submission_id: hastagaras-halu-8b-llama_3098_v1
developer_uid: Hastagaras
status: inactive
model_repo: Hastagaras/Halu-8B-Llama3-v0.5
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 1.05, 'top_p': 1.0, 'min_p': 0.08, 'top_k': 80, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '<|eot_id|>'], 'max_input_tokens': 512, 'best_of': 16, '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: 2024-06-12T23:47:15+00:00
model_name: no-b
model_eval_status: success
model_group: Hastagaras/Halu-8B-Llama
num_battles: 56808
num_wins: 30358
celo_rating: 1201.95
propriety_score: 0.6846056613168882
propriety_total_count: 26107.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: no-b
ineligible_reason: None
language_model: Hastagaras/Halu-8B-Llama3-v0.5
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-06-12
win_ratio: 0.5343965638642445
Resubmit model
Running pipeline stage MKMLizer
Starting job with name hastagaras-halu-8b-llama-3098-v1-mkmlizer
Waiting for job on hastagaras-halu-8b-llama-3098-v1-mkmlizer to finish
hastagaras-halu-8b-llama-3098-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
hastagaras-halu-8b-llama-3098-v1-mkmlizer: ║ _____ __ __ ║
hastagaras-halu-8b-llama-3098-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
hastagaras-halu-8b-llama-3098-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
hastagaras-halu-8b-llama-3098-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
hastagaras-halu-8b-llama-3098-v1-mkmlizer: ║ /___/ ║
hastagaras-halu-8b-llama-3098-v1-mkmlizer: ║ ║
hastagaras-halu-8b-llama-3098-v1-mkmlizer: ║ Version: 0.8.14 ║
hastagaras-halu-8b-llama-3098-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
hastagaras-halu-8b-llama-3098-v1-mkmlizer: ║ https://mk1.ai ║
hastagaras-halu-8b-llama-3098-v1-mkmlizer: ║ ║
hastagaras-halu-8b-llama-3098-v1-mkmlizer: ║ The license key for the current software has been verified as ║
hastagaras-halu-8b-llama-3098-v1-mkmlizer: ║ belonging to: ║
hastagaras-halu-8b-llama-3098-v1-mkmlizer: ║ ║
hastagaras-halu-8b-llama-3098-v1-mkmlizer: ║ Chai Research Corp. ║
hastagaras-halu-8b-llama-3098-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
hastagaras-halu-8b-llama-3098-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
hastagaras-halu-8b-llama-3098-v1-mkmlizer: ║ ║
hastagaras-halu-8b-llama-3098-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
hastagaras-halu-8b-llama-3098-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.
hastagaras-halu-8b-llama-3098-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
hastagaras-halu-8b-llama-3098-v1-mkmlizer: Downloaded to shared memory in 39.376s
hastagaras-halu-8b-llama-3098-v1-mkmlizer: quantizing model to /dev/shm/model_cache
hastagaras-halu-8b-llama-3098-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
hastagaras-halu-8b-llama-3098-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 4%|▍ | 12/291 [00:00<00:02, 114.64it/s] Loading 0: 8%|▊ | 24/291 [00:00<00:02, 111.20it/s] Loading 0: 13%|█▎ | 39/291 [00:00<00:01, 126.68it/s] Loading 0: 18%|█▊ | 52/291 [00:00<00:01, 123.53it/s] Loading 0: 23%|██▎ | 66/291 [00:00<00:01, 129.00it/s] Loading 0: 27%|██▋ | 79/291 [00:00<00:01, 124.56it/s] Loading 0: 32%|███▏ | 92/291 [00:01<00:03, 63.75it/s] Loading 0: 35%|███▌ | 103/291 [00:01<00:02, 71.20it/s] Loading 0: 40%|████ | 117/291 [00:01<00:02, 85.01it/s] Loading 0: 45%|████▍ | 130/291 [00:01<00:01, 93.88it/s] Loading 0: 50%|████▉ | 145/291 [00:01<00:01, 107.21it/s] Loading 0: 54%|█████▍ | 158/291 [00:01<00:01, 106.84it/s] Loading 0: 60%|█████▉ | 174/291 [00:01<00:00, 120.17it/s] Loading 0: 65%|██████▍ | 188/291 [00:02<00:01, 70.81it/s] Loading 0: 69%|██████▉ | 201/291 [00:02<00:01, 80.55it/s] Loading 0: 73%|███████▎ | 212/291 [00:02<00:00, 85.19it/s] Loading 0: 78%|███████▊ | 228/291 [00:02<00:00, 101.29it/s] Loading 0: 83%|████████▎ | 241/291 [00:02<00:00, 105.66it/s] Loading 0: 88%|████████▊ | 256/291 [00:02<00:00, 111.89it/s] Loading 0: 93%|█████████▎| 272/291 [00:02<00:00, 124.14it/s] Loading 0: 98%|█████████▊| 286/291 [00:02<00:00, 121.78it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
hastagaras-halu-8b-llama-3098-v1-mkmlizer: quantized model in 19.539s
hastagaras-halu-8b-llama-3098-v1-mkmlizer: Processed model Hastagaras/Halu-8B-Llama3-v0.5 in 59.955s
hastagaras-halu-8b-llama-3098-v1-mkmlizer: creating bucket guanaco-mkml-models
hastagaras-halu-8b-llama-3098-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
hastagaras-halu-8b-llama-3098-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/hastagaras-halu-8b-llama-3098-v1
hastagaras-halu-8b-llama-3098-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/hastagaras-halu-8b-llama-3098-v1/tokenizer_config.json
hastagaras-halu-8b-llama-3098-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/hastagaras-halu-8b-llama-3098-v1/config.json
hastagaras-halu-8b-llama-3098-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/hastagaras-halu-8b-llama-3098-v1/special_tokens_map.json
hastagaras-halu-8b-llama-3098-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/hastagaras-halu-8b-llama-3098-v1/tokenizer.json
hastagaras-halu-8b-llama-3098-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/hastagaras-halu-8b-llama-3098-v1/flywheel_model.0.safetensors
hastagaras-halu-8b-llama-3098-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
hastagaras-halu-8b-llama-3098-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.
hastagaras-halu-8b-llama-3098-v1-mkmlizer: warnings.warn(
hastagaras-halu-8b-llama-3098-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.
hastagaras-halu-8b-llama-3098-v1-mkmlizer: warnings.warn(
hastagaras-halu-8b-llama-3098-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.
hastagaras-halu-8b-llama-3098-v1-mkmlizer: warnings.warn(
hastagaras-halu-8b-llama-3098-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()
hastagaras-halu-8b-llama-3098-v1-mkmlizer: return self.fget.__get__(instance, owner)()
hastagaras-halu-8b-llama-3098-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
hastagaras-halu-8b-llama-3098-v1-mkmlizer: Saving duration: 0.254s
hastagaras-halu-8b-llama-3098-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 3.775s
hastagaras-halu-8b-llama-3098-v1-mkmlizer: creating bucket guanaco-reward-models
hastagaras-halu-8b-llama-3098-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
hastagaras-halu-8b-llama-3098-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/hastagaras-halu-8b-llama-3098-v1_reward
hastagaras-halu-8b-llama-3098-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/hastagaras-halu-8b-llama-3098-v1_reward/special_tokens_map.json
hastagaras-halu-8b-llama-3098-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/hastagaras-halu-8b-llama-3098-v1_reward/tokenizer_config.json
hastagaras-halu-8b-llama-3098-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/hastagaras-halu-8b-llama-3098-v1_reward/config.json
hastagaras-halu-8b-llama-3098-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/hastagaras-halu-8b-llama-3098-v1_reward/vocab.json
hastagaras-halu-8b-llama-3098-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/hastagaras-halu-8b-llama-3098-v1_reward/merges.txt
hastagaras-halu-8b-llama-3098-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/hastagaras-halu-8b-llama-3098-v1_reward/tokenizer.json
hastagaras-halu-8b-llama-3098-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/hastagaras-halu-8b-llama-3098-v1_reward/reward.tensors
Job hastagaras-halu-8b-llama-3098-v1-mkmlizer completed after 83.16s with status: succeeded
Stopping job with name hastagaras-halu-8b-llama-3098-v1-mkmlizer
Pipeline stage MKMLizer completed in 86.76s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service hastagaras-halu-8b-llama-3098-v1
Waiting for inference service hastagaras-halu-8b-llama-3098-v1 to be ready
Inference service hastagaras-halu-8b-llama-3098-v1 ready after 30.19798231124878s
Pipeline stage ISVCDeployer completed in 37.32s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.1404671669006348s
Received healthy response to inference request in 1.3554306030273438s
Received healthy response to inference request in 1.315969705581665s
Received healthy response to inference request in 1.2898073196411133s
Received healthy response to inference request in 1.3752925395965576s
5 requests
0 failed requests
5th percentile: 1.2950397968292235
10th percentile: 1.300272274017334
20th percentile: 1.3107372283935548
30th percentile: 1.3238618850708008
40th percentile: 1.3396462440490722
50th percentile: 1.3554306030273438
60th percentile: 1.3633753776550293
70th percentile: 1.3713201522827148
80th percentile: 1.5283274650573733
90th percentile: 1.834397315979004
95th percentile: 1.9874322414398191
99th percentile: 2.1098601818084717
mean time: 1.4953934669494628
Pipeline stage StressChecker completed in 8.07s
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 0.03s
M-Eval Dataset for topic stay_in_character is loaded
hastagaras-halu-8b-llama_3098_v1 status is now deployed due to DeploymentManager action
hastagaras-halu-8b-llama_3098_v1 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics