submission_id: rica40325-feedback-2000_v2
developer_uid: rica40325
best_of: 16
celo_rating: 1253.19
display_name: rica40325-feedback-2000_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/feedback_2000
latencies: [{'batch_size': 1, 'throughput': 0.916923882174189, 'latency_mean': 1.090544570684433, 'latency_p50': 1.0868268013000488, 'latency_p90': 1.2288292169570922}, {'batch_size': 4, 'throughput': 1.8245076046666364, 'latency_mean': 2.1866460371017458, 'latency_p50': 2.2015960216522217, 'latency_p90': 2.45121488571167}, {'batch_size': 5, 'throughput': 1.9349828750972582, 'latency_mean': 2.5704763984680175, 'latency_p50': 2.59058678150177, 'latency_p90': 2.8708750724792482}, {'batch_size': 8, 'throughput': 2.025677679829065, 'latency_mean': 3.9217180931568145, 'latency_p50': 3.906338691711426, 'latency_p90': 4.409326791763306}, {'batch_size': 10, 'throughput': 2.041926266391794, 'latency_mean': 4.84262819647789, 'latency_p50': 4.818591952323914, 'latency_p90': 5.543906664848327}, {'batch_size': 12, 'throughput': 2.0500895547797295, 'latency_mean': 5.781850374937058, 'latency_p50': 5.838121175765991, 'latency_p90': 6.530162763595581}, {'batch_size': 15, 'throughput': 2.056996577859914, 'latency_mean': 7.139580699205399, 'latency_p50': 7.220590949058533, 'latency_p90': 7.929987621307373}]
max_input_tokens: 512
max_output_tokens: 64
model_architecture: LlamaForCausalLM
model_group: rica40325/feedback_2000
model_name: rica40325-feedback-2000_v1
model_num_parameters: 8030261248.0
model_repo: rica40325/feedback_2000
model_size: 8B
num_battles: 14655
num_wins: 7571
ranking_group: single
status: torndown
submission_type: basic
throughput_3p7s: 2.03
timestamp: 2024-09-07T16:10:34+00:00
us_pacific_date: 2024-09-07
win_ratio: 0.5166154895939952
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-feedback-2000-v2-mkmlizer
Waiting for job on rica40325-feedback-2000-v2-mkmlizer to finish
Failed to get response for submission blend_remul_2024-08-22: ('http://chaiml-llama-8b-pairwis-8189-v19-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'readfrom tcp 127.0.0.1:45320->127.0.0.1:8080: write tcp 127.0.0.1:45320->127.0.0.1:8080: use of closed network connection\n')
rica40325-feedback-2000-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rica40325-feedback-2000-v2-mkmlizer: ║ _____ __ __ ║
rica40325-feedback-2000-v2-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
rica40325-feedback-2000-v2-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
rica40325-feedback-2000-v2-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
rica40325-feedback-2000-v2-mkmlizer: ║ /___/ ║
rica40325-feedback-2000-v2-mkmlizer: ║ ║
rica40325-feedback-2000-v2-mkmlizer: ║ Version: 0.10.1 ║
rica40325-feedback-2000-v2-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
rica40325-feedback-2000-v2-mkmlizer: ║ https://mk1.ai ║
rica40325-feedback-2000-v2-mkmlizer: ║ ║
rica40325-feedback-2000-v2-mkmlizer: ║ The license key for the current software has been verified as ║
rica40325-feedback-2000-v2-mkmlizer: ║ belonging to: ║
rica40325-feedback-2000-v2-mkmlizer: ║ ║
rica40325-feedback-2000-v2-mkmlizer: ║ Chai Research Corp. ║
rica40325-feedback-2000-v2-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
rica40325-feedback-2000-v2-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
rica40325-feedback-2000-v2-mkmlizer: ║ ║
rica40325-feedback-2000-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
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
rica40325-feedback-2000-v2-mkmlizer: Downloaded to shared memory in 63.604s
rica40325-feedback-2000-v2-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpzeva3f13, device:0
rica40325-feedback-2000-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
rica40325-feedback-2000-v2-mkmlizer: quantized model in 29.570s
rica40325-feedback-2000-v2-mkmlizer: Processed model rica40325/feedback_2000 in 93.174s
rica40325-feedback-2000-v2-mkmlizer: creating bucket guanaco-mkml-models
rica40325-feedback-2000-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rica40325-feedback-2000-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rica40325-feedback-2000-v2
rica40325-feedback-2000-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rica40325-feedback-2000-v2/config.json
rica40325-feedback-2000-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rica40325-feedback-2000-v2/special_tokens_map.json
rica40325-feedback-2000-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rica40325-feedback-2000-v2/tokenizer_config.json
rica40325-feedback-2000-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rica40325-feedback-2000-v2/tokenizer.json
rica40325-feedback-2000-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rica40325-feedback-2000-v2/flywheel_model.0.safetensors
rica40325-feedback-2000-v2-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 2%|▏ | 5/291 [00:00<00:10, 26.70it/s] Loading 0: 4%|▍ | 12/291 [00:00<00:07, 36.96it/s] Loading 0: 5%|▌ | 16/291 [00:00<00:08, 33.99it/s] Loading 0: 7%|▋ | 21/291 [00:00<00:07, 37.52it/s] Loading 0: 9%|▊ | 25/291 [00:00<00:07, 33.79it/s] Loading 0: 10%|▉ | 29/291 [00:00<00:07, 35.22it/s] Loading 0: 11%|█▏ | 33/291 [00:01<00:09, 26.52it/s] Loading 0: 12%|█▏ | 36/291 [00:01<00:10, 23.59it/s] Loading 0: 14%|█▍ | 41/291 [00:01<00:09, 25.07it/s] Loading 0: 16%|█▌ | 46/291 [00:01<00:08, 30.03it/s] Loading 0: 17%|█▋ | 50/291 [00:01<00:08, 28.01it/s] Loading 0: 20%|█▉ | 57/291 [00:01<00:06, 35.55it/s] Loading 0: 21%|██ | 61/291 [00:01<00:06, 33.03it/s] Loading 0: 23%|██▎ | 66/291 [00:02<00:06, 34.80it/s] Loading 0: 24%|██▍ | 70/291 [00:02<00:06, 32.51it/s] Loading 0: 25%|██▌ | 74/291 [00:02<00:06, 32.50it/s] Loading 0: 27%|██▋ | 78/291 [00:02<00:06, 32.94it/s] Loading 0: 28%|██▊ | 82/291 [00:02<00:08, 24.14it/s] Loading 0: 29%|██▉ | 85/291 [00:02<00:08, 24.44it/s] Loading 0: 31%|███ | 90/291 [00:02<00:07, 28.59it/s] Loading 0: 32%|███▏ | 94/291 [00:03<00:06, 28.18it/s] Loading 0: 34%|███▍ | 99/291 [00:03<00:06, 31.30it/s] Loading 0: 35%|███▌ | 103/291 [00:03<00:06, 30.64it/s] Loading 0: 37%|███▋ | 108/291 [00:03<00:05, 33.91it/s] Loading 0: 38%|███▊ | 112/291 [00:03<00:05, 32.63it/s] Loading 0: 40%|███▉ | 116/291 [00:03<00:05, 31.16it/s] Loading 0: 42%|████▏ | 122/291 [00:03<00:05, 30.65it/s] Loading 0: 44%|████▎ | 127/291 [00:04<00:05, 30.13it/s] Loading 0: 46%|████▌ | 133/291 [00:04<00:05, 27.52it/s] Loading 0: 47%|████▋ | 136/291 [00:04<00:05, 26.10it/s] Loading 0: 48%|████▊ | 140/291 [00:04<00:06, 25.11it/s] Loading 0: 51%|█████ | 147/291 [00:04<00:04, 32.21it/s] Loading 0: 52%|█████▏ | 151/291 [00:04<00:04, 31.88it/s] Loading 0: 54%|█████▎ | 156/291 [00:05<00:03, 35.04it/s] Loading 0: 55%|█████▍ | 160/291 [00:05<00:03, 33.39it/s] Loading 0: 57%|█████▋ | 165/291 [00:05<00:03, 34.91it/s] Loading 0: 58%|█████▊ | 169/291 [00:05<00:03, 33.51it/s] Loading 0: 60%|█████▉ | 174/291 [00:05<00:03, 34.75it/s] Loading 0: 61%|██████ | 178/291 [00:05<00:03, 33.06it/s] Loading 0: 63%|██████▎ | 183/291 [00:05<00:02, 36.21it/s] Loading 0: 64%|██████▍ | 187/291 [00:06<00:04, 25.73it/s] Loading 0: 66%|██████▌ | 191/291 [00:06<00:03, 26.48it/s] Loading 0: 67%|██████▋ | 194/291 [00:06<00:04, 23.66it/s] Loading 0: 69%|██████▉ | 201/291 [00:06<00:02, 30.88it/s] Loading 0: 70%|███████ | 205/291 [00:06<00:02, 29.92it/s] Loading 0: 72%|███████▏ | 210/291 [00:06<00:02, 32.69it/s] Loading 0: 74%|███████▎ | 214/291 [00:07<00:02, 31.53it/s] Loading 0: 75%|███████▌ | 219/291 [00:07<00:02, 34.05it/s] Loading 0: 77%|███████▋ | 223/291 [00:07<00:02, 32.51it/s] Loading 0: 78%|███████▊ | 227/291 [00:07<00:02, 31.53it/s] Loading 0: 79%|███████▉ | 231/291 [00:07<00:01, 31.60it/s] Loading 0: 81%|████████ | 235/291 [00:07<00:02, 24.56it/s] Loading 0: 82%|████████▏ | 239/291 [00:07<00:02, 24.23it/s] Loading 0: 85%|████████▍ | 246/291 [00:08<00:01, 32.25it/s] Loading 0: 86%|████████▌ | 250/291 [00:08<00:01, 31.73it/s] Loading 0: 88%|████████▊ | 255/291 [00:08<00:01, 34.38it/s] Loading 0: 89%|████████▉ | 259/291 [00:08<00:00, 32.87it/s] Loading 0: 91%|█████████ | 264/291 [00:08<00:00, 35.96it/s] Loading 0: 92%|█████████▏| 268/291 [00:08<00:00, 33.80it/s] Loading 0: 94%|█████████▍| 273/291 [00:08<00:00, 35.72it/s] Loading 0: 95%|█████████▌| 277/291 [00:08<00:00, 33.52it/s] Loading 0: 97%|█████████▋| 281/291 [00:09<00:00, 32.71it/s] Loading 0: 98%|█████████▊| 285/291 [00:09<00:00, 34.46it/s] Loading 0: 99%|█████████▉| 289/291 [00:14<00:00, 2.34it/s]
Job rica40325-feedback-2000-v2-mkmlizer completed after 249.43s with status: succeeded
Stopping job with name rica40325-feedback-2000-v2-mkmlizer
Pipeline stage MKMLizer completed in 251.11s
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-feedback-2000-v2
Waiting for inference service rica40325-feedback-2000-v2 to be ready
Failed to get response for submission blend_jerun_2024-08-22: ('http://neversleep-noromaid-v0-8068-v150-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:43988->127.0.0.1:8080: read: connection reset by peer\n')
Inference service rica40325-feedback-2000-v2 ready after 150.57476997375488s
Pipeline stage MKMLDeployer completed in 151.26s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.400123357772827s
Received healthy response to inference request in 2.064882278442383s
Received healthy response to inference request in 1.896287202835083s
Received healthy response to inference request in 2.2857930660247803s
Received healthy response to inference request in 2.0150694847106934s
5 requests
0 failed requests
5th percentile: 1.920043659210205
10th percentile: 1.943800115585327
20th percentile: 1.9913130283355713
30th percentile: 2.0250320434570312
40th percentile: 2.044957160949707
50th percentile: 2.064882278442383
60th percentile: 2.153246593475342
70th percentile: 2.2416109085083007
80th percentile: 2.3086591243743895
90th percentile: 2.3543912410736083
95th percentile: 2.377257299423218
99th percentile: 2.3955501461029054
mean time: 2.132431077957153
Pipeline stage StressChecker completed in 11.62s
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 4.70s
Shutdown handler de-registered
rica40325-feedback-2000_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 0.14s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.11s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service rica40325-feedback-2000-v2-profiler
Waiting for inference service rica40325-feedback-2000-v2-profiler to be ready
Tearing down inference service rica40325-feedback-2000-v2-profiler
%s, retrying in %s seconds...
Creating inference service rica40325-feedback-2000-v2-profiler
Waiting for inference service rica40325-feedback-2000-v2-profiler to be ready
Inference service rica40325-feedback-2000-v2-profiler ready after 170.4684329032898s
Pipeline stage MKMLProfilerDeployer completed in 772.74s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/rica40325-feedback-2000-v2-profiler-predictor-00001-deploylgn8r:/code/chaiverse_profiler_1725726666 --namespace tenant-chaiml-guanaco
kubectl exec -it rica40325-feedback-2000-v2-profiler-predictor-00001-deploylgn8r --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1725726666 && 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_1725726666/summary.json'
kubectl exec -it rica40325-feedback-2000-v2-profiler-predictor-00001-deploylgn8r --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1725726666/summary.json'
Pipeline stage MKMLProfilerRunner completed in 828.68s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service rica40325-feedback-2000-v2-profiler is running
Tearing down inference service rica40325-feedback-2000-v2-profiler
Service rica40325-feedback-2000-v2-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.75s
Shutdown handler de-registered
rica40325-feedback-2000_v2 status is now inactive due to auto deactivation removed underperforming models
rica40325-feedback-2000_v2 status is now torndown due to DeploymentManager action