trace2333-pippa-llama2-v0

developer_uid: Trace2333

submission_id: trace2333-pippa-llama2-v0_v1

model_name: trace2333-pippa-llama2-v0_v1

model_group: Trace2333/pippa_llama2_v

status: torndown

timestamp: 2024-06-12T04:10:56+00:00

num_battles: 14320

num_wins: 5198

celo_rating: 1081.97

family_friendly_score: 0.0

submission_type: basic

model_repo: Trace2333/pippa_llama2_v0

model_architecture: LlamaForCausalLM

reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2

model_num_parameters: 6738415616.0

best_of: 4

max_input_tokens: 512

max_output_tokens: 64

display_name: trace2333-pippa-llama2-v0_v1

is_internal_developer: False

language_model: Trace2333/pippa_llama2_v0

model_size: 7B

ranking_group: single

us_pacific_date: 2024-06-11

win_ratio: 0.3629888268156425

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': 4, '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}

model_eval_status: success

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'}

Resubmit model

Running pipeline stage MKMLizer
Starting job with name trace2333-pippa-llama2-v0-v1-mkmlizer
Waiting for job on trace2333-pippa-llama2-v0-v1-mkmlizer to finish
trace2333-pippa-llama2-v0-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
trace2333-pippa-llama2-v0-v1-mkmlizer: ║     _____            __           __                                ║
trace2333-pippa-llama2-v0-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
trace2333-pippa-llama2-v0-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
trace2333-pippa-llama2-v0-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
trace2333-pippa-llama2-v0-v1-mkmlizer: ║       /___/                                                         ║
trace2333-pippa-llama2-v0-v1-mkmlizer: ║                                                                     ║
trace2333-pippa-llama2-v0-v1-mkmlizer: ║  Version: 0.8.14                                                    ║
trace2333-pippa-llama2-v0-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
trace2333-pippa-llama2-v0-v1-mkmlizer: ║  https://mk1.ai                                                     ║
trace2333-pippa-llama2-v0-v1-mkmlizer: ║                                                                     ║
trace2333-pippa-llama2-v0-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
trace2333-pippa-llama2-v0-v1-mkmlizer: ║  belonging to:                                                      ║
trace2333-pippa-llama2-v0-v1-mkmlizer: ║                                                                     ║
trace2333-pippa-llama2-v0-v1-mkmlizer: ║  Chai Research Corp.                                                ║
trace2333-pippa-llama2-v0-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
trace2333-pippa-llama2-v0-v1-mkmlizer: ║  Expiration: 2024-07-15 23:59:59                                    ║
trace2333-pippa-llama2-v0-v1-mkmlizer: ║                                                                     ║
trace2333-pippa-llama2-v0-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
trace2333-pippa-llama2-v0-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.
trace2333-pippa-llama2-v0-v1-mkmlizer:   warnings.warn(warning_message, FutureWarning)
trace2333-pippa-llama2-v0-v1-mkmlizer: Downloaded to shared memory in 55.091s
trace2333-pippa-llama2-v0-v1-mkmlizer: quantizing model to /dev/shm/model_cache
trace2333-pippa-llama2-v0-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
trace2333-pippa-llama2-v0-v1-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   3%|▎         | 10/291 [00:00<00:02, 93.92it/s]
Loading 0:   7%|▋         | 20/291 [00:00<00:02, 96.17it/s]
Loading 0:  10%|█         | 30/291 [00:00<00:02, 89.32it/s]
Loading 0:  14%|█▎        | 40/291 [00:00<00:02, 86.66it/s]
Loading 0:  18%|█▊        | 51/291 [00:00<00:04, 48.94it/s]
Loading 0:  20%|██        | 59/291 [00:00<00:04, 54.62it/s]
Loading 0:  23%|██▎       | 67/291 [00:01<00:03, 60.12it/s]
Loading 0:  26%|██▋       | 77/291 [00:01<00:03, 69.53it/s]
Loading 0:  30%|███       | 88/291 [00:01<00:02, 75.80it/s]
Loading 0:  34%|███▎      | 98/291 [00:01<00:02, 78.25it/s]
Loading 0:  37%|███▋      | 107/291 [00:01<00:03, 49.07it/s]
Loading 0:  40%|███▉      | 116/291 [00:01<00:03, 56.10it/s]
Loading 0:  43%|████▎     | 126/291 [00:01<00:02, 64.43it/s]
Loading 0:  47%|████▋     | 136/291 [00:02<00:02, 71.58it/s]
Loading 0:  51%|█████     | 147/291 [00:02<00:01, 77.59it/s]
Loading 0:  54%|█████▍    | 158/291 [00:02<00:01, 84.35it/s]
Loading 0:  58%|█████▊    | 168/291 [00:02<00:02, 51.11it/s]
Loading 0:  61%|██████    | 178/291 [00:02<00:01, 59.25it/s]
Loading 0:  65%|██████▍   | 188/291 [00:02<00:01, 67.26it/s]
Loading 0:  68%|██████▊   | 198/291 [00:02<00:01, 74.37it/s]
Loading 0:  71%|███████▏  | 208/291 [00:03<00:01, 80.01it/s]
Loading 0:  75%|███████▍  | 218/291 [00:03<00:01, 53.18it/s]
Loading 0:  78%|███████▊  | 228/291 [00:03<00:01, 60.21it/s]
Loading 0:  82%|████████▏ | 238/291 [00:03<00:00, 66.48it/s]
Loading 0:  85%|████████▌ | 248/291 [00:03<00:00, 71.74it/s]
Loading 0:  89%|████████▉ | 259/291 [00:03<00:00, 80.45it/s]
Loading 0:  92%|█████████▏| 269/291 [00:05<00:01, 17.39it/s]
Loading 0:  95%|█████████▍| 276/291 [00:05<00:00, 20.87it/s]
Loading 0:  99%|█████████▊| 287/291 [00:05<00:00, 28.57it/s]
                                                            
You set `add_prefix_space`. The tokenizer needs to be converted from the slow tokenizers
trace2333-pippa-llama2-v0-v1-mkmlizer: quantized model in 15.316s
trace2333-pippa-llama2-v0-v1-mkmlizer: Processed model Trace2333/pippa_llama2_v0 in 74.752s
trace2333-pippa-llama2-v0-v1-mkmlizer: creating bucket guanaco-mkml-models
trace2333-pippa-llama2-v0-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
trace2333-pippa-llama2-v0-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/trace2333-pippa-llama2-v0-v1
trace2333-pippa-llama2-v0-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/trace2333-pippa-llama2-v0-v1/tokenizer_config.json
trace2333-pippa-llama2-v0-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/trace2333-pippa-llama2-v0-v1/special_tokens_map.json
trace2333-pippa-llama2-v0-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.model s3://guanaco-mkml-models/trace2333-pippa-llama2-v0-v1/tokenizer.model
trace2333-pippa-llama2-v0-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/trace2333-pippa-llama2-v0-v1/config.json
trace2333-pippa-llama2-v0-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/trace2333-pippa-llama2-v0-v1/tokenizer.json
trace2333-pippa-llama2-v0-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/trace2333-pippa-llama2-v0-v1/flywheel_model.0.safetensors
trace2333-pippa-llama2-v0-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
trace2333-pippa-llama2-v0-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.
trace2333-pippa-llama2-v0-v1-mkmlizer:   warnings.warn(
trace2333-pippa-llama2-v0-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.
trace2333-pippa-llama2-v0-v1-mkmlizer:   warnings.warn(
trace2333-pippa-llama2-v0-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.
trace2333-pippa-llama2-v0-v1-mkmlizer:   warnings.warn(
trace2333-pippa-llama2-v0-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()
trace2333-pippa-llama2-v0-v1-mkmlizer:   return self.fget.__get__(instance, owner)()
trace2333-pippa-llama2-v0-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
trace2333-pippa-llama2-v0-v1-mkmlizer: Saving duration: 0.445s
trace2333-pippa-llama2-v0-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 4.139s
trace2333-pippa-llama2-v0-v1-mkmlizer: creating bucket guanaco-reward-models
trace2333-pippa-llama2-v0-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
trace2333-pippa-llama2-v0-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/trace2333-pippa-llama2-v0-v1_reward
trace2333-pippa-llama2-v0-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/trace2333-pippa-llama2-v0-v1_reward/config.json
trace2333-pippa-llama2-v0-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/trace2333-pippa-llama2-v0-v1_reward/tokenizer_config.json
trace2333-pippa-llama2-v0-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/trace2333-pippa-llama2-v0-v1_reward/vocab.json
trace2333-pippa-llama2-v0-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/trace2333-pippa-llama2-v0-v1_reward/special_tokens_map.json
trace2333-pippa-llama2-v0-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/trace2333-pippa-llama2-v0-v1_reward/merges.txt
trace2333-pippa-llama2-v0-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/trace2333-pippa-llama2-v0-v1_reward/tokenizer.json
trace2333-pippa-llama2-v0-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/trace2333-pippa-llama2-v0-v1_reward/reward.tensors
Job trace2333-pippa-llama2-v0-v1-mkmlizer completed after 96.36s with status: succeeded
Stopping job with name trace2333-pippa-llama2-v0-v1-mkmlizer
Pipeline stage MKMLizer completed in 101.08s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service trace2333-pippa-llama2-v0-v1
Waiting for inference service trace2333-pippa-llama2-v0-v1 to be ready
Inference service trace2333-pippa-llama2-v0-v1 ready after 30.199641942977905s
Pipeline stage ISVCDeployer completed in 37.55s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.894315481185913s
Received healthy response to inference request in 0.7732219696044922s
Received healthy response to inference request in 0.8305540084838867s
Received healthy response to inference request in 0.8264427185058594s
Received healthy response to inference request in 0.9021553993225098s
5 requests
0 failed requests
5th percentile: 0.7838661193847656
10th percentile: 0.7945102691650391
20th percentile: 0.815798568725586
30th percentile: 0.8272649765014648
40th percentile: 0.8289094924926758
50th percentile: 0.8305540084838867
60th percentile: 0.859194564819336
70th percentile: 0.8878351211547851
80th percentile: 1.1005874156951907
90th percentile: 1.497451448440552
95th percentile: 1.6958834648132322
99th percentile: 1.854629077911377
mean time: 1.0453379154205322
Pipeline stage StressChecker completed in 5.89s
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
trace2333-pippa-llama2-v0_v1 status is now deployed due to DeploymentManager action
trace2333-pippa-llama2-v0_v1 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of trace2333-pippa-llama2-v0_v1
Running pipeline stage ISVCDeleter
Checking if service trace2333-pippa-llama2-v0-v1 is running
Skipping teardown as no inference service was found
Pipeline stage ISVCDeleter completed in 4.18s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key trace2333-pippa-llama2-v0-v1/config.json from bucket guanaco-mkml-models
Deleting key trace2333-pippa-llama2-v0-v1/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key trace2333-pippa-llama2-v0-v1/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key trace2333-pippa-llama2-v0-v1/tokenizer.json from bucket guanaco-mkml-models
Deleting key trace2333-pippa-llama2-v0-v1/tokenizer.model from bucket guanaco-mkml-models
Deleting key trace2333-pippa-llama2-v0-v1/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key trace2333-pippa-llama2-v0-v1_reward/config.json from bucket guanaco-reward-models
Deleting key trace2333-pippa-llama2-v0-v1_reward/merges.txt from bucket guanaco-reward-models
Deleting key trace2333-pippa-llama2-v0-v1_reward/reward.tensors from bucket guanaco-reward-models
Deleting key trace2333-pippa-llama2-v0-v1_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key trace2333-pippa-llama2-v0-v1_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key trace2333-pippa-llama2-v0-v1_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key trace2333-pippa-llama2-v0-v1_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 6.20s
trace2333-pippa-llama2-v0_v1 status is now torndown due to DeploymentManager action