developer_uid: NischayDnk
submission_id: chaiml-sft-magnum-28b-v2try_v1
model_name: chaiml-sft-magnum-28b-v2try_v1
model_group: ChaiML/sft-magnum-28b-v2
status: torndown
timestamp: 2025-04-01T00:15:15+00:00
num_battles: 7820
num_wins: 3770
celo_rating: 1271.1
family_friendly_score: 0.5682
family_friendly_standard_error: 0.0070049805138915275
submission_type: basic
model_repo: ChaiML/sft-magnum-28b-v2try
model_architecture: Gemma2ForCausalLM
model_num_parameters: 28731935232.0
best_of: 8
max_input_tokens: 1024
max_output_tokens: 64
reward_model: default
latencies: [{'batch_size': 1, 'throughput': 0.2764328148790405, 'latency_mean': 3.617419971227646, 'latency_p50': 3.631057024002075, 'latency_p90': 4.026276731491089}, {'batch_size': 2, 'throughput': 0.41804986056785126, 'latency_mean': 4.772444721460342, 'latency_p50': 4.805798888206482, 'latency_p90': 5.245583391189575}, {'batch_size': 3, 'throughput': 0.5302626183891663, 'latency_mean': 5.63118944644928, 'latency_p50': 5.613843202590942, 'latency_p90': 6.213445472717285}, {'batch_size': 5, 'throughput': 0.6755440927042107, 'latency_mean': 7.355484598875046, 'latency_p50': 7.365422248840332, 'latency_p90': 8.16287648677826}]
gpu_counts: {'NVIDIA RTX A6000': 1}
display_name: chaiml-sft-magnum-28b-v2try_v1
is_internal_developer: False
language_model: ChaiML/sft-magnum-28b-v2try
model_size: 29B
ranking_group: single
throughput_3p7s: 0.29
us_pacific_date: 2025-03-31
win_ratio: 0.4820971867007673
generation_params: {'temperature': 0.5, 'top_p': 0.9, 'min_p': 0.05, 'top_k': 80, 'presence_penalty': 0.1, 'frequency_penalty': 0.1, 'stopping_words': ['\n', '<|im_end|>'], 'max_input_tokens': 1024, 'best_of': 8, 'max_output_tokens': 64}
formatter: {'memory_template': '', 'prompt_template': '', 'bot_template': '<|im_start|>assistant\n{message}<|im_end|>\n', 'user_template': '<|im_start|>user\n{message}<|im_end|>\n', 'response_template': '<|im_start|>assistant\n', '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-sft-magnum-28b-v2try-v1-mkmlizer
Waiting for job on chaiml-sft-magnum-28b-v2try-v1-mkmlizer to finish
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: ║ _____ __ __ ║
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: ║ /___/ ║
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: ║ ║
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: ║ Version: 0.12.8 ║
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: ║ https://mk1.ai ║
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: ║ ║
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: ║ The license key for the current software has been verified as ║
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: ║ belonging to: ║
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: ║ ║
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: ║ Chai Research Corp. ║
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: ║ Expiration: 2025-04-15 23:59:59 ║
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: ║ ║
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: Downloaded to shared memory in 101.762s
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp6z6o008h, device:0
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
Retrying (%r) after connection broken by '%r': %s
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: quantized model in 67.199s
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: Processed model ChaiML/sft-magnum-28b-v2try in 168.961s
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: creating bucket guanaco-mkml-models
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-sft-magnum-28b-v2try-v1
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.model s3://guanaco-mkml-models/chaiml-sft-magnum-28b-v2try-v1/tokenizer.model
