arcee-ai-llama-3-1-super_4127

submission_id: arcee-ai-llama-3-1-super_4127_v1

developer_uid: Meliodia

alignment_samples: 10611

alignment_score: 1.6232932356162366

best_of: 1

celo_rating: 1159.58

display_name: arcee-ai-llama-3-1-super_4127_v1

formatter: {'memory_template': '<|im_start|>system\n{memory}<|im_end|>\n', 'prompt_template': '<|im_start|>user\n{prompt}<|im_end|>\n', 'bot_template': '<|im_start|>assistant\n{bot_name}: {message}<|im_end|>\n', 'user_template': '<|im_start|>user\n{user_name}: {message}<|im_end|>\n', 'response_template': '<|im_start|>assistant\n{bot_name}:', 'truncate_by_message': True}

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': 1024, 'best_of': 1, 'max_output_tokens': 64}

gpu_counts: {'NVIDIA RTX A5000': 1}

is_internal_developer: True

language_model: arcee-ai/Llama-3.1-SuperNova-Lite

latencies: [{'batch_size': 1, 'throughput': 0.9376956182004872, 'latency_mean': 1.0663320910930634, 'latency_p50': 1.0741099119186401, 'latency_p90': 1.1801533937454223}, {'batch_size': 5, 'throughput': 2.591231207942671, 'latency_mean': 1.9212498843669892, 'latency_p50': 1.930618166923523, 'latency_p90': 2.164945888519287}, {'batch_size': 10, 'throughput': 3.386692318395994, 'latency_mean': 2.9319711148738863, 'latency_p50': 2.9372304677963257, 'latency_p90': 3.3167243003845215}, {'batch_size': 15, 'throughput': 3.7644668086789768, 'latency_mean': 3.9432926201820373, 'latency_p50': 3.928479552268982, 'latency_p90': 4.47234456539154}, {'batch_size': 20, 'throughput': 3.8945623755613736, 'latency_mean': 5.063257586956024, 'latency_p50': 5.056084513664246, 'latency_p90': 5.652053356170654}]

max_input_tokens: 1024

max_output_tokens: 64

model_architecture: LlamaForCausalLM

model_group: arcee-ai/Llama-3.1-Super

model_name: arcee-ai-llama-3-1-super_4127_v1

model_num_parameters: 8030261248.0

model_repo: arcee-ai/Llama-3.1-SuperNova-Lite

model_size: 8B

num_battles: 10605

num_wins: 4061

propriety_score: 0.7588294651866802

propriety_total_count: 991.0

ranking_group: single

status: inactive

submission_type: basic

throughput_3p7s: 3.73

timestamp: 2024-09-10T21:36:42+00:00

us_pacific_date: 2024-09-10

win_ratio: 0.3829325789721829

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 arcee-ai-llama-3-1-super-4127-v1-mkmlizer
Waiting for job on arcee-ai-llama-3-1-super-4127-v1-mkmlizer to finish
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: ║     _____            __           __                                ║
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: ║       /___/                                                         ║
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: ║                                                                     ║
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: ║  Version: 0.10.1                                                    ║
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: ║  https://mk1.ai                                                     ║
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: ║                                                                     ║
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: ║  belonging to:                                                      ║
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: ║                                                                     ║
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: ║  Chai Research Corp.                                                ║
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: ║                                                                     ║
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
Failed to get response for submission blend_puheb_2024-09-09: ('http://chaiml-llama-8b-pairwis-8189-v19-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:58580->127.0.0.1:8080: read: connection reset by peer\n')
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: Downloaded to shared memory in 33.017s
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpd4i1ic_9, device:0
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: quantized model in 26.822s
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: Processed model arcee-ai/Llama-3.1-SuperNova-Lite in 59.839s
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: creating bucket guanaco-mkml-models
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/arcee-ai-llama-3-1-super-4127-v1
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/arcee-ai-llama-3-1-super-4127-v1/config.json
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/arcee-ai-llama-3-1-super-4127-v1/special_tokens_map.json
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/arcee-ai-llama-3-1-super-4127-v1/tokenizer_config.json
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/arcee-ai-llama-3-1-super-4127-v1/tokenizer.json
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/arcee-ai-llama-3-1-super-4127-v1/flywheel_model.0.safetensors
arcee-ai-llama-3-1-super-4127-v1-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 5/291 [00:00<00:08, 33.92it/s]
Loading 0:   4%|▍         | 12/291 [00:00<00:05, 51.23it/s]
Loading 0:   6%|▌         | 18/291 [00:00<00:05, 51.96it/s]
Loading 0:   8%|▊         | 24/291 [00:00<00:06, 43.52it/s]
Loading 0:  11%|█         | 31/291 [00:00<00:05, 50.09it/s]
Loading 0:  13%|█▎        | 37/291 [00:00<00:05, 48.49it/s]
Loading 0:  15%|█▍        | 43/291 [00:00<00:05, 48.26it/s]
Loading 0:  17%|█▋        | 49/291 [00:01<00:04, 49.62it/s]
Loading 0:  19%|█▉        | 55/291 [00:01<00:05, 45.53it/s]
Loading 0:  21%|██        | 60/291 [00:01<00:05, 43.72it/s]
Loading 0:  23%|██▎       | 66/291 [00:01<00:04, 47.04it/s]
Loading 0:  24%|██▍       | 71/291 [00:01<00:04, 47.42it/s]
Loading 0:  26%|██▌       | 76/291 [00:01<00:04, 46.72it/s]
Loading 0:  28%|██▊       | 82/291 [00:01<00:04, 44.62it/s]
Loading 0:  30%|██▉       | 87/291 [00:02<00:06, 30.99it/s]
Loading 0:  32%|███▏      | 94/291 [00:02<00:05, 38.19it/s]
Loading 0:  34%|███▍      | 99/291 [00:02<00:04, 40.26it/s]
Loading 0:  36%|███▌      | 104/291 [00:02<00:05, 35.16it/s]
Loading 0:  38%|███▊      | 111/291 [00:02<00:04, 40.98it/s]
Loading 0:  40%|███▉      | 116/291 [00:02<00:04, 41.99it/s]
Loading 0:  42%|████▏     | 121/291 [00:02<00:03, 43.44it/s]
Loading 0:  44%|████▎     | 127/291 [00:02<00:03, 42.23it/s]
Loading 0:  45%|████▌     | 132/291 [00:03<00:03, 43.42it/s]
Loading 0:  48%|████▊     | 139/291 [00:03<00:03, 49.79it/s]
Loading 0:  50%|████▉     | 145/291 [00:03<00:03, 44.81it/s]
Loading 0:  52%|█████▏    | 150/291 [00:03<00:03, 42.97it/s]
Loading 0:  54%|█████▎    | 156/291 [00:03<00:02, 46.54it/s]
Loading 0:  55%|█████▌    | 161/291 [00:03<00:02, 45.18it/s]
Loading 0:  57%|█████▋    | 166/291 [00:03<00:02, 43.41it/s]
Loading 0:  59%|█████▉    | 172/291 [00:03<00:02, 41.98it/s]
Loading 0:  61%|██████    | 177/291 [00:04<00:02, 43.55it/s]
Loading 0:  63%|██████▎   | 182/291 [00:04<00:02, 41.41it/s]
Loading 0:  64%|██████▍   | 187/291 [00:04<00:03, 31.76it/s]
Loading 0:  66%|██████▌   | 191/291 [00:04<00:03, 32.67it/s]
Loading 0:  67%|██████▋   | 195/291 [00:04<00:02, 32.50it/s]
Loading 0:  69%|██████▉   | 201/291 [00:04<00:02, 38.08it/s]
Loading 0:  71%|███████   | 206/291 [00:04<00:02, 38.78it/s]
Loading 0:  73%|███████▎  | 211/291 [00:05<00:01, 40.30it/s]
Loading 0:  75%|███████▍  | 217/291 [00:05<00:01, 39.95it/s]
Loading 0:  76%|███████▋  | 222/291 [00:05<00:01, 40.54it/s]
Loading 0:  79%|███████▊  | 229/291 [00:05<00:01, 47.10it/s]
Loading 0:  81%|████████  | 235/291 [00:05<00:01, 45.03it/s]
Loading 0:  82%|████████▏ | 240/291 [00:05<00:01, 42.97it/s]
Loading 0:  85%|████████▍ | 246/291 [00:05<00:00, 46.06it/s]
Loading 0:  86%|████████▋ | 251/291 [00:05<00:00, 45.37it/s]
Loading 0:  88%|████████▊ | 256/291 [00:06<00:00, 43.55it/s]
Loading 0:  90%|█████████ | 262/291 [00:06<00:00, 42.73it/s]
Loading 0:  92%|█████████▏| 267/291 [00:06<00:00, 42.60it/s]
Loading 0:  94%|█████████▍| 273/291 [00:06<00:00, 45.41it/s]
Loading 0:  96%|█████████▌| 278/291 [00:06<00:00, 45.71it/s]
Loading 0:  97%|█████████▋| 283/291 [00:06<00:00, 38.41it/s]
Loading 0:  99%|█████████▉| 288/291 [00:12<00:01,  3.00it/s]
                                                            
Job arcee-ai-llama-3-1-super-4127-v1-mkmlizer completed after 86.13s with status: succeeded
Stopping job with name arcee-ai-llama-3-1-super-4127-v1-mkmlizer
Pipeline stage MKMLizer completed in 87.36s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.08s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service arcee-ai-llama-3-1-super-4127-v1
Waiting for inference service arcee-ai-llama-3-1-super-4127-v1 to be ready
Failed to get response for submission chaiml-llama-8b-pairwis_8189_v19: ('http://chaiml-elo-alignment-run-3-v36-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'request timeout')
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 arcee-ai-llama-3-1-super-4127-v1 ready after 160.54927325248718s
Pipeline stage MKMLDeployer completed in 162.15s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.360941171646118s
Received healthy response to inference request in 1.8075048923492432s
Received healthy response to inference request in 1.4406993389129639s
Received healthy response to inference request in 1.3432343006134033s
Received healthy response to inference request in 1.8813962936401367s
5 requests
0 failed requests
5th percentile: 1.3627273082733153
10th percentile: 1.3822203159332276
20th percentile: 1.4212063312530518
30th percentile: 1.5140604496002197
40th percentile: 1.6607826709747315
50th percentile: 1.8075048923492432
60th percentile: 1.8370614528656006
70th percentile: 1.866618013381958
80th percentile: 1.977305269241333
90th percentile: 2.1691232204437254
95th percentile: 2.265032196044922
99th percentile: 2.341759376525879
mean time: 1.7667551994323731
Pipeline stage StressChecker completed in 13.19s
run pipeline stage %s
Running pipeline stage TriggerMKMLProfilingPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Pipeline stage TriggerMKMLProfilingPipeline completed in 6.94s
Shutdown handler de-registered
arcee-ai-llama-3-1-super_4127_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 0.13s
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 arcee-ai-llama-3-1-super-4127-v1-profiler
Waiting for inference service arcee-ai-llama-3-1-super-4127-v1-profiler to be ready
Inference service arcee-ai-llama-3-1-super-4127-v1-profiler ready after 150.39101886749268s
Pipeline stage MKMLProfilerDeployer completed in 150.73s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/arcee-ai-llama-3-1-s5a3a92784e40a226b49001edd1172163-deplo5zbtf:/code/chaiverse_profiler_1726004666 --namespace tenant-chaiml-guanaco
kubectl exec -it arcee-ai-llama-3-1-s5a3a92784e40a226b49001edd1172163-deplo5zbtf --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1726004666 && python profiles.py profile --best_of_n 1 --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_1726004666/summary.json'
kubectl exec -it arcee-ai-llama-3-1-s5a3a92784e40a226b49001edd1172163-deplo5zbtf --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1726004666/summary.json'
Pipeline stage MKMLProfilerRunner completed in 461.80s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service arcee-ai-llama-3-1-super-4127-v1-profiler is running
Tearing down inference service arcee-ai-llama-3-1-super-4127-v1-profiler
Service arcee-ai-llama-3-1-super-4127-v1-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.72s
Shutdown handler de-registered
arcee-ai-llama-3-1-super_4127_v1 status is now inactive due to auto deactivation removed underperforming models