submission_id: rica40325-sftdpo-9-15-3_v2
developer_uid: rica40325
best_of: 16
celo_rating: 1250.33
display_name: rica40325-sftdpo-9-15-3_v1
family_friendly_score: 0.0
formatter: {'memory_template': "<|start_header_id|>system<|end_header_id|>\n\n{bot_name}'s Persona: {memory}\n\n", 'prompt_template': '{prompt}<|eot_id|>', 'bot_template': '<|start_header_id|>assistant<|end_header_id|>\n\n{bot_name}: {message}<|eot_id|>', 'user_template': '<|start_header_id|>user<|end_header_id|>\n\n{user_name}: {message}<|eot_id|>', 'response_template': '<|start_header_id|>assistant<|end_header_id|>\n\n{bot_name}:', 'truncate_by_message': False}
generation_params: {'temperature': 1.0, 'top_p': 0.95, 'min_p': 0.05, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '<|end_header_id|>', '<|eot_id|>'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64}
gpu_counts: {'NVIDIA RTX A5000': 1}
is_internal_developer: False
language_model: rica40325/sftdpo_9_15_3
latencies: [{'batch_size': 1, 'throughput': 0.9012173709043176, 'latency_mean': 1.1095100009441377, 'latency_p50': 1.1124494075775146, 'latency_p90': 1.227229332923889}, {'batch_size': 4, 'throughput': 1.8157931610133429, 'latency_mean': 2.19416229724884, 'latency_p50': 2.1611311435699463, 'latency_p90': 2.4780476331710815}, {'batch_size': 5, 'throughput': 1.8989076763924944, 'latency_mean': 2.617088048458099, 'latency_p50': 2.5950149297714233, 'latency_p90': 2.932372546195984}, {'batch_size': 8, 'throughput': 2.0281006536840738, 'latency_mean': 3.9122334945201875, 'latency_p50': 3.908984661102295, 'latency_p90': 4.383821821212768}, {'batch_size': 10, 'throughput': 2.0450941788058516, 'latency_mean': 4.843518341779709, 'latency_p50': 4.80112612247467, 'latency_p90': 5.603288578987121}, {'batch_size': 12, 'throughput': 2.0647859231622325, 'latency_mean': 5.7382333433628085, 'latency_p50': 5.747334241867065, 'latency_p90': 6.647723126411438}, {'batch_size': 15, 'throughput': 2.042121949400363, 'latency_mean': 7.196369972229004, 'latency_p50': 7.276387095451355, 'latency_p90': 8.004699110984802}]
max_input_tokens: 512
max_output_tokens: 64
model_architecture: LlamaForCausalLM
model_group: rica40325/sftdpo_9_15_3
model_name: rica40325-sftdpo-9-15-3_v1
model_num_parameters: 8030261248.0
model_repo: rica40325/sftdpo_9_15_3
model_size: 8B
num_battles: 12197
num_wins: 6032
ranking_group: single
status: torndown
submission_type: basic
throughput_3p7s: 2.03
timestamp: 2024-09-18T05:46:40+00:00
us_pacific_date: 2024-09-17
win_ratio: 0.49454783963269655
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-sftdpo-9-15-3-v2-mkmlizer
Waiting for job on rica40325-sftdpo-9-15-3-v2-mkmlizer to finish
rica40325-sftdpo-9-15-3-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rica40325-sftdpo-9-15-3-v2-mkmlizer: ║ _____ __ __ ║
rica40325-sftdpo-9-15-3-v2-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
rica40325-sftdpo-9-15-3-v2-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
rica40325-sftdpo-9-15-3-v2-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
rica40325-sftdpo-9-15-3-v2-mkmlizer: ║ /___/ ║
rica40325-sftdpo-9-15-3-v2-mkmlizer: ║ ║
rica40325-sftdpo-9-15-3-v2-mkmlizer: ║ Version: 0.10.1 ║
rica40325-sftdpo-9-15-3-v2-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
rica40325-sftdpo-9-15-3-v2-mkmlizer: ║ https://mk1.ai ║
rica40325-sftdpo-9-15-3-v2-mkmlizer: ║ ║
rica40325-sftdpo-9-15-3-v2-mkmlizer: ║ The license key for the current software has been verified as ║
rica40325-sftdpo-9-15-3-v2-mkmlizer: ║ belonging to: ║
rica40325-sftdpo-9-15-3-v2-mkmlizer: ║ ║
rica40325-sftdpo-9-15-3-v2-mkmlizer: ║ Chai Research Corp. ║
rica40325-sftdpo-9-15-3-v2-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
rica40325-sftdpo-9-15-3-v2-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
rica40325-sftdpo-9-15-3-v2-mkmlizer: ║ ║
rica40325-sftdpo-9-15-3-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
Connection pool is full, discarding connection: %s. Connection pool size: %s
rica40325-sftdpo-9-15-3-v2-mkmlizer: Downloaded to shared memory in 33.784s
rica40325-sftdpo-9-15-3-v2-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpuo4dgisy, device:0
rica40325-sftdpo-9-15-3-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
rica40325-sftdpo-9-15-3-v2-mkmlizer: quantized model in 26.422s
rica40325-sftdpo-9-15-3-v2-mkmlizer: Processed model rica40325/sftdpo_9_15_3 in 60.207s
rica40325-sftdpo-9-15-3-v2-mkmlizer: creating bucket guanaco-mkml-models
rica40325-sftdpo-9-15-3-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rica40325-sftdpo-9-15-3-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rica40325-sftdpo-9-15-3-v2
rica40325-sftdpo-9-15-3-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rica40325-sftdpo-9-15-3-v2/config.json
rica40325-sftdpo-9-15-3-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rica40325-sftdpo-9-15-3-v2/special_tokens_map.json
rica40325-sftdpo-9-15-3-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rica40325-sftdpo-9-15-3-v2/tokenizer_config.json
rica40325-sftdpo-9-15-3-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rica40325-sftdpo-9-15-3-v2/tokenizer.json
rica40325-sftdpo-9-15-3-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rica40325-sftdpo-9-15-3-v2/flywheel_model.0.safetensors
rica40325-sftdpo-9-15-3-v2-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 1%|▏ | 4/291 [00:00<00:08, 34.79it/s] Loading 0: 4%|▍ | 13/291 [00:00<00:04, 59.59it/s] Loading 0: 9%|▊ | 25/291 [00:00<00:03, 75.21it/s] Loading 0: 14%|█▎ | 40/291 [00:00<00:02, 88.77it/s] Loading 0: 18%|█▊ | 52/291 [00:00<00:02, 86.11it/s] Loading 0: 21%|██ | 61/291 [00:00<00:02, 79.07it/s] Loading 0: 24%|██▍ | 70/291 [00:00<00:02, 77.66it/s] Loading 0: 27%|██▋ | 79/291 [00:01<00:02, 76.36it/s] Loading 0: 30%|██▉ | 87/291 [00:02<00:09, 20.78it/s] Loading 0: 32%|███▏ | 94/291 [00:02<00:07, 24.66it/s] Loading 0: 35%|███▌ | 103/291 [00:02<00:05, 31.49it/s] Loading 0: 38%|███▊ | 112/291 [00:02<00:04, 38.77it/s] Loading 0: 43%|████▎ | 124/291 [00:02<00:03, 47.81it/s] Loading 0: 48%|████▊ | 139/291 [00:02<00:02, 60.65it/s] Loading 0: 51%|█████ | 148/291 [00:02<00:02, 63.67it/s] Loading 0: 54%|█████▍ | 157/291 [00:03<00:02, 64.17it/s] Loading 0: 57%|█████▋ | 166/291 [00:03<00:01, 68.98it/s] Loading 0: 60%|██████ | 175/291 [00:03<00:01, 70.87it/s] Loading 0: 64%|██████▍ | 187/291 [00:04<00:04, 23.27it/s] Loading 0: 69%|██████▉ | 202/291 [00:04<00:02, 33.47it/s] Loading 0: 74%|███████▎ | 214/291 [00:04<00:01, 41.21it/s] Loading 0: 77%|███████▋ | 223/291 [00:04<00:01, 46.28it/s] Loading 0: 80%|███████▉ | 232/291 [00:04<00:01, 52.70it/s] Loading 0: 83%|████████▎ | 241/291 [00:05<00:00, 57.26it/s] Loading 0: 86%|████████▌ | 250/291 [00:05<00:00, 59.19it/s] Loading 0: 91%|█████████ | 265/291 [00:05<00:00, 71.85it/s] Loading 0: 94%|█████████▍| 274/291 [00:05<00:00, 72.52it/s] Loading 0: 97%|█████████▋| 283/291 [00:05<00:00, 71.30it/s] Loading 0: 100%|██████████| 291/291 [00:11<00:00, 5.53it/s]
Job rica40325-sftdpo-9-15-3-v2-mkmlizer completed after 84.4s with status: succeeded
Stopping job with name rica40325-sftdpo-9-15-3-v2-mkmlizer
Pipeline stage MKMLizer completed in 86.31s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.33s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service rica40325-sftdpo-9-15-3-v2
Waiting for inference service rica40325-sftdpo-9-15-3-v2 to be ready
Connection pool is full, discarding connection: %s. Connection pool size: %s
Failed to get response for submission blend_hokok_2024-09-09: ('http://neversleep-noromaid-v0-8068-v150-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', '')
Failed to get response for submission blend_hokok_2024-09-09: ('http://neversleep-noromaid-v0-8068-v150-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', '')
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
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
Inference service rica40325-sftdpo-9-15-3-v2 ready after 191.39858508110046s
Pipeline stage MKMLDeployer completed in 192.13s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.3118703365325928s
Received healthy response to inference request in 1.6476891040802002s
Received healthy response to inference request in 1.5221483707427979s
Received healthy response to inference request in 1.5743465423583984s
Received healthy response to inference request in 1.7720575332641602s
5 requests
0 failed requests
5th percentile: 1.532588005065918
10th percentile: 1.543027639389038
20th percentile: 1.5639069080352783
30th percentile: 1.5890150547027588
40th percentile: 1.6183520793914794
50th percentile: 1.6476891040802002
60th percentile: 1.697436475753784
70th percentile: 1.7471838474273682
80th percentile: 1.8800200939178469
90th percentile: 2.09594521522522
95th percentile: 2.203907775878906
99th percentile: 2.2902778244018553
mean time: 1.7656223773956299
Pipeline stage StressChecker completed in 9.85s
run pipeline stage %s
Running pipeline stage TriggerMKMLProfilingPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%s
Failed to get response for submission blend_hokok_2024-09-09: ('http://neversleep-noromaid-v0-8068-v150-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', '')
Pipeline stage TriggerMKMLProfilingPipeline completed in 5.38s
Shutdown handler de-registered
rica40325-sftdpo-9-15-3_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 0.13s
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 rica40325-sftdpo-9-15-3-v2-profiler
Waiting for inference service rica40325-sftdpo-9-15-3-v2-profiler to be ready
Inference service rica40325-sftdpo-9-15-3-v2-profiler ready after 190.42183232307434s
Pipeline stage MKMLProfilerDeployer completed in 190.79s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/rica40325-sftdpo-9-15-3-v2-profiler-predictor-00001-deploy7blr2:/code/chaiverse_profiler_1726638932 --namespace tenant-chaiml-guanaco
kubectl exec -it rica40325-sftdpo-9-15-3-v2-profiler-predictor-00001-deploy7blr2 --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1726638932 && 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_1726638932/summary.json'
kubectl exec -it rica40325-sftdpo-9-15-3-v2-profiler-predictor-00001-deploy7blr2 --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1726638932/summary.json'
Pipeline stage MKMLProfilerRunner completed in 834.63s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service rica40325-sftdpo-9-15-3-v2-profiler is running
Tearing down inference service rica40325-sftdpo-9-15-3-v2-profiler
Service rica40325-sftdpo-9-15-3-v2-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 2.06s
Shutdown handler de-registered
rica40325-sftdpo-9-15-3_v2 status is now inactive due to auto deactivation removed underperforming models
rica40325-sftdpo-9-15-3_v2 status is now torndown due to DeploymentManager action