submission_id: sao10k-l3-rp-v3-2_v4
developer_uid: sao10k
status: inactive
model_repo: Sao10K/L3-RP-v3.2
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 1.0, 'top_p': 0.95, 'min_p': 0.05, 'top_k': 80, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '<|end_header_id|>,', '<|eot_id|>,', '\n\n{user_name}'], 'max_input_tokens': 1024, 'best_of': 8, 'max_output_tokens': 64}
formatter: {'memory_template': "<|begin_of_text|><|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: 2024-06-05T16:58:35+00:00
model_name: V3-Expr1-Beta
model_eval_status: success
model_group: Sao10K/L3-RP-v3.2
num_battles: 21045
num_wins: 12018
celo_rating: 1221.63
propriety_score: 0.6753688989784336
propriety_total_count: 881.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 8
max_input_tokens: 1024
max_output_tokens: 64
display_name: V3-Expr1-Beta
ineligible_reason: propriety_total_count < 5000
language_model: Sao10K/L3-RP-v3.2
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-06-05
win_ratio: 0.5710620099786172
Resubmit model
Running pipeline stage MKMLizer
Starting job with name sao10k-l3-rp-v3-2-v4-mkmlizer
Waiting for job on sao10k-l3-rp-v3-2-v4-mkmlizer to finish
Stopping job with name sao10k-l3-rp-v3-2-v4-mkmlizer
%s, retrying in %s seconds...
Starting job with name sao10k-l3-rp-v3-2-v4-mkmlizer
Waiting for job on sao10k-l3-rp-v3-2-v4-mkmlizer to finish
Stopping job with name sao10k-l3-rp-v3-2-v4-mkmlizer
%s, retrying in %s seconds...
Starting job with name sao10k-l3-rp-v3-2-v4-mkmlizer
Waiting for job on sao10k-l3-rp-v3-2-v4-mkmlizer to finish
sao10k-l3-rp-v3-2-v4-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
sao10k-l3-rp-v3-2-v4-mkmlizer: ║ _____ __ __ ║
sao10k-l3-rp-v3-2-v4-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
sao10k-l3-rp-v3-2-v4-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
sao10k-l3-rp-v3-2-v4-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
sao10k-l3-rp-v3-2-v4-mkmlizer: ║ /___/ ║
sao10k-l3-rp-v3-2-v4-mkmlizer: ║ ║
sao10k-l3-rp-v3-2-v4-mkmlizer: ║ Version: 0.8.14 ║
sao10k-l3-rp-v3-2-v4-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
sao10k-l3-rp-v3-2-v4-mkmlizer: ║ https://mk1.ai ║
sao10k-l3-rp-v3-2-v4-mkmlizer: ║ ║
sao10k-l3-rp-v3-2-v4-mkmlizer: ║ The license key for the current software has been verified as ║
sao10k-l3-rp-v3-2-v4-mkmlizer: ║ belonging to: ║
sao10k-l3-rp-v3-2-v4-mkmlizer: ║ ║
sao10k-l3-rp-v3-2-v4-mkmlizer: ║ Chai Research Corp. ║
sao10k-l3-rp-v3-2-v4-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
sao10k-l3-rp-v3-2-v4-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
sao10k-l3-rp-v3-2-v4-mkmlizer: ║ ║
sao10k-l3-rp-v3-2-v4-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
sao10k-l3-rp-v3-2-v4-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.
sao10k-l3-rp-v3-2-v4-mkmlizer: warnings.warn(warning_message, FutureWarning)
sao10k-l3-rp-v3-2-v4-mkmlizer: Downloaded to shared memory in 16.347s
sao10k-l3-rp-v3-2-v4-mkmlizer: quantizing model to /dev/shm/model_cache
sao10k-l3-rp-v3-2-v4-mkmlizer: Saving flywheel model at /dev/shm/model_cache
sao10k-l3-rp-v3-2-v4-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 1%| | 2/291 [00:04<11:36, 2.41s/it] Loading 0: 4%|▍ | 13/291 [00:04<01:18, 3.56it/s] Loading 0: 8%|▊ | 22/291 [00:05<00:38, 6.94it/s] Loading 0: 11%|█ | 31/291 [00:05<00:23, 11.30it/s] Loading 0: 14%|█▎ | 40/291 [00:05<00:14, 16.73it/s] Loading 0: 17%|█▋ | 49/291 [00:05<00:10, 23.28it/s] Loading 0: 20%|█▉ | 58/291 [00:05<00:07, 30.73it/s] Loading 0: 23%|██▎ | 67/291 [00:06<00:09, 23.12it/s] Loading 0: 26%|██▌ | 76/291 [00:06<00:07, 29.94it/s] Loading 0: 29%|██▉ | 85/291 [00:06<00:05, 37.60it/s] Loading 0: 32%|███▏ | 94/291 [00:06<00:04, 45.70it/s] Loading 0: 35%|███▌ | 103/291 [00:06<00:03, 52.88it/s] Loading 0: 38%|███▊ | 112/291 [00:06<00:02, 59.76it/s] Loading 0: 42%|████▏ | 121/291 [00:06<00:02, 66.09it/s] Loading 0: 45%|████▍ | 130/291 [00:06<00:02, 71.22it/s] Loading 0: 48%|████▊ | 139/291 [00:06<00:02, 75.49it/s] Loading 0: 51%|█████ | 148/291 [00:06<00:01, 78.45it/s] Loading 0: 54%|█████▍ | 157/291 [00:07<00:01, 81.26it/s] Loading 0: 57%|█████▋ | 166/291 [00:07<00:02, 42.87it/s] Loading 0: 60%|██████ | 176/291 [00:07<00:02, 51.25it/s] Loading 0: 64%|██████▍ | 186/291 [00:07<00:01, 58.24it/s] Loading 0: 67%|██████▋ | 195/291 [00:07<00:01, 64.09it/s] Loading 0: 73%|███████▎ | 211/291 [00:07<00:00, 83.68it/s] Loading 0: 76%|███████▋ | 222/291 [00:08<00:00, 89.47it/s] Loading 0: 82%|████████▏ | 239/291 [00:08<00:00, 104.99it/s] Loading 0: 87%|████████▋ | 253/291 [00:08<00:00, 113.29it/s] Loading 0: 91%|█████████▏| 266/291 [00:08<00:00, 62.64it/s] Loading 0: 95%|█████████▍| 276/291 [00:08<00:00, 65.27it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
sao10k-l3-rp-v3-2-v4-mkmlizer: quantized model in 22.839s
sao10k-l3-rp-v3-2-v4-mkmlizer: Processed model Sao10K/L3-RP-v3.2 in 40.469s
sao10k-l3-rp-v3-2-v4-mkmlizer: creating bucket guanaco-mkml-models
sao10k-l3-rp-v3-2-v4-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
sao10k-l3-rp-v3-2-v4-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/sao10k-l3-rp-v3-2-v4
sao10k-l3-rp-v3-2-v4-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/sao10k-l3-rp-v3-2-v4/special_tokens_map.json
sao10k-l3-rp-v3-2-v4-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/sao10k-l3-rp-v3-2-v4/config.json
sao10k-l3-rp-v3-2-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/sao10k-l3-rp-v3-2-v4/tokenizer_config.json
sao10k-l3-rp-v3-2-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/sao10k-l3-rp-v3-2-v4/tokenizer.json
sao10k-l3-rp-v3-2-v4-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/sao10k-l3-rp-v3-2-v4/flywheel_model.0.safetensors
sao10k-l3-rp-v3-2-v4-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
sao10k-l3-rp-v3-2-v4-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.
sao10k-l3-rp-v3-2-v4-mkmlizer: warnings.warn(
sao10k-l3-rp-v3-2-v4-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.
sao10k-l3-rp-v3-2-v4-mkmlizer: warnings.warn(
sao10k-l3-rp-v3-2-v4-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.
sao10k-l3-rp-v3-2-v4-mkmlizer: warnings.warn(
sao10k-l3-rp-v3-2-v4-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()
sao10k-l3-rp-v3-2-v4-mkmlizer: return self.fget.__get__(instance, owner)()
sao10k-l3-rp-v3-2-v4-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
sao10k-l3-rp-v3-2-v4-mkmlizer: Saving duration: 0.307s
sao10k-l3-rp-v3-2-v4-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 6.937s
sao10k-l3-rp-v3-2-v4-mkmlizer: creating bucket guanaco-reward-models
sao10k-l3-rp-v3-2-v4-mkmlizer: Bucket 's3://guanaco-reward-models/' created
sao10k-l3-rp-v3-2-v4-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/sao10k-l3-rp-v3-2-v4_reward
sao10k-l3-rp-v3-2-v4-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/sao10k-l3-rp-v3-2-v4_reward/special_tokens_map.json
sao10k-l3-rp-v3-2-v4-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/sao10k-l3-rp-v3-2-v4_reward/tokenizer_config.json
sao10k-l3-rp-v3-2-v4-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/sao10k-l3-rp-v3-2-v4_reward/config.json
sao10k-l3-rp-v3-2-v4-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/sao10k-l3-rp-v3-2-v4_reward/merges.txt
sao10k-l3-rp-v3-2-v4-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/sao10k-l3-rp-v3-2-v4_reward/vocab.json
sao10k-l3-rp-v3-2-v4-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/sao10k-l3-rp-v3-2-v4_reward/tokenizer.json
sao10k-l3-rp-v3-2-v4-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/sao10k-l3-rp-v3-2-v4_reward/reward.tensors
Job sao10k-l3-rp-v3-2-v4-mkmlizer completed after 63.33s with status: succeeded
Stopping job with name sao10k-l3-rp-v3-2-v4-mkmlizer
Pipeline stage MKMLizer completed in 77.94s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service sao10k-l3-rp-v3-2-v4
Waiting for inference service sao10k-l3-rp-v3-2-v4 to be ready
Inference service sao10k-l3-rp-v3-2-v4 ready after 70.60356068611145s
Pipeline stage ISVCDeployer completed in 78.63s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.1446032524108887s
Received healthy response to inference request in 1.2319555282592773s
Received healthy response to inference request in 1.2604146003723145s
Received healthy response to inference request in 1.266988754272461s
Received healthy response to inference request in 1.2458217144012451s
5 requests
0 failed requests
5th percentile: 1.2347287654876709
10th percentile: 1.2375020027160644
20th percentile: 1.2430484771728516
30th percentile: 1.248740291595459
40th percentile: 1.2545774459838868
50th percentile: 1.2604146003723145
60th percentile: 1.263044261932373
70th percentile: 1.2656739234924317
80th percentile: 1.4425116539001466
90th percentile: 1.7935574531555176
95th percentile: 1.969080352783203
99th percentile: 2.1094986724853517
mean time: 1.4299567699432374
Pipeline stage StressChecker completed in 7.79s
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
sao10k-l3-rp-v3-2_v4 status is now deployed due to DeploymentManager action
sao10k-l3-rp-v3-2_v4 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics