submission_id: zeuslabs-l3-aethora-15b-v2_v13
developer_uid: elinas
status: inactive
model_repo: ZeusLabs/L3-Aethora-15B-V2
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 1.0, 'top_p': 1.0, 'min_p': 0.1, 'top_k': 40, 'presence_penalty': 0.7, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '<|eot_id|>'], 'max_input_tokens': 512, 'best_of': 4, 'max_output_tokens': 64}
formatter: {'memory_template': "<|begin_of_text|><|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}
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: 2024-06-29T02:48:45+00:00
model_name: l3-aethora-15b-v2
model_group: ZeusLabs/L3-Aethora-15B-
num_battles: 28614
num_wins: 12524
celo_rating: 1130.42
propriety_score: 0.7164614151424615
propriety_total_count: 13723.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 15009845248.0
best_of: 4
max_input_tokens: 512
max_output_tokens: 64
display_name: l3-aethora-15b-v2
ineligible_reason: None
language_model: ZeusLabs/L3-Aethora-15B-V2
model_size: 15B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-06-28
win_ratio: 0.43768784511078496
Resubmit model
Running pipeline stage MKMLizer
Starting job with name zeuslabs-l3-aethora-15b-v2-v13-mkmlizer
Waiting for job on zeuslabs-l3-aethora-15b-v2-v13-mkmlizer to finish
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: ║ _____ __ __ ║
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: ║ /___/ ║
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: ║ ║
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: ║ Version: 0.8.14 ║
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: ║ https://mk1.ai ║
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: ║ ║
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: ║ The license key for the current software has been verified as ║
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: ║ belonging to: ║
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: ║ ║
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: ║ Chai Research Corp. ║
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: ║ ║
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
zeuslabs-l3-aethora-15b-v2-v13-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.
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: warnings.warn(warning_message, FutureWarning)
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: Downloaded to shared memory in 29.802s
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: quantizing model to /dev/shm/model_cache
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: Saving flywheel model at /dev/shm/model_cache
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: Loading 0: 0%| | 0/579 [00:00<?, ?it/s] Loading 0: 2%|▏ | 14/579 [00:00<00:04, 136.73it/s] Loading 0: 6%|▌ | 32/579 [00:00<00:03, 152.85it/s] Loading 0: 9%|▊ | 50/579 [00:00<00:03, 157.06it/s] Loading 0: 12%|█▏ | 68/579 [00:00<00:03, 158.51it/s] Loading 0: 15%|█▍ | 84/579 [00:00<00:06, 82.12it/s] Loading 0: 18%|█▊ | 102/579 [00:00<00:04, 99.64it/s] Loading 0: 20%|██ | 118/579 [00:01<00:04, 112.68it/s] Loading 0: 23%|██▎ | 133/579 [00:01<00:03, 120.77it/s] Loading 0: 26%|██▌ | 149/579 [00:01<00:03, 126.70it/s] Loading 0: 29%|██▉ | 167/579 [00:01<00:03, 136.57it/s] Loading 0: 32%|███▏ | 187/579 [00:01<00:04, 89.23it/s] Loading 0: 34%|███▍ | 199/579 [00:01<00:04, 93.42it/s] Loading 0: 37%|███▋ | 212/579 [00:01<00:03, 100.11it/s] Loading 0: 40%|███▉ | 230/579 [00:02<00:03, 115.50it/s] Loading 0: 43%|████▎ | 247/579 [00:02<00:02, 126.31it/s] Loading 0: 46%|████▌ | 265/579 [00:02<00:02, 135.05it/s] Loading 0: 49%|████▊ | 281/579 [00:02<00:02, 137.90it/s] Loading 0: 51%|█████ | 296/579 [00:02<00:03, 84.09it/s] Loading 0: 54%|█████▎ | 311/579 [00:02<00:02, 95.98it/s] Loading 0: 57%|█████▋ | 329/579 [00:02<00:02, 111.02it/s] Loading 0: 60%|█████▉ | 347/579 [00:03<00:01, 124.39it/s] Loading 0: 63%|██████▎ | 364/579 [00:03<00:01, 134.40it/s] Loading 0: 66%|██████▌ | 382/579 [00:03<00:01, 143.27it/s] Loading 0: 69%|██████▊ | 398/579 [00:03<00:02, 89.07it/s] Loading 0: 71%|███████ | 411/579 [00:03<00:01, 96.60it/s] Loading 0: 74%|███████▍ | 428/579 [00:03<00:01, 108.07it/s] Loading 0: 77%|███████▋ | 445/579 [00:03<00:01, 121.17it/s] Loading 0: 80%|███████▉ | 462/579 [00:04<00:00, 128.26it/s] Loading 0: 82%|████████▏ | 477/579 [00:04<00:00, 128.95it/s] Loading 0: 85%|████████▌ | 493/579 [00:04<00:00, 86.40it/s] Loading 0: 88%|████████▊ | 508/579 [00:04<00:00, 96.30it/s] Loading 0: 89%|████████▉ | 516/579 [00:20<00:00, 96.30it/s] Loading 0: 89%|████████▉ | 517/579 [00:20<00:21, 2.88it/s] Loading 0: 92%|█████████▏| 535/579 [00:20<00:09, 4.45it/s] Loading 0: 96%|█████████▌| 553/579 [00:20<00:03, 6.66it/s] Loading 0: 99%|█████████▊| 571/579 [00:20<00:00, 9.70it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: quantized model in 33.347s
