submission_id: kaoeiri-keiana-l3-test5-_3680_v1
developer_uid: helkero
status: inactive
model_repo: Kaoeiri/Keiana-L3-Test5.4-8B-10
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 0.9, 'top_p': 1.0, 'min_p': 0.01, '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}
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}
reward_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}
timestamp: 2024-06-09T02:33:22+00:00
model_name: kaoeiri-keiana-l3-test5-_3680_v1
model_eval_status: success
model_group: Kaoeiri/Keiana-L3-Test5.
num_battles: 11602
num_wins: 6243
celo_rating: 1208.49
safety_score: 0.94
propriety_score: 0.653179190751445
propriety_total_count: 865.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: kaoeiri-keiana-l3-test5-_3680_v1
ineligible_reason: propriety_total_count < 5000
language_model: Kaoeiri/Keiana-L3-Test5.4-8B-10
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-06-08
win_ratio: 0.538096879848302
Resubmit model
Running pipeline stage MKMLizer
Starting job with name kaoeiri-keiana-l3-test5-3680-v1-mkmlizer
Waiting for job on kaoeiri-keiana-l3-test5-3680-v1-mkmlizer to finish
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: ║ _____ __ __ ║
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: ║ /___/ ║
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: ║ ║
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: ║ Version: 0.8.14 ║
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: ║ https://mk1.ai ║
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: ║ ║
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: ║ The license key for the current software has been verified as ║
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: ║ belonging to: ║
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: ║ ║
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: ║ Chai Research Corp. ║
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: ║ ║
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/huggingface_hub/utils/_deprecation.py:131: FutureWarning: 'list_files_info' (from 'huggingface_hub.hf_api') is deprecated and will be removed from version '0.23'. Use `list_repo_tree` and `get_paths_info` instead.
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: Downloaded to shared memory in 67.874s
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: quantizing model to /dev/shm/model_cache
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 1%| | 2/291 [00:04<10:49, 2.25s/it] Loading 0: 5%|▌ | 15/291 [00:04<01:02, 4.42it/s] Loading 0: 8%|▊ | 22/291 [00:04<00:38, 6.95it/s] Loading 0: 14%|█▎ | 40/291 [00:04<00:15, 16.36it/s] Loading 0: 20%|█▉ | 58/291 [00:05<00:08, 28.01it/s] Loading 0: 24%|██▍ | 71/291 [00:05<00:06, 33.80it/s] Loading 0: 30%|██▉ | 87/291 [00:05<00:04, 46.86it/s] Loading 0: 35%|███▌ | 103/291 [00:05<00:03, 54.07it/s] Loading 0: 42%|████▏ | 121/291 [00:05<00:02, 70.98it/s] Loading 0: 48%|████▊ | 139/291 [00:05<00:01, 88.11it/s] Loading 0: 53%|█████▎ | 153/291 [00:05<00:01, 80.28it/s] Loading 0: 58%|█████▊ | 170/291 [00:06<00:01, 96.57it/s] Loading 0: 63%|██████▎ | 184/291 [00:06<00:01, 86.47it/s] Loading 0: 69%|██████▉ | 202/291 [00:06<00:00, 103.83it/s] Loading 0: 76%|███████▌ | 220/291 [00:06<00:00, 119.42it/s] Loading 0: 81%|████████ | 235/291 [00:06<00:00, 99.51it/s] Loading 0: 87%|████████▋ | 253/291 [00:06<00:00, 115.40it/s] Loading 0: 92%|█████████▏| 267/291 [00:07<00:00, 94.37it/s] Loading 0: 98%|█████████▊| 285/291 [00:07<00:00, 110.68it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: quantized model in 22.942s
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: Processed model Kaoeiri/Keiana-L3-Test5.4-8B-10 in 93.313s
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: creating bucket guanaco-mkml-models
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/kaoeiri-keiana-l3-test5-3680-v1
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/kaoeiri-keiana-l3-test5-3680-v1/config.json
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/kaoeiri-keiana-l3-test5-3680-v1/tokenizer_config.json
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/kaoeiri-keiana-l3-test5-3680-v1/special_tokens_map.json
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/kaoeiri-keiana-l3-test5-3680-v1/tokenizer.json
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/kaoeiri-keiana-l3-test5-3680-v1/flywheel_model.0.safetensors
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:913: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: warnings.warn(
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:757: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: warnings.warn(
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/auto_factory.py:468: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: warnings.warn(
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/torch/_utils.py:831: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage()
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: return self.fget.__get__(instance, owner)()
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: Saving duration: 0.406s
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 4.774s
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: creating bucket guanaco-reward-models
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/kaoeiri-keiana-l3-test5-3680-v1_reward
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/kaoeiri-keiana-l3-test5-3680-v1_reward/special_tokens_map.json
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/kaoeiri-keiana-l3-test5-3680-v1_reward/config.json
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/kaoeiri-keiana-l3-test5-3680-v1_reward/tokenizer_config.json
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/kaoeiri-keiana-l3-test5-3680-v1_reward/vocab.json
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/kaoeiri-keiana-l3-test5-3680-v1_reward/merges.txt
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/kaoeiri-keiana-l3-test5-3680-v1_reward/tokenizer.json
kaoeiri-keiana-l3-test5-3680-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/kaoeiri-keiana-l3-test5-3680-v1_reward/reward.tensors
Job kaoeiri-keiana-l3-test5-3680-v1-mkmlizer completed after 124.31s with status: succeeded
Stopping job with name kaoeiri-keiana-l3-test5-3680-v1-mkmlizer
Pipeline stage MKMLizer completed in 126.81s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.11s
Running pipeline stage ISVCDeployer
Creating inference service kaoeiri-keiana-l3-test5-3680-v1
Waiting for inference service kaoeiri-keiana-l3-test5-3680-v1 to be ready
Inference service kaoeiri-keiana-l3-test5-3680-v1 ready after 50.33088159561157s
Pipeline stage ISVCDeployer completed in 57.21s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.237959384918213s
Received healthy response to inference request in 2.319681406021118s
Received healthy response to inference request in 1.2838685512542725s
Received healthy response to inference request in 1.2803850173950195s
Received healthy response to inference request in 1.2659025192260742s
5 requests
0 failed requests
5th percentile: 1.2687990188598632
10th percentile: 1.2716955184936523
20th percentile: 1.2774885177612305
30th percentile: 1.2810817241668702
40th percentile: 1.2824751377105712
50th percentile: 1.2838685512542725
60th percentile: 1.6655048847198486
70th percentile: 2.0471412181854247
80th percentile: 2.254303789138794
90th percentile: 2.286992597579956
95th percentile: 2.303337001800537
99th percentile: 2.316412525177002
mean time: 1.6775593757629395
Pipeline stage StressChecker completed in 9.06s
Running pipeline stage DaemonicModelEvalScorer
Pipeline stage DaemonicModelEvalScorer completed in 0.03s
Running pipeline stage DaemonicSafetyScorer
Running M-Eval for topic stay_in_character
Pipeline stage DaemonicSafetyScorer completed in 0.04s
M-Eval Dataset for topic stay_in_character is loaded
kaoeiri-keiana-l3-test5-_3680_v1 status is now deployed due to DeploymentManager action
kaoeiri-keiana-l3-test5-_3680_v1 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics