chaiml-workshopsft

developer_uid: cycy233

submission_id: chaiml-workshopsft_v6

model_name: 4894

model_group: ChaiML/WorkshopSFT

status: torndown

timestamp: 2025-03-07T07:54:18+00:00

num_battles: 9292

num_wins: 4253

celo_rating: 1245.62

family_friendly_score: 0.5292

family_friendly_standard_error: 0.007058999362515909

submission_type: basic

model_repo: ChaiML/WorkshopSFT

model_architecture: MistralForCausalLM

model_num_parameters: 12772070400.0

best_of: 8

max_input_tokens: 1024

max_output_tokens: 64

reward_model: default

latencies: [{'batch_size': 1, 'throughput': 0.6059638922984328, 'latency_mean': 1.6501964163780212, 'latency_p50': 1.6581897735595703, 'latency_p90': 1.815079402923584}, {'batch_size': 3, 'throughput': 1.108830228237137, 'latency_mean': 2.6904048991203307, 'latency_p50': 2.683133840560913, 'latency_p90': 2.9882728099822997}, {'batch_size': 5, 'throughput': 1.355315014163017, 'latency_mean': 3.676606684923172, 'latency_p50': 3.6761003732681274, 'latency_p90': 4.087197852134705}, {'batch_size': 6, 'throughput': 1.4040913555470598, 'latency_mean': 4.2479753124713895, 'latency_p50': 4.265997529029846, 'latency_p90': 4.778967046737671}, {'batch_size': 8, 'throughput': 1.4845024461448568, 'latency_mean': 5.360135772228241, 'latency_p50': 5.411736845970154, 'latency_p90': 5.976767539978027}, {'batch_size': 10, 'throughput': 1.5145592989888308, 'latency_mean': 6.538348662853241, 'latency_p50': 6.5307042598724365, 'latency_p90': 7.409029388427735}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: 4894

is_internal_developer: False

language_model: ChaiML/WorkshopSFT

model_size: 13B

ranking_group: single

throughput_3p7s: 1.36

us_pacific_date: 2025-03-06

win_ratio: 0.45770555316401207

generation_params: {'temperature': 1.0, 'top_p': 1.0, 'min_p': 0.0, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n'], 'max_input_tokens': 1024, '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': 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-workshopsft-v6-mkmlizer
Waiting for job on chaiml-workshopsft-v6-mkmlizer to finish
chaiml-workshopsft-v6-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-workshopsft-v6-mkmlizer: ║     _____            __           __                                ║
chaiml-workshopsft-v6-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
chaiml-workshopsft-v6-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
chaiml-workshopsft-v6-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
chaiml-workshopsft-v6-mkmlizer: ║       /___/                                                         ║
chaiml-workshopsft-v6-mkmlizer: ║                                                                     ║
chaiml-workshopsft-v6-mkmlizer: ║  Version: 0.12.8                                                    ║
chaiml-workshopsft-v6-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
chaiml-workshopsft-v6-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-workshopsft-v6-mkmlizer: ║                                                                     ║
chaiml-workshopsft-v6-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-workshopsft-v6-mkmlizer: ║  belonging to:                                                      ║
chaiml-workshopsft-v6-mkmlizer: ║                                                                     ║
chaiml-workshopsft-v6-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-workshopsft-v6-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-workshopsft-v6-mkmlizer: ║  Expiration: 2025-04-15 23:59:59                                    ║
chaiml-workshopsft-v6-mkmlizer: ║                                                                     ║
chaiml-workshopsft-v6-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-workshopsft-v6-mkmlizer: Downloaded to shared memory in 32.726s
chaiml-workshopsft-v6-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp4gqg2dns, device:0
chaiml-workshopsft-v6-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-workshopsft-v6-mkmlizer: quantized model in 36.021s
chaiml-workshopsft-v6-mkmlizer: Processed model ChaiML/WorkshopSFT in 68.748s
chaiml-workshopsft-v6-mkmlizer: creating bucket guanaco-mkml-models
chaiml-workshopsft-v6-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-workshopsft-v6-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-workshopsft-v6
chaiml-workshopsft-v6-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-workshopsft-v6/special_tokens_map.json
chaiml-workshopsft-v6-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-workshopsft-v6/config.json
chaiml-workshopsft-v6-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-workshopsft-v6/tokenizer_config.json
chaiml-workshopsft-v6-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-workshopsft-v6/tokenizer.json
chaiml-workshopsft-v6-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-workshopsft-v6/flywheel_model.0.safetensors
chaiml-workshopsft-v6-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|▏         | 5/363 [00:00<00:12, 28.44it/s]
Loading 0:   4%|▎         | 13/363 [00:00<00:06, 50.10it/s]
Loading 0:   5%|▌         | 19/363 [00:00<00:07, 47.88it/s]
Loading 0:   7%|▋         | 25/363 [00:00<00:06, 49.06it/s]
Loading 0:   9%|▊         | 31/363 [00:00<00:06, 51.87it/s]
Loading 0:  10%|█         | 37/363 [00:00<00:06, 49.40it/s]
Loading 0:  12%|█▏        | 43/363 [00:00<00:06, 50.39it/s]
Loading 0:  13%|█▎        | 49/363 [00:00<00:05, 52.41it/s]
Loading 0:  15%|█▌        | 55/363 [00:01<00:06, 48.70it/s]
Loading 0:  17%|█▋        | 61/363 [00:01<00:08, 36.84it/s]
Loading 0:  18%|█▊        | 66/363 [00:01<00:08, 36.70it/s]
Loading 0:  20%|█▉        | 72/363 [00:01<00:07, 40.97it/s]
Loading 0:  21%|██▏       | 78/363 [00:01<00:06, 42.12it/s]
Loading 0:  23%|██▎       | 83/363 [00:01<00:06, 41.44it/s]
Loading 0:  25%|██▍       | 90/363 [00:02<00:05, 45.97it/s]
Loading 0:  26%|██▌       | 95/363 [00:02<00:05, 46.72it/s]
Loading 0:  28%|██▊       | 100/363 [00:02<00:06, 40.48it/s]
Loading 0:  30%|██▉       | 108/363 [00:02<00:05, 49.77it/s]
Loading 0:  31%|███▏      | 114/363 [00:02<00:05, 44.60it/s]
Loading 0:  33%|███▎      | 119/363 [00:02<00:05, 43.88it/s]
Loading 0:  35%|███▍      | 126/363 [00:02<00:04, 48.92it/s]
Loading 0:  36%|███▋      | 132/363 [00:02<00:04, 47.02it/s]
Loading 0:  38%|███▊      | 137/363 [00:03<00:04, 46.13it/s]
Loading 0:  39%|███▉      | 142/363 [00:03<00:06, 32.94it/s]
Loading 0:  40%|████      | 146/363 [00:03<00:06, 33.36it/s]
Loading 0:  41%|████▏     | 150/363 [00:03<00:06, 33.36it/s]
Loading 0:  43%|████▎     | 157/363 [00:03<00:05, 40.72it/s]
Loading 0:  45%|████▍     | 163/363 [00:03<00:04, 41.69it/s]
Loading 0:  46%|████▋     | 168/363 [00:03<00:04, 42.20it/s]
Loading 0:  48%|████▊     | 175/363 [00:04<00:03, 47.58it/s]
Loading 0:  50%|████▉     | 180/363 [00:04<00:03, 47.23it/s]
Loading 0:  51%|█████     | 185/363 [00:04<00:04, 37.95it/s]
Loading 0:  53%|█████▎    | 193/363 [00:04<00:03, 45.61it/s]
Loading 0:  55%|█████▍    | 199/363 [00:04<00:03, 44.92it/s]
Loading 0:  56%|█████▌    | 204/363 [00:04<00:03, 43.63it/s]
Loading 0:  58%|█████▊    | 210/363 [00:04<00:03, 46.39it/s]
Loading 0:  59%|█████▉    | 215/363 [00:04<00:03, 43.64it/s]
Loading 0:  61%|██████    | 220/363 [00:05<00:03, 43.54it/s]
Loading 0:  62%|██████▏   | 225/363 [00:05<00:05, 25.52it/s]
Loading 0:  63%|██████▎   | 230/363 [00:05<00:04, 27.74it/s]
Loading 0:  66%|██████▌   | 238/363 [00:05<00:03, 36.00it/s]
Loading 0:  67%|██████▋   | 244/363 [00:05<00:03, 37.43it/s]
Loading 0:  69%|██████▊   | 249/363 [00:05<00:02, 38.50it/s]
Loading 0:  71%|███████   | 256/363 [00:06<00:02, 43.64it/s]
Loading 0:  72%|███████▏  | 262/363 [00:06<00:02, 42.58it/s]
Loading 0:  74%|███████▎  | 267/363 [00:06<00:02, 42.75it/s]
Loading 0:  75%|███████▌  | 274/363 [00:06<00:01, 47.52it/s]
Loading 0:  77%|███████▋  | 280/363 [00:06<00:01, 45.10it/s]
Loading 0:  79%|███████▊  | 285/363 [00:06<00:01, 42.78it/s]
Loading 0:  80%|████████  | 291/363 [00:06<00:01, 46.43it/s]
Loading 0:  82%|████████▏ | 296/363 [00:06<00:01, 45.80it/s]
Loading 0:  83%|████████▎ | 301/363 [00:07<00:01, 46.44it/s]
Loading 0:  84%|████████▍ | 306/363 [00:13<00:22,  2.51it/s]
Loading 0:  85%|████████▌ | 310/363 [00:13<00:16,  3.25it/s]
Loading 0:  87%|████████▋ | 315/363 [00:14<00:10,  4.54it/s]
Loading 0:  88%|████████▊ | 321/363 [00:14<00:06,  6.47it/s]
Loading 0:  91%|█████████ | 329/363 [00:14<00:03, 10.12it/s]
Loading 0:  92%|█████████▏| 335/363 [00:14<00:02, 13.03it/s]
Loading 0:  94%|█████████▎| 340/363 [00:14<00:01, 15.95it/s]
Loading 0:  96%|█████████▌| 347/363 [00:14<00:00, 21.52it/s]
Loading 0:  97%|█████████▋| 353/363 [00:14<00:00, 25.44it/s]
Loading 0:  99%|█████████▊| 358/363 [00:14<00:00, 28.70it/s]
                                                            
Job chaiml-workshopsft-v6-mkmlizer completed after 104.06s with status: succeeded
Stopping job with name chaiml-workshopsft-v6-mkmlizer
Pipeline stage MKMLizer completed in 104.51s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.18s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service chaiml-workshopsft-v6
Waiting for inference service chaiml-workshopsft-v6 to be ready
Inference service chaiml-workshopsft-v6 ready after 90.3548367023468s
Pipeline stage MKMLDeployer completed in 90.90s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.5076301097869873s
Received healthy response to inference request in 1.558929681777954s
Received healthy response to inference request in 1.6466319561004639s
Received healthy response to inference request in 1.7306129932403564s
Received healthy response to inference request in 1.541116714477539s
5 requests
0 failed requests
5th percentile: 1.544679307937622
10th percentile: 1.5482419013977051
20th percentile: 1.5553670883178712
30th percentile: 1.576470136642456
40th percentile: 1.61155104637146
50th percentile: 1.6466319561004639
60th percentile: 1.6802243709564209
70th percentile: 1.7138167858123778
80th percentile: 1.8860164165496827
90th percentile: 2.196823263168335
95th percentile: 2.352226686477661
99th percentile: 2.476549425125122
mean time: 1.7969842910766602
Pipeline stage StressChecker completed in 10.40s
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-workshopsft_v6 status is now deployed due to DeploymentManager action
Shutdown handler registered
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Skipping teardown as no inference service was successfully deployed
Pipeline stage MKMLProfilerDeleter completed in 0.09s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.08s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service chaiml-workshopsft-v6-profiler
Waiting for inference service chaiml-workshopsft-v6-profiler to be ready
Inference service chaiml-workshopsft-v6-profiler ready after 100.42175674438477s
Pipeline stage MKMLProfilerDeployer completed in 100.84s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/chaiml-workshopsft-v6-profiler-predictor-00001-deployment-r2cwb:/code/chaiverse_profiler_1741334411 --namespace tenant-chaiml-guanaco
kubectl exec -it chaiml-workshopsft-v6-profiler-predictor-00001-deployment-r2cwb --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1741334411 && python profiles.py profile --best_of_n 8 --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 1024 --output_tokens 64 --summary /code/chaiverse_profiler_1741334411/summary.json'
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 2326.82s
Shutdown handler de-registered
chaiml-workshopsft_v6 status is now inactive due to auto deactivation removed underperforming models
chaiml-workshopsft_v6 status is now torndown due to DeploymentManager action