submission_id: hastagaras-waduh-m-llama-3-8b_v1
developer_uid: Hastagaras
status: inactive
model_repo: Hastagaras/Waduh-M-llama-3-8b
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 1.0, 'top_p': 1.0, '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': "<|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: 2024-05-19T02:47:23+00:00
model_name: mwrge-test
model_eval_status: success
model_group: Hastagaras/Waduh-M-llama
num_battles: 15868
num_wins: 8742
celo_rating: 1206.61
safety_score: 0.95
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: mwrge-test
ineligible_reason: propriety_total_count < 5000
language_model: Hastagaras/Waduh-M-llama-3-8b
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-05-18
win_ratio: 0.5509200907486765
Resubmit model
Running pipeline stage MKMLizer
Starting job with name hastagaras-waduh-m-llama-3-8b-v1-mkmlizer
Waiting for job on hastagaras-waduh-m-llama-3-8b-v1-mkmlizer to finish
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: ║ _____ __ __ ║
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: ║ /___/ ║
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: ║ ║
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: ║ Version: 0.8.14 ║
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: ║ https://mk1.ai ║
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: ║ ║
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: ║ The license key for the current software has been verified as ║
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: ║ belonging to: ║
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: ║ ║
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: ║ Chai Research Corp. ║
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: ║ ║
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
hastagaras-waduh-m-llama-3-8b-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-waduh-m-llama-3-8b-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: Downloaded to shared memory in 36.276s
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: quantizing model to /dev/shm/model_cache
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 1%| | 2/291 [00:04<09:38, 2.00s/it] Loading 0: 8%|▊ | 22/291 [00:04<00:36, 7.36it/s] Loading 0: 14%|█▍ | 41/291 [00:04<00:15, 15.89it/s] Loading 0: 21%|██ | 60/291 [00:04<00:10, 22.14it/s] Loading 0: 29%|██▉ | 85/291 [00:04<00:05, 37.54it/s] Loading 0: 36%|███▌ | 104/291 [00:04<00:03, 50.58it/s] Loading 0: 42%|████▏ | 123/291 [00:04<00:02, 65.62it/s] Loading 0: 49%|████▉ | 144/291 [00:05<00:01, 84.98it/s] Loading 0: 56%|█████▌ | 162/291 [00:05<00:01, 99.95it/s] Loading 0: 62%|██████▏ | 180/291 [00:05<00:01, 70.98it/s] Loading 0: 68%|██████▊ | 199/291 [00:05<00:01, 87.77it/s] Loading 0: 75%|███████▍ | 217/291 [00:05<00:00, 103.29it/s] Loading 0: 82%|████████▏ | 238/291 [00:05<00:00, 121.25it/s] Loading 0: 88%|████████▊ | 257/291 [00:06<00:00, 134.68it/s] Loading 0: 95%|█████████▍| 275/291 [00:06<00:00, 88.14it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: quantized model in 17.987s
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: Processed model Hastagaras/Waduh-M-llama-3-8b in 55.362s
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: creating bucket guanaco-mkml-models
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/hastagaras-waduh-m-llama-3-8b-v1
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/hastagaras-waduh-m-llama-3-8b-v1/special_tokens_map.json
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/hastagaras-waduh-m-llama-3-8b-v1/config.json
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/hastagaras-waduh-m-llama-3-8b-v1/tokenizer_config.json
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/hastagaras-waduh-m-llama-3-8b-v1/tokenizer.json
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/hastagaras-waduh-m-llama-3-8b-v1/flywheel_model.0.safetensors
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
hastagaras-waduh-m-llama-3-8b-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-waduh-m-llama-3-8b-v1-mkmlizer: warnings.warn(
hastagaras-waduh-m-llama-3-8b-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-waduh-m-llama-3-8b-v1-mkmlizer: warnings.warn(
hastagaras-waduh-m-llama-3-8b-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-waduh-m-llama-3-8b-v1-mkmlizer: warnings.warn(
hastagaras-waduh-m-llama-3-8b-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-waduh-m-llama-3-8b-v1-mkmlizer: return self.fget.__get__(instance, owner)()
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: Saving duration: 0.257s
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 5.888s
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: creating bucket guanaco-reward-models
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/hastagaras-waduh-m-llama-3-8b-v1_reward
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/hastagaras-waduh-m-llama-3-8b-v1_reward/config.json
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/hastagaras-waduh-m-llama-3-8b-v1_reward/vocab.json
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/hastagaras-waduh-m-llama-3-8b-v1_reward/merges.txt
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/hastagaras-waduh-m-llama-3-8b-v1_reward/tokenizer_config.json
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/hastagaras-waduh-m-llama-3-8b-v1_reward/special_tokens_map.json
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/hastagaras-waduh-m-llama-3-8b-v1_reward/tokenizer.json
hastagaras-waduh-m-llama-3-8b-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/hastagaras-waduh-m-llama-3-8b-v1_reward/reward.tensors
Job hastagaras-waduh-m-llama-3-8b-v1-mkmlizer completed after 83.19s with status: succeeded
Stopping job with name hastagaras-waduh-m-llama-3-8b-v1-mkmlizer
Pipeline stage MKMLizer completed in 86.28s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.11s
Running pipeline stage ISVCDeployer
Creating inference service hastagaras-waduh-m-llama-3-8b-v1
Waiting for inference service hastagaras-waduh-m-llama-3-8b-v1 to be ready
Inference service hastagaras-waduh-m-llama-3-8b-v1 ready after 40.22740149497986s
Pipeline stage ISVCDeployer completed in 47.09s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.271522283554077s
Received healthy response to inference request in 1.3455493450164795s
Received healthy response to inference request in 1.3469061851501465s
Received healthy response to inference request in 1.306917667388916s
Received healthy response to inference request in 1.3768694400787354s
5 requests
0 failed requests
5th percentile: 1.3146440029144286
10th percentile: 1.3223703384399415
20th percentile: 1.3378230094909669
30th percentile: 1.345820713043213
40th percentile: 1.3463634490966796
50th percentile: 1.3469061851501465
60th percentile: 1.358891487121582
70th percentile: 1.3708767890930176
80th percentile: 1.555800008773804
90th percentile: 1.9136611461639406
95th percentile: 2.0925917148590085
99th percentile: 2.2357361698150635
mean time: 1.529552984237671
Pipeline stage StressChecker completed in 8.24s
Running pipeline stage DaemonicModelEvalScorer
Pipeline stage DaemonicModelEvalScorer completed in 0.03s
Running pipeline stage DaemonicSafetyScorer
Pipeline stage DaemonicSafetyScorer completed in 0.03s
Running M-Eval for topic stay_in_character
hastagaras-waduh-m-llama-3-8b_v1 status is now deployed due to DeploymentManager action
M-Eval Dataset for topic stay_in_character is loaded
hastagaras-waduh-m-llama-3-8b_v1 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics