep1_v1 status is now inactive due to auto deactivation removed underperforming models
chaiml-dpo-debug-ep1_v1 status is now torndown due to DeploymentManager action