submission_id: hastagaras-halu-oas-8b-l_4311_v2
developer_uid: Hastagaras
status: inactive
model_repo: Hastagaras/Halu-OAS-8B-Llama3-TUNED
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 0.8, 'top_p': 0.95, 'min_p': 0.05, '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: 2024-05-28T06:10:02+00:00
model_name: this-thing-is-crazy
model_eval_status: success
model_group: Hastagaras/Halu-OAS-8B-L
num_battles: 12228
num_wins: 6690
celo_rating: 1203.32
safety_score: 0.78
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: this-thing-is-crazy
ineligible_reason: propriety_total_count < 5000
language_model: Hastagaras/Halu-OAS-8B-Llama3-TUNED
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-05-27
win_ratio: 0.5471050049067714
Resubmit model
Running pipeline stage MKMLizer
Starting job with name hastagaras-halu-oas-8b-l-4311-v2-mkmlizer
Waiting for job on hastagaras-halu-oas-8b-l-4311-v2-mkmlizer to finish
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: ║ _____ __ __ ║
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: ║ /___/ ║
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: ║ ║
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: ║ Version: 0.8.14 ║
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: ║ https://mk1.ai ║
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: ║ ║
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: ║ The license key for the current software has been verified as ║
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: ║ belonging to: ║
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: ║ ║
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: ║ Chai Research Corp. ║
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: ║ ║
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
hastagaras-halu-oas-8b-l-4311-v2-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-halu-oas-8b-l-4311-v2-mkmlizer: warnings.warn(warning_message, FutureWarning)
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: Downloaded to shared memory in 21.554s
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: quantizing model to /dev/shm/model_cache
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 5%|▍ | 14/291 [00:00<00:02, 136.45it/s] Loading 0: 12%|█▏ | 35/291 [00:00<00:01, 175.79it/s] Loading 0: 19%|█▊ | 54/291 [00:00<00:01, 180.13it/s] Loading 0: 25%|██▌ | 73/291 [00:00<00:01, 167.41it/s] Loading 0: 31%|███ | 90/291 [00:00<00:02, 75.72it/s] Loading 0: 38%|███▊ | 111/291 [00:01<00:01, 97.75it/s] Loading 0: 45%|████▍ | 130/291 [00:01<00:01, 114.60it/s] Loading 0: 51%|█████ | 149/291 [00:01<00:01, 129.28it/s] Loading 0: 59%|█████▉ | 171/291 [00:01<00:00, 149.52it/s] Loading 0: 65%|██████▍ | 189/291 [00:01<00:01, 84.39it/s] Loading 0: 72%|███████▏ | 210/291 [00:01<00:00, 103.39it/s] Loading 0: 79%|███████▊ | 229/291 [00:01<00:00, 118.40it/s] Loading 0: 85%|████████▌ | 248/291 [00:02<00:00, 131.43it/s] Loading 0: 92%|█████████▏| 269/291 [00:02<00:00, 149.20it/s] Loading 0: 99%|█████████▊| 287/291 [00:07<00:00, 11.38it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: quantized model in 19.432s
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: Processed model Hastagaras/Halu-OAS-8B-Llama3-TUNED in 42.228s
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: creating bucket guanaco-mkml-models
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/hastagaras-halu-oas-8b-l-4311-v2
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/hastagaras-halu-oas-8b-l-4311-v2/config.json
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/hastagaras-halu-oas-8b-l-4311-v2/special_tokens_map.json
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/hastagaras-halu-oas-8b-l-4311-v2/tokenizer_config.json
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/hastagaras-halu-oas-8b-l-4311-v2/tokenizer.json
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/hastagaras-halu-oas-8b-l-4311-v2/flywheel_model.0.safetensors
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
hastagaras-halu-oas-8b-l-4311-v2-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-halu-oas-8b-l-4311-v2-mkmlizer: warnings.warn(
hastagaras-halu-oas-8b-l-4311-v2-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-halu-oas-8b-l-4311-v2-mkmlizer: warnings.warn(
hastagaras-halu-oas-8b-l-4311-v2-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-halu-oas-8b-l-4311-v2-mkmlizer: warnings.warn(
hastagaras-halu-oas-8b-l-4311-v2-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-halu-oas-8b-l-4311-v2-mkmlizer: return self.fget.__get__(instance, owner)()
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: Saving duration: 0.277s
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 6.330s
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: Bucket 's3://guanaco-reward-models/' created
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/hastagaras-halu-oas-8b-l-4311-v2_reward
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/hastagaras-halu-oas-8b-l-4311-v2_reward/config.json
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/hastagaras-halu-oas-8b-l-4311-v2_reward/special_tokens_map.json
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/hastagaras-halu-oas-8b-l-4311-v2_reward/tokenizer_config.json
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/hastagaras-halu-oas-8b-l-4311-v2_reward/merges.txt
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/hastagaras-halu-oas-8b-l-4311-v2_reward/vocab.json
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/hastagaras-halu-oas-8b-l-4311-v2_reward/tokenizer.json
hastagaras-halu-oas-8b-l-4311-v2-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/hastagaras-halu-oas-8b-l-4311-v2_reward/reward.tensors
Job hastagaras-halu-oas-8b-l-4311-v2-mkmlizer completed after 176.74s with status: succeeded
Stopping job with name hastagaras-halu-oas-8b-l-4311-v2-mkmlizer
Pipeline stage MKMLizer completed in 179.92s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service hastagaras-halu-oas-8b-l-4311-v2
Waiting for inference service hastagaras-halu-oas-8b-l-4311-v2 to be ready
Inference service hastagaras-halu-oas-8b-l-4311-v2 ready after 60.28717875480652s
Pipeline stage ISVCDeployer completed in 67.39s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.1612212657928467s
Received healthy response to inference request in 1.3499441146850586s
Received healthy response to inference request in 1.3229057788848877s
Received healthy response to inference request in 1.319610595703125s
Received healthy response to inference request in 1.2395994663238525s
5 requests
0 failed requests
5th percentile: 1.255601692199707
10th percentile: 1.2716039180755616
20th percentile: 1.3036083698272705
30th percentile: 1.3202696323394776
40th percentile: 1.3215877056121825
50th percentile: 1.3229057788848877
60th percentile: 1.333721113204956
70th percentile: 1.3445364475250243
80th percentile: 1.5121995449066163
90th percentile: 1.8367104053497316
95th percentile: 1.998965835571289
99th percentile: 2.128770179748535
mean time: 1.478656244277954
Pipeline stage StressChecker completed in 7.96s
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
hastagaras-halu-oas-8b-l_4311_v2 status is now deployed due to DeploymentManager action
hastagaras-halu-oas-8b-l_4311_v2 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics