submission_id: chaiml-albert-dpo-0912-v_7175_v1
developer_uid: chai_backend_admin
best_of: 8
celo_rating: 1251.64
display_name: edit_top_bot_personality
family_friendly_score: 0.0
formatter: {'memory_template': '', 'prompt_template': '', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '{bot_name}:', 'truncate_by_message': False}
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>', '###', 'Bot:', 'User:', 'You:', '<|im_end|>'], 'max_input_tokens': 1024, 'best_of': 8, 'max_output_tokens': 64}
gpu_counts: {'NVIDIA RTX A5000': 1}
is_internal_developer: True
language_model: ChaiML/albert_dpo_0912_virgo_edit_variety_top_bot_personality
latencies: [{'batch_size': 1, 'throughput': 0.6120975183735723, 'latency_mean': 1.6335249519348145, 'latency_p50': 1.6327568292617798, 'latency_p90': 1.7887208938598633}, {'batch_size': 3, 'throughput': 1.062931416256256, 'latency_mean': 2.8147723495960237, 'latency_p50': 2.7867064476013184, 'latency_p90': 3.1204746961593623}, {'batch_size': 5, 'throughput': 1.2043103029462174, 'latency_mean': 4.1309771466255185, 'latency_p50': 4.164233088493347, 'latency_p90': 4.7083744764328}, {'batch_size': 6, 'throughput': 1.2362148365088574, 'latency_mean': 4.82392427444458, 'latency_p50': 4.807650089263916, 'latency_p90': 5.493195533752441}, {'batch_size': 8, 'throughput': 1.216128079612076, 'latency_mean': 6.540775756835938, 'latency_p50': 6.5560160875320435, 'latency_p90': 7.433024477958679}, {'batch_size': 10, 'throughput': 1.182543081832756, 'latency_mean': 8.406432594060897, 'latency_p50': 8.45857560634613, 'latency_p90': 9.425828456878662}]
max_input_tokens: 1024
max_output_tokens: 64
model_architecture: MistralForCausalLM
model_group: ChaiML/albert_dpo_0912_v
model_name: edit_top_bot_personality
model_num_parameters: 12772070400.0
model_repo: ChaiML/albert_dpo_0912_virgo_edit_variety_top_bot_personality
model_size: 13B
num_battles: 11133
num_wins: 5654
ranking_group: single
status: torndown
submission_type: basic
throughput_3p7s: 1.17
timestamp: 2024-09-13T19:19:49+00:00
us_pacific_date: 2024-09-13
win_ratio: 0.5078595167519986
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-albert-dpo-0912-v-7175-v1-mkmlizer
Waiting for job on chaiml-albert-dpo-0912-v-7175-v1-mkmlizer to finish
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: ║ _____ __ __ ║
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: ║ /___/ ║
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: ║ ║
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: ║ Version: 0.10.1 ║
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: ║ https://mk1.ai ║
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: ║ ║
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: ║ The license key for the current software has been verified as ║
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: ║ belonging to: ║
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: ║ ║
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: ║ Chai Research Corp. ║
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: ║ ║
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: Downloaded to shared memory in 97.087s
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp7va7yng7, device:0
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: quantized model in 40.371s
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: Processed model ChaiML/albert_dpo_0912_virgo_edit_variety_top_bot_personality in 137.458s
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: creating bucket guanaco-mkml-models
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-7175-v1
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-7175-v1/config.json
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-7175-v1/special_tokens_map.json
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-7175-v1/tokenizer_config.json
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-7175-v1/flywheel_model.0.safetensors
chaiml-albert-dpo-0912-v-7175-v1-mkmlizer: Loading 0: 0%| | 0/363 [00:00<?, ?it/s] Loading 0: 1%|▏ | 5/363 [00:00<00:15, 23.37it/s] Loading 0: 3%|▎ | 10/363 [00:00<00:11, 29.90it/s] Loading 0: 4%|▍ | 14/363 [00:00<00:13, 25.20it/s] Loading 0: 6%|▌ | 21/363 [00:00<00:09, 35.96it/s] Loading 0: 7%|▋ | 26/363 [00:01<00:16, 20.28it/s] Loading 0: 9%|▊ | 31/363 [00:01<00:13, 25.05it/s] Loading 0: 10%|▉ | 35/363 [00:01<00:12, 25.84it/s] Loading 0: 11%|█ | 39/363 [00:01<00:11, 27.39it/s] Loading 0: 12%|█▏ | 43/363 [00:01<00:11, 26.96it/s] Loading 0: 13%|█▎ | 48/363 [00:01<00:10, 29.79it/s] Loading 0: 14%|█▍ | 52/363 [00:01<00:11, 28.09it/s] Loading 0: 15%|█▌ | 56/363 [00:02<00:10, 28.66it/s] Loading 0: 17%|█▋ | 61/363 [00:02<00:12, 25.16it/s] Loading 0: 18%|█▊ | 64/363 [00:02<00:13, 22.22it/s] Loading 0: 20%|█▉ | 71/363 [00:02<00:10, 28.75it/s] Loading 0: 21%|██ | 75/363 [00:02<00:10, 28.26it/s] Loading 0: 21%|██▏ | 78/363 [00:02<00:10, 26.08it/s] Loading 0: 23%|██▎ | 84/363 [00:03<00:09, 30.16it/s] Loading 0: 24%|██▍ | 88/363 [00:03<00:09, 28.78it/s] Loading 0: 26%|██▌ | 93/363 [00:03<00:08, 31.22it/s] Loading 0: 27%|██▋ | 97/363 [00:03<00:09, 29.35it/s] Loading 0: 28%|██▊ | 101/363 [00:03<00:11, 23.79it/s] Loading 0: 29%|██▊ | 104/363 [00:03<00:12, 21.38it/s] Loading 0: 30%|███ | 109/363 [00:04<00:09, 26.57it/s] Loading 0: 31%|███ | 113/363 [00:04<00:10, 24.61it/s] Loading 0: 33%|███▎ | 120/363 [00:04<00:07, 31.66it/s] Loading 0: 34%|███▍ | 124/363 [00:04<00:07, 30.42it/s] Loading 0: 36%|███▌ | 129/363 [00:04<00:07, 33.15it/s] Loading 0: 37%|███▋ | 133/363 [00:04<00:07, 31.16it/s] Loading 0: 38%|███▊ | 137/363 [00:04<00:07, 31.22it/s] Loading 0: 39%|███▉ | 142/363 [00:05<00:08, 27.08it/s] Loading 0: 40%|███▉ | 145/363 [00:05<00:08, 24.68it/s] Loading 0: 41%|████ | 149/363 [00:05<00:09, 23.08it/s] Loading 0: 43%|████▎ | 156/363 [00:05<00:07, 29.57it/s] Loading 0: 44%|████▍ | 160/363 [00:05<00:07, 28.47it/s] Loading 0: 45%|████▌ | 165/363 [00:05<00:06, 29.98it/s] Loading 0: 47%|████▋ | 169/363 [00:06<00:06, 28.89it/s] Loading 0: 48%|████▊ | 174/363 [00:06<00:06, 30.67it/s] Loading 0: 49%|████▉ | 178/363 [00:06<00:06, 29.24it/s] Loading 0: 50%|█████ | 182/363 [00:06<00:07, 24.95it/s] Loading 0: 51%|█████ | 185/363 [00:06<00:08, 22.20it/s] Loading 0: 53%|█████▎ | 192/363 [00:06<00:05, 29.17it/s] Loading 0: 54%|█████▍ | 196/363 [00:07<00:05, 28.48it/s] Loading 0: 55%|█████▌ | 201/363 [00:07<00:05, 31.06it/s] Loading 0: 56%|█████▋ | 205/363 [00:07<00:05, 29.54it/s] Loading 0: 58%|█████▊ | 210/363 [00:07<00:04, 30.82it/s] Loading 0: 59%|█████▉ | 214/363 [00:07<00:05, 29.07it/s] Loading 0: 60%|█████▉ | 217/363 [00:07<00:05, 29.09it/s] Loading 0: 61%|██████ | 222/363 [00:07<00:04, 30.41it/s] Loading 0: 62%|██████▏ | 226/363 [00:08<00:06, 21.60it/s] Loading 0: 63%|██████▎ | 230/363 [00:08<00:06, 21.06it/s] Loading 0: 65%|██████▌ | 237/363 [00:08<00:04, 27.79it/s] Loading 0: 66%|██████▋ | 241/363 [00:08<00:04, 27.36it/s] Loading 0: 68%|██████▊ | 246/363 [00:08<00:03, 29.73it/s] Loading 0: 69%|██████▉ | 250/363 [00:09<00:03, 28.90it/s] Loading 0: 70%|███████ | 255/363 [00:09<00:03, 31.02it/s] Loading 0: 71%|███████▏ | 259/363 [00:09<00:03, 28.69it/s] Loading 0: 72%|███████▏ | 263/363 [00:09<00:04, 24.65it/s] Loading 0: 73%|███████▎ | 266/363 [00:09<00:04, 21.41it/s] Loading 0: 75%|███████▍ | 271/363 [00:09<00:03, 26.44it/s] Loading 0: 76%|███████▌ | 275/363 [00:10<00:03, 24.41it/s] Loading 0: 78%|███████▊ | 282/363 [00:10<00:02, 30.64it/s] Loading 0: 79%|███████▉ | 286/363 [00:10<00:02, 28.99it/s] Loading 0: 80%|████████ | 291/363 [00:10<00:02, 30.59it/s] Loading 0: 81%|████████▏ | 295/363 [00:10<00:02, 28.99it/s] Loading 0: 82%|████████▏ | 299/363 [00:10<00:02, 29.21it/s] Loading 0: 84%|████████▎ | 304/363 [00:11<00:02, 26.61it/s] Loading 0: 85%|████████▍ | 307/363 [00:11<00:02, 24.98it/s] Loading 0: 86%|████████▌ | 311/363 [00:11<00:02, 23.76it/s] Loading 0: 88%|████████▊ | 318/363 [00:11<00:01, 30.73it/s] Loading 0: 89%|████████▊ | 322/363 [00:11<00:01, 29.58it/s] Loading 0: 90%|█████████ | 327/363 [00:11<00:01, 32.11it/s] Loading 0: 91%|█████████ | 331/363 [00:11<00:01, 29.75it/s] Loading 0: 93%|█████████▎| 336/363 [00:12<00:00, 31.53it/s] Loading 0: 94%|█████████▎| 340/363 [00:12<00:00, 30.25it/s] Loading 0: 95%|█████████▍| 344/363 [00:19<00:09, 1.98it/s] Loading 0: 96%|█████████▌| 348/363 [00:19<00:05, 2.67it/s] Loading 0: 97%|█████████▋| 353/363 [00:19<00:02, 3.87it/s] Loading 0: 98%|█████████▊| 357/363 [00:19<00:01, 5.03it/s]
Job chaiml-albert-dpo-0912-v-7175-v1-mkmlizer completed after 157.35s with status: succeeded
Stopping job with name chaiml-albert-dpo-0912-v-7175-v1-mkmlizer
Pipeline stage MKMLizer completed in 158.27s
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-albert-dpo-0912-v-7175-v1
Waiting for inference service chaiml-albert-dpo-0912-v-7175-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
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Inference service chaiml-albert-dpo-0912-v-7175-v1 ready after 191.28570556640625s
Pipeline stage MKMLDeployer completed in 193.04s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.008103847503662s
Received healthy response to inference request in 2.0451717376708984s
Received healthy response to inference request in 1.8224327564239502s
Received healthy response to inference request in 1.5549147129058838s
Received healthy response to inference request in 2.712552070617676s
5 requests
0 failed requests
5th percentile: 1.6084183216094972
10th percentile: 1.6619219303131103
20th percentile: 1.7689291477203368
30th percentile: 1.8595669746398926
40th percentile: 1.9338354110717773
50th percentile: 2.008103847503662
60th percentile: 2.0229310035705566
70th percentile: 2.037758159637451
80th percentile: 2.178647804260254
90th percentile: 2.4455999374389648
95th percentile: 2.57907600402832
99th percentile: 2.6858568572998047
mean time: 2.028635025024414
Pipeline stage StressChecker completed in 15.46s
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 5.79s
Shutdown handler de-registered
chaiml-albert-dpo-0912-v_7175_v1 status is now deployed due to DeploymentManager action
Connection pool is full, discarding connection: %s. Connection pool size: %s
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 chaiml-albert-dpo-0912-v-7175-v1-profiler
Waiting for inference service chaiml-albert-dpo-0912-v-7175-v1-profiler to be ready
Inference service chaiml-albert-dpo-0912-v-7175-v1-profiler ready after 180.41897106170654s
Pipeline stage MKMLProfilerDeployer completed in 180.75s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/chaiml-albert-dpo-09fe0cf74ab653c86419b9505070868405-deplobxjkw:/code/chaiverse_profiler_1726255785 --namespace tenant-chaiml-guanaco
kubectl exec -it chaiml-albert-dpo-09fe0cf74ab653c86419b9505070868405-deplobxjkw --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1726255785 && python profiles.py profile --best_of_n 8 --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_1726255785/summary.json'
kubectl exec -it chaiml-albert-dpo-09fe0cf74ab653c86419b9505070868405-deplobxjkw --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1726255785/summary.json'
Pipeline stage MKMLProfilerRunner completed in 1181.91s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service chaiml-albert-dpo-0912-v-7175-v1-profiler is running
Tearing down inference service chaiml-albert-dpo-0912-v-7175-v1-profiler
Service chaiml-albert-dpo-0912-v-7175-v1-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 2.08s
Shutdown handler de-registered
chaiml-albert-dpo-0912-v_7175_v1 status is now inactive due to auto deactivation removed underperforming models
chaiml-albert-dpo-0912-v_7175_v1 status is now torndown due to DeploymentManager action
run pipeline stage %s
Pipeline stage MKMLModelDeleter completed in 3.54s
Pipeline stage %s skipped, reason=%s
run pipeline %s
Shutdown handler not registered because Python interpreter is not running in the main thread
run pipeline stage %s
admin requested tearing down of chaiml-chaiml-sft-horror_v5
Pipeline stage MKMLDeleter completed in 3.91s
chaiml-albert-dpo-0912-v_7175_v1 status is now torndown due to DeploymentManager action
Pipeline stage %s skipped, reason=%s
Running pipeline stage MKMLDeleter
Shutdown handler de-registered
admin requested tearing down of chaiml-chaiml-sft-horror_v4
chaiml-albert-dpo-0912-v_5665_v2 status is now torndown due to DeploymentManager action
Pipeline stage MKMLDeleter completed in 5.30s
Pipeline stage %s skipped, reason=%s
Pipeline stage MKMLModelDeleter completed in 4.92s
Running pipeline stage MKMLDeleter
run pipeline stage %s
run pipeline %s
chaiml-albert-dpo-0912-v_7175_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 chaiml-chaiml-sft-horror_v5
run pipeline stage %s
Pipeline stage %s skipped, reason=%s
run pipeline %s
Running pipeline stage MKMLModelDeleter
Shutdown handler not registered because Python interpreter is not running in the main thread
Pipeline stage MKMLDeleter completed in 4.89s
admin requested tearing down of chaiml-chaiml-sft-horror_v5
Shutdown handler de-registered
chaiml-albert-dpo-0912-v_7175_v1 status is now torndown due to DeploymentManager action
Running pipeline stage MKMLDeleter
Pipeline stage %s skipped, reason=%s
run pipeline stage %s