submission_id: rica40325-sftdpo-9-15-2_v1
developer_uid: rica40325
best_of: 16
celo_rating: 1263.4
display_name: rica40325-sftdpo-9-15-2_v1
family_friendly_score: 0.0
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}
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': 512, 'best_of': 16, 'max_output_tokens': 64}
gpu_counts: {'NVIDIA RTX A5000': 1}
is_internal_developer: False
language_model: rica40325/sftdpo_9_15_2
latencies: [{'batch_size': 1, 'throughput': 0.9093487264637758, 'latency_mean': 1.0995919549465178, 'latency_p50': 1.0861119031906128, 'latency_p90': 1.2332220792770385}, {'batch_size': 4, 'throughput': 1.7766242577362699, 'latency_mean': 2.2416108548641205, 'latency_p50': 2.25016713142395, 'latency_p90': 2.493992066383362}, {'batch_size': 5, 'throughput': 1.8654010593220165, 'latency_mean': 2.66528434753418, 'latency_p50': 2.644783854484558, 'latency_p90': 3.012795662879944}, {'batch_size': 8, 'throughput': 1.9767159921944513, 'latency_mean': 4.015633993148803, 'latency_p50': 4.017249822616577, 'latency_p90': 4.558178544044495}, {'batch_size': 10, 'throughput': 1.994554291344638, 'latency_mean': 4.96679762840271, 'latency_p50': 4.9912614822387695, 'latency_p90': 5.691798686981201}, {'batch_size': 12, 'throughput': 2.0139353118189423, 'latency_mean': 5.870930643081665, 'latency_p50': 5.967160701751709, 'latency_p90': 6.770223402976989}, {'batch_size': 15, 'throughput': 2.011821081601813, 'latency_mean': 7.3038315367698665, 'latency_p50': 7.419245481491089, 'latency_p90': 8.130823349952697}]
max_input_tokens: 512
max_output_tokens: 64
model_architecture: LlamaForCausalLM
model_group: rica40325/sftdpo_9_15_2
model_name: rica40325-sftdpo-9-15-2_v1
model_num_parameters: 8030261248.0
model_repo: rica40325/sftdpo_9_15_2
model_size: 8B
num_battles: 10932
num_wins: 5632
ranking_group: single
status: torndown
submission_type: basic
throughput_3p7s: 1.97
timestamp: 2024-09-17T04:08:11+00:00
us_pacific_date: 2024-09-16
win_ratio: 0.5151847786315404
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 rica40325-sftdpo-9-15-2-v1-mkmlizer
Waiting for job on rica40325-sftdpo-9-15-2-v1-mkmlizer to finish
rica40325-sftdpo-9-15-2-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rica40325-sftdpo-9-15-2-v1-mkmlizer: ║ _____ __ __ ║
rica40325-sftdpo-9-15-2-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
rica40325-sftdpo-9-15-2-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
rica40325-sftdpo-9-15-2-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
rica40325-sftdpo-9-15-2-v1-mkmlizer: ║ /___/ ║
rica40325-sftdpo-9-15-2-v1-mkmlizer: ║ ║
rica40325-sftdpo-9-15-2-v1-mkmlizer: ║ Version: 0.10.1 ║
rica40325-sftdpo-9-15-2-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
rica40325-sftdpo-9-15-2-v1-mkmlizer: ║ https://mk1.ai ║
rica40325-sftdpo-9-15-2-v1-mkmlizer: ║ ║
rica40325-sftdpo-9-15-2-v1-mkmlizer: ║ The license key for the current software has been verified as ║
rica40325-sftdpo-9-15-2-v1-mkmlizer: ║ belonging to: ║
rica40325-sftdpo-9-15-2-v1-mkmlizer: ║ ║
rica40325-sftdpo-9-15-2-v1-mkmlizer: ║ Chai Research Corp. ║
rica40325-sftdpo-9-15-2-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
rica40325-sftdpo-9-15-2-v1-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
rica40325-sftdpo-9-15-2-v1-mkmlizer: ║ ║
rica40325-sftdpo-9-15-2-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
rica40325-sftdpo-9-15-2-v1-mkmlizer: Downloaded to shared memory in 34.782s
rica40325-sftdpo-9-15-2-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp50b_6ff_, device:0
rica40325-sftdpo-9-15-2-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
rica40325-sftdpo-9-15-2-v1-mkmlizer: quantized model in 25.788s
rica40325-sftdpo-9-15-2-v1-mkmlizer: Processed model rica40325/sftdpo_9_15_2 in 60.571s
rica40325-sftdpo-9-15-2-v1-mkmlizer: creating bucket guanaco-mkml-models
rica40325-sftdpo-9-15-2-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rica40325-sftdpo-9-15-2-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rica40325-sftdpo-9-15-2-v1
rica40325-sftdpo-9-15-2-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rica40325-sftdpo-9-15-2-v1/config.json
rica40325-sftdpo-9-15-2-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rica40325-sftdpo-9-15-2-v1/special_tokens_map.json
rica40325-sftdpo-9-15-2-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rica40325-sftdpo-9-15-2-v1/tokenizer_config.json
rica40325-sftdpo-9-15-2-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rica40325-sftdpo-9-15-2-v1/flywheel_model.0.safetensors
rica40325-sftdpo-9-15-2-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 1%|▏ | 4/291 [00:00<00:08, 35.78it/s] Loading 0: 5%|▌ | 16/291 [00:00<00:04, 66.01it/s] Loading 0: 11%|█ | 31/291 [00:00<00:02, 88.95it/s] Loading 0: 14%|█▎ | 40/291 [00:00<00:03, 80.90it/s] Loading 0: 18%|█▊ | 52/291 [00:00<00:02, 81.85it/s] Loading 0: 21%|██ | 61/291 [00:00<00:02, 80.76it/s] Loading 0: 26%|██▌ | 76/291 [00:00<00:02, 87.08it/s] Loading 0: 29%|██▉ | 85/291 [00:02<00:09, 22.80it/s] Loading 0: 33%|███▎ | 97/291 [00:02<00:06, 30.05it/s] Loading 0: 36%|███▋ | 106/291 [00:02<00:05, 35.28it/s] Loading 0: 42%|████▏ | 121/291 [00:02<00:03, 47.38it/s] Loading 0: 45%|████▍ | 130/291 [00:02<00:03, 50.99it/s] Loading 0: 48%|████▊ | 139/291 [00:02<00:02, 55.42it/s] Loading 0: 51%|█████ | 148/291 [00:02<00:02, 60.23it/s] Loading 0: 54%|█████▍ | 157/291 [00:03<00:02, 66.06it/s] Loading 0: 58%|█████▊ | 169/291 [00:03<00:01, 72.77it/s] Loading 0: 63%|██████▎ | 184/291 [00:03<00:01, 83.58it/s] Loading 0: 67%|██████▋ | 194/291 [00:04<00:03, 25.74it/s] Loading 0: 70%|███████ | 205/291 [00:04<00:02, 31.58it/s] Loading 0: 76%|███████▌ | 220/291 [00:04<00:01, 42.73it/s] Loading 0: 79%|███████▊ | 229/291 [00:04<00:01, 47.35it/s] Loading 0: 83%|████████▎ | 241/291 [00:05<00:00, 54.25it/s] Loading 0: 86%|████████▌ | 250/291 [00:05<00:00, 57.37it/s] Loading 0: 89%|████████▉ | 259/291 [00:05<00:00, 62.78it/s] Loading 0: 92%|█████████▏| 267/291 [00:05<00:00, 63.21it/s] Loading 0: 95%|█████████▌| 277/291 [00:05<00:00, 67.21it/s] Loading 0: 99%|█████████▊| 287/291 [00:05<00:00, 44.20it/s]
Job rica40325-sftdpo-9-15-2-v1-mkmlizer completed after 83.35s with status: succeeded
Stopping job with name rica40325-sftdpo-9-15-2-v1-mkmlizer
Pipeline stage MKMLizer completed in 84.48s
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 rica40325-sftdpo-9-15-2-v1
Waiting for inference service rica40325-sftdpo-9-15-2-v1 to be ready
Inference service rica40325-sftdpo-9-15-2-v1 ready after 170.45654106140137s
Pipeline stage MKMLDeployer completed in 170.82s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.3343989849090576s
Received healthy response to inference request in 2.175405979156494s
Received healthy response to inference request in 1.9316258430480957s
Received healthy response to inference request in 1.7514429092407227s
Received healthy response to inference request in 2.0208499431610107s
5 requests
0 failed requests
5th percentile: 1.7874794960021974
10th percentile: 1.8235160827636718
20th percentile: 1.895589256286621
30th percentile: 1.9494706630706786
40th percentile: 1.9851603031158447
50th percentile: 2.0208499431610107
60th percentile: 2.082672357559204
70th percentile: 2.1444947719573975
80th percentile: 2.207204580307007
90th percentile: 2.2708017826080322
95th percentile: 2.302600383758545
99th percentile: 2.328039264678955
mean time: 2.0427447319030763
Pipeline stage StressChecker completed in 10.80s
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 7.97s
Shutdown handler de-registered
rica40325-sftdpo-9-15-2_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 0.12s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service rica40325-sftdpo-9-15-2-v1-profiler
Waiting for inference service rica40325-sftdpo-9-15-2-v1-profiler to be ready
Inference service rica40325-sftdpo-9-15-2-v1-profiler ready after 170.3852825164795s
Pipeline stage MKMLProfilerDeployer completed in 170.73s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/rica40325-sftdpo-9-15-2-v1-profiler-predictor-00001-deploy9jqfr:/code/chaiverse_profiler_1726546587 --namespace tenant-chaiml-guanaco
kubectl exec -it rica40325-sftdpo-9-15-2-v1-profiler-predictor-00001-deploy9jqfr --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1726546587 && 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_1726546587/summary.json'
kubectl exec -it rica40325-sftdpo-9-15-2-v1-profiler-predictor-00001-deploy9jqfr --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1726546587/summary.json'
Pipeline stage MKMLProfilerRunner completed in 846.88s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service rica40325-sftdpo-9-15-2-v1-profiler is running
Tearing down inference service rica40325-sftdpo-9-15-2-v1-profiler
Service rica40325-sftdpo-9-15-2-v1-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.96s
Shutdown handler de-registered
rica40325-sftdpo-9-15-2_v1 status is now inactive due to auto deactivation removed underperforming models
rica40325-sftdpo-9-15-2_v1 status is now torndown due to DeploymentManager action