chaiml-2fe5-v1-dpo-cosine

developer_uid: chai_backend_admin

submission_id: chaiml-2fe5-v1-dpo-cosine_v1

model_name: training123

model_group: ChaiML/2fe5-v1-dpo_cosin

status: torndown

timestamp: 2025-11-21T07:31:23+00:00

num_battles: 5630

num_wins: 2941

celo_rating: 1303.51

family_friendly_score: 0.5292

family_friendly_standard_error: 0.007058999362515909

submission_type: basic

model_repo: ChaiML/2fe5-v1-dpo_cosine

model_architecture: MistralForCausalLM

model_num_parameters: 12772070400.0

best_of: 8

max_input_tokens: 2048

max_output_tokens: 64

reward_model: default

latencies: [{'batch_size': 1, 'throughput': 0.6884785540776391, 'latency_mean': 1.4523569548130035, 'latency_p50': 1.450685739517212, 'latency_p90': 1.6127763509750366}, {'batch_size': 3, 'throughput': 1.2923965909542006, 'latency_mean': 2.312327188253403, 'latency_p50': 2.3204352855682373, 'latency_p90': 2.522824120521545}, {'batch_size': 5, 'throughput': 1.5853725098339644, 'latency_mean': 3.142801042795181, 'latency_p50': 3.1340038776397705, 'latency_p90': 3.447249174118042}, {'batch_size': 6, 'throughput': 1.6826343099937104, 'latency_mean': 3.5519750940799715, 'latency_p50': 3.5272886753082275, 'latency_p90': 3.9796488523483275}, {'batch_size': 8, 'throughput': 1.8003334539048672, 'latency_mean': 4.420652085542679, 'latency_p50': 4.443403720855713, 'latency_p90': 4.979939651489258}, {'batch_size': 10, 'throughput': 1.8868905456587544, 'latency_mean': 5.268258663415909, 'latency_p50': 5.285846948623657, 'latency_p90': 5.916210865974426}]

gpu_counts: {'NVIDIA L40S': 1}

display_name: training123

is_internal_developer: True

language_model: ChaiML/2fe5-v1-dpo_cosine

model_size: 13B

ranking_group: single

throughput_3p7s: 1.71

us_pacific_date: 2025-11-17

win_ratio: 0.522380106571936

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': ['User:', '\n', 'You:', '</s>'], 'max_input_tokens': 2048, 'best_of': 8, 'max_output_tokens': 64}

formatter: {'memory_template': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '{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-2fe5-v1-dpo-cosine-v1-mkmlizer
Waiting for job on chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer to finish
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: ║                                                                     ║
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: ║   ██████   ██████  █████   ████  ████                               ║
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: ║  ░░██████ ██████  ░░███   ███░  ░░███                               ║
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: ║   ░███░█████░███   ░███  ███     ░███                               ║
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: ║   ░███░░███ ░███   ░███████      ░███                               ║
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: ║   ░███ ░░░  ░███   ░███░░███     ░███                               ║
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: ║   ░███      ░███   ░███ ░░███    ░███                               ║
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: ║   █████     █████  █████ ░░████  █████                              ║
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: ║  ░░░░░     ░░░░░  ░░░░░   ░░░░  ░░░░░                               ║
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: ║                                                                     ║
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: ║  Version: 0.30.2                                                    ║
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: ║  Features: FLYWHEEL, CUDA                                           ║
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: ║  Copyright 2023-2025 MK ONE TECHNOLOGIES Inc.                       ║
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: ║                                                                     ║
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: ║  belonging to:                                                      ║
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: ║                                                                     ║
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: ║                                                                     ║
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: Downloaded to shared memory in 42.195s
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: Checking if ChaiML/2fe5-v1-dpo_cosine already exists in ChaiML
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp2rx2r4es, device:0
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
Failed to get response for submission mistralai-mistral-nem_93303_v569: ('http://mistralai-mistral-nem-93303-v569-predictor.tenant-chaiml-guanaco.k2.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', '')
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: quantized model in 26.966s
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: Processed model ChaiML/2fe5-v1-dpo_cosine in 69.162s
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: creating bucket guanaco-mkml-models
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-2fe5-v1-dpo-cosine-v1/nvidia
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-2fe5-v1-dpo-cosine-v1/nvidia/config.json
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-2fe5-v1-dpo-cosine-v1/nvidia/special_tokens_map.json
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-2fe5-v1-dpo-cosine-v1/nvidia/tokenizer_config.json
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-2fe5-v1-dpo-cosine-v1/nvidia/tokenizer.json
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-2fe5-v1-dpo-cosine-v1/nvidia/flywheel_model.0.safetensors
chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|▏         | 5/363 [00:00<00:10, 32.88it/s]
Loading 0:   4%|▍         | 14/363 [00:00<00:07, 46.37it/s]
Loading 0:   6%|▋         | 23/363 [00:00<00:06, 50.54it/s]
Loading 0:   9%|▉         | 32/363 [00:00<00:06, 50.90it/s]
Loading 0:  11%|█▏        | 41/363 [00:00<00:06, 52.81it/s]
Loading 0:  14%|█▍        | 50/363 [00:00<00:05, 53.78it/s]
Loading 0:  17%|█▋        | 60/363 [00:01<00:05, 59.38it/s]
Loading 0:  18%|█▊        | 67/363 [00:01<00:06, 43.25it/s]
Loading 0:  20%|██        | 73/363 [00:01<00:06, 41.58it/s]
Loading 0:  22%|██▏       | 80/363 [00:01<00:06, 44.37it/s]
Loading 0:  23%|██▎       | 85/363 [00:01<00:06, 41.00it/s]
Loading 0:  25%|██▍       | 90/363 [00:01<00:06, 39.37it/s]
Loading 0:  26%|██▌       | 95/363 [00:02<00:06, 39.40it/s]
Loading 0:  28%|██▊       | 100/363 [00:02<00:08, 31.98it/s]
Loading 0:  29%|██▉       | 105/363 [00:02<00:07, 35.34it/s]
Loading 0:  30%|███       | 110/363 [00:02<00:07, 36.12it/s]
Loading 0:  31%|███▏      | 114/363 [00:02<00:07, 34.70it/s]
Loading 0:  33%|███▎      | 118/363 [00:02<00:07, 34.62it/s]
Loading 0:  35%|███▍      | 126/363 [00:02<00:05, 44.22it/s]
Loading 0:  36%|███▋      | 132/363 [00:03<00:05, 44.77it/s]
Loading 0:  38%|███▊      | 137/363 [00:03<00:04, 45.22it/s]
Loading 0:  39%|███▉      | 142/363 [00:03<00:06, 36.55it/s]
Loading 0:  40%|████      | 147/363 [00:03<00:05, 37.16it/s]
Loading 0:  42%|████▏     | 153/363 [00:03<00:04, 42.25it/s]
Loading 0:  44%|████▎     | 158/363 [00:03<00:05, 39.37it/s]
Loading 0:  46%|████▌     | 166/363 [00:03<00:04, 48.58it/s]
Loading 0:  47%|████▋     | 172/363 [00:03<00:03, 47.97it/s]
Loading 0:  49%|████▉     | 178/363 [00:04<00:03, 50.15it/s]
Loading 0:  51%|█████     | 185/363 [00:04<00:03, 47.86it/s]
Loading 0:  53%|█████▎    | 193/363 [00:04<00:03, 55.02it/s]
Loading 0:  55%|█████▍    | 199/363 [00:04<00:03, 52.15it/s]
Loading 0:  56%|█████▋    | 205/363 [00:04<00:02, 52.83it/s]
Loading 0:  58%|█████▊    | 212/363 [00:04<00:03, 48.86it/s]
Loading 0:  61%|██████    | 222/363 [00:04<00:02, 54.95it/s]
Loading 0:  63%|██████▎   | 228/363 [00:05<00:03, 41.67it/s]
Loading 0:  64%|██████▍   | 233/363 [00:05<00:04, 31.01it/s]
Loading 0:  66%|██████▌   | 238/363 [00:05<00:03, 32.23it/s]
Loading 0:  67%|██████▋   | 244/363 [00:05<00:03, 35.53it/s]
Loading 0:  69%|██████▊   | 249/363 [00:05<00:02, 38.13it/s]
Loading 0:  71%|███████   | 256/363 [00:05<00:02, 44.87it/s]
Loading 0:  72%|███████▏  | 262/363 [00:06<00:02, 45.32it/s]
Loading 0:  74%|███████▎  | 267/363 [00:06<00:02, 45.57it/s]
Loading 0:  75%|███████▌  | 274/363 [00:06<00:01, 50.96it/s]
Loading 0:  77%|███████▋  | 280/363 [00:06<00:01, 49.63it/s]
Loading 0:  79%|███████▉  | 286/363 [00:06<00:01, 50.77it/s]
Loading 0:  80%|████████  | 292/363 [00:06<00:01, 52.77it/s]
Loading 0:  82%|████████▏ | 298/363 [00:06<00:01, 49.86it/s]
Loading 0:  84%|████████▎ | 304/363 [00:07<00:01, 29.53it/s]
Loading 0:  85%|████████▌ | 310/363 [00:07<00:01, 32.68it/s]
Loading 0:  87%|████████▋ | 317/363 [00:07<00:01, 36.85it/s]
Loading 0:  89%|████████▊ | 322/363 [00:07<00:01, 39.24it/s]
Loading 0:  91%|█████████ | 330/363 [00:07<00:00, 43.24it/s]
Loading 0:  93%|█████████▎| 339/363 [00:07<00:00, 47.16it/s]
Loading 0:  96%|█████████▌| 348/363 [00:08<00:00, 50.03it/s]
Loading 0:  98%|█████████▊| 357/363 [00:08<00:00, 51.91it/s]
                                                            
Job chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer completed after 113.97s with status: succeeded
Stopping job with name chaiml-2fe5-v1-dpo-cosine-v1-mkmlizer
Pipeline stage MKMLizer completed in 114.91s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.22s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service chaiml-2fe5-v1-dpo-cosine-v1
Waiting for inference service chaiml-2fe5-v1-dpo-cosine-v1 to be ready
Inference service chaiml-2fe5-v1-dpo-cosine-v1 ready after 150.65268898010254s
Pipeline stage MKMLDeployer completed in 151.39s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.608086109161377s
Received healthy response to inference request in 1.4998753070831299s
Received healthy response to inference request in 1.2649626731872559s
Received healthy response to inference request in 1.5809087753295898s
Received healthy response to inference request in 1.316521406173706s
5 requests
0 failed requests
5th percentile: 1.2752744197845458
10th percentile: 1.285586166381836
20th percentile: 1.306209659576416
30th percentile: 1.353192186355591
40th percentile: 1.4265337467193604
50th percentile: 1.4998753070831299
60th percentile: 1.5322886943817138
70th percentile: 1.564702081680298
80th percentile: 1.5863442420959473
90th percentile: 1.597215175628662
95th percentile: 1.6026506423950195
99th percentile: 1.6069990158081056
mean time: 1.4540708541870118
Pipeline stage StressChecker completed in 8.84s
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 0.78s
Shutdown handler de-registered
chaiml-2fe5-v1-dpo-cosine_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 2606.24s
Shutdown handler de-registered
chaiml-2fe5-v1-dpo-cosine_v1 status is now inactive due to auto deactivation removed underperforming models
chaiml-2fe5-v1-dpo-cosine_v1 status is now torndown due to DeploymentManager action