submission_id: cycy233-l3-p-v1-c4_v1
developer_uid: shiroe40
best_of: 16
celo_rating: 1255.46
display_name: auto
family_friendly_score: 0.0
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}
generation_params: {'temperature': 1.0, 'top_p': 0.9, 'min_p': 0.05, 'top_k': 80, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['<|end_header_id|>', '<|eot_id|>'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64}
gpu_counts: {'NVIDIA RTX A5000': 1}
is_internal_developer: False
language_model: cycy233/L3-p-v1-c4
latencies: [{'batch_size': 1, 'throughput': 0.8999603240457869, 'latency_mean': 1.1110937345027923, 'latency_p50': 1.124469518661499, 'latency_p90': 1.2322253704071044}, {'batch_size': 4, 'throughput': 1.7980640296536456, 'latency_mean': 2.2167507660388948, 'latency_p50': 2.223546862602234, 'latency_p90': 2.481722187995911}, {'batch_size': 5, 'throughput': 1.8844079444717539, 'latency_mean': 2.6437380361557006, 'latency_p50': 2.622496008872986, 'latency_p90': 2.9454896688461303}, {'batch_size': 8, 'throughput': 2.000995568650382, 'latency_mean': 3.9738320899009705, 'latency_p50': 4.026252031326294, 'latency_p90': 4.500487518310547}, {'batch_size': 10, 'throughput': 2.029597153052243, 'latency_mean': 4.883272210359573, 'latency_p50': 4.881932258605957, 'latency_p90': 5.723604393005371}, {'batch_size': 12, 'throughput': 2.033177338032037, 'latency_mean': 5.816126766204834, 'latency_p50': 5.805496454238892, 'latency_p90': 6.736151576042175}, {'batch_size': 15, 'throughput': 2.043013670005117, 'latency_mean': 7.203268933296203, 'latency_p50': 7.303053259849548, 'latency_p90': 8.10137152671814}]
max_input_tokens: 512
max_output_tokens: 64
model_architecture: LlamaForCausalLM
model_group: cycy233/L3-p-v1-c4
model_name: auto
model_num_parameters: 8030261248.0
model_repo: cycy233/L3-p-v1-c4
model_size: 8B
num_battles: 11304
num_wins: 5815
ranking_group: single
status: torndown
submission_type: basic
throughput_3p7s: 2.0
timestamp: 2024-09-09T03:23:11+00:00
us_pacific_date: 2024-09-08
win_ratio: 0.5144196744515216
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 cycy233-l3-p-v1-c4-v1-mkmlizer
Waiting for job on cycy233-l3-p-v1-c4-v1-mkmlizer to finish
cycy233-l3-p-v1-c4-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
cycy233-l3-p-v1-c4-v1-mkmlizer: ║ _____ __ __ ║
cycy233-l3-p-v1-c4-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
cycy233-l3-p-v1-c4-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
cycy233-l3-p-v1-c4-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
cycy233-l3-p-v1-c4-v1-mkmlizer: ║ /___/ ║
cycy233-l3-p-v1-c4-v1-mkmlizer: ║ ║
cycy233-l3-p-v1-c4-v1-mkmlizer: ║ Version: 0.10.1 ║
cycy233-l3-p-v1-c4-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
cycy233-l3-p-v1-c4-v1-mkmlizer: ║ https://mk1.ai ║
cycy233-l3-p-v1-c4-v1-mkmlizer: ║ ║
cycy233-l3-p-v1-c4-v1-mkmlizer: ║ The license key for the current software has been verified as ║
cycy233-l3-p-v1-c4-v1-mkmlizer: ║ belonging to: ║
cycy233-l3-p-v1-c4-v1-mkmlizer: ║ ║
cycy233-l3-p-v1-c4-v1-mkmlizer: ║ Chai Research Corp. ║
cycy233-l3-p-v1-c4-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
cycy233-l3-p-v1-c4-v1-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
cycy233-l3-p-v1-c4-v1-mkmlizer: ║ ║
cycy233-l3-p-v1-c4-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
cycy233-l3-p-v1-c4-v1-mkmlizer: Downloaded to shared memory in 36.029s
cycy233-l3-p-v1-c4-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp9znn7s8m, device:0
cycy233-l3-p-v1-c4-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
cycy233-l3-p-v1-c4-v1-mkmlizer: quantized model in 25.801s
cycy233-l3-p-v1-c4-v1-mkmlizer: Processed model cycy233/L3-p-v1-c4 in 61.830s
cycy233-l3-p-v1-c4-v1-mkmlizer: creating bucket guanaco-mkml-models
cycy233-l3-p-v1-c4-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
cycy233-l3-p-v1-c4-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/cycy233-l3-p-v1-c4-v1
cycy233-l3-p-v1-c4-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/cycy233-l3-p-v1-c4-v1/config.json
cycy233-l3-p-v1-c4-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/cycy233-l3-p-v1-c4-v1/special_tokens_map.json
cycy233-l3-p-v1-c4-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/cycy233-l3-p-v1-c4-v1/tokenizer_config.json
cycy233-l3-p-v1-c4-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/cycy233-l3-p-v1-c4-v1/flywheel_model.0.safetensors
cycy233-l3-p-v1-c4-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 1%|▏ | 4/291 [00:00<00:07, 39.29it/s] Loading 0: 5%|▌ | 16/291 [00:00<00:04, 68.30it/s] Loading 0: 11%|█ | 31/291 [00:00<00:03, 86.55it/s] Loading 0: 15%|█▍ | 43/291 [00:00<00:02, 87.31it/s] Loading 0: 20%|█▉ | 58/291 [00:00<00:02, 94.03it/s] Loading 0: 23%|██▎ | 68/291 [00:00<00:02, 95.38it/s] Loading 0: 27%|██▋ | 79/291 [00:00<00:02, 87.45it/s] Loading 0: 30%|███ | 88/291 [00:02<00:08, 23.53it/s] Loading 0: 35%|███▌ | 103/291 [00:02<00:05, 33.58it/s] Loading 0: 39%|███▉ | 114/291 [00:02<00:04, 41.94it/s] Loading 0: 43%|████▎ | 124/291 [00:02<00:03, 46.85it/s] Loading 0: 48%|████▊ | 139/291 [00:02<00:02, 58.58it/s] Loading 0: 52%|█████▏ | 151/291 [00:02<00:02, 64.40it/s] Loading 0: 55%|█████▍ | 160/291 [00:02<00:01, 68.65it/s] Loading 0: 60%|██████ | 175/291 [00:03<00:01, 78.23it/s] Loading 0: 64%|██████▍ | 187/291 [00:04<00:03, 26.83it/s] Loading 0: 67%|██████▋ | 196/291 [00:04<00:02, 32.17it/s] Loading 0: 70%|███████ | 205/291 [00:04<00:02, 38.47it/s] Loading 0: 76%|███████▌ | 220/291 [00:04<00:01, 50.76it/s] Loading 0: 79%|███████▊ | 229/291 [00:04<00:01, 56.54it/s] Loading 0: 82%|████████▏ | 238/291 [00:04<00:00, 61.89it/s] Loading 0: 86%|████████▌ | 250/291 [00:04<00:00, 67.74it/s] Loading 0: 91%|█████████ | 265/291 [00:04<00:00, 78.39it/s] Loading 0: 95%|█████████▌| 277/291 [00:05<00:00, 79.77it/s] Loading 0: 99%|█████████▊| 287/291 [00:05<00:00, 48.44it/s]
Job cycy233-l3-p-v1-c4-v1-mkmlizer completed after 84.04s with status: succeeded
Stopping job with name cycy233-l3-p-v1-c4-v1-mkmlizer
Pipeline stage MKMLizer completed in 84.82s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.80s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service cycy233-l3-p-v1-c4-v1
Waiting for inference service cycy233-l3-p-v1-c4-v1 to be ready
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Inference service cycy233-l3-p-v1-c4-v1 ready after 151.79049038887024s
Pipeline stage MKMLDeployer completed in 152.26s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.0386667251586914s
Received healthy response to inference request in 2.1806859970092773s
Received healthy response to inference request in 1.9352214336395264s
Received healthy response to inference request in 2.370138645172119s
Received healthy response to inference request in 2.699394702911377s
5 requests
0 failed requests
5th percentile: 1.9559104919433594
10th percentile: 1.9765995502471925
20th percentile: 2.0179776668548586
30th percentile: 2.0670705795288087
40th percentile: 2.123878288269043
50th percentile: 2.1806859970092773
60th percentile: 2.2564670562744142
70th percentile: 2.3322481155395507
80th percentile: 2.4359898567199707
90th percentile: 2.567692279815674
95th percentile: 2.6335434913635254
99th percentile: 2.6862244606018066
mean time: 2.2448215007781984
Pipeline stage StressChecker completed in 12.06s
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 4.55s
Shutdown handler de-registered
cycy233-l3-p-v1-c4_v1 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 cycy233-l3-p-v1-c4-v1-profiler
Waiting for inference service cycy233-l3-p-v1-c4-v1-profiler to be ready
Inference service cycy233-l3-p-v1-c4-v1-profiler ready after 150.35387563705444s
Pipeline stage MKMLProfilerDeployer completed in 151.56s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/cycy233-l3-p-v1-c4-v1-profiler-predictor-00001-deployment-lfxvj:/code/chaiverse_profiler_1725852635 --namespace tenant-chaiml-guanaco
kubectl exec -it cycy233-l3-p-v1-c4-v1-profiler-predictor-00001-deployment-lfxvj --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1725852635 && python profiles.py profile --best_of_n 16 --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_1725852635/summary.json'
kubectl exec -it cycy233-l3-p-v1-c4-v1-profiler-predictor-00001-deployment-lfxvj --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1725852635/summary.json'
Pipeline stage MKMLProfilerRunner completed in 839.71s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service cycy233-l3-p-v1-c4-v1-profiler is running
Tearing down inference service cycy233-l3-p-v1-c4-v1-profiler
Service cycy233-l3-p-v1-c4-v1-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.72s
Shutdown handler de-registered
cycy233-l3-p-v1-c4_v1 status is now inactive due to auto deactivation removed underperforming models
cycy233-l3-p-v1-c4_v1 status is now torndown due to DeploymentManager action
run pipeline %s
Running pipeline stage MKMLModelDeleter
run pipeline stage %s
Pipeline stage MKMLModelDeleter completed in 3.51s
Running pipeline stage MKMLDeleter
Pipeline stage MKMLDeleter completed in 3.26s
admin requested tearing down of cycy233-l3-p-v3-c2_v1
Shutdown handler not registered because Python interpreter is not running in the main thread
Pipeline stage %s skipped, reason=%s
cycy233-l3-p-v1-c4_v1 status is now torndown due to DeploymentManager action
Pipeline stage MKMLModelDeleter completed in 3.42s
run pipeline stage %s
Pipeline stage %s skipped, reason=%s
Running pipeline stage MKMLModelDeleter
Running pipeline stage MKMLModelDeleter
chaiml-virgo-edit-v1-1e5_v3 status is now torndown due to DeploymentManager action
Pipeline stage %s skipped, reason=%s
Pipeline stage MKMLDeleter completed in 4.89s
Shutdown handler de-registered
run pipeline stage %s
Pipeline stage MKMLModelDeleter completed in 4.56s
run pipeline %s
Shutdown handler not registered because Python interpreter is not running in the main thread
admin requested tearing down of cycy233-l3-p-v1-c4_v1
Pipeline stage %s skipped, reason=%s
Running pipeline stage MKMLModelDeleter
Pipeline stage MKMLDeleter completed in 4.70s
run pipeline stage %s
chaiml-virgo-edit-v1-1e5_v4 status is now torndown due to DeploymentManager action
Pipeline stage %s skipped, reason=%s
Running pipeline stage MKMLDeleter
Shutdown handler de-registered
run pipeline stage %s
run pipeline %s
Shutdown handler not registered because Python interpreter is not running in the main thread
Pipeline stage MKMLModelDeleter completed in 5.47s
Pipeline stage %s skipped, reason=%s
run pipeline stage %s
Running pipeline stage MKMLModelDeleter
Pipeline stage MKMLDeleter completed in 5.06s
Pipeline stage %s skipped, reason=%s
chaiml-virgo-edit-v1-1e5_v5 status is now torndown due to DeploymentManager action
Running pipeline stage MKMLDeleter
run pipeline stage %s
Shutdown handler not registered because Python interpreter is not running in the main thread
run pipeline %s
admin requested tearing down of cycy233-l3-p-v1-c4_v3
Shutdown handler de-registered
Pipeline stage MKMLModelDeleter completed in 5.76s
Running pipeline stage MKMLModelDeleter
Pipeline stage %s skipped, reason=%s
Pipeline stage MKMLDeleter completed in 5.67s
Pipeline stage %s skipped, reason=%s
Running pipeline stage MKMLDeleter
run pipeline stage %s
run pipeline %s
Shutdown handler not registered because Python interpreter is not running in the main thread
chaiml-virgo-edit-v1-1e5_v6 status is now torndown due to DeploymentManager action
Shutdown handler de-registered
admin requested tearing down of cycy233-l3-p-v1-c4_v4
Pipeline stage %s skipped, reason=%s
Pipeline stage MKMLModelDeleter completed in 5.86s
Running pipeline stage MKMLModelDeleter
run pipeline stage %s
Running pipeline stage MKMLDeleter
run pipeline stage %s
run pipeline %s
Shutdown handler de-registered
Shutdown handler not registered because Python interpreter is not running in the main thread
Pipeline stage MKMLModelDeleter completed in 5.64s
chaiml-virgo-top-bots3x100-hr_v1 status is now torndown due to DeploymentManager action
admin requested tearing down of cycy233-l3-p-v2-c1_v1
Pipeline stage %s skipped, reason=%s
Running pipeline stage MKMLModelDeleter
run pipeline stage %s
Pipeline stage MKMLDeleter completed in 5.94s
Pipeline stage %s skipped, reason=%s
Running pipeline stage MKMLDeleter
run pipeline stage %s
chaiml-virgo-top-bots5x100-hr_v1 status is now torndown due to DeploymentManager action
run pipeline %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Shutdown handler de-registered
Shutdown handler not registered because Python interpreter is not running in the main thread
Pipeline stage MKMLModelDeleter completed in 7.49s
admin requested tearing down of cycy233-l3-p-v2-c2_v1
Pipeline stage %s skipped, reason=%s
Running pipeline stage MKMLModelDeleter
run pipeline stage %s
Pipeline stage MKMLDeleter completed in 7.38s
Pipeline stage %s skipped, reason=%s
Running pipeline stage MKMLDeleter
run pipeline %s
Shutdown handler de-registered
run pipeline stage %s
chaiml-virgo-top-bots6x50-hr_v1 status is now torndown due to DeploymentManager action
Shutdown handler not registered because Python interpreter is not running in the main thread
admin requested tearing down of cycy233-l3-p-v2-c3_v2
Pipeline stage MKMLModelDeleter completed in 6.68s
Pipeline stage %s skipped, reason=%s
Pipeline stage MKMLDeleter completed in 6.83s
Pipeline stage %s skipped, reason=%s
run pipeline stage %s
cycy233-l3-e-v0-c4_v5 status is now torndown due to DeploymentManager action
Running pipeline stage MKMLDeleter
run pipeline %s
Shutdown handler not registered because Python interpreter is not running in the main thread
Shutdown handler de-registered
admin requested tearing down of cycy233-l3-p-v2-c4_v2
Pipeline stage MKMLModelDeleter completed in 5.03s
Pipeline stage %s skipped, reason=%s
run pipeline stage %s
Running pipeline stage MKMLModelDeleter
Pipeline stage MKMLDeleter completed in 5.06s
Running pipeline stage MKMLDeleter
run pipeline stage %s
run pipeline %s
cycy233-l3-p-v1-c1_v1 status is now torndown due to DeploymentManager action
Shutdown handler not registered because Python interpreter is not running in the main thread
Shutdown handler de-registered
admin requested tearing down of cycy233-l3-p-v2_v1
Pipeline stage MKMLModelDeleter completed in 5.53s
Running pipeline stage MKMLModelDeleter
Pipeline stage %s skipped, reason=%s
run pipeline stage %s
Pipeline stage %s skipped, reason=%s
Pipeline stage MKMLDeleter completed in 5.44s
Running pipeline stage MKMLDeleter
run pipeline stage %s
run pipeline %s
cycy233-l3-p-v1-c2_v1 status is now torndown due to DeploymentManager action
Shutdown handler not registered because Python interpreter is not running in the main thread
Shutdown handler de-registered
Pipeline stage %s skipped, reason=%s
admin requested tearing down of cycy233-l3-p-v3-c1_v1
Running pipeline stage MKMLModelDeleter
Pipeline stage MKMLModelDeleter completed in 6.33s
Pipeline stage MKMLDeleter completed in 6.78s
run pipeline stage %s
Pipeline stage %s skipped, reason=%s
Running pipeline stage MKMLDeleter
run pipeline stage %s
Pipeline stage MKMLModelDeleter completed in 5.18s
Shutdown handler not registered because Python interpreter is not running in the main thread
cycy233-l3-p-v1-c3_v1 status is now torndown due to DeploymentManager action
Pipeline stage %s skipped, reason=%s
admin requested tearing down of cycy233-l3-p-v3-c2_v1
Shutdown handler de-registered
run pipeline stage %s
Running pipeline stage MKMLModelDeleter
Pipeline stage MKMLDeleter completed in 6.29s
Pipeline stage %s skipped, reason=%s
Shutdown handler de-registered
Running pipeline stage MKMLDeleter
run pipeline stage %s
run pipeline %s
Pipeline stage MKMLModelDeleter completed in 5.65s
Shutdown handler not registered because Python interpreter is not running in the main thread
admin requested tearing down of cycy233-l3-p-v3-c3_v1
Running pipeline stage MKMLModelDeleter
Pipeline stage MKMLDeleter completed in 6.29s
Pipeline stage %s skipped, reason=%s
Shutdown handler de-registered
run pipeline stage %s
run pipeline %s
Pipeline stage MKMLModelDeleter completed in 5.65s
Shutdown handler not registered because Python interpreter is not running in the main thread
admin requested tearing down of cycy233-l3-p-v3-c3_v1
Running pipeline stage MKMLModelDeleter
Pipeline stage %s skipped, reason=%s
run pipeline stage %s
Pipeline stage MKMLDeleter completed in 5.66s
Running pipeline stage MKMLDeleter