submission_id: hastagaras-rebahan-8b-l3_6682_v1
developer_uid: Hastagaras
status: inactive
model_repo: Hastagaras/Rebahan-8B-L3-MK.II-ALT
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 1.05, 'top_p': 1.0, 'min_p': 0.1, 'top_k': 45, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '<|eot_id|>'], 'max_input_tokens': 512, 'best_of': 16, '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: 2024-06-21T10:14:02+00:00
model_name: alt1
model_group: Hastagaras/Rebahan-8B-L3
num_battles: 16998
num_wins: 9367
celo_rating: 1222.43
propriety_score: 0.7090886820186183
propriety_total_count: 8164.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: alt1
ineligible_reason: None
language_model: Hastagaras/Rebahan-8B-L3-MK.II-ALT
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-06-21
win_ratio: 0.5510648311566066
Resubmit model
Running pipeline stage MKMLizer
Starting job with name hastagaras-rebahan-8b-l3-6682-v1-mkmlizer
Waiting for job on hastagaras-rebahan-8b-l3-6682-v1-mkmlizer to finish
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: ║ _____ __ __ ║
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: ║ /___/ ║
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: ║ ║
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: ║ Version: 0.8.14 ║
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: ║ https://mk1.ai ║
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: ║ ║
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: ║ The license key for the current software has been verified as ║
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: ║ belonging to: ║
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: ║ ║
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: ║ Chai Research Corp. ║
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: ║ ║
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
hastagaras-rebahan-8b-l3-6682-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-rebahan-8b-l3-6682-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: Downloaded to shared memory in 28.710s
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: quantizing model to /dev/shm/model_cache
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 4%|▍ | 13/291 [00:00<00:02, 128.31it/s] Loading 0: 11%|█ | 31/291 [00:00<00:01, 151.75it/s] Loading 0: 17%|█▋ | 49/291 [00:00<00:01, 160.18it/s] Loading 0: 23%|██▎ | 67/291 [00:00<00:01, 163.58it/s] Loading 0: 29%|██▉ | 84/291 [00:00<00:02, 85.16it/s] Loading 0: 34%|███▍ | 100/291 [00:00<00:01, 100.32it/s] Loading 0: 39%|███▉ | 114/291 [00:01<00:01, 107.82it/s] Loading 0: 45%|████▌ | 131/291 [00:01<00:01, 120.30it/s] Loading 0: 51%|█████ | 149/291 [00:01<00:01, 133.60it/s] Loading 0: 57%|█████▋ | 167/291 [00:01<00:00, 144.01it/s] Loading 0: 64%|██████▍ | 187/291 [00:01<00:01, 97.93it/s] Loading 0: 69%|██████▉ | 202/291 [00:01<00:00, 106.14it/s] Loading 0: 76%|███████▌ | 220/291 [00:01<00:00, 120.14it/s] Loading 0: 82%|████████▏ | 238/291 [00:01<00:00, 132.80it/s] Loading 0: 88%|████████▊ | 256/291 [00:02<00:00, 143.60it/s] Loading 0: 94%|█████████▍| 274/291 [00:02<00:00, 152.18it/s] Loading 0: 100%|██████████| 291/291 [00:07<00:00, 10.34it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: quantized model in 22.494s
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: Processed model Hastagaras/Rebahan-8B-L3-MK.II-ALT in 53.737s
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: creating bucket guanaco-mkml-models
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/hastagaras-rebahan-8b-l3-6682-v1
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/hastagaras-rebahan-8b-l3-6682-v1/config.json
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/hastagaras-rebahan-8b-l3-6682-v1/tokenizer_config.json
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/hastagaras-rebahan-8b-l3-6682-v1/special_tokens_map.json
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/hastagaras-rebahan-8b-l3-6682-v1/tokenizer.json
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/hastagaras-rebahan-8b-l3-6682-v1/flywheel_model.0.safetensors
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
hastagaras-rebahan-8b-l3-6682-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-rebahan-8b-l3-6682-v1-mkmlizer: warnings.warn(
hastagaras-rebahan-8b-l3-6682-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-rebahan-8b-l3-6682-v1-mkmlizer: warnings.warn(
hastagaras-rebahan-8b-l3-6682-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-rebahan-8b-l3-6682-v1-mkmlizer: warnings.warn(
hastagaras-rebahan-8b-l3-6682-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-rebahan-8b-l3-6682-v1-mkmlizer: return self.fget.__get__(instance, owner)()
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: Saving duration: 0.415s
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 3.783s
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: creating bucket guanaco-reward-models
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/hastagaras-rebahan-8b-l3-6682-v1_reward
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/hastagaras-rebahan-8b-l3-6682-v1_reward/tokenizer_config.json
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/hastagaras-rebahan-8b-l3-6682-v1_reward/vocab.json
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/hastagaras-rebahan-8b-l3-6682-v1_reward/config.json
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/hastagaras-rebahan-8b-l3-6682-v1_reward/special_tokens_map.json
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/hastagaras-rebahan-8b-l3-6682-v1_reward/merges.txt
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/hastagaras-rebahan-8b-l3-6682-v1_reward/tokenizer.json
hastagaras-rebahan-8b-l3-6682-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/hastagaras-rebahan-8b-l3-6682-v1_reward/reward.tensors
Job hastagaras-rebahan-8b-l3-6682-v1-mkmlizer completed after 82.99s with status: succeeded
Stopping job with name hastagaras-rebahan-8b-l3-6682-v1-mkmlizer
Pipeline stage MKMLizer completed in 83.36s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.44s
Running pipeline stage ISVCDeployer
Creating inference service hastagaras-rebahan-8b-l3-6682-v1
Waiting for inference service hastagaras-rebahan-8b-l3-6682-v1 to be ready
Inference service hastagaras-rebahan-8b-l3-6682-v1 ready after 40.19825029373169s
Pipeline stage ISVCDeployer completed in 45.78s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.1772336959838867s
Received healthy response to inference request in 1.2233202457427979s
Received healthy response to inference request in 1.3266892433166504s
Received healthy response to inference request in 1.2978332042694092s
Received healthy response to inference request in 1.366081714630127s
5 requests
0 failed requests
5th percentile: 1.2382228374481201
10th percentile: 1.2531254291534424
20th percentile: 1.282930612564087
30th percentile: 1.3036044120788575
40th percentile: 1.3151468276977538
50th percentile: 1.3266892433166504
60th percentile: 1.342446231842041
70th percentile: 1.3582032203674317
80th percentile: 1.528312110900879
90th percentile: 1.852772903442383
95th percentile: 2.0150032997131344
99th percentile: 2.144787616729736
mean time: 1.4782316207885742
Pipeline stage StressChecker completed in 8.10s
Running pipeline stage DaemonicSafetyScorer
Pipeline stage DaemonicSafetyScorer completed in 0.03s
hastagaras-rebahan-8b-l3_6682_v1 status is now deployed due to DeploymentManager action
hastagaras-rebahan-8b-l3_6682_v1 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics