submission_id: hastagaras-halu-8b-llama_5820_v2
developer_uid: Hastagaras
status: inactive
model_repo: Hastagaras/HALU-8B-LLAMA3-EXP-3
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 0.85, 'top_p': 0.95, 'min_p': 0.08, 'top_k': 120, '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': "<|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|>{bot_name}<|end_header_id|>\n\n{bot_name}: {message}<|eot_id|>', 'user_template': '<|start_header_id|>{user_name}<|end_header_id|>\n\n{user_name}: {message}<|eot_id|>', 'response_template': '<|start_header_id|>{bot_name}<|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: 2024-06-01T02:34:03+00:00
model_name: ins-name-low-temp
model_eval_status: success
model_group: Hastagaras/HALU-8B-LLAMA
num_battles: 13120
num_wins: 7120
celo_rating: 1212.98
safety_score: 0.97
propriety_score: 0.6666666666666666
propriety_total_count: 6.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: ins-name-low-temp
ineligible_reason: propriety_total_count < 5000
language_model: Hastagaras/HALU-8B-LLAMA3-EXP-3
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-05-31
win_ratio: 0.5426829268292683
Resubmit model
Running pipeline stage MKMLizer
Starting job with name hastagaras-halu-8b-llama-5820-v2-mkmlizer
Waiting for job on hastagaras-halu-8b-llama-5820-v2-mkmlizer to finish
hastagaras-halu-8b-llama-5820-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
hastagaras-halu-8b-llama-5820-v2-mkmlizer: ║ _____ __ __ ║
hastagaras-halu-8b-llama-5820-v2-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
hastagaras-halu-8b-llama-5820-v2-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
hastagaras-halu-8b-llama-5820-v2-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
hastagaras-halu-8b-llama-5820-v2-mkmlizer: ║ /___/ ║
hastagaras-halu-8b-llama-5820-v2-mkmlizer: ║ ║
hastagaras-halu-8b-llama-5820-v2-mkmlizer: ║ Version: 0.8.14 ║
hastagaras-halu-8b-llama-5820-v2-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
hastagaras-halu-8b-llama-5820-v2-mkmlizer: ║ https://mk1.ai ║
hastagaras-halu-8b-llama-5820-v2-mkmlizer: ║ ║
hastagaras-halu-8b-llama-5820-v2-mkmlizer: ║ The license key for the current software has been verified as ║
hastagaras-halu-8b-llama-5820-v2-mkmlizer: ║ belonging to: ║
hastagaras-halu-8b-llama-5820-v2-mkmlizer: ║ ║
hastagaras-halu-8b-llama-5820-v2-mkmlizer: ║ Chai Research Corp. ║
hastagaras-halu-8b-llama-5820-v2-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
hastagaras-halu-8b-llama-5820-v2-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
hastagaras-halu-8b-llama-5820-v2-mkmlizer: ║ ║
hastagaras-halu-8b-llama-5820-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
hastagaras-halu-8b-llama-5820-v2-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.
hastagaras-halu-8b-llama-5820-v2-mkmlizer: warnings.warn(warning_message, FutureWarning)
hastagaras-halu-8b-llama-5820-v2-mkmlizer: Downloaded to shared memory in 20.150s
hastagaras-halu-8b-llama-5820-v2-mkmlizer: quantizing model to /dev/shm/model_cache
hastagaras-halu-8b-llama-5820-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
hastagaras-halu-8b-llama-5820-v2-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 5%|▍ | 14/291 [00:00<00:02, 135.04it/s] Loading 0: 11%|█ | 32/291 [00:00<00:01, 151.00it/s] Loading 0: 17%|█▋ | 50/291 [00:00<00:01, 154.26it/s] Loading 0: 23%|██▎ | 68/291 [00:00<00:01, 156.23it/s] Loading 0: 29%|██▉ | 84/291 [00:00<00:02, 84.80it/s] Loading 0: 35%|███▌ | 102/291 [00:00<00:01, 101.80it/s] Loading 0: 41%|████ | 120/291 [00:01<00:01, 115.76it/s] Loading 0: 47%|████▋ | 138/291 [00:01<00:01, 128.17it/s] Loading 0: 54%|█████▎ | 156/291 [00:01<00:00, 138.24it/s] Loading 0: 59%|█████▉ | 173/291 [00:01<00:00, 142.34it/s] Loading 0: 65%|██████▍ | 189/291 [00:01<00:01, 90.05it/s] Loading 0: 71%|███████ | 206/291 [00:01<00:00, 104.32it/s] Loading 0: 76%|███████▌ | 221/291 [00:01<00:00, 113.50it/s] Loading 0: 82%|████████▏ | 239/291 [00:02<00:00, 127.25it/s] Loading 0: 88%|████████▊ | 257/291 [00:02<00:00, 137.33it/s] Loading 0: 95%|█████████▍| 275/291 [00:02<00:00, 145.33it/s] Loading 0: 100%|██████████| 291/291 [00:07<00:00, 9.95it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
hastagaras-halu-8b-llama-5820-v2-mkmlizer: quantized model in 22.789s
hastagaras-halu-8b-llama-5820-v2-mkmlizer: Processed model Hastagaras/HALU-8B-LLAMA3-EXP-3 in 45.503s
hastagaras-halu-8b-llama-5820-v2-mkmlizer: creating bucket guanaco-mkml-models
hastagaras-halu-8b-llama-5820-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
hastagaras-halu-8b-llama-5820-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/hastagaras-halu-8b-llama-5820-v2
hastagaras-halu-8b-llama-5820-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/hastagaras-halu-8b-llama-5820-v2/config.json
hastagaras-halu-8b-llama-5820-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/hastagaras-halu-8b-llama-5820-v2/tokenizer.json
hastagaras-halu-8b-llama-5820-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/hastagaras-halu-8b-llama-5820-v2/special_tokens_map.json
hastagaras-halu-8b-llama-5820-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/hastagaras-halu-8b-llama-5820-v2/tokenizer_config.json
hastagaras-halu-8b-llama-5820-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/hastagaras-halu-8b-llama-5820-v2/flywheel_model.0.safetensors
hastagaras-halu-8b-llama-5820-v2-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
hastagaras-halu-8b-llama-5820-v2-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.
hastagaras-halu-8b-llama-5820-v2-mkmlizer: warnings.warn(
hastagaras-halu-8b-llama-5820-v2-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.
hastagaras-halu-8b-llama-5820-v2-mkmlizer: warnings.warn(
hastagaras-halu-8b-llama-5820-v2-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.
hastagaras-halu-8b-llama-5820-v2-mkmlizer: warnings.warn(
hastagaras-halu-8b-llama-5820-v2-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()
hastagaras-halu-8b-llama-5820-v2-mkmlizer: return self.fget.__get__(instance, owner)()
hastagaras-halu-8b-llama-5820-v2-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
hastagaras-halu-8b-llama-5820-v2-mkmlizer: Saving duration: 0.405s
hastagaras-halu-8b-llama-5820-v2-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 4.347s
hastagaras-halu-8b-llama-5820-v2-mkmlizer: creating bucket guanaco-reward-models
hastagaras-halu-8b-llama-5820-v2-mkmlizer: Bucket 's3://guanaco-reward-models/' created
hastagaras-halu-8b-llama-5820-v2-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/hastagaras-halu-8b-llama-5820-v2_reward
hastagaras-halu-8b-llama-5820-v2-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/hastagaras-halu-8b-llama-5820-v2_reward/special_tokens_map.json
hastagaras-halu-8b-llama-5820-v2-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/hastagaras-halu-8b-llama-5820-v2_reward/config.json
hastagaras-halu-8b-llama-5820-v2-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/hastagaras-halu-8b-llama-5820-v2_reward/tokenizer_config.json
hastagaras-halu-8b-llama-5820-v2-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/hastagaras-halu-8b-llama-5820-v2_reward/vocab.json
hastagaras-halu-8b-llama-5820-v2-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/hastagaras-halu-8b-llama-5820-v2_reward/merges.txt
hastagaras-halu-8b-llama-5820-v2-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/hastagaras-halu-8b-llama-5820-v2_reward/tokenizer.json
hastagaras-halu-8b-llama-5820-v2-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/hastagaras-halu-8b-llama-5820-v2_reward/reward.tensors
Job hastagaras-halu-8b-llama-5820-v2-mkmlizer completed after 73.92s with status: succeeded
Stopping job with name hastagaras-halu-8b-llama-5820-v2-mkmlizer
Pipeline stage MKMLizer completed in 74.77s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service hastagaras-halu-8b-llama-5820-v2
Waiting for inference service hastagaras-halu-8b-llama-5820-v2 to be ready
Inference service hastagaras-halu-8b-llama-5820-v2 ready after 40.26931118965149s
Pipeline stage ISVCDeployer completed in 46.18s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.213893175125122s
Received healthy response to inference request in 1.3727622032165527s
Received healthy response to inference request in 1.3716001510620117s
Received healthy response to inference request in 1.2890114784240723s
Received healthy response to inference request in 1.2818734645843506s
5 requests
0 failed requests
5th percentile: 1.2833010673522949
10th percentile: 1.2847286701202392
20th percentile: 1.287583875656128
30th percentile: 1.3055292129516602
40th percentile: 1.338564682006836
50th percentile: 1.3716001510620117
60th percentile: 1.372064971923828
70th percentile: 1.3725297927856446
80th percentile: 1.5409883975982668
90th percentile: 1.8774407863616944
95th percentile: 2.045666980743408
99th percentile: 2.1802479362487794
mean time: 1.5058280944824218
Pipeline stage StressChecker completed in 8.14s
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
hastagaras-halu-8b-llama_5820_v2 status is now deployed due to DeploymentManager action
hastagaras-halu-8b-llama_5820_v2 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics