developer_uid: zonemercy
submission_id: chaiml-horror-sft-v1-5e6_v5
model_name: chaiml-horror-sft-v1-1e5_v1
model_group: ChaiML/Horror-SFT-v1-5e6
status: torndown
timestamp: 2024-09-05T21:21:07+00:00
num_battles: 11492
num_wins: 5118
celo_rating: 1198.29
family_friendly_score: 0.0
submission_type: basic
model_repo: ChaiML/Horror-SFT-v1-5e6
model_architecture: MistralForCausalLM
model_num_parameters: 12772070400.0
best_of: 4
max_input_tokens: 1024
max_output_tokens: 64
latencies: [{'batch_size': 1, 'throughput': 0.6272588444427792, 'latency_mean': 1.594137442111969, 'latency_p50': 1.6036230325698853, 'latency_p90': 1.7522860765457153}, {'batch_size': 3, 'throughput': 1.217449516635225, 'latency_mean': 2.4565743243694307, 'latency_p50': 2.449995279312134, 'latency_p90': 2.7456589221954344}, {'batch_size': 5, 'throughput': 1.5236631005623062, 'latency_mean': 3.2732128274440764, 'latency_p50': 3.275214195251465, 'latency_p90': 3.717589831352234}, {'batch_size': 6, 'throughput': 1.6037923843545983, 'latency_mean': 3.721743433475494, 'latency_p50': 3.7399739027023315, 'latency_p90': 4.250715637207032}, {'batch_size': 8, 'throughput': 1.6974974521070367, 'latency_mean': 4.680763937234879, 'latency_p50': 4.685117602348328, 'latency_p90': 5.243572044372558}, {'batch_size': 10, 'throughput': 1.7340862887054946, 'latency_mean': 5.731855192184448, 'latency_p50': 5.745940208435059, 'latency_p90': 6.378151082992553}]
gpu_counts: {'NVIDIA RTX A5000': 1}
display_name: chaiml-horror-sft-v1-1e5_v1
is_internal_developer: True
language_model: ChaiML/Horror-SFT-v1-5e6
model_size: 13B
ranking_group: single
throughput_3p7s: 1.61
us_pacific_date: 2024-09-05
win_ratio: 0.445353289244692
generation_params: {'temperature': 0.9, 'top_p': 1.0, 'min_p': 0.05, 'top_k': 80, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '</s>', 'You:', 'Bot:', 'User:'], 'max_input_tokens': 1024, 'best_of': 4, 'max_output_tokens': 64}
formatter: {'memory_template': "{bot_name}'s Persona: {memory}\nMake the conversation as scary as possible while keep in character\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '{bot_name}[Extreme horror edition]:', 'truncate_by_message': False}
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 chaiml-horror-sft-v1-5e6-v5-mkmlizer
Waiting for job on chaiml-horror-sft-v1-5e6-v5-mkmlizer to finish
chaiml-horror-sft-v1-5e6-v5-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-horror-sft-v1-5e6-v5-mkmlizer: ║ _____ __ __ ║
chaiml-horror-sft-v1-5e6-v5-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
chaiml-horror-sft-v1-5e6-v5-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
chaiml-horror-sft-v1-5e6-v5-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
chaiml-horror-sft-v1-5e6-v5-mkmlizer: ║ /___/ ║
chaiml-horror-sft-v1-5e6-v5-mkmlizer: ║ ║
chaiml-horror-sft-v1-5e6-v5-mkmlizer: ║ Version: 0.10.1 ║
chaiml-horror-sft-v1-5e6-v5-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
chaiml-horror-sft-v1-5e6-v5-mkmlizer: ║ https://mk1.ai ║
chaiml-horror-sft-v1-5e6-v5-mkmlizer: ║ ║
chaiml-horror-sft-v1-5e6-v5-mkmlizer: ║ The license key for the current software has been verified as ║
chaiml-horror-sft-v1-5e6-v5-mkmlizer: ║ belonging to: ║
chaiml-horror-sft-v1-5e6-v5-mkmlizer: ║ ║
chaiml-horror-sft-v1-5e6-v5-mkmlizer: ║ Chai Research Corp. ║
chaiml-horror-sft-v1-5e6-v5-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
chaiml-horror-sft-v1-5e6-v5-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
chaiml-horror-sft-v1-5e6-v5-mkmlizer: ║ ║
chaiml-horror-sft-v1-5e6-v5-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
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
chaiml-horror-sft-v1-5e6-v5-mkmlizer: Downloaded to shared memory in 30.691s
chaiml-horror-sft-v1-5e6-v5-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpx9q2uyrq, device:0
chaiml-horror-sft-v1-5e6-v5-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-horror-sft-v1-5e6-v5-mkmlizer: quantized model in 35.507s
chaiml-horror-sft-v1-5e6-v5-mkmlizer: Processed model ChaiML/Horror-SFT-v1-5e6 in 66.199s
chaiml-horror-sft-v1-5e6-v5-mkmlizer: creating bucket guanaco-mkml-models
chaiml-horror-sft-v1-5e6-v5-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-horror-sft-v1-5e6-v5-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-horror-sft-v1-5e6-v5
chaiml-horror-sft-v1-5e6-v5-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-horror-sft-v1-5e6-v5/tokenizer.json
chaiml-horror-sft-v1-5e6-v5-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-horror-sft-v1-5e6-v5/flywheel_model.0.safetensors
chaiml-horror-sft-v1-5e6-v5-mkmlizer: Loading 0: 0%| | 0/363 [00:00<?, ?it/s] Loading 0: 1%|▏ | 5/363 [00:00<00:11, 30.10it/s] Loading 0: 3%|▎ | 12/363 [00:00<00:07, 47.29it/s] Loading 0: 5%|▍ | 18/363 [00:00<00:07, 48.14it/s] Loading 0: 7%|▋ | 24/363 [00:00<00:08, 39.32it/s] Loading 0: 9%|▊ | 31/363 [00:00<00:07, 45.71it/s] Loading 0: 10%|█ | 37/363 [00:00<00:07, 42.65it/s] Loading 0: 12%|█▏ | 42/363 [00:00<00:07, 41.54it/s] Loading 0: 13%|█▎ | 49/363 [00:01<00:06, 47.75it/s] Loading 0: 15%|█▌ | 55/363 [00:01<00:06, 45.61it/s] Loading 0: 17%|█▋ | 61/363 [00:01<00:08, 35.44it/s] Loading 0: 18%|█▊ | 66/363 [00:01<00:08, 36.56it/s] Loading 0: 20%|█▉ | 72/363 [00:01<00:07, 40.42it/s] Loading 0: 21%|██ | 77/363 [00:01<00:06, 42.49it/s] Loading 0: 23%|██▎ | 82/363 [00:02<00:07, 36.94it/s] Loading 0: 25%|██▍ | 90/363 [00:02<00:06, 44.69it/s] Loading 0: 26%|██▌ | 95/363 [00:02<00:05, 45.75it/s] Loading 0: 28%|██▊ | 100/363 [00:02<00:06, 38.37it/s] Loading 0: 30%|██▉ | 108/363 [00:02<00:05, 47.52it/s] Loading 0: 31%|███▏ | 114/363 [00:02<00:06, 41.16it/s] Loading 0: 33%|███▎ | 119/363 [00:02<00:06, 40.25it/s] Loading 0: 35%|███▍ | 126/363 [00:02<00:05, 45.29it/s] Loading 0: 36%|███▋ | 132/363 [00:03<00:05, 43.05it/s] Loading 0: 38%|███▊ | 137/363 [00:03<00:05, 41.13it/s] Loading 0: 39%|███▉ | 142/363 [00:03<00:06, 32.18it/s] Loading 0: 40%|████ | 146/363 [00:03<00:06, 33.52it/s] Loading 0: 41%|████▏ | 150/363 [00:03<00:06, 33.41it/s] Loading 0: 43%|████▎ | 157/363 [00:03<00:05, 40.69it/s] Loading 0: 45%|████▍ | 163/363 [00:04<00:04, 40.83it/s] Loading 0: 46%|████▋ | 168/363 [00:04<00:04, 41.13it/s] Loading 0: 48%|████▊ | 175/363 [00:04<00:04, 46.43it/s] Loading 0: 50%|████▉ | 180/363 [00:04<00:03, 47.31it/s] Loading 0: 51%|█████ | 185/363 [00:04<00:04, 38.54it/s] Loading 0: 53%|█████▎ | 193/363 [00:04<00:03, 46.59it/s] Loading 0: 55%|█████▍ | 199/363 [00:04<00:03, 43.09it/s] Loading 0: 56%|█████▌ | 204/363 [00:04<00:03, 42.20it/s] Loading 0: 58%|█████▊ | 210/363 [00:05<00:03, 45.78it/s] Loading 0: 59%|█████▉ | 215/363 [00:05<00:03, 45.14it/s] Loading 0: 61%|██████ | 220/363 [00:05<00:03, 44.04it/s] Loading 0: 62%|██████▏ | 225/363 [00:05<00:04, 28.19it/s] Loading 0: 63%|██████▎ | 230/363 [00:05<00:04, 30.63it/s] Loading 0: 66%|██████▌ | 238/363 [00:05<00:03, 39.36it/s] Loading 0: 67%|██████▋ | 244/363 [00:06<00:02, 39.90it/s] Loading 0: 69%|██████▊ | 249/363 [00:06<00:02, 40.55it/s] Loading 0: 70%|███████ | 255/363 [00:06<00:02, 45.07it/s] Loading 0: 72%|███████▏ | 260/363 [00:06<00:02, 45.65it/s] Loading 0: 73%|███████▎ | 265/363 [00:06<00:02, 46.77it/s] Loading 0: 75%|███████▍ | 271/363 [00:06<00:02, 43.32it/s] Loading 0: 76%|███████▌ | 276/363 [00:06<00:02, 41.73it/s] Loading 0: 78%|███████▊ | 282/363 [00:06<00:01, 44.57it/s] Loading 0: 79%|███████▉ | 287/363 [00:06<00:01, 41.97it/s] Loading 0: 80%|████████ | 292/363 [00:07<00:01, 42.32it/s] Loading 0: 82%|████████▏ | 298/363 [00:07<00:01, 41.72it/s] Loading 0: 84%|████████▎ | 304/363 [00:13<00:22, 2.68it/s] Loading 0: 85%|████████▍ | 307/363 [00:14<00:17, 3.22it/s] Loading 0: 86%|████████▌ | 312/363 [00:14<00:11, 4.49it/s] Loading 0: 88%|████████▊ | 320/363 [00:14<00:05, 7.32it/s] Loading 0: 90%|████████▉ | 325/363 [00:14<00:04, 9.49it/s] Loading 0: 91%|█████████ | 330/363 [00:14<00:02, 11.56it/s] Loading 0: 93%|█████████▎| 338/363 [00:14<00:01, 17.08it/s] Loading 0: 94%|█████████▍| 343/363 [00:14<00:00, 20.49it/s] Loading 0: 96%|█████████▌| 348/363 [00:14<00:00, 22.34it/s] Loading 0: 98%|█████████▊| 356/363 [00:15<00:00, 30.18it/s] Loading 0: 100%|█████████▉| 362/363 [00:15<00:00, 32.44it/s]
Job chaiml-horror-sft-v1-5e6-v5-mkmlizer completed after 95.87s with status: succeeded
Stopping job with name chaiml-horror-sft-v1-5e6-v5-mkmlizer
Pipeline stage MKMLizer completed in 96.91s
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 chaiml-horror-sft-v1-5e6-v5
Waiting for inference service chaiml-horror-sft-v1-5e6-v5 to be ready
Failed to get response for submission blend_remul_2024-08-22: ('http://zonemercy-lexical-nemo-1518-v18-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:35632->127.0.0.1:8080: read: connection reset by peer\n')
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
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
Failed to get response for submission blend_susol_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\')]"}')
Inference service chaiml-horror-sft-v1-5e6-v5 ready after 151.38251900672913s
Pipeline stage MKMLDeployer completed in 152.26s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.761270523071289s
Received healthy response to inference request in 1.7726068496704102s
Received healthy response to inference request in 1.7464637756347656s
Received healthy response to inference request in 2.083228826522827s
Received healthy response to inference request in 2.1610138416290283s
5 requests
0 failed requests
5th percentile: 1.7516923904418946
10th percentile: 1.7569210052490234
20th percentile: 1.7673782348632812
30th percentile: 1.8347312450408935
40th percentile: 1.9589800357818603
50th percentile: 2.083228826522827
60th percentile: 2.1143428325653075
70th percentile: 2.145456838607788
80th percentile: 2.2810651779174806
90th percentile: 2.521167850494385
95th percentile: 2.6412191867828367
99th percentile: 2.7372602558135988
mean time: 2.1049167633056642
Pipeline stage StressChecker completed in 12.72s
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 5.39s
Shutdown handler de-registered
chaiml-horror-sft-v1-5e6_v5 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 0.11s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service chaiml-horror-sft-v1-5e6-v5-profiler
Waiting for inference service chaiml-horror-sft-v1-5e6-v5-profiler to be ready
Inference service chaiml-horror-sft-v1-5e6-v5-profiler ready after 140.33080315589905s
Pipeline stage MKMLProfilerDeployer completed in 140.72s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/chaiml-horror-sft-v1-5e6-v5-profiler-predictor-00001-deplokw8c8:/code/chaiverse_profiler_1725571709 --namespace tenant-chaiml-guanaco
kubectl exec -it chaiml-horror-sft-v1-5e6-v5-profiler-predictor-00001-deplokw8c8 --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1725571709 && python profiles.py profile --best_of_n 4 --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 1024 --output_tokens 64 --summary /code/chaiverse_profiler_1725571709/summary.json'
kubectl exec -it chaiml-horror-sft-v1-5e6-v5-profiler-predictor-00001-deplokw8c8 --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1725571709/summary.json'
Pipeline stage MKMLProfilerRunner completed in 977.11s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service chaiml-horror-sft-v1-5e6-v5-profiler is running
Tearing down inference service chaiml-horror-sft-v1-5e6-v5-profiler
Service chaiml-horror-sft-v1-5e6-v5-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.77s
Shutdown handler de-registered
chaiml-horror-sft-v1-5e6_v5 status is now inactive due to auto deactivation removed underperforming models
chaiml-horror-sft-v1-5e6_v5 status is now torndown due to DeploymentManager action