submission_id: riverise-feedback-dpo-merged_v1
developer_uid: Riverise
alignment_samples: 10963
alignment_score: 1.0946511490526172
best_of: 16
celo_rating: 1061.28
display_name: riverise-feedback-dpo-merged_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': 1.0, '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: Riverise/feedback_dpo_merged
latencies: [{'batch_size': 1, 'throughput': 0.8981030187545916, 'latency_mean': 1.1133566474914551, 'latency_p50': 1.1194324493408203, 'latency_p90': 1.2448061943054198}, {'batch_size': 4, 'throughput': 1.7981744673093818, 'latency_mean': 2.2189607203006743, 'latency_p50': 2.243447184562683, 'latency_p90': 2.4765761613845827}, {'batch_size': 5, 'throughput': 1.8745378882816082, 'latency_mean': 2.647966785430908, 'latency_p50': 2.6424962282180786, 'latency_p90': 2.9734248399734495}, {'batch_size': 8, 'throughput': 1.9949388872484262, 'latency_mean': 3.977656475305557, 'latency_p50': 4.017573356628418, 'latency_p90': 4.475029420852661}, {'batch_size': 10, 'throughput': 2.0183863060903953, 'latency_mean': 4.911407153606415, 'latency_p50': 4.921507835388184, 'latency_p90': 5.59112331867218}, {'batch_size': 12, 'throughput': 2.0178312599972137, 'latency_mean': 5.872041561603546, 'latency_p50': 5.8955196142196655, 'latency_p90': 6.800777244567871}, {'batch_size': 15, 'throughput': 2.0149311913159895, 'latency_mean': 7.289120084047317, 'latency_p50': 7.386255979537964, 'latency_p90': 8.09388837814331}]
max_input_tokens: 512
max_output_tokens: 64
model_architecture: LlamaForCausalLM
model_group: Riverise/feedback_dpo_me
model_name: riverise-feedback-dpo-merged_v1
model_num_parameters: 8030261248.0
model_repo: Riverise/feedback_dpo_merged
model_size: 8B
num_battles: 10963
num_wins: 3048
propriety_score: 0.6772486772486772
propriety_total_count: 1134.0
ranking_group: single
status: inactive
submission_type: basic
throughput_3p7s: 1.99
timestamp: 2024-09-06T12:20:56+00:00
us_pacific_date: 2024-09-06
win_ratio: 0.27802608774970355
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 riverise-feedback-dpo-merged-v1-mkmlizer
Waiting for job on riverise-feedback-dpo-merged-v1-mkmlizer to finish
riverise-feedback-dpo-merged-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
riverise-feedback-dpo-merged-v1-mkmlizer: ║     _____            __           __                                ║
riverise-feedback-dpo-merged-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
riverise-feedback-dpo-merged-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
riverise-feedback-dpo-merged-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
riverise-feedback-dpo-merged-v1-mkmlizer: ║       /___/                                                         ║
riverise-feedback-dpo-merged-v1-mkmlizer: ║                                                                     ║
riverise-feedback-dpo-merged-v1-mkmlizer: ║  Version: 0.10.1                                                    ║
riverise-feedback-dpo-merged-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
riverise-feedback-dpo-merged-v1-mkmlizer: ║  https://mk1.ai                                                     ║
riverise-feedback-dpo-merged-v1-mkmlizer: ║                                                                     ║
riverise-feedback-dpo-merged-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
riverise-feedback-dpo-merged-v1-mkmlizer: ║  belonging to:                                                      ║
riverise-feedback-dpo-merged-v1-mkmlizer: ║                                                                     ║
riverise-feedback-dpo-merged-v1-mkmlizer: ║  Chai Research Corp.                                                ║
riverise-feedback-dpo-merged-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
riverise-feedback-dpo-merged-v1-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
riverise-feedback-dpo-merged-v1-mkmlizer: ║                                                                     ║
riverise-feedback-dpo-merged-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
Failed to get response for submission zonemercy-base-story-v1_v3: ('http://zonemercy-base-story-v1-v3-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\')]"}')
Failed to get response for submission zonemercy-base-story-v1_v3: ('http://zonemercy-base-story-v1-v3-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'upstream connect error or disconnect/reset before headers. reset reason: connection timeout')
riverise-feedback-dpo-merged-v1-mkmlizer: Downloaded to shared memory in 68.009s
riverise-feedback-dpo-merged-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpkx6k086c, device:0
riverise-feedback-dpo-merged-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
riverise-feedback-dpo-merged-v1-mkmlizer: quantized model in 28.783s
riverise-feedback-dpo-merged-v1-mkmlizer: Processed model Riverise/feedback_dpo_merged in 96.792s
riverise-feedback-dpo-merged-v1-mkmlizer: creating bucket guanaco-mkml-models
riverise-feedback-dpo-merged-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
riverise-feedback-dpo-merged-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/riverise-feedback-dpo-merged-v1
Failed to get response for submission zonemercy-base-story-v1_v3: ('http://zonemercy-base-story-v1-v3-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'upstream connect error or disconnect/reset before headers. reset reason: connection timeout')
Failed to get response for submission zonemercy-base-story-v1_v2: ('http://zonemercy-base-story-v1-v2-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\')]"}')
Failed to get response for submission zonemercy-base-story-v1_v3: ('http://zonemercy-base-story-v1-v3-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\')]"}')
riverise-feedback-dpo-merged-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/riverise-feedback-dpo-merged-v1/tokenizer.json
riverise-feedback-dpo-merged-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/riverise-feedback-dpo-merged-v1/flywheel_model.0.safetensors
riverise-feedback-dpo-merged-v1-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 5/291 [00:00<00:10, 26.44it/s]
Loading 0:   4%|▍         | 12/291 [00:00<00:07, 36.83it/s]
Loading 0:   5%|▌         | 16/291 [00:00<00:08, 33.76it/s]
Loading 0:   7%|▋         | 21/291 [00:00<00:07, 36.93it/s]
Loading 0:   9%|▊         | 25/291 [00:00<00:07, 33.55it/s]
Loading 0:  11%|█         | 31/291 [00:00<00:06, 40.31it/s]
Loading 0:  12%|█▏        | 36/291 [00:01<00:10, 23.55it/s]
Loading 0:  14%|█▍        | 41/291 [00:01<00:09, 25.03it/s]
Loading 0:  16%|█▋        | 48/291 [00:01<00:07, 31.85it/s]
Loading 0:  18%|█▊        | 52/291 [00:01<00:07, 30.71it/s]
Loading 0:  20%|█▉        | 57/291 [00:01<00:07, 32.84it/s]
Loading 0:  21%|██        | 61/291 [00:01<00:07, 32.32it/s]
Loading 0:  23%|██▎       | 66/291 [00:02<00:06, 35.34it/s]
Loading 0:  24%|██▍       | 70/291 [00:02<00:06, 33.42it/s]
Loading 0:  25%|██▌       | 74/291 [00:02<00:06, 33.35it/s]
Loading 0:  27%|██▋       | 78/291 [00:02<00:06, 33.83it/s]
Loading 0:  28%|██▊       | 82/291 [00:02<00:08, 23.99it/s]
Loading 0:  29%|██▉       | 85/291 [00:02<00:08, 24.75it/s]
Loading 0:  31%|███       | 90/291 [00:02<00:06, 29.31it/s]
Loading 0:  32%|███▏      | 94/291 [00:03<00:06, 29.42it/s]
Loading 0:  34%|███▍      | 99/291 [00:03<00:05, 33.55it/s]
Loading 0:  35%|███▌      | 103/291 [00:03<00:05, 31.40it/s]
Loading 0:  37%|███▋      | 108/291 [00:03<00:05, 33.83it/s]
Loading 0:  38%|███▊      | 112/291 [00:03<00:05, 31.65it/s]
Loading 0:  40%|███▉      | 116/291 [00:03<00:05, 31.14it/s]
Loading 0:  42%|████▏     | 122/291 [00:03<00:04, 35.86it/s]
Loading 0:  44%|████▎     | 127/291 [00:04<00:04, 33.69it/s]
Loading 0:  46%|████▌     | 133/291 [00:04<00:05, 29.87it/s]
Loading 0:  47%|████▋     | 137/291 [00:04<00:05, 30.29it/s]
Loading 0:  48%|████▊     | 141/291 [00:04<00:05, 28.92it/s]
Loading 0:  51%|█████     | 147/291 [00:04<00:04, 34.24it/s]
Loading 0:  52%|█████▏    | 151/291 [00:04<00:04, 32.83it/s]
Loading 0:  54%|█████▎    | 156/291 [00:04<00:03, 35.21it/s]
Loading 0:  55%|█████▍    | 160/291 [00:05<00:03, 34.49it/s]
Loading 0:  57%|█████▋    | 165/291 [00:05<00:03, 37.70it/s]
Loading 0:  58%|█████▊    | 169/291 [00:05<00:03, 34.26it/s]
Loading 0:  60%|█████▉    | 174/291 [00:05<00:03, 36.13it/s]
Loading 0:  61%|██████    | 178/291 [00:05<00:03, 35.12it/s]
Loading 0:  63%|██████▎   | 182/291 [00:05<00:03, 36.32it/s]
Loading 0:  64%|██████▍   | 186/291 [00:05<00:03, 26.77it/s]
Loading 0:  65%|██████▌   | 190/291 [00:06<00:03, 25.82it/s]
Loading 0:  67%|██████▋   | 194/291 [00:06<00:03, 24.51it/s]
Loading 0:  69%|██████▉   | 201/291 [00:06<00:02, 31.67it/s]
Loading 0:  70%|███████   | 205/291 [00:06<00:02, 31.50it/s]
Loading 0:  72%|███████▏  | 210/291 [00:06<00:02, 34.80it/s]
Loading 0:  74%|███████▎  | 214/291 [00:06<00:02, 33.56it/s]
Loading 0:  75%|███████▌  | 219/291 [00:06<00:01, 36.62it/s]
Loading 0:  77%|███████▋  | 223/291 [00:06<00:01, 34.75it/s]
Loading 0:  78%|███████▊  | 227/291 [00:07<00:01, 35.20it/s]
Loading 0:  79%|███████▉  | 231/291 [00:07<00:01, 35.03it/s]
Loading 0:  81%|████████  | 235/291 [00:07<00:02, 25.90it/s]
Loading 0:  82%|████████▏ | 239/291 [00:07<00:02, 25.52it/s]
Loading 0:  85%|████████▍ | 246/291 [00:07<00:01, 33.79it/s]
Loading 0:  86%|████████▌ | 250/291 [00:07<00:01, 32.02it/s]
Loading 0:  88%|████████▊ | 255/291 [00:08<00:01, 34.68it/s]
Loading 0:  89%|████████▉ | 259/291 [00:08<00:00, 32.89it/s]
Loading 0:  91%|█████████ | 264/291 [00:08<00:00, 36.45it/s]
Loading 0:  92%|█████████▏| 268/291 [00:08<00:00, 33.59it/s]
Loading 0:  94%|█████████▍| 273/291 [00:08<00:00, 34.65it/s]
Loading 0:  95%|█████████▌| 277/291 [00:08<00:00, 33.63it/s]
Loading 0:  97%|█████████▋| 281/291 [00:08<00:00, 34.52it/s]
Loading 0:  98%|█████████▊| 286/291 [00:14<00:01,  2.61it/s]
Loading 0:  99%|█████████▉| 289/291 [00:14<00:00,  3.24it/s]
                                                            

Job riverise-feedback-dpo-merged-v1-mkmlizer completed after 115.66s with status: succeeded
Stopping job with name riverise-feedback-dpo-merged-v1-mkmlizer
Pipeline stage MKMLizer completed in 117.62s
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 riverise-feedback-dpo-merged-v1
Waiting for inference service riverise-feedback-dpo-merged-v1 to be ready
Failed to get response for submission zonemercy-lexical-nemo-_1518_v23: ('http://zonemercy-lexical-nemo-1518-v23-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\')]"}')
Failed to get response for submission zonemercy-lexical-nemo-_1518_v23: ('http://zonemercy-lexical-nemo-1518-v23-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 riverise-feedback-dpo-merged-v1 ready after 150.6088285446167s
Pipeline stage MKMLDeployer completed in 151.14s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.017017126083374s
Received healthy response to inference request in 0.8930490016937256s
Received healthy response to inference request in 1.2840838432312012s
Received healthy response to inference request in 1.800973653793335s
Received healthy response to inference request in 1.7653710842132568s
5 requests
0 failed requests
5th percentile: 0.9712559700012207
10th percentile: 1.0494629383087157
20th percentile: 1.205876874923706
30th percentile: 1.3803412914276123
40th percentile: 1.5728561878204346
50th percentile: 1.7653710842132568
60th percentile: 1.7796121120452881
70th percentile: 1.7938531398773194
80th percentile: 1.8441823482513429
90th percentile: 1.9305997371673584
95th percentile: 1.9738084316253661
99th percentile: 2.0083753871917724
mean time: 1.5520989418029785
Pipeline stage StressChecker completed in 10.20s
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 5.98s
Shutdown handler de-registered
riverise-feedback-dpo-merged_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 riverise-feedback-dpo-merged-v1-profiler
Waiting for inference service riverise-feedback-dpo-merged-v1-profiler to be ready
Inference service riverise-feedback-dpo-merged-v1-profiler ready after 140.40798115730286s
Pipeline stage MKMLProfilerDeployer completed in 140.75s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/riverise-feedback-dp64040ed02d9d1301c21b96cb41269068-deploc8pl5:/code/chaiverse_profiler_1725625720 --namespace tenant-chaiml-guanaco
kubectl exec -it riverise-feedback-dp64040ed02d9d1301c21b96cb41269068-deploc8pl5 --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1725625720 && 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_1725625720/summary.json'
kubectl exec -it riverise-feedback-dp64040ed02d9d1301c21b96cb41269068-deploc8pl5 --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1725625720/summary.json'
Pipeline stage MKMLProfilerRunner completed in 843.67s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service riverise-feedback-dpo-merged-v1-profiler is running
Tearing down inference service riverise-feedback-dpo-merged-v1-profiler
Service riverise-feedback-dpo-merged-v1-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.56s
Shutdown handler de-registered
riverise-feedback-dpo-merged_v1 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics