submission_id: hastagaras-halu-8b-llam_2564_v10
developer_uid: Hastagaras
status: inactive
model_repo: Hastagaras/Halu-8B-Llama3-v0.3
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 0.95, 'top_p': 1.0, 'min_p': 0.1, '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: 2024-05-28T01:42:42+00:00
model_name: debug-for-dpo
model_eval_status: success
model_group: Hastagaras/Halu-8B-Llama
num_battles: 8538
num_wins: 4650
celo_rating: 1207.85
safety_score: 0.96
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: debug-for-dpo
ineligible_reason: propriety_total_count < 5000
language_model: Hastagaras/Halu-8B-Llama3-v0.3
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-05-27
win_ratio: 0.5446240337315531
Resubmit model
Running pipeline stage MKMLizer
Starting job with name hastagaras-halu-8b-llam-2564-v10-mkmlizer
Waiting for job on hastagaras-halu-8b-llam-2564-v10-mkmlizer to finish
hastagaras-halu-8b-llam-2564-v10-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
hastagaras-halu-8b-llam-2564-v10-mkmlizer: ║ _____ __ __ ║
hastagaras-halu-8b-llam-2564-v10-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
hastagaras-halu-8b-llam-2564-v10-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
hastagaras-halu-8b-llam-2564-v10-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
hastagaras-halu-8b-llam-2564-v10-mkmlizer: ║ /___/ ║
hastagaras-halu-8b-llam-2564-v10-mkmlizer: ║ ║
hastagaras-halu-8b-llam-2564-v10-mkmlizer: ║ Version: 0.8.14 ║
hastagaras-halu-8b-llam-2564-v10-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
hastagaras-halu-8b-llam-2564-v10-mkmlizer: ║ https://mk1.ai ║
hastagaras-halu-8b-llam-2564-v10-mkmlizer: ║ ║
hastagaras-halu-8b-llam-2564-v10-mkmlizer: ║ The license key for the current software has been verified as ║
hastagaras-halu-8b-llam-2564-v10-mkmlizer: ║ belonging to: ║
hastagaras-halu-8b-llam-2564-v10-mkmlizer: ║ ║
hastagaras-halu-8b-llam-2564-v10-mkmlizer: ║ Chai Research Corp. ║
hastagaras-halu-8b-llam-2564-v10-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
hastagaras-halu-8b-llam-2564-v10-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
hastagaras-halu-8b-llam-2564-v10-mkmlizer: ║ ║
hastagaras-halu-8b-llam-2564-v10-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
hastagaras-halu-8b-llam-2564-v10-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-llam-2564-v10-mkmlizer: warnings.warn(warning_message, FutureWarning)
hastagaras-halu-8b-llam-2564-v10-mkmlizer: Downloaded to shared memory in 29.104s
hastagaras-halu-8b-llam-2564-v10-mkmlizer: quantizing model to /dev/shm/model_cache
hastagaras-halu-8b-llam-2564-v10-mkmlizer: Saving flywheel model at /dev/shm/model_cache
hastagaras-halu-8b-llam-2564-v10-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 6%|▌ | 17/291 [00:00<00:01, 165.10it/s] Loading 0: 12%|█▏ | 34/291 [00:00<00:01, 166.96it/s] Loading 0: 19%|█▊ | 54/291 [00:00<00:01, 181.45it/s] Loading 0: 26%|██▌ | 75/291 [00:00<00:01, 187.36it/s] Loading 0: 32%|███▏ | 94/291 [00:00<00:02, 77.22it/s] Loading 0: 39%|███▉ | 113/291 [00:01<00:01, 95.91it/s] Loading 0: 47%|████▋ | 136/291 [00:01<00:01, 122.05it/s] Loading 0: 54%|█████▎ | 156/291 [00:01<00:00, 138.24it/s] Loading 0: 60%|██████ | 176/291 [00:01<00:00, 151.96it/s] Loading 0: 67%|██████▋ | 195/291 [00:01<00:01, 82.26it/s] Loading 0: 74%|███████▍ | 216/291 [00:01<00:00, 101.87it/s] Loading 0: 81%|████████▏ | 237/291 [00:02<00:00, 120.51it/s] Loading 0: 88%|████████▊ | 256/291 [00:02<00:00, 132.29it/s] Loading 0: 94%|█████████▍| 274/291 [00:02<00:00, 141.95it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
hastagaras-halu-8b-llam-2564-v10-mkmlizer: quantized model in 22.344s
hastagaras-halu-8b-llam-2564-v10-mkmlizer: Processed model Hastagaras/Halu-8B-Llama3-v0.3 in 52.761s
hastagaras-halu-8b-llam-2564-v10-mkmlizer: creating bucket guanaco-mkml-models
hastagaras-halu-8b-llam-2564-v10-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
hastagaras-halu-8b-llam-2564-v10-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/hastagaras-halu-8b-llam-2564-v10
