submission_id: turboderp-llama3-turbca_4336_v15
developer_uid: Lina09
status: inactive
model_repo: turboderp/llama3-turbcat-instruct-8b
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 0.9, 'top_p': 0.8, 'min_p': 0.0, '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': '<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\n{memory}<|eot_id|>', 'prompt_template': '<|start_header_id|>system<|end_header_id|>\n\nThe following message provides the necessary information about the below conversation and the characters in the conversation.\n{prompt}\nThe conversation below will be carried out according to information in the above text.<|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: 2024-07-03T09:35:26+00:00
model_name: turboderp-llama3-turbcat
model_group: turboderp/llama3-turbcat
num_battles: 18712
num_wins: 9205
celo_rating: 1179.57
propriety_score: 0.7029580474637274
propriety_total_count: 8891.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: turboderp-llama3-turbcat
ineligible_reason: None
language_model: turboderp/llama3-turbcat-instruct-8b
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-07-03
win_ratio: 0.4919303120991877
Resubmit model
Running pipeline stage MKMLizer
Starting job with name turboderp-llama3-turbca-4336-v15-mkmlizer
Waiting for job on turboderp-llama3-turbca-4336-v15-mkmlizer to finish
turboderp-llama3-turbca-4336-v15-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
turboderp-llama3-turbca-4336-v15-mkmlizer: ║ _____ __ __ ║
turboderp-llama3-turbca-4336-v15-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
turboderp-llama3-turbca-4336-v15-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
turboderp-llama3-turbca-4336-v15-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
turboderp-llama3-turbca-4336-v15-mkmlizer: ║ /___/ ║
turboderp-llama3-turbca-4336-v15-mkmlizer: ║ ║
turboderp-llama3-turbca-4336-v15-mkmlizer: ║ Version: 0.8.14 ║
turboderp-llama3-turbca-4336-v15-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
turboderp-llama3-turbca-4336-v15-mkmlizer: ║ https://mk1.ai ║
turboderp-llama3-turbca-4336-v15-mkmlizer: ║ ║
turboderp-llama3-turbca-4336-v15-mkmlizer: ║ The license key for the current software has been verified as ║
turboderp-llama3-turbca-4336-v15-mkmlizer: ║ belonging to: ║
turboderp-llama3-turbca-4336-v15-mkmlizer: ║ ║
turboderp-llama3-turbca-4336-v15-mkmlizer: ║ Chai Research Corp. ║
turboderp-llama3-turbca-4336-v15-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
turboderp-llama3-turbca-4336-v15-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
turboderp-llama3-turbca-4336-v15-mkmlizer: ║ ║
turboderp-llama3-turbca-4336-v15-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
turboderp-llama3-turbca-4336-v15-mkmlizer: Downloaded to shared memory in 34.234s
turboderp-llama3-turbca-4336-v15-mkmlizer: quantizing model to /dev/shm/model_cache
turboderp-llama3-turbca-4336-v15-mkmlizer: Saving flywheel model at /dev/shm/model_cache
turboderp-llama3-turbca-4336-v15-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 4%|▍ | 12/291 [00:00<00:02, 111.06it/s] Loading 0: 8%|▊ | 24/291 [00:00<00:02, 111.62it/s] Loading 0: 13%|█▎ | 39/291 [00:00<00:01, 127.07it/s] Loading 0: 18%|█▊ | 52/291 [00:00<00:01, 123.43it/s] Loading 0: 23%|██▎ | 67/291 [00:00<00:01, 124.69it/s] Loading 0: 28%|██▊ | 81/291 [00:00<00:01, 127.41it/s] Loading 0: 32%|███▏ | 94/291 [00:01<00:03, 58.42it/s] Loading 0: 37%|███▋ | 108/291 [00:01<00:02, 71.59it/s] Loading 0: 42%|████▏ | 121/291 [00:01<00:02, 80.80it/s] Loading 0: 46%|████▋ | 135/291 [00:01<00:01, 92.55it/s] Loading 0: 51%|█████ | 148/291 [00:01<00:01, 99.14it/s] Loading 0: 56%|█████▌ | 162/291 [00:01<00:01, 108.90it/s] Loading 0: 60%|██████ | 176/291 [00:01<00:00, 115.40it/s] Loading 0: 65%|██████▍ | 189/291 [00:02<00:01, 59.30it/s] Loading 0: 69%|██████▉ | 202/291 [00:02<00:01, 69.78it/s] Loading 0: 74%|███████▍ | 216/291 [00:02<00:00, 81.87it/s] Loading 0: 79%|███████▊ | 229/291 [00:02<00:00, 89.94it/s] Loading 0: 84%|████████▎ | 243/291 [00:02<00:00, 100.72it/s] Loading 0: 88%|████████▊ | 256/291 [00:02<00:00, 105.42it/s] Loading 0: 93%|█████████▎| 270/291 [00:02<00:00, 113.34it/s] Loading 0: 97%|█████████▋| 283/291 [00:03<00:00, 113.02it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
turboderp-llama3-turbca-4336-v15-mkmlizer: quantized model in 28.990s
turboderp-llama3-turbca-4336-v15-mkmlizer: Processed model turboderp/llama3-turbcat-instruct-8b in 63.224s
turboderp-llama3-turbca-4336-v15-mkmlizer: creating bucket guanaco-mkml-models
turboderp-llama3-turbca-4336-v15-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
turboderp-llama3-turbca-4336-v15-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/turboderp-llama3-turbca-4336-v15
turboderp-llama3-turbca-4336-v15-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/turboderp-llama3-turbca-4336-v15/special_tokens_map.json
turboderp-llama3-turbca-4336-v15-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/turboderp-llama3-turbca-4336-v15/tokenizer_config.json
turboderp-llama3-turbca-4336-v15-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/turboderp-llama3-turbca-4336-v15/config.json
turboderp-llama3-turbca-4336-v15-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/turboderp-llama3-turbca-4336-v15/tokenizer.json
turboderp-llama3-turbca-4336-v15-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/turboderp-llama3-turbca-4336-v15/flywheel_model.0.safetensors
turboderp-llama3-turbca-4336-v15-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
turboderp-llama3-turbca-4336-v15-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:919: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
turboderp-llama3-turbca-4336-v15-mkmlizer: warnings.warn(
turboderp-llama3-turbca-4336-v15-mkmlizer: /opt/conda/lib/python3.10/site-packages/huggingface_hub/file_download.py:1132: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.
turboderp-llama3-turbca-4336-v15-mkmlizer: warnings.warn(
turboderp-llama3-turbca-4336-v15-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:769: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
turboderp-llama3-turbca-4336-v15-mkmlizer: warnings.warn(
turboderp-llama3-turbca-4336-v15-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.
turboderp-llama3-turbca-4336-v15-mkmlizer: warnings.warn(
turboderp-llama3-turbca-4336-v15-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()
turboderp-llama3-turbca-4336-v15-mkmlizer: return self.fget.__get__(instance, owner)()
turboderp-llama3-turbca-4336-v15-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
turboderp-llama3-turbca-4336-v15-mkmlizer: Saving duration: 0.497s
turboderp-llama3-turbca-4336-v15-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 7.324s
turboderp-llama3-turbca-4336-v15-mkmlizer: creating bucket guanaco-reward-models
turboderp-llama3-turbca-4336-v15-mkmlizer: Bucket 's3://guanaco-reward-models/' created
turboderp-llama3-turbca-4336-v15-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/turboderp-llama3-turbca-4336-v15_reward
turboderp-llama3-turbca-4336-v15-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/turboderp-llama3-turbca-4336-v15_reward/config.json
turboderp-llama3-turbca-4336-v15-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/turboderp-llama3-turbca-4336-v15_reward/tokenizer_config.json
turboderp-llama3-turbca-4336-v15-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/turboderp-llama3-turbca-4336-v15_reward/merges.txt
turboderp-llama3-turbca-4336-v15-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/turboderp-llama3-turbca-4336-v15_reward/vocab.json
turboderp-llama3-turbca-4336-v15-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/turboderp-llama3-turbca-4336-v15_reward/special_tokens_map.json
turboderp-llama3-turbca-4336-v15-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/turboderp-llama3-turbca-4336-v15_reward/tokenizer.json
turboderp-llama3-turbca-4336-v15-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/turboderp-llama3-turbca-4336-v15_reward/reward.tensors
Job turboderp-llama3-turbca-4336-v15-mkmlizer completed after 94.0s with status: succeeded
Stopping job with name turboderp-llama3-turbca-4336-v15-mkmlizer
Pipeline stage MKMLizer completed in 94.95s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.12s
Running pipeline stage ISVCDeployer
Creating inference service turboderp-llama3-turbca-4336-v15
Waiting for inference service turboderp-llama3-turbca-4336-v15 to be ready
Inference service turboderp-llama3-turbca-4336-v15 ready after 40.172165632247925s
Pipeline stage ISVCDeployer completed in 47.34s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.09765625s
Received healthy response to inference request in 1.3317222595214844s
Received healthy response to inference request in 1.3291680812835693s
Received healthy response to inference request in 1.2555780410766602s
Received healthy response to inference request in 1.4306256771087646s
5 requests
0 failed requests
5th percentile: 1.270296049118042
10th percentile: 1.2850140571594237
20th percentile: 1.3144500732421875
30th percentile: 1.3296789169311523
40th percentile: 1.3307005882263183
50th percentile: 1.3317222595214844
60th percentile: 1.3712836265563966
70th percentile: 1.4108449935913085
80th percentile: 1.564031791687012
90th percentile: 1.830844020843506
95th percentile: 1.9642501354217528
99th percentile: 2.0709750270843506
mean time: 1.4889500617980957
Pipeline stage StressChecker completed in 8.20s
turboderp-llama3-turbca_4336_v15 status is now deployed due to DeploymentManager action
turboderp-llama3-turbca_4336_v15 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics