hastagaras-halu-8b-llama_9226

submission_id: hastagaras-halu-8b-llama_9226_v3

developer_uid: Hastagaras

best_of: 16

celo_rating: 1206.92

display_name: experiment-2-flat-test

family_friendly_score: 0.0

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}

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}

is_internal_developer: False

language_model: Hastagaras/HALU-8B-LLAMA3-EXP-2

max_input_tokens: 512

max_output_tokens: 64

model_architecture: LlamaForCausalLM

model_eval_status: success

model_group: Hastagaras/HALU-8B-LLAMA

model_name: experiment-2-flat-test

model_num_parameters: 8030261248.0

model_repo: Hastagaras/HALU-8B-LLAMA3-EXP-2

model_size: 8B

num_battles: 8484

num_wins: 4400

ranking_group: single

reward_formatter: {'bot_template': '{bot_name}: {message}\n', 'memory_template': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'response_template': '{bot_name}:', 'truncate_by_message': False, 'user_template': '{user_name}: {message}\n'}

reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2

status: torndown

submission_type: basic

timestamp: 2024-05-31T12:49:30+00:00

us_pacific_date: 2024-05-31

win_ratio: 0.5186232909005186

Resubmit model

Running pipeline stage MKMLizer
Starting job with name hastagaras-halu-8b-llama-9226-v3-mkmlizer
Waiting for job on hastagaras-halu-8b-llama-9226-v3-mkmlizer to finish
hastagaras-halu-8b-llama-9226-v3-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
hastagaras-halu-8b-llama-9226-v3-mkmlizer: ║     _____            __           __                                ║
hastagaras-halu-8b-llama-9226-v3-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
hastagaras-halu-8b-llama-9226-v3-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
hastagaras-halu-8b-llama-9226-v3-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
hastagaras-halu-8b-llama-9226-v3-mkmlizer: ║       /___/                                                         ║
hastagaras-halu-8b-llama-9226-v3-mkmlizer: ║                                                                     ║
hastagaras-halu-8b-llama-9226-v3-mkmlizer: ║  Version: 0.8.14                                                    ║
hastagaras-halu-8b-llama-9226-v3-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
hastagaras-halu-8b-llama-9226-v3-mkmlizer: ║  https://mk1.ai                                                     ║
hastagaras-halu-8b-llama-9226-v3-mkmlizer: ║                                                                     ║
hastagaras-halu-8b-llama-9226-v3-mkmlizer: ║  The license key for the current software has been verified as      ║
hastagaras-halu-8b-llama-9226-v3-mkmlizer: ║  belonging to:                                                      ║
hastagaras-halu-8b-llama-9226-v3-mkmlizer: ║                                                                     ║
hastagaras-halu-8b-llama-9226-v3-mkmlizer: ║  Chai Research Corp.                                                ║
hastagaras-halu-8b-llama-9226-v3-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
hastagaras-halu-8b-llama-9226-v3-mkmlizer: ║  Expiration: 2024-07-15 23:59:59                                    ║
hastagaras-halu-8b-llama-9226-v3-mkmlizer: ║                                                                     ║
hastagaras-halu-8b-llama-9226-v3-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
hastagaras-halu-8b-llama-9226-v3-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-8b-llama-9226-v3-mkmlizer:   warnings.warn(warning_message, FutureWarning)
hastagaras-halu-8b-llama-9226-v3-mkmlizer: Downloaded to shared memory in 17.415s
hastagaras-halu-8b-llama-9226-v3-mkmlizer: quantizing model to /dev/shm/model_cache
hastagaras-halu-8b-llama-9226-v3-mkmlizer: Saving flywheel model at /dev/shm/model_cache
hastagaras-halu-8b-llama-9226-v3-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   1%|          | 2/291 [00:04<10:42,  2.22s/it]
Loading 0:   6%|▌         | 18/291 [00:04<00:50,  5.42it/s]
Loading 0:  12%|█▏        | 36/291 [00:04<00:19, 12.83it/s]
Loading 0:  18%|█▊        | 51/291 [00:04<00:11, 20.68it/s]
Loading 0:  22%|██▏       | 65/291 [00:05<00:08, 25.51it/s]
Loading 0:  28%|██▊       | 82/291 [00:05<00:05, 37.69it/s]
Loading 0:  34%|███▍      | 99/291 [00:05<00:03, 51.76it/s]
Loading 0:  39%|███▉      | 114/291 [00:05<00:02, 64.44it/s]
Loading 0:  45%|████▌     | 132/291 [00:05<00:01, 81.75it/s]
Loading 0:  51%|█████     | 148/291 [00:05<00:01, 94.60it/s]
Loading 0:  56%|█████▌    | 163/291 [00:05<00:01, 103.87it/s]
Loading 0:  61%|██████    | 178/291 [00:06<00:01, 75.13it/s] 
Loading 0:  67%|██████▋   | 195/291 [00:06<00:01, 91.05it/s]
Loading 0:  73%|███████▎  | 213/291 [00:06<00:00, 106.03it/s]
Loading 0:  79%|███████▊  | 229/291 [00:06<00:00, 116.81it/s]
Loading 0:  85%|████████▍ | 247/291 [00:06<00:00, 128.57it/s]
Loading 0:  91%|█████████ | 265/291 [00:06<00:00, 138.48it/s]
Loading 0:  97%|█████████▋| 281/291 [00:06<00:00, 89.73it/s] 
                                                            
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
hastagaras-halu-8b-llama-9226-v3-mkmlizer: quantized model in 22.737s
hastagaras-halu-8b-llama-9226-v3-mkmlizer: Processed model Hastagaras/HALU-8B-LLAMA3-EXP-2 in 42.626s
hastagaras-halu-8b-llama-9226-v3-mkmlizer: creating bucket guanaco-mkml-models
hastagaras-halu-8b-llama-9226-v3-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
hastagaras-halu-8b-llama-9226-v3-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/hastagaras-halu-8b-llama-9226-v3
hastagaras-halu-8b-llama-9226-v3-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/hastagaras-halu-8b-llama-9226-v3/special_tokens_map.json
hastagaras-halu-8b-llama-9226-v3-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/hastagaras-halu-8b-llama-9226-v3/config.json
hastagaras-halu-8b-llama-9226-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/hastagaras-halu-8b-llama-9226-v3/tokenizer_config.json
hastagaras-halu-8b-llama-9226-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/hastagaras-halu-8b-llama-9226-v3/tokenizer.json
hastagaras-halu-8b-llama-9226-v3-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/hastagaras-halu-8b-llama-9226-v3/flywheel_model.0.safetensors
hastagaras-halu-8b-llama-9226-v3-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-8b-llama-9226-v3-mkmlizer:   warnings.warn(
hastagaras-halu-8b-llama-9226-v3-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
hastagaras-halu-8b-llama-9226-v3-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-8b-llama-9226-v3-mkmlizer:   warnings.warn(
hastagaras-halu-8b-llama-9226-v3-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-8b-llama-9226-v3-mkmlizer:   warnings.warn(
hastagaras-halu-8b-llama-9226-v3-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-8b-llama-9226-v3-mkmlizer:   return self.fget.__get__(instance, owner)()
hastagaras-halu-8b-llama-9226-v3-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
hastagaras-halu-8b-llama-9226-v3-mkmlizer: Saving duration: 0.395s
hastagaras-halu-8b-llama-9226-v3-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 4.717s
hastagaras-halu-8b-llama-9226-v3-mkmlizer: creating bucket guanaco-reward-models
hastagaras-halu-8b-llama-9226-v3-mkmlizer: Bucket 's3://guanaco-reward-models/' created
hastagaras-halu-8b-llama-9226-v3-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/hastagaras-halu-8b-llama-9226-v3_reward
hastagaras-halu-8b-llama-9226-v3-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/hastagaras-halu-8b-llama-9226-v3_reward/tokenizer_config.json
hastagaras-halu-8b-llama-9226-v3-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/hastagaras-halu-8b-llama-9226-v3_reward/special_tokens_map.json
hastagaras-halu-8b-llama-9226-v3-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/hastagaras-halu-8b-llama-9226-v3_reward/config.json
hastagaras-halu-8b-llama-9226-v3-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/hastagaras-halu-8b-llama-9226-v3_reward/vocab.json
hastagaras-halu-8b-llama-9226-v3-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/hastagaras-halu-8b-llama-9226-v3_reward/merges.txt
hastagaras-halu-8b-llama-9226-v3-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/hastagaras-halu-8b-llama-9226-v3_reward/tokenizer.json
hastagaras-halu-8b-llama-9226-v3-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/hastagaras-halu-8b-llama-9226-v3_reward/reward.tensors
Job hastagaras-halu-8b-llama-9226-v3-mkmlizer completed after 64.22s with status: succeeded
Stopping job with name hastagaras-halu-8b-llama-9226-v3-mkmlizer
Pipeline stage MKMLizer completed in 65.04s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service hastagaras-halu-8b-llama-9226-v3
Waiting for inference service hastagaras-halu-8b-llama-9226-v3 to be ready
Inference service hastagaras-halu-8b-llama-9226-v3 ready after 100.72380089759827s
Pipeline stage ISVCDeployer completed in 107.15s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.174808979034424s
Received healthy response to inference request in 1.3321619033813477s
Received healthy response to inference request in 1.2962300777435303s
Received healthy response to inference request in 1.285282850265503s
Received healthy response to inference request in 1.2350547313690186s
5 requests
0 failed requests
5th percentile: 1.2451003551483155
10th percentile: 1.2551459789276123
20th percentile: 1.275237226486206
30th percentile: 1.2874722957611084
40th percentile: 1.2918511867523192
50th percentile: 1.2962300777435303
60th percentile: 1.3106028079986571
70th percentile: 1.3249755382537842
80th percentile: 1.500691318511963
90th percentile: 1.8377501487731935
95th percentile: 2.0062795639038082
99th percentile: 2.1411030960083006
mean time: 1.4647077083587647
Pipeline stage StressChecker completed in 8.06s
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-8b-llama_9226_v3 status is now deployed due to DeploymentManager action
hastagaras-halu-8b-llama_9226_v3 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of hastagaras-halu-8b-llama_9226_v3
Running pipeline stage ISVCDeleter
Checking if service hastagaras-halu-8b-llama-9226-v3 is running
Skipping teardown as no inference service was found
Pipeline stage ISVCDeleter completed in 2.30s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key hastagaras-halu-8b-llama-9226-v3/config.json from bucket guanaco-mkml-models
Deleting key hastagaras-halu-8b-llama-9226-v3/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key hastagaras-halu-8b-llama-9226-v3/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key hastagaras-halu-8b-llama-9226-v3/tokenizer.json from bucket guanaco-mkml-models
Deleting key hastagaras-halu-8b-llama-9226-v3/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key hastagaras-halu-8b-llama-9226-v3_reward/config.json from bucket guanaco-reward-models
Deleting key hastagaras-halu-8b-llama-9226-v3_reward/merges.txt from bucket guanaco-reward-models
Deleting key hastagaras-halu-8b-llama-9226-v3_reward/reward.tensors from bucket guanaco-reward-models
Deleting key hastagaras-halu-8b-llama-9226-v3_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key hastagaras-halu-8b-llama-9226-v3_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key hastagaras-halu-8b-llama-9226-v3_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key hastagaras-halu-8b-llama-9226-v3_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 3.69s
hastagaras-halu-8b-llama_9226_v3 status is now torndown due to DeploymentManager action