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-sft-magnum-28b-v2try-v1/tokenizer.json
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.2.safetensors s3://guanaco-mkml-models/chaiml-sft-magnum-28b-v2try-v1/flywheel_model.2.safetensors
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/chaiml-sft-magnum-28b-v2try-v1/flywheel_model.1.safetensors
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-sft-magnum-28b-v2try-v1/flywheel_model.0.safetensors
chaiml-sft-magnum-28b-v2try-v1-mkmlizer: Loading 0: 0%| | 0/508 [00:00<?, ?it/s] Loading 0: 1%| | 5/508 [00:00<00:15, 33.45it/s] Loading 0: 2%|▏ | 12/508 [00:00<00:10, 46.34it/s] Loading 0: 3%|▎ | 17/508 [00:00<00:14, 34.52it/s] Loading 0: 5%|▌ | 27/508 [00:00<00:10, 46.73it/s] Loading 0: 6%|▋ | 32/508 [00:00<00:15, 29.86it/s] Loading 0: 7%|▋ | 38/508 [00:01<00:14, 32.71it/s] Loading 0: 9%|▉ | 47/508 [00:01<00:11, 39.10it/s] Loading 0: 10%|█ | 52/508 [00:01<00:11, 40.93it/s] Loading 0: 11%|█▏ | 58/508 [00:01<00:10, 41.15it/s] Loading 0: 12%|█▏ | 63/508 [00:01<00:10, 42.96it/s] Loading 0: 13%|█▎ | 68/508 [00:01<00:11, 39.47it/s] Loading 0: 14%|█▍ | 73/508 [00:02<00:15, 28.86it/s] Loading 0: 15%|█▌ | 78/508 [00:02<00:13, 30.81it/s] Loading 0: 16%|█▌ | 82/508 [00:02<00:13, 30.77it/s] Loading 0: 17%|█▋ | 87/508 [00:02<00:12, 34.85it/s] Loading 0: 19%|█▊ | 95/508 [00:02<00:10, 40.93it/s] Loading 0: 20%|█▉ | 100/508 [00:02<00:11, 36.96it/s] Loading 0: 21%|██▏ | 108/508 [00:02<00:09, 40.92it/s] Loading 0: 22%|██▏ | 113/508 [00:03<00:09, 42.69it/s] Loading 0: 23%|██▎ | 118/508 [00:03<00:11, 33.29it/s] Loading 0: 24%|██▍ | 124/508 [00:03<00:10, 38.39it/s] Loading 0: 25%|██▌ | 129/508 [00:03<00:09, 40.65it/s] Loading 0: 27%|██▋ | 135/508 [00:03<00:09, 39.95it/s] Loading 0: 28%|██▊ | 140/508 [00:03<00:08, 41.81it/s] Loading 0: 29%|██▊ | 146/508 [00:03<00:08, 40.90it/s] Loading 0: 30%|██▉ | 151/508 [00:03<00:08, 43.04it/s] Loading 0: 31%|███ | 157/508 [00:04<00:08, 40.68it/s] Loading 0: 32%|███▏ | 162/508 [00:04<00:08, 42.80it/s] Loading 0: 33%|███▎ | 168/508 [00:04<00:07, 45.01it/s] Loading 0: 34%|███▍ | 173/508 [00:04<00:11, 29.73it/s] Loading 0: 35%|███▌ | 179/508 [00:04<00:09, 35.45it/s] Loading 0: 36%|███▌ | 184/508 [00:04<00:08, 38.45it/s] Loading 0: 37%|███▋ | 190/508 [00:05<00:08, 39.00it/s] Loading 0: 38%|███▊ | 195/508 [00:19<04:10, 1.25it/s] Loading 0: 39%|███▉ | 199/508 [00:19<03:10, 1.63it/s] Loading 0: 40%|████ | 204/508 [00:19<02:17, 2.21it/s] Loading 0: 42%|████▏ | 211/508 [00:19<01:26, 3.43it/s] Loading 0: 43%|████▎ | 216/508 [00:20<01:05, 4.46it/s] Loading 0: 43%|████▎ | 220/508 [00:20<00:50, 5.66it/s] Loading 0: 44%|████▍ | 225/508 [00:20<00:37, 7.55it/s] Loading 0: 46%|████▌ | 234/508 [00:20<00:22, 12.08it/s] Loading 0: 47%|████▋ | 238/508 [00:20<00:18, 14.21it/s] Loading 0: 48%|████▊ | 245/508 [00:20<00:13, 18.83it/s] Loading 0: 49%|████▉ | 250/508 [00:20<00:11, 22.46it/s] Loading 0: 50%|█████ | 255/508 [00:20<00:10, 24.35it/s] Loading 0: 51%|█████ | 260/508 [00:21<00:08, 27.60it/s] Loading 0: 52%|█████▏ | 265/508 [00:21<00:09, 25.99it/s] Loading 0: 53%|█████▎ | 269/508 [00:21<00:09, 26.29it/s] Loading 0: 55%|█████▍ | 278/508 [00:21<00:06, 34.92it/s] Loading 0: 56%|█████▌ | 283/508 [00:21<00:05, 37.76it/s] Loading 0: 57%|█████▋ | 289/508 [00:21<00:05, 37.94it/s] Loading 0: 58%|█████▊ | 295/508 [00:21<00:05, 42.49it/s] Loading 0: 59%|█████▉ | 300/508 [00:22<00:05, 40.51it/s] Loading 0: 60%|██████ | 305/508 [00:22<00:04, 42.70it/s] Loading 0: 61%|██████ | 311/508 [00:22<00:04, 46.89it/s] Loading 0: 62%|██████▏ | 316/508 [00:22<00:06, 30.42it/s] Loading 0: 63%|██████▎ | 322/508 [00:22<00:05, 36.00it/s] Loading 0: 64%|██████▍ | 327/508 [00:22<00:04, 38.71it/s] Loading 0: 66%|██████▌ | 333/508 [00:22<00:04, 39.05it/s] Loading 0: 67%|██████▋ | 338/508 [00:23<00:04, 41.47it/s] Loading 0: 68%|██████▊ | 344/508 [00:23<00:04, 39.99it/s] Loading 0: 69%|██████▊ | 349/508 [00:23<00:03, 42.33it/s] Loading 0: 70%|██████▉ | 354/508 [00:23<00:04, 38.47it/s] Loading 0: 71%|███████ | 359/508 [00:23<00:04, 32.07it/s] Loading 0: 71%|███████▏ | 363/508 [00:23<00:04, 33.24it/s] Loading 0: 72%|███████▏ | 368/508 [00:23<00:04, 33.53it/s] Loading 0: 74%|███████▍ | 377/508 [00:24<00:03, 40.74it/s] Loading 0: 75%|███████▌ | 383/508 [00:24<00:02, 44.78it/s] Loading 0: 76%|███████▋ | 388/508 [00:24<00:02, 41.41it/s] Loading 0: 77%|███████▋ | 393/508 [00:24<00:02, 43.38it/s] Loading 0: 78%|███████▊ | 398/508 [00:24<00:02, 39.84it/s] Loading 0: 79%|███████▉ | 403/508 [00:24<00:02, 40.14it/s] Loading 0: 80%|████████ | 408/508 [00:24<00:02, 33.89it/s] Loading 0: 81%|████████ | 412/508 [00:25<00:03, 31.98it/s] Loading 0: 83%|████████▎ | 421/508 [00:25<00:02, 39.31it/s] Loading 0: 84%|████████▍ | 427/508 [00:25<00:01, 43.61it/s] Loading 0: 85%|████████▌ | 432/508 [00:25<00:01, 40.32it/s] Loading 0: 86%|████████▌ | 437/508 [00:25<00:01, 42.47it/s] Loading 0: 87%|████████▋ | 443/508 [00:25<00:01, 41.20it/s] Loading 0: 88%|████████▊ | 448/508 [00:25<00:01, 43.28it/s] Loading 0: 89%|████████▉ | 451/508 [00:40<00:01, 43.28it/s] Loading 0: 89%|████████▉ | 452/508 [00:40<00:48, 1.16it/s] Loading 0: 90%|████████▉ | 456/508 [00:40<00:33, 1.53it/s] Loading 0: 91%|█████████ | 461/508 [00:40<00:22, 2.13it/s] Loading 0: 92%|█████████▏| 467/508 [00:40<00:12, 3.18it/s] Loading 0: 93%|█████████▎| 474/508 [00:40<00:06, 4.88it/s] Loading 0: 94%|█████████▍| 479/508 [00:40<00:04, 6.29it/s] Loading 0: 96%|█████████▌| 487/508 [00:41<00:02, 9.53it/s] Loading 0: 97%|█████████▋| 492/508 [00:41<00:01, 11.99it/s] Loading 0: 98%|█████████▊| 497/508 [00:41<00:00, 14.24it/s] Loading 0: 99%|█████████▉| 502/508 [00:41<00:00, 15.38it/s] Loading 0: 100%|█████████▉| 506/508 [00:41<00:00, 17.86it/s]
Job chaiml-sft-magnum-28b-v2try-v1-mkmlizer completed after 201.41s with status: succeeded
Stopping job with name chaiml-sft-magnum-28b-v2try-v1-mkmlizer
Pipeline stage MKMLizer completed in 202.55s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.14s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service chaiml-sft-magnum-28b-v2try-v1
Waiting for inference service chaiml-sft-magnum-28b-v2try-v1 to be ready
Inference service chaiml-sft-magnum-28b-v2try-v1 ready after 120.5001266002655s
Pipeline stage MKMLDeployer completed in 121.00s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 4.012522220611572s
Received healthy response to inference request in 3.1905975341796875s
Received healthy response to inference request in 3.181701183319092s
Received healthy response to inference request in 3.652682304382324s
Received healthy response to inference request in 3.2492642402648926s
5 requests
0 failed requests
5th percentile: 3.183480453491211
10th percentile: 3.18525972366333
20th percentile: 3.188818264007568
30th percentile: 3.2023308753967283
40th percentile: 3.2257975578308105
50th percentile: 3.2492642402648926
60th percentile: 3.410631465911865
70th percentile: 3.571998691558838
80th percentile: 3.7246502876281737
90th percentile: 3.8685862541198732
95th percentile: 3.9405542373657227
99th percentile: 3.9981286239624025
mean time: 3.4573534965515136
Pipeline stage StressChecker completed in 18.55s
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 0.61s
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 0.59s
Shutdown handler de-registered
chaiml-sft-magnum-28b-v2try_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 3167.82s
Shutdown handler de-registered
chaiml-sft-magnum-28b-v2try_v1 status is now inactive due to auto deactivation removed underperforming models
chaiml-sft-magnum-28b-v2try_v1 status is now torndown due to DeploymentManager action