submission_id: saishf-sovl-mega-mash-l3-8b_v1
developer_uid: saishf
status: torndown
model_repo: saishf/SOVL-Mega-Mash-L3-8B
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': "{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}
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: 2024-05-12T13:43:56+00:00
model_name: saishf-sovl-mega-mash-l3-8b_v1
model_eval_status: success
model_group: saishf/SOVL-Mega-Mash-L3
num_battles: 7267
num_wins: 3637
celo_rating: 1172.12
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: saishf-sovl-mega-mash-l3-8b_v1
ineligible_reason: propriety_total_count < 800
language_model: saishf/SOVL-Mega-Mash-L3-8B
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-05-12
win_ratio: 0.5004816292830604
preference_data_url: None
Resubmit model
Running pipeline stage MKMLizer
Starting job with name saishf-sovl-mega-mash-l3-8b-v1-mkmlizer
Waiting for job on saishf-sovl-mega-mash-l3-8b-v1-mkmlizer to finish
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: ║ _____ __ __ ║
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: ║ /___/ ║
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: ║ ║
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: ║ Version: 0.8.10 ║
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: ║ ║
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: ║ The license key for the current software has been verified as ║
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: ║ belonging to: ║
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: ║ ║
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: ║ Chai Research Corp. ║
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: ║ ║
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
saishf-sovl-mega-mash-l3-8b-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.
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: Downloaded to shared memory in 28.112s
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: quantizing model to /dev/shm/model_cache
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 1%| | 2/291 [00:04<10:12, 2.12s/it] Loading 0: 42%|████▏ | 122/291 [00:05<00:05, 29.93it/s] Loading 0: 93%|█████████▎| 271/291 [00:06<00:00, 60.94it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: quantized model in 18.378s
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: Processed model saishf/SOVL-Mega-Mash-L3-8B in 47.571s
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: creating bucket guanaco-mkml-models
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/saishf-sovl-mega-mash-l3-8b-v1
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/saishf-sovl-mega-mash-l3-8b-v1/tokenizer_config.json
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/saishf-sovl-mega-mash-l3-8b-v1/config.json
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/saishf-sovl-mega-mash-l3-8b-v1/special_tokens_map.json
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/saishf-sovl-mega-mash-l3-8b-v1/tokenizer.json
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/saishf-sovl-mega-mash-l3-8b-v1/flywheel_model.0.safetensors
saishf-sovl-mega-mash-l3-8b-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.
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: warnings.warn(
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
saishf-sovl-mega-mash-l3-8b-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.
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: warnings.warn(
saishf-sovl-mega-mash-l3-8b-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()
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: return self.fget.__get__(instance, owner)()
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: Saving duration: 0.252s
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 7.368s
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: creating bucket guanaco-reward-models
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/saishf-sovl-mega-mash-l3-8b-v1_reward
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/saishf-sovl-mega-mash-l3-8b-v1_reward/config.json
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/saishf-sovl-mega-mash-l3-8b-v1_reward/tokenizer_config.json
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/saishf-sovl-mega-mash-l3-8b-v1_reward/special_tokens_map.json
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/saishf-sovl-mega-mash-l3-8b-v1_reward/merges.txt
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/saishf-sovl-mega-mash-l3-8b-v1_reward/vocab.json
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/saishf-sovl-mega-mash-l3-8b-v1_reward/tokenizer.json
saishf-sovl-mega-mash-l3-8b-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/saishf-sovl-mega-mash-l3-8b-v1_reward/reward.tensors
Job saishf-sovl-mega-mash-l3-8b-v1-mkmlizer completed after 72.88s with status: succeeded
Stopping job with name saishf-sovl-mega-mash-l3-8b-v1-mkmlizer
Pipeline stage MKMLizer completed in 75.22s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service saishf-sovl-mega-mash-l3-8b-v1
Waiting for inference service saishf-sovl-mega-mash-l3-8b-v1 to be ready
Inference service saishf-sovl-mega-mash-l3-8b-v1 ready after 50.28426742553711s
Pipeline stage ISVCDeployer completed in 56.79s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.3670032024383545s
Received healthy response to inference request in 1.2984397411346436s
Received healthy response to inference request in 1.2973828315734863s
Received healthy response to inference request in 1.2779037952423096s
Received healthy response to inference request in 1.2803890705108643s
5 requests
0 failed requests
5th percentile: 1.2784008502960205
10th percentile: 1.2788979053497314
20th percentile: 1.2798920154571534
30th percentile: 1.2837878227233888
40th percentile: 1.2905853271484375
50th percentile: 1.2973828315734863
60th percentile: 1.2978055953979493
70th percentile: 1.2982283592224122
80th percentile: 1.512152433395386
90th percentile: 1.9395778179168701
95th percentile: 2.1532905101776123
99th percentile: 2.3242606639862062
mean time: 1.5042237281799316
Pipeline stage StressChecker completed in 8.08s
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 0.03s
M-Eval Dataset for topic stay_in_character is loaded
saishf-sovl-mega-mash-l3-8b_v1 status is now deployed due to DeploymentManager action
saishf-sovl-mega-mash-l3-8b_v1 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of saishf-sovl-mega-mash-l3-8b_v1
Running pipeline stage ISVCDeleter
Checking if service saishf-sovl-mega-mash-l3-8b-v1 is running
Tearing down inference service saishf-sovl-mega-mash-l3-8b-v1
Toredown service saishf-sovl-mega-mash-l3-8b-v1
Pipeline stage ISVCDeleter completed in 5.23s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key saishf-sovl-mega-mash-l3-8b-v1/config.json from bucket guanaco-mkml-models
Deleting key saishf-sovl-mega-mash-l3-8b-v1/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key saishf-sovl-mega-mash-l3-8b-v1/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key saishf-sovl-mega-mash-l3-8b-v1/tokenizer.json from bucket guanaco-mkml-models
Deleting key saishf-sovl-mega-mash-l3-8b-v1/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key saishf-sovl-mega-mash-l3-8b-v1_reward/config.json from bucket guanaco-reward-models
Deleting key saishf-sovl-mega-mash-l3-8b-v1_reward/merges.txt from bucket guanaco-reward-models
Deleting key saishf-sovl-mega-mash-l3-8b-v1_reward/reward.tensors from bucket guanaco-reward-models
Deleting key saishf-sovl-mega-mash-l3-8b-v1_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key saishf-sovl-mega-mash-l3-8b-v1_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key saishf-sovl-mega-mash-l3-8b-v1_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key saishf-sovl-mega-mash-l3-8b-v1_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 3.23s
saishf-sovl-mega-mash-l3-8b_v1 status is now torndown due to DeploymentManager action

Usage Metrics

Latency Metrics