submission_id: rica40325-alignment-first_v2
developer_uid: rica40325
alignment_samples: 14761
alignment_score: 0.1456918353903548
best_of: 16
celo_rating: 1258.19
display_name: rica40325-alignment-first_v1
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': 1.15, 'top_p': 0.95, 'min_p': 0.05, 'top_k': 80, '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/alignment-first
latencies: [{'batch_size': 1, 'throughput': 0.9050312835083357, 'latency_mean': 1.1048712873458861, 'latency_p50': 1.1013438701629639, 'latency_p90': 1.234945034980774}, {'batch_size': 4, 'throughput': 1.7704281659281194, 'latency_mean': 2.250933474302292, 'latency_p50': 2.243991732597351, 'latency_p90': 2.5389283657073975}, {'batch_size': 5, 'throughput': 1.8521440997204772, 'latency_mean': 2.6819391441345215, 'latency_p50': 2.678904891014099, 'latency_p90': 3.0057841539382935}, {'batch_size': 8, 'throughput': 1.9518920928551176, 'latency_mean': 4.062639164924621, 'latency_p50': 4.081029415130615, 'latency_p90': 4.600485777854919}, {'batch_size': 10, 'throughput': 1.996497319844254, 'latency_mean': 4.963616088628769, 'latency_p50': 4.901096701622009, 'latency_p90': 5.698187160491943}, {'batch_size': 12, 'throughput': 2.0082952526855262, 'latency_mean': 5.900218398571014, 'latency_p50': 5.964988350868225, 'latency_p90': 6.704848027229309}, {'batch_size': 15, 'throughput': 2.0024495430150773, 'latency_mean': 7.3471551716327665, 'latency_p50': 7.497701406478882, 'latency_p90': 8.24213969707489}]
max_input_tokens: 512
max_output_tokens: 64
model_architecture: LlamaForCausalLM
model_group: rica40325/alignment-firs
model_name: rica40325-alignment-first_v1
model_num_parameters: 8030261248.0
model_repo: rica40325/alignment-first
model_size: 8B
num_battles: 14761
num_wins: 7734
propriety_score: 0.7509727626459144
propriety_total_count: 1285.0
ranking_group: single
status: inactive
submission_type: basic
throughput_3p7s: 1.95
timestamp: 2024-09-08T09:07:36+00:00
us_pacific_date: 2024-09-08
win_ratio: 0.5239482419890251
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-alignment-first-v2-mkmlizer
Waiting for job on rica40325-alignment-first-v2-mkmlizer to finish
rica40325-alignment-first-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rica40325-alignment-first-v2-mkmlizer: ║ _____ __ __ ║
rica40325-alignment-first-v2-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
rica40325-alignment-first-v2-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
rica40325-alignment-first-v2-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
rica40325-alignment-first-v2-mkmlizer: ║ /___/ ║
rica40325-alignment-first-v2-mkmlizer: ║ ║
rica40325-alignment-first-v2-mkmlizer: ║ Version: 0.10.1 ║
rica40325-alignment-first-v2-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
rica40325-alignment-first-v2-mkmlizer: ║ https://mk1.ai ║
rica40325-alignment-first-v2-mkmlizer: ║ ║
rica40325-alignment-first-v2-mkmlizer: ║ The license key for the current software has been verified as ║
rica40325-alignment-first-v2-mkmlizer: ║ belonging to: ║
rica40325-alignment-first-v2-mkmlizer: ║ ║
rica40325-alignment-first-v2-mkmlizer: ║ Chai Research Corp. ║
rica40325-alignment-first-v2-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
rica40325-alignment-first-v2-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
rica40325-alignment-first-v2-mkmlizer: ║ ║
rica40325-alignment-first-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
rica40325-alignment-first-v2-mkmlizer: Downloaded to shared memory in 22.757s
rica40325-alignment-first-v2-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmps0zz7b46, device:0
rica40325-alignment-first-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
rica40325-alignment-first-v2-mkmlizer: quantized model in 26.621s
rica40325-alignment-first-v2-mkmlizer: Processed model rica40325/alignment-first in 49.378s
rica40325-alignment-first-v2-mkmlizer: creating bucket guanaco-mkml-models
rica40325-alignment-first-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rica40325-alignment-first-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rica40325-alignment-first-v2
rica40325-alignment-first-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rica40325-alignment-first-v2/config.json
rica40325-alignment-first-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rica40325-alignment-first-v2/special_tokens_map.json
rica40325-alignment-first-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rica40325-alignment-first-v2/tokenizer_config.json
rica40325-alignment-first-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rica40325-alignment-first-v2/tokenizer.json
rica40325-alignment-first-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rica40325-alignment-first-v2/flywheel_model.0.safetensors
rica40325-alignment-first-v2-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 2%|▏ | 6/291 [00:00<00:04, 59.70it/s] Loading 0: 4%|▍ | 13/291 [00:00<00:04, 62.74it/s] Loading 0: 8%|▊ | 22/291 [00:00<00:03, 70.76it/s] Loading 0: 11%|█ | 31/291 [00:00<00:03, 70.62it/s] Loading 0: 14%|█▎ | 40/291 [00:00<00:03, 76.43it/s] Loading 0: 17%|█▋ | 49/291 [00:00<00:03, 79.68it/s] Loading 0: 21%|██ | 60/291 [00:00<00:02, 88.90it/s] Loading 0: 24%|██▍ | 70/291 [00:00<00:02, 80.74it/s] Loading 0: 27%|██▋ | 79/291 [00:01<00:02, 83.23it/s] Loading 0: 30%|███ | 88/291 [00:02<00:10, 20.28it/s] Loading 0: 33%|███▎ | 97/291 [00:02<00:07, 25.66it/s] Loading 0: 36%|███▋ | 106/291 [00:02<00:05, 32.36it/s] Loading 0: 40%|███▉ | 115/291 [00:02<00:04, 39.84it/s] Loading 0: 43%|████▎ | 124/291 [00:02<00:03, 45.61it/s] Loading 0: 46%|████▌ | 133/291 [00:02<00:03, 51.15it/s] Loading 0: 49%|████▉ | 142/291 [00:02<00:02, 58.02it/s] Loading 0: 52%|█████▏ | 151/291 [00:03<00:02, 64.66it/s] Loading 0: 55%|█████▍ | 160/291 [00:03<00:01, 68.19it/s] Loading 0: 58%|█████▊ | 169/291 [00:03<00:01, 72.34it/s] Loading 0: 61%|██████ | 178/291 [00:03<00:01, 72.09it/s] Loading 0: 64%|██████▍ | 187/291 [00:04<00:05, 20.35it/s] Loading 0: 67%|██████▋ | 196/291 [00:04<00:03, 25.70it/s] Loading 0: 70%|███████ | 205/291 [00:04<00:02, 32.40it/s] Loading 0: 74%|███████▎ | 214/291 [00:04<00:02, 38.49it/s] Loading 0: 77%|███████▋ | 223/291 [00:05<00:01, 45.32it/s] Loading 0: 80%|███████▉ | 232/291 [00:05<00:01, 52.72it/s] Loading 0: 85%|████████▍ | 247/291 [00:05<00:00, 66.20it/s] Loading 0: 88%|████████▊ | 256/291 [00:05<00:00, 70.85it/s] Loading 0: 91%|█████████▏| 266/291 [00:05<00:00, 77.42it/s] Loading 0: 95%|█████████▍| 275/291 [00:05<00:00, 78.07it/s] Loading 0: 98%|█████████▊| 284/291 [00:05<00:00, 71.98it/s]
Job rica40325-alignment-first-v2-mkmlizer completed after 77.2s with status: succeeded
Stopping job with name rica40325-alignment-first-v2-mkmlizer
Pipeline stage MKMLizer completed in 78.47s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.08s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service rica40325-alignment-first-v2
Waiting for inference service rica40325-alignment-first-v2 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
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 rica40325-alignment-first-v2 ready after 150.64192581176758s
Pipeline stage MKMLDeployer completed in 150.97s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.9329168796539307s
Received healthy response to inference request in 1.9334797859191895s
Received healthy response to inference request in 1.563814640045166s
Received healthy response to inference request in 1.6662952899932861s
Received healthy response to inference request in 1.953803300857544s
5 requests
0 failed requests
5th percentile: 1.58431077003479
10th percentile: 1.604806900024414
20th percentile: 1.645799160003662
30th percentile: 1.7196196079254151
40th percentile: 1.826268243789673
50th percentile: 1.9329168796539307
60th percentile: 1.9331420421600343
70th percentile: 1.9333672046661377
80th percentile: 1.9375444889068603
90th percentile: 1.945673894882202
95th percentile: 1.9497385978698731
99th percentile: 1.9529903602600098
mean time: 1.8100619792938233
Pipeline stage StressChecker completed in 10.53s
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 8.25s
Shutdown handler de-registered
rica40325-alignment-first_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 0.13s
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-alignment-first-v2-profiler
Waiting for inference service rica40325-alignment-first-v2-profiler to be ready
Inference service rica40325-alignment-first-v2-profiler ready after 150.36093997955322s
Pipeline stage MKMLProfilerDeployer completed in 150.78s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/rica40325-alignment-8b2a44f0a04eb521a8a7b9164427ce51-deplop49r5:/code/chaiverse_profiler_1725786900 --namespace tenant-chaiml-guanaco
kubectl exec -it rica40325-alignment-8b2a44f0a04eb521a8a7b9164427ce51-deplop49r5 --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1725786900 && 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_1725786900/summary.json'
kubectl exec -it rica40325-alignment-8b2a44f0a04eb521a8a7b9164427ce51-deplop49r5 --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1725786900/summary.json'
Pipeline stage MKMLProfilerRunner completed in 851.98s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service rica40325-alignment-first-v2-profiler is running
Tearing down inference service rica40325-alignment-first-v2-profiler
Service rica40325-alignment-first-v2-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.75s
Shutdown handler de-registered
rica40325-alignment-first_v2 status is now inactive due to auto deactivation removed underperforming models