submission_id: turboderp-llama3-turbcat_4336_v1
developer_uid: kaltcit
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': 4, 'max_output_tokens': 64}
formatter: {'memory_template': '### Instruction:\\n{memory}\\n', 'prompt_template': '### Input:\\n{prompt}\\n', 'bot_template': '{bot_name}: {message}\\n', 'user_template': '{user_name}: {message}\\n', 'response_template': '### Response:\\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: 2024-06-25T20:41:49+00:00
model_name: turboderp-llama3-turbcat_4336_v1
model_group: turboderp/llama3-turbcat
num_battles: 17707
num_wins: 8175
celo_rating: 1156.14
propriety_score: 0.7313804554079696
propriety_total_count: 8432.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 4
max_input_tokens: 512
max_output_tokens: 64
display_name: turboderp-llama3-turbcat_4336_v1
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: 2024-06-25
win_ratio: 0.4616818207488564
Resubmit model
Running pipeline stage MKMLizer
Starting job with name turboderp-llama3-turbcat-4336-v1-mkmlizer
Waiting for job on turboderp-llama3-turbcat-4336-v1-mkmlizer to finish
turboderp-llama3-turbcat-4336-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
turboderp-llama3-turbcat-4336-v1-mkmlizer: ║ _____ __ __ ║
turboderp-llama3-turbcat-4336-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
turboderp-llama3-turbcat-4336-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
turboderp-llama3-turbcat-4336-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
turboderp-llama3-turbcat-4336-v1-mkmlizer: ║ /___/ ║
turboderp-llama3-turbcat-4336-v1-mkmlizer: ║ ║
turboderp-llama3-turbcat-4336-v1-mkmlizer: ║ Version: 0.8.14 ║
turboderp-llama3-turbcat-4336-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
turboderp-llama3-turbcat-4336-v1-mkmlizer: ║ https://mk1.ai ║
turboderp-llama3-turbcat-4336-v1-mkmlizer: ║ ║
turboderp-llama3-turbcat-4336-v1-mkmlizer: ║ The license key for the current software has been verified as ║
turboderp-llama3-turbcat-4336-v1-mkmlizer: ║ belonging to: ║
turboderp-llama3-turbcat-4336-v1-mkmlizer: ║ ║
turboderp-llama3-turbcat-4336-v1-mkmlizer: ║ Chai Research Corp. ║
turboderp-llama3-turbcat-4336-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
turboderp-llama3-turbcat-4336-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
turboderp-llama3-turbcat-4336-v1-mkmlizer: ║ ║
turboderp-llama3-turbcat-4336-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
turboderp-llama3-turbcat-4336-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.
turboderp-llama3-turbcat-4336-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
turboderp-llama3-turbcat-4336-v1-mkmlizer: Downloaded to shared memory in 48.752s
turboderp-llama3-turbcat-4336-v1-mkmlizer: quantizing model to /dev/shm/model_cache
turboderp-llama3-turbcat-4336-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
turboderp-llama3-turbcat-4336-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 4%|▍ | 13/291 [00:00<00:02, 129.61it/s] Loading 0: 11%|█ | 31/291 [00:00<00:01, 146.46it/s] Loading 0: 17%|█▋ | 49/291 [00:00<00:01, 153.91it/s] Loading 0: 23%|██▎ | 67/291 [00:00<00:01, 156.58it/s] Loading 0: 29%|██▊ | 83/291 [00:00<00:02, 82.33it/s] Loading 0: 33%|███▎ | 95/291 [00:00<00:02, 89.79it/s] Loading 0: 39%|███▉ | 113/291 [00:01<00:01, 107.37it/s] Loading 0: 45%|████▌ | 131/291 [00:01<00:01, 121.16it/s] Loading 0: 51%|█████ | 149/291 [00:01<00:01, 132.14it/s] Loading 0: 57%|█████▋ | 167/291 [00:01<00:00, 140.67it/s] Loading 0: 64%|██████▍ | 187/291 [00:01<00:01, 91.07it/s] Loading 0: 69%|██████▉ | 201/291 [00:01<00:00, 99.42it/s] Loading 0: 75%|███████▌ | 219/291 [00:01<00:00, 113.14it/s] Loading 0: 81%|████████▏ | 237/291 [00:02<00:00, 124.53it/s] Loading 0: 88%|████████▊ | 255/291 [00:02<00:00, 134.00it/s] Loading 0: 94%|█████████▍| 273/291 [00:02<00:00, 141.17it/s] Loading 0: 99%|█████████▉| 289/291 [00:07<00:00, 10.94it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
turboderp-llama3-turbcat-4336-v1-mkmlizer: quantized model in 18.374s
turboderp-llama3-turbcat-4336-v1-mkmlizer: Processed model turboderp/llama3-turbcat-instruct-8b in 68.084s
turboderp-llama3-turbcat-4336-v1-mkmlizer: creating bucket guanaco-mkml-models
turboderp-llama3-turbcat-4336-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
turboderp-llama3-turbcat-4336-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/turboderp-llama3-turbcat-4336-v1
turboderp-llama3-turbcat-4336-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/turboderp-llama3-turbcat-4336-v1/special_tokens_map.json
turboderp-llama3-turbcat-4336-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/turboderp-llama3-turbcat-4336-v1/config.json
turboderp-llama3-turbcat-4336-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/turboderp-llama3-turbcat-4336-v1/tokenizer_config.json
turboderp-llama3-turbcat-4336-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/turboderp-llama3-turbcat-4336-v1/tokenizer.json
turboderp-llama3-turbcat-4336-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/turboderp-llama3-turbcat-4336-v1/flywheel_model.0.safetensors
turboderp-llama3-turbcat-4336-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
turboderp-llama3-turbcat-4336-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.
turboderp-llama3-turbcat-4336-v1-mkmlizer: warnings.warn(
turboderp-llama3-turbcat-4336-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.
turboderp-llama3-turbcat-4336-v1-mkmlizer: warnings.warn(
turboderp-llama3-turbcat-4336-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.
turboderp-llama3-turbcat-4336-v1-mkmlizer: warnings.warn(
turboderp-llama3-turbcat-4336-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()
turboderp-llama3-turbcat-4336-v1-mkmlizer: return self.fget.__get__(instance, owner)()
turboderp-llama3-turbcat-4336-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
turboderp-llama3-turbcat-4336-v1-mkmlizer: Saving duration: 0.294s
turboderp-llama3-turbcat-4336-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 14.810s
turboderp-llama3-turbcat-4336-v1-mkmlizer: creating bucket guanaco-reward-models
turboderp-llama3-turbcat-4336-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
turboderp-llama3-turbcat-4336-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/turboderp-llama3-turbcat-4336-v1_reward
turboderp-llama3-turbcat-4336-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/turboderp-llama3-turbcat-4336-v1_reward/config.json
turboderp-llama3-turbcat-4336-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/turboderp-llama3-turbcat-4336-v1_reward/tokenizer_config.json
turboderp-llama3-turbcat-4336-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/turboderp-llama3-turbcat-4336-v1_reward/special_tokens_map.json
turboderp-llama3-turbcat-4336-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/turboderp-llama3-turbcat-4336-v1_reward/merges.txt
turboderp-llama3-turbcat-4336-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/turboderp-llama3-turbcat-4336-v1_reward/vocab.json
turboderp-llama3-turbcat-4336-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/turboderp-llama3-turbcat-4336-v1_reward/tokenizer.json
turboderp-llama3-turbcat-4336-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/turboderp-llama3-turbcat-4336-v1_reward/reward.tensors
Job turboderp-llama3-turbcat-4336-v1-mkmlizer completed after 105.47s with status: succeeded
Stopping job with name turboderp-llama3-turbcat-4336-v1-mkmlizer
Pipeline stage MKMLizer completed in 105.91s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.11s
Running pipeline stage ISVCDeployer
Creating inference service turboderp-llama3-turbcat-4336-v1
Waiting for inference service turboderp-llama3-turbcat-4336-v1 to be ready
Inference service turboderp-llama3-turbcat-4336-v1 ready after 30.20691752433777s
Pipeline stage ISVCDeployer completed in 35.97s
Running pipeline stage StressChecker
Received healthy response to inference request in 7.442602872848511s
Received healthy response to inference request in 1.209460973739624s
Received healthy response to inference request in 1.1213116645812988s
Received healthy response to inference request in 1.124342679977417s
Received healthy response to inference request in 1.1617398262023926s
5 requests
0 failed requests
5th percentile: 1.1219178676605224
10th percentile: 1.1225240707397461
20th percentile: 1.1237364768981934
30th percentile: 1.131822109222412
40th percentile: 1.1467809677124023
50th percentile: 1.1617398262023926
60th percentile: 1.1808282852172851
70th percentile: 1.1999167442321776
80th percentile: 2.456089353561403
90th percentile: 4.949346113204957
95th percentile: 6.195974493026732
99th percentile: 7.193277196884155
mean time: 2.4118916034698485
Pipeline stage StressChecker completed in 12.89s
turboderp-llama3-turbcat_4336_v1 status is now deployed due to DeploymentManager action
turboderp-llama3-turbcat_4336_v1 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics