submission_id: sao10k-mn-12b-lyra-v4a1_v3
developer_uid: sao10k
best_of: 8
celo_rating: 1259.96
display_name: lyra41
family_friendly_score: 0.5509291577790705
family_friendly_standard_error: 0.004860130668386469
formatter: {'memory_template': '<|im_start|>system\n{memory}<|im_end|>\n', 'prompt_template': '<|im_start|>user\n{prompt}<|im_end|>\n', 'bot_template': '<|im_start|>assistant\n{bot_name}: {message}<|im_end|>\n', 'user_template': '<|im_start|>user\n{user_name}: {message}<|im_end|>\n', 'response_template': '<|im_start|>assistant\n{bot_name}:', 'truncate_by_message': False}
generation_params: {'temperature': 0.75, 'top_p': 1.0, 'min_p': 0.1, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '\n\n', '\nYou:', '[/INST]', '<|im_end|>', '</s>'], 'max_input_tokens': 512, 'best_of': 8, 'max_output_tokens': 64}
gpu_counts: {'NVIDIA RTX A5000': 1}
is_internal_developer: False
language_model: Sao10K/MN-12B-Lyra-v4a1
latencies: [{'batch_size': 1, 'throughput': 0.6926061585594797, 'latency_mean': 1.443722903728485, 'latency_p50': 1.4491052627563477, 'latency_p90': 1.5991499423980713}, {'batch_size': 3, 'throughput': 1.3266055396022032, 'latency_mean': 2.259152697324753, 'latency_p50': 2.2799019813537598, 'latency_p90': 2.501718282699585}, {'batch_size': 5, 'throughput': 1.5670494297909603, 'latency_mean': 3.176794725656509, 'latency_p50': 3.1908438205718994, 'latency_p90': 3.5684115648269654}, {'batch_size': 6, 'throughput': 1.6107447985176984, 'latency_mean': 3.7004927265644074, 'latency_p50': 3.6850993633270264, 'latency_p90': 4.2070800304412845}, {'batch_size': 8, 'throughput': 1.5834694838118468, 'latency_mean': 5.0247733175754545, 'latency_p50': 5.0697479248046875, 'latency_p90': 5.690066027641296}, {'batch_size': 10, 'throughput': 1.533016294988604, 'latency_mean': 6.485439442396164, 'latency_p50': 6.527211427688599, 'latency_p90': 7.2984956979751585}]
max_input_tokens: 512
max_output_tokens: 64
model_architecture: MistralForCausalLM
model_group: Sao10K/MN-12B-Lyra-v4a1
model_name: lyra41
model_num_parameters: 12772070400.0
model_repo: Sao10K/MN-12B-Lyra-v4a1
model_size: 13B
num_battles: 5147942
num_wins: 2641132
ranking_group: single
status: deployed
submission_type: basic
throughput_3p7s: 1.62
timestamp: 2024-09-06T03:03:46+00:00
us_pacific_date: 2024-09-05
win_ratio: 0.5130461842810194
Download Preference Data
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-mn-12b-lyra-v4a1-v3-mkmlizer
Waiting for job on sao10k-mn-12b-lyra-v4a1-v3-mkmlizer to finish
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: ║ _____ __ __ ║
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: ║ /___/ ║
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: ║ ║
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: ║ Version: 0.10.1 ║
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: ║ https://mk1.ai ║
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: ║ ║
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: ║ The license key for the current software has been verified as ║
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: ║ belonging to: ║
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: ║ ║
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: ║ Chai Research Corp. ║
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: ║ ║
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: Downloaded to shared memory in 33.326s
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpnynanxbm, device:0
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: Saving flywheel model at /dev/shm/model_cache
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: quantized model in 35.288s
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: Processed model Sao10K/MN-12B-Lyra-v4a1 in 68.614s
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: creating bucket guanaco-mkml-models
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/sao10k-mn-12b-lyra-v4a1-v3
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/sao10k-mn-12b-lyra-v4a1-v3/config.json
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/sao10k-mn-12b-lyra-v4a1-v3/special_tokens_map.json
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/sao10k-mn-12b-lyra-v4a1-v3/tokenizer_config.json
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/sao10k-mn-12b-lyra-v4a1-v3/tokenizer.json
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/sao10k-mn-12b-lyra-v4a1-v3/flywheel_model.0.safetensors
sao10k-mn-12b-lyra-v4a1-v3-mkmlizer: Loading 0: 0%| | 0/363 [00:00<?, ?it/s] Loading 0: 1%| | 2/363 [00:06<18:07, 3.01s/it] Loading 0: 2%|▏ | 6/363 [00:06<04:48, 1.24it/s] Loading 0: 4%|▎ | 13/363 [00:06<01:42, 3.40it/s] Loading 0: 5%|▌ | 19/363 [00:06<00:58, 5.84it/s] Loading 0: 7%|▋ | 24/363 [00:06<00:41, 8.17it/s] Loading 0: 9%|▉ | 32/363 [00:06<00:24, 13.50it/s] Loading 0: 10%|█ | 38/363 [00:06<00:18, 17.46it/s] Loading 0: 12%|█▏ | 44/363 [00:06<00:14, 21.86it/s] Loading 0: 14%|█▍ | 50/363 [00:06<00:11, 27.04it/s] Loading 0: 15%|█▌ | 56/363 [00:07<00:10, 30.61it/s] Loading 0: 17%|█▋ | 61/363 [00:07<00:12, 24.16it/s] Loading 0: 18%|█▊ | 67/363 [00:07<00:09, 29.67it/s] Loading 0: 20%|█▉ | 72/363 [00:07<00:08, 32.53it/s] Loading 0: 21%|██ | 77/363 [00:07<00:07, 35.96it/s] Loading 0: 23%|██▎ | 83/363 [00:07<00:07, 37.92it/s] Loading 0: 24%|██▍ | 88/363 [00:08<00:07, 38.37it/s] Loading 0: 26%|██▌ | 95/363 [00:08<00:06, 44.10it/s] Loading 0: 28%|██▊ | 101/363 [00:08<00:06, 43.18it/s] Loading 0: 29%|██▉ | 106/363 [00:08<00:06, 40.79it/s] Loading 0: 31%|███ | 112/363 [00:08<00:05, 43.82it/s] Loading 0: 32%|███▏ | 117/363 [00:08<00:05, 43.33it/s] Loading 0: 34%|███▎ | 122/363 [00:08<00:05, 44.56it/s] Loading 0: 35%|███▌ | 128/363 [00:08<00:05, 42.73it/s] Loading 0: 37%|███▋ | 133/363 [00:09<00:05, 41.64it/s] Loading 0: 39%|███▊ | 140/363 [00:09<00:04, 46.54it/s] Loading 0: 40%|████ | 146/363 [00:09<00:04, 45.26it/s] Loading 0: 42%|████▏ | 151/363 [00:09<00:04, 43.75it/s] Loading 0: 43%|████▎ | 157/363 [00:09<00:06, 31.20it/s] Loading 0: 44%|████▍ | 161/363 [00:09<00:06, 30.06it/s] Loading 0: 46%|████▌ | 166/363 [00:10<00:05, 33.77it/s] Loading 0: 47%|████▋ | 170/363 [00:10<00:05, 34.32it/s] Loading 0: 48%|████▊ | 176/363 [00:10<00:04, 38.66it/s] Loading 0: 50%|█████ | 182/363 [00:10<00:04, 40.14it/s] Loading 0: 52%|█████▏ | 187/363 [00:10<00:04, 39.06it/s] Loading 0: 53%|█████▎ | 193/363 [00:10<00:03, 42.90it/s] Loading 0: 55%|█████▍ | 198/363 [00:10<00:03, 42.63it/s] Loading 0: 56%|█████▌ | 203/363 [00:10<00:03, 43.57it/s] Loading 0: 58%|█████▊ | 209/363 [00:10<00:03, 43.45it/s] Loading 0: 59%|█████▉ | 214/363 [00:11<00:03, 43.00it/s] Loading 0: 61%|██████ | 220/363 [00:11<00:03, 45.45it/s] Loading 0: 62%|██████▏ | 225/363 [00:11<00:03, 44.73it/s] Loading 0: 63%|██████▎ | 230/363 [00:11<00:02, 44.76it/s] Loading 0: 65%|██████▌ | 236/363 [00:11<00:02, 44.32it/s] Loading 0: 66%|██████▋ | 241/363 [00:11<00:02, 43.96it/s] Loading 0: 68%|██████▊ | 248/363 [00:11<00:02, 49.24it/s] Loading 0: 70%|██████▉ | 254/363 [00:11<00:02, 48.13it/s] Loading 0: 71%|███████▏ | 259/363 [00:12<00:03, 31.59it/s] Loading 0: 73%|███████▎ | 266/363 [00:12<00:02, 37.48it/s] Loading 0: 75%|███████▍ | 271/363 [00:12<00:02, 39.90it/s] Loading 0: 76%|███████▌ | 276/363 [00:12<00:02, 36.15it/s] Loading 0: 78%|███████▊ | 284/363 [00:12<00:01, 44.49it/s] Loading 0: 80%|███████▉ | 290/363 [00:12<00:01, 45.15it/s] Loading 0: 81%|████████▏ | 295/363 [00:13<00:01, 44.82it/s] Loading 0: 83%|████████▎ | 302/363 [00:13<00:01, 49.67it/s] Loading 0: 85%|████████▍ | 308/363 [00:13<00:01, 48.22it/s] Loading 0: 86%|████████▌ | 313/363 [00:13<00:01, 47.24it/s] Loading 0: 88%|████████▊ | 320/363 [00:13<00:00, 51.65it/s] Loading 0: 90%|████████▉ | 326/363 [00:13<00:00, 49.79it/s] Loading 0: 91%|█████████▏| 332/363 [00:13<00:00, 48.08it/s] Loading 0: 93%|█████████▎| 337/363 [00:13<00:00, 47.26it/s] Loading 0: 94%|█████████▍| 342/363 [00:13<00:00, 45.81it/s] Loading 0: 96%|█████████▌| 347/363 [00:14<00:00, 46.31it/s] Loading 0: 97%|█████████▋| 353/363 [00:14<00:00, 44.96it/s] Loading 0: 99%|█████████▊| 358/363 [00:14<00:00, 29.88it/s]
Job sao10k-mn-12b-lyra-v4a1-v3-mkmlizer completed after 95.12s with status: succeeded
Stopping job with name sao10k-mn-12b-lyra-v4a1-v3-mkmlizer
Pipeline stage MKMLizer completed in 96.61s
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-mn-12b-lyra-v4a1-v3
Waiting for inference service sao10k-mn-12b-lyra-v4a1-v3 to be ready
Failed to get response for submission mistralai-mixtral-8x7b_3473_v131: ('http://mistralai-mixtral-8x7b-3473-v131-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', '{"error":"ValueError : [TypeError(\\"\'numpy.int64\' object is not iterable\\"), TypeError(\'vars() argument must have __dict__ attribute\')]"}')
Failed to get response for submission blend_jidor_2024-08-22: ('http://mistralai-mixtral-8x7b-3473-v130-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', '{"error":"ValueError : [TypeError(\\"\'numpy.int64\' object is not iterable\\"), TypeError(\'vars() argument must have __dict__ attribute\')]"}')
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Inference service sao10k-mn-12b-lyra-v4a1-v3 ready after 150.62052536010742s
Pipeline stage MKMLDeployer completed in 151.87s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.704251289367676s
Received healthy response to inference request in 2.3711483478546143s
Received healthy response to inference request in 2.185274362564087s
Received healthy response to inference request in 2.82092022895813s
Received healthy response to inference request in 2.136176586151123s
5 requests
0 failed requests
5th percentile: 2.1459961414337156
10th percentile: 2.1558156967163087
20th percentile: 2.1754548072814943
30th percentile: 2.2224491596221925
40th percentile: 2.296798753738403
50th percentile: 2.3711483478546143
60th percentile: 2.504389524459839
70th percentile: 2.6376307010650635
80th percentile: 2.7275850772857666
90th percentile: 2.7742526531219482
95th percentile: 2.797586441040039
99th percentile: 2.8162534713745115
mean time: 2.443554162979126
Pipeline stage StressChecker completed in 12.97s
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 7.64s
Shutdown handler de-registered
sao10k-mn-12b-lyra-v4a1_v3 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 0.12s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.12s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service sao10k-mn-12b-lyra-v4a1-v3-profiler
Waiting for inference service sao10k-mn-12b-lyra-v4a1-v3-profiler to be ready
Inference service sao10k-mn-12b-lyra-v4a1-v3-profiler ready after 150.3465611934662s
Pipeline stage MKMLProfilerDeployer completed in 150.74s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/sao10k-mn-12b-lyra-v4a1-v3-profiler-predictor-00001-deployqksrs:/code/chaiverse_profiler_1725592288 --namespace tenant-chaiml-guanaco
kubectl exec -it sao10k-mn-12b-lyra-v4a1-v3-profiler-predictor-00001-deployqksrs --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1725592288 && 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 512 --output_tokens 64 --summary /code/chaiverse_profiler_1725592288/summary.json'
kubectl exec -it sao10k-mn-12b-lyra-v4a1-v3-profiler-predictor-00001-deployqksrs --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1725592288/summary.json'
Pipeline stage MKMLProfilerRunner completed in 952.76s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service sao10k-mn-12b-lyra-v4a1-v3-profiler is running
Tearing down inference service sao10k-mn-12b-lyra-v4a1-v3-profiler
Service sao10k-mn-12b-lyra-v4a1-v3-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 2.04s
Shutdown handler de-registered
sao10k-mn-12b-lyra-v4a1_v3 status is now inactive due to auto deactivation removed underperforming models
sao10k-mn-12b-lyra-v4a1_v3 status is now deployed due to admin request