submission_id: hastagaras-halu-8b-llama_3029_v1
developer_uid: Hastagaras
status: inactive
model_repo: Hastagaras/Halu-8B-Llama3-v0.4
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 1.0, 'top_p': 0.95, 'min_p': 0.05, 'top_k': 100, '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|>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: 2024-05-25T01:15:00+00:00
model_name: hastagaras-halu-8b-llama_3029_v1
model_eval_status: pending
model_group: Hastagaras/Halu-8B-Llama
num_battles: 8558
num_wins: 4522
celo_rating: 1195.99
safety_score: 0.98
propriety_score: 0.0
propriety_total_count: 0.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: hastagaras-halu-8b-llama_3029_v1
ineligible_reason: propriety_total_count < 5000
language_model: Hastagaras/Halu-8B-Llama3-v0.4
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-05-24
win_ratio: 0.5283944846926852
Resubmit model
Running pipeline stage MKMLizer
Starting job with name hastagaras-halu-8b-llama-3029-v1-mkmlizer
Waiting for job on hastagaras-halu-8b-llama-3029-v1-mkmlizer to finish
hastagaras-halu-8b-llama-3029-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
hastagaras-halu-8b-llama-3029-v1-mkmlizer: ║ _____ __ __ ║
hastagaras-halu-8b-llama-3029-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
hastagaras-halu-8b-llama-3029-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
hastagaras-halu-8b-llama-3029-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
hastagaras-halu-8b-llama-3029-v1-mkmlizer: ║ /___/ ║
hastagaras-halu-8b-llama-3029-v1-mkmlizer: ║ ║
hastagaras-halu-8b-llama-3029-v1-mkmlizer: ║ Version: 0.8.14 ║
hastagaras-halu-8b-llama-3029-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
hastagaras-halu-8b-llama-3029-v1-mkmlizer: ║ https://mk1.ai ║
hastagaras-halu-8b-llama-3029-v1-mkmlizer: ║ ║
hastagaras-halu-8b-llama-3029-v1-mkmlizer: ║ The license key for the current software has been verified as ║
hastagaras-halu-8b-llama-3029-v1-mkmlizer: ║ belonging to: ║
hastagaras-halu-8b-llama-3029-v1-mkmlizer: ║ ║
hastagaras-halu-8b-llama-3029-v1-mkmlizer: ║ Chai Research Corp. ║
hastagaras-halu-8b-llama-3029-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
hastagaras-halu-8b-llama-3029-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
hastagaras-halu-8b-llama-3029-v1-mkmlizer: ║ ║
hastagaras-halu-8b-llama-3029-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
hastagaras-halu-8b-llama-3029-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-3029-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
hastagaras-halu-8b-llama-3029-v1-mkmlizer: Downloaded to shared memory in 30.399s
hastagaras-halu-8b-llama-3029-v1-mkmlizer: quantizing model to /dev/shm/model_cache
hastagaras-halu-8b-llama-3029-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
Retrying (%r) after connection broken by '%r': %s
hastagaras-halu-8b-llama-3029-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 2%|▏ | 5/291 [00:00<00:06, 45.02it/s] Loading 0: 8%|▊ | 22/291 [00:00<00:02, 114.06it/s] Loading 0: 14%|█▎ | 40/291 [00:00<00:01, 136.23it/s] Loading 0: 20%|█▉ | 58/291 [00:00<00:01, 145.46it/s] Loading 0: 26%|██▌ | 76/291 [00:00<00:01, 149.57it/s] Loading 0: 31%|███▏ | 91/291 [00:00<00:02, 81.57it/s] Loading 0: 35%|███▌ | 103/291 [00:01<00:02, 84.79it/s] Loading 0: 41%|████ | 120/291 [00:01<00:01, 100.73it/s] Loading 0: 46%|████▌ | 134/291 [00:01<00:01, 109.47it/s] Loading 0: 51%|█████ | 149/291 [00:01<00:01, 117.94it/s] Loading 0: 57%|█████▋ | 167/291 [00:01<00:00, 130.50it/s] Loading 0: 64%|██████▍ | 187/291 [00:01<00:01, 90.74it/s] Loading 0: 69%|██████▉ | 202/291 [00:01<00:00, 99.42it/s] Loading 0: 76%|███████▌ | 220/291 [00:02<00:00, 114.08it/s] Loading 0: 81%|████████▏ | 237/291 [00:02<00:00, 126.22it/s] Loading 0: 88%|████████▊ | 255/291 [00:02<00:00, 136.30it/s] Loading 0: 94%|█████████▍| 273/291 [00:02<00:00, 143.80it/s] Loading 0: 99%|█████████▉| 289/291 [00:07<00:00, 9.85it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
hastagaras-halu-8b-llama-3029-v1-mkmlizer: quantized model in 23.056s
hastagaras-halu-8b-llama-3029-v1-mkmlizer: Processed model Hastagaras/Halu-8B-Llama3-v0.4 in 56.058s
hastagaras-halu-8b-llama-3029-v1-mkmlizer: creating bucket guanaco-mkml-models
hastagaras-halu-8b-llama-3029-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
hastagaras-halu-8b-llama-3029-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/hastagaras-halu-8b-llama-3029-v1
hastagaras-halu-8b-llama-3029-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/hastagaras-halu-8b-llama-3029-v1/special_tokens_map.json
hastagaras-halu-8b-llama-3029-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/hastagaras-halu-8b-llama-3029-v1/config.json
hastagaras-halu-8b-llama-3029-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/hastagaras-halu-8b-llama-3029-v1/tokenizer_config.json
hastagaras-halu-8b-llama-3029-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/hastagaras-halu-8b-llama-3029-v1/tokenizer.json
hastagaras-halu-8b-llama-3029-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/hastagaras-halu-8b-llama-3029-v1/flywheel_model.0.safetensors
hastagaras-halu-8b-llama-3029-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
hastagaras-halu-8b-llama-3029-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-3029-v1-mkmlizer: warnings.warn(
hastagaras-halu-8b-llama-3029-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-3029-v1-mkmlizer: warnings.warn(
Retrying (%r) after connection broken by '%r': %s
hastagaras-halu-8b-llama-3029-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-3029-v1-mkmlizer: return self.fget.__get__(instance, owner)()
hastagaras-halu-8b-llama-3029-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
hastagaras-halu-8b-llama-3029-v1-mkmlizer: Saving duration: 0.415s
hastagaras-halu-8b-llama-3029-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 3.886s
hastagaras-halu-8b-llama-3029-v1-mkmlizer: creating bucket guanaco-reward-models
hastagaras-halu-8b-llama-3029-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
hastagaras-halu-8b-llama-3029-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/hastagaras-halu-8b-llama-3029-v1_reward
hastagaras-halu-8b-llama-3029-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/hastagaras-halu-8b-llama-3029-v1_reward/special_tokens_map.json
hastagaras-halu-8b-llama-3029-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/hastagaras-halu-8b-llama-3029-v1_reward/config.json
hastagaras-halu-8b-llama-3029-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/hastagaras-halu-8b-llama-3029-v1_reward/tokenizer_config.json
hastagaras-halu-8b-llama-3029-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/hastagaras-halu-8b-llama-3029-v1_reward/merges.txt
hastagaras-halu-8b-llama-3029-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/hastagaras-halu-8b-llama-3029-v1_reward/vocab.json
hastagaras-halu-8b-llama-3029-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/hastagaras-halu-8b-llama-3029-v1_reward/tokenizer.json
hastagaras-halu-8b-llama-3029-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/hastagaras-halu-8b-llama-3029-v1_reward/reward.tensors
Job hastagaras-halu-8b-llama-3029-v1-mkmlizer completed after 83.05s with status: succeeded
Stopping job with name hastagaras-halu-8b-llama-3029-v1-mkmlizer
Pipeline stage MKMLizer completed in 87.58s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service hastagaras-halu-8b-llama-3029-v1
Retrying (%r) after connection broken by '%r': %s
Waiting for inference service hastagaras-halu-8b-llama-3029-v1 to be ready
Retrying (%r) after connection broken by '%r': %s
Retrying (%r) after connection broken by '%r': %s
Retrying (%r) after connection broken by '%r': %s
Inference service hastagaras-halu-8b-llama-3029-v1 ready after 70.43177247047424s
Pipeline stage ISVCDeployer completed in 78.09s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.28554105758667s
Received healthy response to inference request in 1.3346304893493652s
Received healthy response to inference request in 1.3279008865356445s
Received healthy response to inference request in 1.2832047939300537s
Received healthy response to inference request in 1.2499971389770508s
5 requests
0 failed requests
5th percentile: 1.2566386699676513
10th percentile: 1.263280200958252
20th percentile: 1.2765632629394532
30th percentile: 1.2921440124511718
40th percentile: 1.3100224494934083
50th percentile: 1.3279008865356445
60th percentile: 1.330592727661133
70th percentile: 1.333284568786621
80th percentile: 1.5248126029968263
90th percentile: 1.9051768302917482
95th percentile: 2.0953589439392086
99th percentile: 2.2475046348571777
mean time: 1.4962548732757568
Pipeline stage StressChecker completed in 8.06s
Running pipeline stage DaemonicModelEvalScorer
Pipeline stage DaemonicModelEvalScorer completed in 0.03s
Running M-Eval for topic stay_in_character
Running pipeline stage DaemonicSafetyScorer
M-Eval Dataset for topic stay_in_character is loaded
Pipeline stage DaemonicSafetyScorer completed in 0.07s
hastagaras-halu-8b-llama_3029_v1 status is now deployed due to DeploymentManager action
hastagaras-halu-8b-llama_3029_v1 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics