submission_id: chaiml-albert-dpo-0912-v_9673_v1
developer_uid: chai_backend_admin
alignment_samples: 11272
alignment_score: -0.1774157303171639
best_of: 8
celo_rating: 1252.03
display_name: ai-disagree-with-user
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_ai_disagree_with_user
latencies: [{'batch_size': 1, 'throughput': 0.6079225562218308, 'latency_mean': 1.6448674631118774, 'latency_p50': 1.657462477684021, 'latency_p90': 1.8023300647735596}, {'batch_size': 3, 'throughput': 1.0787628245948764, 'latency_mean': 2.777008732557297, 'latency_p50': 2.7734360694885254, 'latency_p90': 3.0463658094406125}, {'batch_size': 5, 'throughput': 1.2352297575002273, 'latency_mean': 4.044561815261841, 'latency_p50': 4.008437752723694, 'latency_p90': 4.608502650260926}, {'batch_size': 6, 'throughput': 1.2582546068103975, 'latency_mean': 4.744158926010132, 'latency_p50': 4.813750624656677, 'latency_p90': 5.272963976860046}, {'batch_size': 8, 'throughput': 1.2321653142125635, 'latency_mean': 6.450395438671112, 'latency_p50': 6.4829347133636475, 'latency_p90': 7.3358025550842285}, {'batch_size': 10, 'throughput': 1.1960127409011645, 'latency_mean': 8.311051261425018, 'latency_p50': 8.330715417861938, 'latency_p90': 9.511400485038758}]
max_input_tokens: 1024
max_output_tokens: 64
model_architecture: MistralForCausalLM
model_group: ChaiML/albert_dpo_0912_v
model_name: ai-disagree-with-user
model_num_parameters: 12772070400.0
model_repo: ChaiML/albert_dpo_0912_virgo_edit_variety_ai_disagree_with_user
model_size: 13B
num_battles: 11272
num_wins: 5735
propriety_score: 0.7205284552845529
propriety_total_count: 984.0
ranking_group: single
status: inactive
submission_type: basic
throughput_3p7s: 1.2
timestamp: 2024-09-13T20:02:06+00:00
us_pacific_date: 2024-09-13
win_ratio: 0.5087828246983677
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 chaiml-albert-dpo-0912-v-9673-v1-mkmlizer
Waiting for job on chaiml-albert-dpo-0912-v-9673-v1-mkmlizer to finish
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: ║     _____            __           __                                ║
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: ║       /___/                                                         ║
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: ║                                                                     ║
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: ║  Version: 0.10.1                                                    ║
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: ║                                                                     ║
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: ║  belonging to:                                                      ║
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: ║                                                                     ║
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: ║                                                                     ║
chaiml-albert-dpo-0912-v-9673-v1-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
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: Downloaded to shared memory in 91.400s
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp6lg1zw2b, device:0
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
Connection pool is full, discarding connection: %s. Connection pool size: %s
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: quantized model in 40.375s
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: Processed model ChaiML/albert_dpo_0912_virgo_edit_variety_ai_disagree_with_user in 131.775s
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: creating bucket guanaco-mkml-models
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-9673-v1
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-9673-v1/config.json
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-9673-v1/special_tokens_map.json
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-9673-v1/tokenizer_config.json
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-9673-v1/tokenizer.json
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-9673-v1/flywheel_model.0.safetensors
chaiml-albert-dpo-0912-v-9673-v1-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|▏         | 5/363 [00:00<00:16, 21.95it/s]
Loading 0:   3%|▎         | 10/363 [00:00<00:12, 27.82it/s]
Loading 0:   4%|▍         | 14/363 [00:00<00:14, 24.47it/s]
Loading 0:   6%|▌         | 21/363 [00:00<00:09, 35.23it/s]
Loading 0:   7%|▋         | 26/363 [00:01<00:16, 20.14it/s]
Loading 0:   9%|▊         | 31/363 [00:01<00:13, 24.75it/s]
Loading 0:  10%|▉         | 35/363 [00:01<00:12, 26.11it/s]
Loading 0:  11%|█         | 39/363 [00:01<00:12, 26.99it/s]
Loading 0:  12%|█▏        | 43/363 [00:01<00:12, 26.46it/s]
Loading 0:  13%|█▎        | 48/363 [00:01<00:10, 28.71it/s]
Loading 0:  14%|█▍        | 52/363 [00:01<00:11, 28.11it/s]
Loading 0:  15%|█▌        | 56/363 [00:02<00:10, 29.06it/s]
Loading 0:  17%|█▋        | 61/363 [00:02<00:12, 24.63it/s]
Loading 0:  18%|█▊        | 64/363 [00:02<00:13, 21.98it/s]
Loading 0:  20%|█▉        | 71/363 [00:02<00:10, 29.13it/s]
Loading 0:  21%|██        | 75/363 [00:02<00:10, 28.71it/s]
Loading 0:  22%|██▏       | 79/363 [00:02<00:10, 27.53it/s]
Loading 0:  23%|██▎       | 84/363 [00:03<00:09, 29.75it/s]
Loading 0:  24%|██▍       | 88/363 [00:03<00:09, 27.84it/s]
Loading 0:  26%|██▌       | 93/363 [00:03<00:08, 30.35it/s]
Loading 0:  27%|██▋       | 97/363 [00:03<00:09, 29.47it/s]
Loading 0:  28%|██▊       | 101/363 [00:03<00:11, 23.55it/s]
Loading 0:  29%|██▊       | 104/363 [00:04<00:12, 21.28it/s]
Loading 0:  31%|███       | 111/363 [00:04<00:08, 28.90it/s]
Loading 0:  32%|███▏      | 115/363 [00:04<00:08, 28.83it/s]
Loading 0:  33%|███▎      | 120/363 [00:04<00:07, 30.88it/s]
Loading 0:  34%|███▍      | 124/363 [00:04<00:08, 29.65it/s]
Loading 0:  36%|███▌      | 129/363 [00:04<00:07, 31.85it/s]
Loading 0:  37%|███▋      | 133/363 [00:04<00:07, 30.15it/s]
Loading 0:  38%|███▊      | 137/363 [00:04<00:07, 29.57it/s]
Loading 0:  39%|███▉      | 142/363 [00:05<00:08, 26.04it/s]
Loading 0:  40%|███▉      | 145/363 [00:05<00:08, 24.34it/s]
Loading 0:  41%|████      | 149/363 [00:05<00:09, 22.91it/s]
Loading 0:  43%|████▎     | 156/363 [00:05<00:06, 30.06it/s]
Loading 0:  44%|████▍     | 160/363 [00:05<00:06, 29.14it/s]
Loading 0:  45%|████▌     | 165/363 [00:05<00:06, 31.50it/s]
Loading 0:  47%|████▋     | 169/363 [00:06<00:06, 29.99it/s]
Loading 0:  48%|████▊     | 174/363 [00:06<00:05, 32.32it/s]
Loading 0:  49%|████▉     | 178/363 [00:06<00:06, 30.05it/s]
Loading 0:  50%|█████     | 182/363 [00:06<00:07, 24.53it/s]
Loading 0:  51%|█████     | 185/363 [00:06<00:08, 21.93it/s]
Loading 0:  52%|█████▏    | 190/363 [00:06<00:06, 27.05it/s]
Loading 0:  53%|█████▎    | 194/363 [00:07<00:06, 24.65it/s]
Loading 0:  55%|█████▌    | 201/363 [00:07<00:05, 31.34it/s]
Loading 0:  56%|█████▋    | 205/363 [00:07<00:05, 29.52it/s]
Loading 0:  58%|█████▊    | 210/363 [00:07<00:04, 31.54it/s]
Loading 0:  59%|█████▉    | 214/363 [00:07<00:04, 30.16it/s]
Loading 0:  60%|██████    | 218/363 [00:07<00:04, 29.99it/s]
Loading 0:  61%|██████▏   | 223/363 [00:08<00:05, 25.66it/s]
Loading 0:  62%|██████▏   | 226/363 [00:08<00:05, 24.25it/s]
Loading 0:  63%|██████▎   | 230/363 [00:08<00:05, 23.31it/s]
Loading 0:  65%|██████▌   | 237/363 [00:08<00:04, 30.69it/s]
Loading 0:  66%|██████▋   | 241/363 [00:08<00:04, 28.96it/s]
Loading 0:  68%|██████▊   | 246/363 [00:08<00:03, 31.22it/s]
Loading 0:  69%|██████▉   | 250/363 [00:09<00:03, 29.32it/s]
Loading 0:  70%|███████   | 255/363 [00:09<00:03, 31.33it/s]
Loading 0:  71%|███████▏  | 259/363 [00:09<00:03, 29.43it/s]
Loading 0:  72%|███████▏  | 263/363 [00:09<00:04, 24.94it/s]
Loading 0:  73%|███████▎  | 266/363 [00:09<00:04, 21.48it/s]
Loading 0:  75%|███████▍  | 271/363 [00:09<00:03, 26.70it/s]
Loading 0:  76%|███████▌  | 275/363 [00:10<00:03, 24.83it/s]
Loading 0:  78%|███████▊  | 282/363 [00:10<00:02, 31.79it/s]
Loading 0:  79%|███████▉  | 286/363 [00:10<00:02, 29.94it/s]
Loading 0:  80%|████████  | 291/363 [00:10<00:02, 31.39it/s]
Loading 0:  81%|████████▏ | 295/363 [00:10<00:02, 30.23it/s]
Loading 0:  82%|████████▏ | 299/363 [00:10<00:02, 30.65it/s]
Loading 0:  84%|████████▎ | 304/363 [00:11<00:02, 26.76it/s]
Loading 0:  85%|████████▍ | 307/363 [00:11<00:02, 25.05it/s]
Loading 0:  86%|████████▌ | 311/363 [00:11<00:02, 23.60it/s]
Loading 0:  88%|████████▊ | 318/363 [00:11<00:01, 30.46it/s]
Loading 0:  89%|████████▊ | 322/363 [00:11<00:01, 29.25it/s]
Loading 0:  90%|█████████ | 327/363 [00:11<00:01, 31.95it/s]
Loading 0:  91%|█████████ | 331/363 [00:11<00:01, 30.72it/s]
Loading 0:  93%|█████████▎| 336/363 [00:12<00:00, 32.74it/s]
Loading 0:  94%|█████████▎| 340/363 [00:12<00:00, 31.59it/s]
Loading 0:  95%|█████████▍| 344/363 [00:19<00:09,  1.99it/s]
Loading 0:  96%|█████████▌| 348/363 [00:19<00:05,  2.67it/s]
Loading 0:  97%|█████████▋| 353/363 [00:19<00:02,  3.88it/s]
Loading 0:  98%|█████████▊| 357/363 [00:19<00:01,  5.00it/s]
                                                            

Job chaiml-albert-dpo-0912-v-9673-v1-mkmlizer completed after 156.71s with status: succeeded
Stopping job with name chaiml-albert-dpo-0912-v-9673-v1-mkmlizer
Pipeline stage MKMLizer completed in 157.69s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.12s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service chaiml-albert-dpo-0912-v-9673-v1
Waiting for inference service chaiml-albert-dpo-0912-v-9673-v1 to be ready
Inference service chaiml-albert-dpo-0912-v-9673-v1 ready after 191.3658103942871s
Pipeline stage MKMLDeployer completed in 191.73s
run pipeline stage %s
Running pipeline stage StressChecker
Failed to get response for submission chaiml-elo-alignment-run-3_v41: ('http://chaiml-elo-alignment-run-3-v41-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:53738->127.0.0.1:8080: read: connection reset by peer\n')
Received healthy response to inference request in 3.0198090076446533s
Received healthy response to inference request in 1.3328757286071777s
Received healthy response to inference request in 1.6334125995635986s
Received healthy response to inference request in 1.853745460510254s
Received healthy response to inference request in 1.836479663848877s
5 requests
0 failed requests
5th percentile: 1.3929831027984618
10th percentile: 1.4530904769897461
20th percentile: 1.5733052253723145
30th percentile: 1.6740260124206543
40th percentile: 1.7552528381347656
50th percentile: 1.836479663848877
60th percentile: 1.8433859825134278
70th percentile: 1.8502923011779786
80th percentile: 2.086958169937134
90th percentile: 2.5533835887908936
95th percentile: 2.7865962982177734
99th percentile: 2.9731664657592773
mean time: 1.935264492034912
Pipeline stage StressChecker completed in 10.88s
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 6.53s
Shutdown handler de-registered
chaiml-albert-dpo-0912-v_9673_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 0.12s
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-9673-v1-profiler
Waiting for inference service chaiml-albert-dpo-0912-v-9673-v1-profiler to be ready
Inference service chaiml-albert-dpo-0912-v-9673-v1-profiler ready after 190.4316487312317s
Pipeline stage MKMLProfilerDeployer completed in 190.80s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/chaiml-albert-dpo-096f5bd9e760888272b4e632aa5e4e2b94-deplo7kw4g:/code/chaiverse_profiler_1726258331 --namespace tenant-chaiml-guanaco
kubectl exec -it chaiml-albert-dpo-096f5bd9e760888272b4e632aa5e4e2b94-deplo7kw4g --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1726258331 && 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_1726258331/summary.json'
kubectl exec -it chaiml-albert-dpo-096f5bd9e760888272b4e632aa5e4e2b94-deplo7kw4g --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1726258331/summary.json'
Pipeline stage MKMLProfilerRunner completed in 1169.74s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service chaiml-albert-dpo-0912-v-9673-v1-profiler is running
Tearing down inference service chaiml-albert-dpo-0912-v-9673-v1-profiler
Service chaiml-albert-dpo-0912-v-9673-v1-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 2.22s
Shutdown handler de-registered
chaiml-albert-dpo-0912-v_9673_v1 status is now inactive due to auto deactivation removed underperforming models