submission_id: chaser-cz-qlora-out_v5
developer_uid: chace9580
status: inactive
model_repo: Chaser-cz/qlora-out
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: 2024-07-03T02:52:17+00:00
model_name: chaser-cz-qlora-out_v5
model_group: Chaser-cz/qlora-out
num_battles: 14123
num_wins: 6523
celo_rating: 1155.37
propriety_score: 0.6993764845605701
propriety_total_count: 6736.0
submission_type: basic
model_architecture: MistralForCausalLM
model_num_parameters: 7241732096.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: chaser-cz-qlora-out_v5
ineligible_reason: None
language_model: Chaser-cz/qlora-out
model_size: 7B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-07-02
win_ratio: 0.46187070735679386
Resubmit model
Running pipeline stage MKMLizer
Starting job with name chaser-cz-qlora-out-v5-mkmlizer
Waiting for job on chaser-cz-qlora-out-v5-mkmlizer to finish
chaser-cz-qlora-out-v5-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaser-cz-qlora-out-v5-mkmlizer: ║ _____ __ __ ║
chaser-cz-qlora-out-v5-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
chaser-cz-qlora-out-v5-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
chaser-cz-qlora-out-v5-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
chaser-cz-qlora-out-v5-mkmlizer: ║ /___/ ║
chaser-cz-qlora-out-v5-mkmlizer: ║ ║
chaser-cz-qlora-out-v5-mkmlizer: ║ Version: 0.8.14 ║
chaser-cz-qlora-out-v5-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
chaser-cz-qlora-out-v5-mkmlizer: ║ https://mk1.ai ║
chaser-cz-qlora-out-v5-mkmlizer: ║ ║
chaser-cz-qlora-out-v5-mkmlizer: ║ The license key for the current software has been verified as ║
chaser-cz-qlora-out-v5-mkmlizer: ║ belonging to: ║
chaser-cz-qlora-out-v5-mkmlizer: ║ ║
chaser-cz-qlora-out-v5-mkmlizer: ║ Chai Research Corp. ║
chaser-cz-qlora-out-v5-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
chaser-cz-qlora-out-v5-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
chaser-cz-qlora-out-v5-mkmlizer: ║ ║
chaser-cz-qlora-out-v5-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaser-cz-qlora-out-v5-mkmlizer: Downloaded to shared memory in 22.773s
chaser-cz-qlora-out-v5-mkmlizer: quantizing model to /dev/shm/model_cache
chaser-cz-qlora-out-v5-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaser-cz-qlora-out-v5-mkmlizer: quantized model in 12.641s
chaser-cz-qlora-out-v5-mkmlizer: Processed model Chaser-cz/qlora-out in 35.415s
chaser-cz-qlora-out-v5-mkmlizer: creating bucket guanaco-mkml-models
chaser-cz-qlora-out-v5-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaser-cz-qlora-out-v5-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaser-cz-qlora-out-v5
chaser-cz-qlora-out-v5-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaser-cz-qlora-out-v5/config.json
chaser-cz-qlora-out-v5-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaser-cz-qlora-out-v5/special_tokens_map.json
chaser-cz-qlora-out-v5-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaser-cz-qlora-out-v5/tokenizer_config.json
chaser-cz-qlora-out-v5-mkmlizer: cp /dev/shm/model_cache/tokenizer.model s3://guanaco-mkml-models/chaser-cz-qlora-out-v5/tokenizer.model
chaser-cz-qlora-out-v5-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaser-cz-qlora-out-v5/tokenizer.json
chaser-cz-qlora-out-v5-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
chaser-cz-qlora-out-v5-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 34%|███▎ | 98/291 [00:01<00:02, 82.04it/s] Loading 0: 70%|███████ | 204/291 [00:02<00:00, 88.70it/s] Loading 0: 100%|██████████| 291/291 [00:03<00:00, 73.63it/s] /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.
chaser-cz-qlora-out-v5-mkmlizer: warnings.warn(
chaser-cz-qlora-out-v5-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`.
chaser-cz-qlora-out-v5-mkmlizer: warnings.warn(
chaser-cz-qlora-out-v5-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.
chaser-cz-qlora-out-v5-mkmlizer: warnings.warn(
chaser-cz-qlora-out-v5-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.
chaser-cz-qlora-out-v5-mkmlizer: warnings.warn(
chaser-cz-qlora-out-v5-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()
chaser-cz-qlora-out-v5-mkmlizer: return self.fget.__get__(instance, owner)()
chaser-cz-qlora-out-v5-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
chaser-cz-qlora-out-v5-mkmlizer: Saving duration: 0.284s
chaser-cz-qlora-out-v5-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 3.385s
chaser-cz-qlora-out-v5-mkmlizer: creating bucket guanaco-reward-models
chaser-cz-qlora-out-v5-mkmlizer: Bucket 's3://guanaco-reward-models/' created
chaser-cz-qlora-out-v5-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/chaser-cz-qlora-out-v5_reward
chaser-cz-qlora-out-v5-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/chaser-cz-qlora-out-v5_reward/config.json
chaser-cz-qlora-out-v5-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/chaser-cz-qlora-out-v5_reward/special_tokens_map.json
chaser-cz-qlora-out-v5-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/chaser-cz-qlora-out-v5_reward/vocab.json
chaser-cz-qlora-out-v5-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/chaser-cz-qlora-out-v5_reward/tokenizer_config.json
chaser-cz-qlora-out-v5-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/chaser-cz-qlora-out-v5_reward/merges.txt
chaser-cz-qlora-out-v5-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/chaser-cz-qlora-out-v5_reward/tokenizer.json
chaser-cz-qlora-out-v5-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/chaser-cz-qlora-out-v5_reward/reward.tensors
Connection pool is full, discarding connection: %s
Job chaser-cz-qlora-out-v5-mkmlizer completed after 79.82s with status: succeeded
Stopping job with name chaser-cz-qlora-out-v5-mkmlizer
Pipeline stage MKMLizer completed in 80.69s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service chaser-cz-qlora-out-v5
Waiting for inference service chaser-cz-qlora-out-v5 to be ready
Inference service chaser-cz-qlora-out-v5 ready after 40.25394129753113s
Pipeline stage ISVCDeployer completed in 46.88s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.935368299484253s
Received healthy response to inference request in 1.355947494506836s
Received healthy response to inference request in 1.1645073890686035s
Received healthy response to inference request in 1.282494068145752s
Received healthy response to inference request in 1.0137720108032227s
5 requests
0 failed requests
5th percentile: 1.0439190864562988
10th percentile: 1.074066162109375
20th percentile: 1.1343603134155273
30th percentile: 1.1881047248840333
40th percentile: 1.2352993965148926
50th percentile: 1.282494068145752
60th percentile: 1.3118754386901856
70th percentile: 1.341256809234619
80th percentile: 1.4718316555023194
90th percentile: 1.7035999774932862
95th percentile: 1.8194841384887694
99th percentile: 1.9121914672851563
mean time: 1.3504178524017334
Pipeline stage StressChecker completed in 8.05s
chaser-cz-qlora-out_v5 status is now deployed due to DeploymentManager action
chaser-cz-qlora-out_v5 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics