developer_uid: azuruce
submission_id: chaiml-chaiml-sft-horror_v3
model_name: 0919-eboney-horror
model_group: ChaiML/chaiml-sft-horror
status: torndown
timestamp: 2024-09-19T19:22:35+00:00
num_battles: 13644
num_wins: 6565
celo_rating: 1240.54
family_friendly_score: 0.0
submission_type: basic
model_repo: ChaiML/chaiml-sft-horror
model_architecture: MistralForCausalLM
model_num_parameters: 12772070400.0
best_of: 4
max_input_tokens: 1024
max_output_tokens: 64
latencies: [{'batch_size': 1, 'throughput': 0.635725618253525, 'latency_mean': 1.5729442191123963, 'latency_p50': 1.565726637840271, 'latency_p90': 1.739984679222107}, {'batch_size': 3, 'throughput': 1.2530932563940842, 'latency_mean': 2.3873020339012148, 'latency_p50': 2.3870028257369995, 'latency_p90': 2.609681177139282}, {'batch_size': 5, 'throughput': 1.5538881653753625, 'latency_mean': 3.1993695211410524, 'latency_p50': 3.189504027366638, 'latency_p90': 3.6276423692703244}, {'batch_size': 6, 'throughput': 1.6444576423668154, 'latency_mean': 3.628567770719528, 'latency_p50': 3.6042792797088623, 'latency_p90': 4.05694055557251}, {'batch_size': 8, 'throughput': 1.7563983432015715, 'latency_mean': 4.5178751111030575, 'latency_p50': 4.514602780342102, 'latency_p90': 5.169720983505249}, {'batch_size': 10, 'throughput': 1.8014144012247562, 'latency_mean': 5.510025978088379, 'latency_p50': 5.497288942337036, 'latency_p90': 6.36171588897705}]
gpu_counts: {'NVIDIA RTX A5000': 1}
display_name: 0919-eboney-horror
is_internal_developer: True
language_model: ChaiML/chaiml-sft-horror
model_size: 13B
ranking_group: single
throughput_3p7s: 1.67
us_pacific_date: 2024-09-19
win_ratio: 0.48116388155965994
generation_params: {'temperature': 0.9, 'top_p': 0.9, 'min_p': 0.05, 'top_k': 80, 'presence_penalty': 0.5, 'frequency_penalty': 0.5, 'stopping_words': ['\n', '</s>'], 'max_input_tokens': 1024, 'best_of': 4, 'max_output_tokens': 64}
formatter: {'memory_template': "{bot_name}'s Persona: {memory}\n##Try to make the conversation scary while stay in character##\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}
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-chaiml-sft-horror-v3-mkmlizer
Waiting for job on chaiml-chaiml-sft-horror-v3-mkmlizer to finish
chaiml-chaiml-sft-horror-v3-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-chaiml-sft-horror-v3-mkmlizer: ║     _____            __           __                                ║
chaiml-chaiml-sft-horror-v3-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
chaiml-chaiml-sft-horror-v3-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
chaiml-chaiml-sft-horror-v3-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
chaiml-chaiml-sft-horror-v3-mkmlizer: ║       /___/                                                         ║
chaiml-chaiml-sft-horror-v3-mkmlizer: ║                                                                     ║
chaiml-chaiml-sft-horror-v3-mkmlizer: ║  Version: 0.10.1                                                    ║
chaiml-chaiml-sft-horror-v3-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
chaiml-chaiml-sft-horror-v3-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-chaiml-sft-horror-v3-mkmlizer: ║                                                                     ║
chaiml-chaiml-sft-horror-v3-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-chaiml-sft-horror-v3-mkmlizer: ║  belonging to:                                                      ║
chaiml-chaiml-sft-horror-v3-mkmlizer: ║                                                                     ║
chaiml-chaiml-sft-horror-v3-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-chaiml-sft-horror-v3-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-chaiml-sft-horror-v3-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
chaiml-chaiml-sft-horror-v3-mkmlizer: ║                                                                     ║
chaiml-chaiml-sft-horror-v3-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-chaiml-sft-horror-v3-mkmlizer: Downloaded to shared memory in 92.739s
chaiml-chaiml-sft-horror-v3-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpvs9sp6jc, device:0
chaiml-chaiml-sft-horror-v3-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-chaiml-sft-horror-v3-mkmlizer: quantized model in 40.533s
chaiml-chaiml-sft-horror-v3-mkmlizer: Processed model ChaiML/chaiml-sft-horror in 133.272s
chaiml-chaiml-sft-horror-v3-mkmlizer: creating bucket guanaco-mkml-models
chaiml-chaiml-sft-horror-v3-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-chaiml-sft-horror-v3-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-chaiml-sft-horror-v3
chaiml-chaiml-sft-horror-v3-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-chaiml-sft-horror-v3/config.json
chaiml-chaiml-sft-horror-v3-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-chaiml-sft-horror-v3/special_tokens_map.json
chaiml-chaiml-sft-horror-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-chaiml-sft-horror-v3/tokenizer_config.json
chaiml-chaiml-sft-horror-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-chaiml-sft-horror-v3/tokenizer.json
chaiml-chaiml-sft-horror-v3-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-chaiml-sft-horror-v3/flywheel_model.0.safetensors
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
Job chaiml-chaiml-sft-horror-v3-mkmlizer completed after 159.25s with status: succeeded
Stopping job with name chaiml-chaiml-sft-horror-v3-mkmlizer
Pipeline stage MKMLizer completed in 164.24s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.70s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service chaiml-chaiml-sft-horror-v3
Waiting for inference service chaiml-chaiml-sft-horror-v3 to be ready
Inference service chaiml-chaiml-sft-horror-v3 ready after 201.41521286964417s
Pipeline stage MKMLDeployer completed in 202.06s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.401135206222534s
Received healthy response to inference request in 2.1187493801116943s
Received healthy response to inference request in 2.3247454166412354s
Received healthy response to inference request in 2.2246432304382324s
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
Received healthy response to inference request in 2.1676056385040283s
Connection pool is full, discarding connection: %s. Connection pool size: %s
5 requests
0 failed requests
5th percentile: 2.128520631790161
10th percentile: 2.1382918834686278
20th percentile: 2.1578343868255616
30th percentile: 2.179013156890869
40th percentile: 2.201828193664551
50th percentile: 2.2246432304382324
60th percentile: 2.2646841049194335
70th percentile: 2.3047249794006346
80th percentile: 2.340023374557495
90th percentile: 2.3705792903900145
95th percentile: 2.3858572483062743
99th percentile: 2.398079614639282
mean time: 2.2473757743835447
Pipeline stage StressChecker completed in 12.80s
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 8.14s
Shutdown handler de-registered
chaiml-chaiml-sft-horror_v3 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 0.11s
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 chaiml-chaiml-sft-horror-v3-profiler
Waiting for inference service chaiml-chaiml-sft-horror-v3-profiler to be ready
Inference service chaiml-chaiml-sft-horror-v3-profiler ready after 200.46677207946777s
Pipeline stage MKMLProfilerDeployer completed in 200.83s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/chaiml-chaiml-sft-horror-v3-profiler-predictor-00001-deplocvqrr:/code/chaiverse_profiler_1726774394 --namespace tenant-chaiml-guanaco
kubectl exec -it chaiml-chaiml-sft-horror-v3-profiler-predictor-00001-deplocvqrr --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1726774394 && python profiles.py profile --best_of_n 4 --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_1726774394/summary.json'
kubectl exec -it chaiml-chaiml-sft-horror-v3-profiler-predictor-00001-deplocvqrr --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1726774394/summary.json'
Pipeline stage MKMLProfilerRunner completed in 954.04s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service chaiml-chaiml-sft-horror-v3-profiler is running
Tearing down inference service chaiml-chaiml-sft-horror-v3-profiler
Service chaiml-chaiml-sft-horror-v3-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 2.03s
Shutdown handler de-registered
chaiml-chaiml-sft-horror_v3 status is now inactive due to auto deactivation removed underperforming models
run pipeline stage %s
run pipeline %s
Shutdown handler not registered because Python interpreter is not running in the main thread
admin requested tearing down of chaiml-elite-feed-convo-_7085_v2
chaiml-albert-sft-intent_3850_v1 status is now torndown due to DeploymentManager action
Cleaning model data from model cache
Service chaiml-elite-feed-convo-1137-v2 has been torndown
run pipeline stage %s
Shutdown handler de-registered
Shutdown handler not registered because Python interpreter is not running in the main thread
admin requested tearing down of chaiml-elite-feed-convo-_5606_v3
Pipeline stage %s skipped, reason=%s
Pipeline stage MKMLModelDeleter completed in 6.49s
Running pipeline stage MKMLModelDeleter
Pipeline stage MKMLDeleter completed in 6.16s
run pipeline stage %s
Running pipeline stage MKMLDeleter
Pipeline stage %s skipped, reason=%s
run pipeline stage %s
chaiml-chaiml-sft-horror_v3 status is now torndown due to DeploymentManager action
run pipeline %s
Shutdown handler not registered because Python interpreter is not running in the main thread
Pipeline stage MKMLModelDeleter completed in 5.47s
Pipeline stage %s skipped, reason=%s
Running pipeline stage MKMLDeleter
run pipeline stage %s
run pipeline %s
Pipeline stage %s skipped, reason=%s
Shutdown handler not registered because Python interpreter is not running in the main thread
Pipeline stage MKMLModelDeleter completed in 7.23s
Running pipeline stage MKMLModelDeleter
chaiml-chaiml-sft-horror_v3 status is now torndown due to DeploymentManager action
run pipeline stage %s
admin requested tearing down of chaiml-elite-feed-convo-_5606_v3