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: Processed model ZeusLabs/L3-Aethora-15B-V2 in 67.950s
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: creating bucket guanaco-mkml-models
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/zeuslabs-l3-aethora-15b-v2-v13/tokenizer.json
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/zeuslabs-l3-aethora-15b-v2-v13/flywheel_model.1.safetensors
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/zeuslabs-l3-aethora-15b-v2-v13/flywheel_model.0.safetensors
zeuslabs-l3-aethora-15b-v2-v13-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.
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: warnings.warn(
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
zeuslabs-l3-aethora-15b-v2-v13-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.
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: warnings.warn(
zeuslabs-l3-aethora-15b-v2-v13-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.
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: warnings.warn(
zeuslabs-l3-aethora-15b-v2-v13-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()
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: return self.fget.__get__(instance, owner)()
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: Saving duration: 0.402s
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 7.290s
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: creating bucket guanaco-reward-models
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: Bucket 's3://guanaco-reward-models/' created
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/zeuslabs-l3-aethora-15b-v2-v13_reward
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/zeuslabs-l3-aethora-15b-v2-v13_reward/config.json
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/zeuslabs-l3-aethora-15b-v2-v13_reward/tokenizer_config.json
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/zeuslabs-l3-aethora-15b-v2-v13_reward/merges.txt
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/zeuslabs-l3-aethora-15b-v2-v13_reward/vocab.json
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/zeuslabs-l3-aethora-15b-v2-v13_reward/tokenizer.json
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/zeuslabs-l3-aethora-15b-v2-v13_reward/special_tokens_map.json
zeuslabs-l3-aethora-15b-v2-v13-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/zeuslabs-l3-aethora-15b-v2-v13_reward/reward.tensors
Job zeuslabs-l3-aethora-15b-v2-v13-mkmlizer completed after 103.92s with status: succeeded
Stopping job with name zeuslabs-l3-aethora-15b-v2-v13-mkmlizer
Pipeline stage MKMLizer completed in 105.29s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.13s
Running pipeline stage ISVCDeployer
Creating inference service zeuslabs-l3-aethora-15b-v2-v13
Waiting for inference service zeuslabs-l3-aethora-15b-v2-v13 to be ready
Inference service zeuslabs-l3-aethora-15b-v2-v13 ready after 40.19439744949341s
Pipeline stage ISVCDeployer completed in 47.26s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.290517568588257s
Received healthy response to inference request in 1.0693268775939941s
Received healthy response to inference request in 1.4596235752105713s
Received healthy response to inference request in 1.3202285766601562s
Received healthy response to inference request in 1.17696213722229s
5 requests
0 failed requests
5th percentile: 1.0908539295196533
10th percentile: 1.1123809814453125
20th percentile: 1.1554350852966309
30th percentile: 1.2056154251098632
40th percentile: 1.2629220008850097
50th percentile: 1.3202285766601562
60th percentile: 1.3759865760803223
70th percentile: 1.4317445755004883
80th percentile: 1.6258023738861085
90th percentile: 1.9581599712371827
95th percentile: 2.1243387699127196
99th percentile: 2.257281808853149
mean time: 1.4633317470550538
Pipeline stage StressChecker completed in 8.10s
zeuslabs-l3-aethora-15b-v2_v13 status is now deployed due to DeploymentManager action
zeuslabs-l3-aethora-15b-v2_v13 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics