submission_id: meseca-caspian-0_v1
developer_uid: nguyenzzz
status: inactive
model_repo: meseca/caspian-0
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 1.1, 'top_p': 1.0, 'min_p': 0.08, '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: 2024-06-03T08:15:49+00:00
model_name: experiment-2-high-temp-test
model_eval_status: success
model_group: meseca/caspian-0
num_battles: 67680
num_wins: 36534
celo_rating: 1211.04
safety_score: 0.93
propriety_score: 0.6620594333102972
propriety_total_count: 1447.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: experiment-2-high-temp-test
ineligible_reason: propriety_total_count < 5000
language_model: meseca/caspian-0
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-06-03
win_ratio: 0.5398049645390071
Resubmit model
Running pipeline stage MKMLizer
Starting job with name meseca-caspian-0-v1-mkmlizer
Waiting for job on meseca-caspian-0-v1-mkmlizer to finish
meseca-caspian-0-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
meseca-caspian-0-v1-mkmlizer: ║ _____ __ __ ║
meseca-caspian-0-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
meseca-caspian-0-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
meseca-caspian-0-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
meseca-caspian-0-v1-mkmlizer: ║ /___/ ║
meseca-caspian-0-v1-mkmlizer: ║ ║
meseca-caspian-0-v1-mkmlizer: ║ Version: 0.8.14 ║
meseca-caspian-0-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
meseca-caspian-0-v1-mkmlizer: ║ https://mk1.ai ║
meseca-caspian-0-v1-mkmlizer: ║ ║
meseca-caspian-0-v1-mkmlizer: ║ The license key for the current software has been verified as ║
meseca-caspian-0-v1-mkmlizer: ║ belonging to: ║
meseca-caspian-0-v1-mkmlizer: ║ ║
meseca-caspian-0-v1-mkmlizer: ║ Chai Research Corp. ║
meseca-caspian-0-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
meseca-caspian-0-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
meseca-caspian-0-v1-mkmlizer: ║ ║
meseca-caspian-0-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
meseca-caspian-0-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.
meseca-caspian-0-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
meseca-caspian-0-v1-mkmlizer: Downloaded to shared memory in 28.764s
meseca-caspian-0-v1-mkmlizer: quantizing model to /dev/shm/model_cache
meseca-caspian-0-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
meseca-caspian-0-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 1%| | 2/291 [00:04<10:46, 2.24s/it] Loading 0: 5%|▌ | 15/291 [00:04<01:01, 4.45it/s] Loading 0: 11%|█▏ | 33/291 [00:04<00:21, 11.85it/s] Loading 0: 18%|█▊ | 51/291 [00:04<00:11, 21.35it/s] Loading 0: 22%|██▏ | 65/291 [00:05<00:08, 26.05it/s] Loading 0: 28%|██▊ | 82/291 [00:05<00:05, 38.10it/s] Loading 0: 34%|███▍ | 99/291 [00:05<00:03, 52.24it/s] Loading 0: 40%|███▉ | 115/291 [00:05<00:02, 66.52it/s] Loading 0: 45%|████▌ | 132/291 [00:05<00:01, 82.20it/s] Loading 0: 52%|█████▏ | 150/291 [00:05<00:01, 97.36it/s] Loading 0: 57%|█████▋ | 166/291 [00:05<00:01, 76.71it/s] Loading 0: 63%|██████▎ | 184/291 [00:06<00:01, 92.56it/s] Loading 0: 69%|██████▉ | 202/291 [00:06<00:00, 108.48it/s] Loading 0: 76%|███████▌ | 220/291 [00:06<00:00, 122.22it/s] Loading 0: 82%|████████▏ | 238/291 [00:06<00:00, 131.82it/s] Loading 0: 88%|████████▊ | 256/291 [00:06<00:00, 139.49it/s] Loading 0: 93%|█████████▎| 272/291 [00:06<00:00, 91.49it/s] Loading 0: 99%|█████████▊| 287/291 [00:06<00:00, 102.35it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
meseca-caspian-0-v1-mkmlizer: quantized model in 22.494s
meseca-caspian-0-v1-mkmlizer: Processed model meseca/caspian-0 in 53.700s
meseca-caspian-0-v1-mkmlizer: creating bucket guanaco-mkml-models
meseca-caspian-0-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
meseca-caspian-0-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/meseca-caspian-0-v1
meseca-caspian-0-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/meseca-caspian-0-v1/config.json
meseca-caspian-0-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/meseca-caspian-0-v1/tokenizer_config.json
meseca-caspian-0-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/meseca-caspian-0-v1/tokenizer.json
meseca-caspian-0-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/meseca-caspian-0-v1/special_tokens_map.json
meseca-caspian-0-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/meseca-caspian-0-v1/flywheel_model.0.safetensors
meseca-caspian-0-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
meseca-caspian-0-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.
meseca-caspian-0-v1-mkmlizer: warnings.warn(
meseca-caspian-0-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.
meseca-caspian-0-v1-mkmlizer: warnings.warn(
meseca-caspian-0-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.
meseca-caspian-0-v1-mkmlizer: warnings.warn(
meseca-caspian-0-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()
meseca-caspian-0-v1-mkmlizer: return self.fget.__get__(instance, owner)()
meseca-caspian-0-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
meseca-caspian-0-v1-mkmlizer: Saving duration: 0.391s
meseca-caspian-0-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 12.995s
meseca-caspian-0-v1-mkmlizer: creating bucket guanaco-reward-models
meseca-caspian-0-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
meseca-caspian-0-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/meseca-caspian-0-v1_reward
meseca-caspian-0-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/meseca-caspian-0-v1_reward/tokenizer_config.json
meseca-caspian-0-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/meseca-caspian-0-v1_reward/config.json
meseca-caspian-0-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/meseca-caspian-0-v1_reward/special_tokens_map.json
meseca-caspian-0-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/meseca-caspian-0-v1_reward/merges.txt
meseca-caspian-0-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/meseca-caspian-0-v1_reward/vocab.json
meseca-caspian-0-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/meseca-caspian-0-v1_reward/tokenizer.json
meseca-caspian-0-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/meseca-caspian-0-v1_reward/reward.tensors
Job meseca-caspian-0-v1-mkmlizer completed after 82.7s with status: succeeded
Stopping job with name meseca-caspian-0-v1-mkmlizer
Pipeline stage MKMLizer completed in 86.25s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.08s
Running pipeline stage ISVCDeployer
Creating inference service meseca-caspian-0-v1
Waiting for inference service meseca-caspian-0-v1 to be ready
Inference service meseca-caspian-0-v1 ready after 190.89154314994812s
Pipeline stage ISVCDeployer completed in 197.88s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.1502387523651123s
Received healthy response to inference request in 1.3257484436035156s
Received healthy response to inference request in 1.2974021434783936s
Received healthy response to inference request in 1.2725307941436768s
Received healthy response to inference request in 1.2279510498046875s
5 requests
0 failed requests
5th percentile: 1.2368669986724854
10th percentile: 1.2457829475402833
20th percentile: 1.2636148452758789
30th percentile: 1.27750506401062
40th percentile: 1.2874536037445068
50th percentile: 1.2974021434783936
60th percentile: 1.3087406635284424
70th percentile: 1.3200791835784913
80th percentile: 1.490646505355835
90th percentile: 1.8204426288604738
95th percentile: 1.9853406906127928
99th percentile: 2.1172591400146485
mean time: 1.4547742366790772
Pipeline stage StressChecker completed in 7.91s
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
meseca-caspian-0_v1 status is now deployed due to DeploymentManager action
meseca-caspian-0_v1 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics