meta-llama-meta-llama-3-_5386

submission_id: meta-llama-meta-llama-3-_5386_v7

developer_uid: chai_backend_admin

best_of: 8

celo_rating: 1219.64

display_name: meta-llama-meta-llama-3-_5386_v7

family_friendly_score: 0.5841476655808904

family_friendly_standard_error: 0.011474087714123535

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': 1024, 'best_of': 8, 'max_output_tokens': 64}

gpu_counts: {'NVIDIA RTX A5000': 1}

ineligible_reason: num_battles<5000

is_internal_developer: True

language_model: meta-llama/Meta-Llama-3-8B-Instruct

latencies: [{'batch_size': 1, 'throughput': 0.8766275570825298, 'latency_mean': 1.1406641232967376, 'latency_p50': 1.1441845893859863, 'latency_p90': 1.2602874279022216}, {'batch_size': 4, 'throughput': 1.878021935645135, 'latency_mean': 2.1210211515426636, 'latency_p50': 2.1290464401245117, 'latency_p90': 2.3315988779067993}, {'batch_size': 5, 'throughput': 2.029576160684928, 'latency_mean': 2.4530843138694762, 'latency_p50': 2.4544864892959595, 'latency_p90': 2.7535692691802978}, {'batch_size': 8, 'throughput': 2.244699534813585, 'latency_mean': 3.5351341342926026, 'latency_p50': 3.535127878189087, 'latency_p90': 3.963687539100647}, {'batch_size': 10, 'throughput': 2.304861724831623, 'latency_mean': 4.301239175796509, 'latency_p50': 4.336079955101013, 'latency_p90': 4.781917691230774}, {'batch_size': 12, 'throughput': 2.3338026065817665, 'latency_mean': 5.097603951692581, 'latency_p50': 5.155501484870911, 'latency_p90': 5.829752206802368}, {'batch_size': 15, 'throughput': 2.357655656656277, 'latency_mean': 6.2800760281085966, 'latency_p50': 6.28175675868988, 'latency_p90': 7.035756802558899}]

max_input_tokens: 1024

max_output_tokens: 64

model_architecture: LlamaForCausalLM

model_group: meta-llama/Meta-Llama-3-

model_name: meta-llama-meta-llama-3-_5386_v7

model_num_parameters: 8030261248.0

model_repo: meta-llama/Meta-Llama-3-8B-Instruct

model_size: 8B

num_battles: 4001

num_wins: 1826

ranking_group: single

status: torndown

submission_type: basic

throughput_3p7s: 2.28

timestamp: 2024-09-25T02:45:31+00:00

us_pacific_date: 2024-09-24

win_ratio: 0.45638590352411895

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 meta-llama-meta-llama-3-5386-v7-mkmlizer
Waiting for job on meta-llama-meta-llama-3-5386-v7-mkmlizer to finish
meta-llama-meta-llama-3-5386-v7-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
meta-llama-meta-llama-3-5386-v7-mkmlizer: ║     _____            __           __                                ║
meta-llama-meta-llama-3-5386-v7-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
meta-llama-meta-llama-3-5386-v7-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
meta-llama-meta-llama-3-5386-v7-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
meta-llama-meta-llama-3-5386-v7-mkmlizer: ║       /___/                                                         ║
meta-llama-meta-llama-3-5386-v7-mkmlizer: ║                                                                     ║
meta-llama-meta-llama-3-5386-v7-mkmlizer: ║  Version: 0.11.12                                                   ║
meta-llama-meta-llama-3-5386-v7-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
meta-llama-meta-llama-3-5386-v7-mkmlizer: ║  https://mk1.ai                                                     ║
meta-llama-meta-llama-3-5386-v7-mkmlizer: ║                                                                     ║
meta-llama-meta-llama-3-5386-v7-mkmlizer: ║  The license key for the current software has been verified as      ║
meta-llama-meta-llama-3-5386-v7-mkmlizer: ║  belonging to:                                                      ║
meta-llama-meta-llama-3-5386-v7-mkmlizer: ║                                                                     ║
meta-llama-meta-llama-3-5386-v7-mkmlizer: ║  Chai Research Corp.                                                ║
meta-llama-meta-llama-3-5386-v7-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
meta-llama-meta-llama-3-5386-v7-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
meta-llama-meta-llama-3-5386-v7-mkmlizer: ║                                                                     ║
meta-llama-meta-llama-3-5386-v7-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
Connection pool is full, discarding connection: %s. Connection pool size: %s
meta-llama-meta-llama-3-5386-v7-mkmlizer: Downloaded to shared memory in 49.344s
meta-llama-meta-llama-3-5386-v7-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpo5yqrrpf, device:0
meta-llama-meta-llama-3-5386-v7-mkmlizer: Saving flywheel model at /dev/shm/model_cache
meta-llama-meta-llama-3-5386-v7-mkmlizer: quantized model in 25.949s
meta-llama-meta-llama-3-5386-v7-mkmlizer: Processed model meta-llama/Meta-Llama-3-8B-Instruct in 75.293s
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
meta-llama-meta-llama-3-5386-v7-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
meta-llama-meta-llama-3-5386-v7-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/meta-llama-meta-llama-3-5386-v7
meta-llama-meta-llama-3-5386-v7-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/meta-llama-meta-llama-3-5386-v7/config.json
meta-llama-meta-llama-3-5386-v7-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/meta-llama-meta-llama-3-5386-v7/special_tokens_map.json
meta-llama-meta-llama-3-5386-v7-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/meta-llama-meta-llama-3-5386-v7/tokenizer_config.json
meta-llama-meta-llama-3-5386-v7-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/meta-llama-meta-llama-3-5386-v7/tokenizer.json
meta-llama-meta-llama-3-5386-v7-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/meta-llama-meta-llama-3-5386-v7/flywheel_model.0.safetensors
meta-llama-meta-llama-3-5386-v7-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 5/291 [00:00<00:08, 33.08it/s]
Loading 0:   5%|▍         | 14/291 [00:00<00:05, 46.60it/s]
Loading 0:   8%|▊         | 23/291 [00:00<00:05, 51.34it/s]
Loading 0:  11%|█         | 32/291 [00:00<00:04, 53.67it/s]
Loading 0:  14%|█▍        | 41/291 [00:00<00:04, 54.90it/s]
Loading 0:  17%|█▋        | 49/291 [00:00<00:03, 60.73it/s]
Loading 0:  19%|█▉        | 56/291 [00:01<00:03, 59.52it/s]
Loading 0:  22%|██▏       | 63/291 [00:01<00:03, 59.53it/s]
Loading 0:  24%|██▍       | 70/291 [00:01<00:04, 51.81it/s]
Loading 0:  26%|██▋       | 77/291 [00:01<00:04, 49.11it/s]
Loading 0:  29%|██▊       | 83/291 [00:01<00:05, 37.74it/s]
Loading 0:  30%|███       | 88/291 [00:01<00:05, 38.22it/s]
Loading 0:  33%|███▎      | 95/291 [00:02<00:04, 39.82it/s]
Loading 0:  35%|███▌      | 103/291 [00:02<00:03, 48.09it/s]
Loading 0:  37%|███▋      | 109/291 [00:02<00:03, 47.59it/s]
Loading 0:  40%|███▉      | 115/291 [00:02<00:03, 49.65it/s]
Loading 0:  42%|████▏     | 122/291 [00:02<00:03, 46.14it/s]
Loading 0:  45%|████▍     | 130/291 [00:02<00:03, 51.94it/s]
Loading 0:  47%|████▋     | 136/291 [00:02<00:03, 48.03it/s]
Loading 0:  49%|████▉     | 142/291 [00:02<00:03, 48.26it/s]
Loading 0:  51%|█████     | 149/291 [00:03<00:03, 46.63it/s]
Loading 0:  54%|█████▍    | 157/291 [00:03<00:02, 54.30it/s]
Loading 0:  56%|█████▌    | 163/291 [00:03<00:02, 52.26it/s]
Loading 0:  58%|█████▊    | 169/291 [00:03<00:02, 53.01it/s]
Loading 0:  60%|██████    | 175/291 [00:03<00:02, 54.65it/s]
Loading 0:  62%|██████▏   | 181/291 [00:03<00:02, 47.69it/s]
Loading 0:  64%|██████▍   | 187/291 [00:03<00:02, 35.05it/s]
Loading 0:  66%|██████▌   | 192/291 [00:04<00:02, 36.22it/s]
Loading 0:  68%|██████▊   | 199/291 [00:04<00:02, 39.56it/s]
Loading 0:  70%|███████   | 204/291 [00:04<00:02, 41.62it/s]
Loading 0:  73%|███████▎  | 211/291 [00:04<00:01, 47.62it/s]
Loading 0:  75%|███████▍  | 217/291 [00:04<00:01, 47.34it/s]
Loading 0:  76%|███████▋  | 222/291 [00:04<00:01, 47.48it/s]
Loading 0:  79%|███████▊  | 229/291 [00:04<00:01, 53.02it/s]
Loading 0:  81%|████████  | 235/291 [00:04<00:01, 50.68it/s]
Loading 0:  83%|████████▎ | 241/291 [00:05<00:01, 49.59it/s]
Loading 0:  85%|████████▍ | 247/291 [00:05<00:00, 51.99it/s]
Loading 0:  87%|████████▋ | 253/291 [00:05<00:00, 48.38it/s]
Loading 0:  89%|████████▊ | 258/291 [00:05<00:00, 47.44it/s]
Loading 0:  91%|█████████ | 265/291 [00:05<00:00, 53.21it/s]
Loading 0:  93%|█████████▎| 271/291 [00:05<00:00, 51.45it/s]
Loading 0:  95%|█████████▌| 277/291 [00:05<00:00, 50.16it/s]
Loading 0:  97%|█████████▋| 283/291 [00:05<00:00, 45.63it/s]
Loading 0:  99%|█████████▉| 288/291 [00:11<00:00,  3.34it/s]
                                                            
Job meta-llama-meta-llama-3-5386-v7-mkmlizer completed after 93.34s with status: succeeded
Stopping job with name meta-llama-meta-llama-3-5386-v7-mkmlizer
Pipeline stage MKMLizer completed in 94.10s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.18s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service meta-llama-meta-llama-3-5386-v7
Waiting for inference service meta-llama-meta-llama-3-5386-v7 to be ready
Failed to get response for submission nousresearch-meta-llama_4941_v54: ('http://nousresearch-meta-llama-4941-v54-predictor-default.tenant-chaiml-guanaco.knative.ord1.coreweave.cloud/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:47620->127.0.0.1:8080: read: connection reset by peer\n')
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
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
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 meta-llama-meta-llama-3-5386-v7 ready after 202.73169445991516s
Pipeline stage MKMLDeployer completed in 204.40s
run pipeline stage %s
Running pipeline stage StressChecker
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
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
HTTPSConnectionPool(host='guanaco-submitter.chai-research.com', port=443): Read timed out. (read timeout=20)
Received unhealthy response to inference request!
Received healthy response to inference request in 1.8120996952056885s
Received healthy response to inference request in 3.273338794708252s
Received healthy response to inference request in 1.8502771854400635s
Received healthy response to inference request in 2.4707698822021484s
5 requests
1 failed requests
5th percentile: 1.8197351932525634
10th percentile: 1.8273706912994385
20th percentile: 1.8426416873931886
30th percentile: 1.9743757247924805
40th percentile: 2.2225728034973145
50th percentile: 2.4707698822021484
60th percentile: 2.7917974472045897
70th percentile: 3.112825012207031
80th percentile: 6.833129072189334
90th percentile: 13.952709627151492
95th percentile: 17.512499904632566
99th percentile: 20.360332126617433
mean time: 6.09575514793396
%s, retrying in %s seconds...
Received healthy response to inference request in 2.104536771774292s
Received healthy response to inference request in 1.615889549255371s
Received healthy response to inference request in 1.312192678451538s
Received healthy response to inference request in 1.7392055988311768s
Received healthy response to inference request in 1.3251583576202393s
5 requests
0 failed requests
5th percentile: 1.3147858142852784
10th percentile: 1.3173789501190185
20th percentile: 1.322565221786499
30th percentile: 1.3833045959472656
40th percentile: 1.4995970726013184
50th percentile: 1.615889549255371
60th percentile: 1.6652159690856934
70th percentile: 1.7145423889160156
80th percentile: 1.8122718334197998
90th percentile: 1.958404302597046
95th percentile: 2.031470537185669
99th percentile: 2.0899235248565673
mean time: 1.6193965911865233
Pipeline stage StressChecker completed in 41.58s
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 6.50s
Shutdown handler de-registered
meta-llama-meta-llama-3-_5386_v7 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 0.19s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.15s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service meta-llama-meta-llama-3-5386-v7-profiler
Waiting for inference service meta-llama-meta-llama-3-5386-v7-profiler to be ready
Inference service meta-llama-meta-llama-3-5386-v7-profiler ready after 220.49423122406006s
Pipeline stage MKMLProfilerDeployer completed in 220.94s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/meta-llama-meta-llamee4a1b48e2aecc374fda6b791d5edcf5-deplojd45s:/code/chaiverse_profiler_1727232947 --namespace tenant-chaiml-guanaco
kubectl exec -it meta-llama-meta-llamee4a1b48e2aecc374fda6b791d5edcf5-deplojd45s --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1727232947 && 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_1727232947/summary.json'
kubectl exec -it meta-llama-meta-llamee4a1b48e2aecc374fda6b791d5edcf5-deplojd45s --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1727232947/summary.json'
Pipeline stage MKMLProfilerRunner completed in 785.81s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service meta-llama-meta-llama-3-5386-v7-profiler is running
Tearing down inference service meta-llama-meta-llama-3-5386-v7-profiler
Service meta-llama-meta-llama-3-5386-v7-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 2.15s
Shutdown handler de-registered
meta-llama-meta-llama-3-_5386_v7 status is now inactive due to auto deactivation removed underperforming models
meta-llama-meta-llama-3-_5386_v7 status is now inactive due to auto deactivation removed underperforming models
meta-llama-meta-llama-3-_5386_v7 status is now torndown due to DeploymentManager action