sao10k-hanami-1-t2

developer_uid: sao10k

submission_id: sao10k-hanami-1-t2_v2

model_name: Hanami1t2

model_group: Sao10K/Hanami-1-t2

status: torndown

timestamp: 2024-09-14T10:08:19+00:00

num_battles: 11467

num_wins: 5451

celo_rating: 1235.53

family_friendly_score: 0.0

submission_type: basic

model_repo: Sao10K/Hanami-1-t2

model_architecture: LlamaForCausalLM

model_num_parameters: 8030261248.0

best_of: 8

max_input_tokens: 1024

max_output_tokens: 64

latencies: [{'batch_size': 1, 'throughput': 0.8690677832711078, 'latency_mean': 1.1505954277515411, 'latency_p50': 1.1479309797286987, 'latency_p90': 1.2842433452606201}, {'batch_size': 4, 'throughput': 1.834049922696835, 'latency_mean': 2.1774490058422087, 'latency_p50': 2.1784881353378296, 'latency_p90': 2.4101496934890747}, {'batch_size': 5, 'throughput': 1.9754893033959564, 'latency_mean': 2.5209620547294618, 'latency_p50': 2.5130088329315186, 'latency_p90': 2.850889205932617}, {'batch_size': 8, 'throughput': 2.1414040504606477, 'latency_mean': 3.711528924703598, 'latency_p50': 3.7131842374801636, 'latency_p90': 4.224092245101929}, {'batch_size': 10, 'throughput': 2.170121129409783, 'latency_mean': 4.576292990446091, 'latency_p50': 4.59443724155426, 'latency_p90': 5.129844975471497}, {'batch_size': 12, 'throughput': 2.165474507219235, 'latency_mean': 5.496030659675598, 'latency_p50': 5.525946378707886, 'latency_p90': 6.241402435302734}, {'batch_size': 15, 'throughput': 2.158077500593512, 'latency_mean': 6.875766993761062, 'latency_p50': 6.919549584388733, 'latency_p90': 7.714779305458069}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: Hanami1t2

is_internal_developer: False

language_model: Sao10K/Hanami-1-t2

model_size: 8B

ranking_group: single

throughput_3p7s: 2.15

us_pacific_date: 2024-09-14

win_ratio: 0.47536408825324844

generation_params: {'temperature': 1.2, 'top_p': 1.0, 'min_p': 0.2, 'top_k': 50, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '\n\n', '<|eot_id|>', '\n\n{user_name}', '\nYou:', '<|end_header_id|>'], 'max_input_tokens': 1024, 'best_of': 8, 'max_output_tokens': 64}

formatter: {'memory_template': "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\n{bot_name}'s Persona: {memory}\n\n", 'prompt_template': '{prompt}<|eot_id|>', 'bot_template': '<|start_header_id|>assistant<|end_header_id|>\n\n{bot_name}: {message}<|eot_id|>', 'user_template': '<|start_header_id|>user<|end_header_id|>\n\n{user_name}: {message}<|eot_id|>', 'response_template': '<|start_header_id|>assistant<|end_header_id|>\n\n{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 sao10k-hanami-1-t2-v2-mkmlizer
Waiting for job on sao10k-hanami-1-t2-v2-mkmlizer to finish
sao10k-hanami-1-t2-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
sao10k-hanami-1-t2-v2-mkmlizer: ║     _____            __           __                                ║
sao10k-hanami-1-t2-v2-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
sao10k-hanami-1-t2-v2-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
sao10k-hanami-1-t2-v2-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
sao10k-hanami-1-t2-v2-mkmlizer: ║       /___/                                                         ║
sao10k-hanami-1-t2-v2-mkmlizer: ║                                                                     ║
sao10k-hanami-1-t2-v2-mkmlizer: ║  Version: 0.10.1                                                    ║
sao10k-hanami-1-t2-v2-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
sao10k-hanami-1-t2-v2-mkmlizer: ║  https://mk1.ai                                                     ║
sao10k-hanami-1-t2-v2-mkmlizer: ║                                                                     ║
sao10k-hanami-1-t2-v2-mkmlizer: ║  The license key for the current software has been verified as      ║
sao10k-hanami-1-t2-v2-mkmlizer: ║  belonging to:                                                      ║
sao10k-hanami-1-t2-v2-mkmlizer: ║                                                                     ║
sao10k-hanami-1-t2-v2-mkmlizer: ║  Chai Research Corp.                                                ║
sao10k-hanami-1-t2-v2-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
sao10k-hanami-1-t2-v2-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
sao10k-hanami-1-t2-v2-mkmlizer: ║                                                                     ║
sao10k-hanami-1-t2-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
sao10k-hanami-1-t2-v2-mkmlizer: Downloaded to shared memory in 19.308s
sao10k-hanami-1-t2-v2-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpufwz8tum, device:0
sao10k-hanami-1-t2-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
sao10k-hanami-1-t2-v2-mkmlizer: quantized model in 26.153s
sao10k-hanami-1-t2-v2-mkmlizer: Processed model Sao10K/Hanami-1-t2 in 45.460s
sao10k-hanami-1-t2-v2-mkmlizer: creating bucket guanaco-mkml-models
sao10k-hanami-1-t2-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
sao10k-hanami-1-t2-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/sao10k-hanami-1-t2-v2
sao10k-hanami-1-t2-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/sao10k-hanami-1-t2-v2/config.json
sao10k-hanami-1-t2-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/sao10k-hanami-1-t2-v2/special_tokens_map.json
sao10k-hanami-1-t2-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/sao10k-hanami-1-t2-v2/tokenizer_config.json
sao10k-hanami-1-t2-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/sao10k-hanami-1-t2-v2/tokenizer.json
sao10k-hanami-1-t2-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/sao10k-hanami-1-t2-v2/flywheel_model.0.safetensors
sao10k-hanami-1-t2-v2-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   1%|          | 2/291 [00:04<11:29,  2.38s/it]
Loading 0:   2%|▏         | 6/291 [00:04<03:03,  1.56it/s]
Loading 0:   4%|▍         | 13/291 [00:05<01:05,  4.24it/s]
Loading 0:   6%|▌         | 17/291 [00:05<00:44,  6.14it/s]
Loading 0:   8%|▊         | 23/291 [00:05<00:27,  9.89it/s]
Loading 0:  10%|▉         | 29/291 [00:05<00:19, 13.65it/s]
Loading 0:  12%|█▏        | 34/291 [00:05<00:14, 17.48it/s]
Loading 0:  14%|█▍        | 42/291 [00:05<00:10, 23.75it/s]
Loading 0:  17%|█▋        | 50/291 [00:05<00:07, 32.03it/s]
Loading 0:  19%|█▉        | 56/291 [00:05<00:06, 33.89it/s]
Loading 0:  21%|██▏       | 62/291 [00:06<00:06, 36.82it/s]
Loading 0:  23%|██▎       | 68/291 [00:06<00:05, 40.55it/s]
Loading 0:  25%|██▌       | 74/291 [00:06<00:05, 40.97it/s]
Loading 0:  27%|██▋       | 79/291 [00:06<00:05, 42.08it/s]
Loading 0:  29%|██▉       | 85/291 [00:06<00:04, 44.79it/s]
Loading 0:  31%|███       | 90/291 [00:06<00:04, 45.49it/s]
Loading 0:  33%|███▎      | 95/291 [00:06<00:04, 41.26it/s]
Loading 0:  35%|███▍      | 101/291 [00:06<00:04, 39.41it/s]
Loading 0:  36%|███▋      | 106/291 [00:07<00:06, 26.98it/s]
Loading 0:  38%|███▊      | 112/291 [00:07<00:05, 32.51it/s]
Loading 0:  40%|████      | 117/291 [00:07<00:04, 34.90it/s]
Loading 0:  42%|████▏     | 123/291 [00:07<00:04, 35.27it/s]
Loading 0:  45%|████▌     | 131/291 [00:07<00:03, 43.43it/s]
Loading 0:  47%|████▋     | 137/291 [00:07<00:03, 42.35it/s]
Loading 0:  49%|████▉     | 142/291 [00:08<00:03, 43.22it/s]
Loading 0:  51%|█████     | 149/291 [00:08<00:02, 49.30it/s]
Loading 0:  53%|█████▎    | 155/291 [00:08<00:03, 44.34it/s]
Loading 0:  55%|█████▍    | 160/291 [00:08<00:03, 43.48it/s]
Loading 0:  57%|█████▋    | 167/291 [00:08<00:02, 49.58it/s]
Loading 0:  59%|█████▉    | 173/291 [00:08<00:02, 47.20it/s]
Loading 0:  61%|██████    | 178/291 [00:08<00:02, 45.97it/s]
Loading 0:  63%|██████▎   | 184/291 [00:08<00:02, 49.35it/s]
Loading 0:  65%|██████▌   | 190/291 [00:09<00:01, 50.70it/s]
Loading 0:  67%|██████▋   | 196/291 [00:09<00:02, 45.12it/s]
Loading 0:  69%|██████▉   | 202/291 [00:09<00:01, 48.45it/s]
Loading 0:  71%|███████▏  | 208/291 [00:09<00:01, 49.72it/s]
Loading 0:  74%|███████▎  | 214/291 [00:09<00:01, 44.10it/s]
Loading 0:  76%|███████▌  | 221/291 [00:09<00:01, 49.29it/s]
Loading 0:  78%|███████▊  | 227/291 [00:09<00:01, 44.84it/s]
Loading 0:  80%|███████▉  | 232/291 [00:09<00:01, 44.54it/s]
Loading 0:  82%|████████▏ | 238/291 [00:10<00:01, 47.50it/s]
Loading 0:  84%|████████▎ | 243/291 [00:10<00:01, 47.75it/s]
Loading 0:  85%|████████▌ | 248/291 [00:10<00:01, 31.36it/s]
Loading 0:  87%|████████▋ | 254/291 [00:10<00:01, 32.60it/s]
Loading 0:  89%|████████▊ | 258/291 [00:10<00:00, 33.59it/s]
Loading 0:  91%|█████████▏| 266/291 [00:10<00:00, 43.42it/s]
Loading 0:  93%|█████████▎| 272/291 [00:10<00:00, 43.03it/s]
Loading 0:  95%|█████████▌| 277/291 [00:11<00:00, 42.80it/s]
Loading 0:  98%|█████████▊| 284/291 [00:11<00:00, 49.13it/s]
Loading 0: 100%|█████████▉| 290/291 [00:11<00:00, 47.02it/s]
                                                            
Job sao10k-hanami-1-t2-v2-mkmlizer completed after 66.04s with status: succeeded
Stopping job with name sao10k-hanami-1-t2-v2-mkmlizer
Pipeline stage MKMLizer completed in 69.33s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.09s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service sao10k-hanami-1-t2-v2
Waiting for inference service sao10k-hanami-1-t2-v2 to be ready
Inference service sao10k-hanami-1-t2-v2 ready after 170.9698359966278s
Pipeline stage MKMLDeployer completed in 172.08s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.510711431503296s
Received healthy response to inference request in 1.9173681735992432s
Received healthy response to inference request in 1.6367316246032715s
Received healthy response to inference request in 1.7042577266693115s
Received healthy response to inference request in 2.2596843242645264s
5 requests
0 failed requests
5th percentile: 1.6502368450164795
10th percentile: 1.6637420654296875
20th percentile: 1.6907525062561035
30th percentile: 1.7468798160552979
40th percentile: 1.8321239948272705
50th percentile: 1.9173681735992432
60th percentile: 2.0542946338653563
70th percentile: 2.19122109413147
80th percentile: 2.3098897457122805
90th percentile: 2.410300588607788
95th percentile: 2.460506010055542
99th percentile: 2.5006703472137453
mean time: 2.0057506561279297
Pipeline stage StressChecker completed in 11.91s
run pipeline stage %s
Running pipeline stage TriggerMKMLProfilingPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%s
Pipeline stage TriggerMKMLProfilingPipeline completed in 6.42s
Shutdown handler de-registered
sao10k-hanami-1-t2_v2 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 0.11s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.10s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service sao10k-hanami-1-t2-v2-profiler
Waiting for inference service sao10k-hanami-1-t2-v2-profiler to be ready
Inference service sao10k-hanami-1-t2-v2-profiler ready after 170.3793501853943s
Pipeline stage MKMLProfilerDeployer completed in 170.74s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/sao10k-hanami-1-t2-v2-profiler-predictor-00001-deployment-htksv:/code/chaiverse_profiler_1726308970 --namespace tenant-chaiml-guanaco
kubectl exec -it sao10k-hanami-1-t2-v2-profiler-predictor-00001-deployment-htksv --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1726308970 && 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_1726308970/summary.json'
kubectl exec -it sao10k-hanami-1-t2-v2-profiler-predictor-00001-deployment-htksv --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1726308970/summary.json'
Pipeline stage MKMLProfilerRunner completed in 816.87s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service sao10k-hanami-1-t2-v2-profiler is running
Tearing down inference service sao10k-hanami-1-t2-v2-profiler
Service sao10k-hanami-1-t2-v2-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.70s
Shutdown handler de-registered
sao10k-hanami-1-t2_v2 status is now inactive due to auto deactivation removed underperforming models
sao10k-hanami-1-t2_v2 status is now torndown due to DeploymentManager action