hastagaras-halu-8b-llam-2564-v10-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/hastagaras-halu-8b-llam-2564-v10/config.json
hastagaras-halu-8b-llam-2564-v10-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/hastagaras-halu-8b-llam-2564-v10/special_tokens_map.json
hastagaras-halu-8b-llam-2564-v10-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/hastagaras-halu-8b-llam-2564-v10/tokenizer_config.json
hastagaras-halu-8b-llam-2564-v10-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/hastagaras-halu-8b-llam-2564-v10/tokenizer.json
hastagaras-halu-8b-llam-2564-v10-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/hastagaras-halu-8b-llam-2564-v10/flywheel_model.0.safetensors
hastagaras-halu-8b-llam-2564-v10-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
hastagaras-halu-8b-llam-2564-v10-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-llam-2564-v10-mkmlizer: warnings.warn(
hastagaras-halu-8b-llam-2564-v10-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-llam-2564-v10-mkmlizer: warnings.warn(
hastagaras-halu-8b-llam-2564-v10-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-llam-2564-v10-mkmlizer: warnings.warn(
hastagaras-halu-8b-llam-2564-v10-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-llam-2564-v10-mkmlizer: return self.fget.__get__(instance, owner)()
hastagaras-halu-8b-llam-2564-v10-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
hastagaras-halu-8b-llam-2564-v10-mkmlizer: Saving duration: 0.285s
hastagaras-halu-8b-llam-2564-v10-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 6.365s
hastagaras-halu-8b-llam-2564-v10-mkmlizer: creating bucket guanaco-reward-models
hastagaras-halu-8b-llam-2564-v10-mkmlizer: Bucket 's3://guanaco-reward-models/' created
hastagaras-halu-8b-llam-2564-v10-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/hastagaras-halu-8b-llam-2564-v10_reward
hastagaras-halu-8b-llam-2564-v10-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/hastagaras-halu-8b-llam-2564-v10_reward/config.json
hastagaras-halu-8b-llam-2564-v10-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/hastagaras-halu-8b-llam-2564-v10_reward/tokenizer_config.json
hastagaras-halu-8b-llam-2564-v10-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/hastagaras-halu-8b-llam-2564-v10_reward/special_tokens_map.json
hastagaras-halu-8b-llam-2564-v10-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/hastagaras-halu-8b-llam-2564-v10_reward/merges.txt
hastagaras-halu-8b-llam-2564-v10-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/hastagaras-halu-8b-llam-2564-v10_reward/vocab.json
hastagaras-halu-8b-llam-2564-v10-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/hastagaras-halu-8b-llam-2564-v10_reward/tokenizer.json
hastagaras-halu-8b-llam-2564-v10-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/hastagaras-halu-8b-llam-2564-v10_reward/reward.tensors
Job hastagaras-halu-8b-llam-2564-v10-mkmlizer completed after 83.89s with status: succeeded
Stopping job with name hastagaras-halu-8b-llam-2564-v10-mkmlizer
Pipeline stage MKMLizer completed in 86.65s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service hastagaras-halu-8b-llam-2564-v10
Waiting for inference service hastagaras-halu-8b-llam-2564-v10 to be ready
Inference service hastagaras-halu-8b-llam-2564-v10 ready after 30.170738458633423s
Pipeline stage ISVCDeployer completed in 37.60s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.2345542907714844s
Received healthy response to inference request in 1.3561556339263916s
Received healthy response to inference request in 1.3251762390136719s
Received healthy response to inference request in 1.3285608291625977s
Received healthy response to inference request in 1.2607967853546143s
5 requests
0 failed requests
5th percentile: 1.2736726760864259
10th percentile: 1.2865485668182373
20th percentile: 1.3123003482818603
30th percentile: 1.325853157043457
40th percentile: 1.3272069931030273
50th percentile: 1.3285608291625977
60th percentile: 1.3395987510681153
70th percentile: 1.3506366729736328
80th percentile: 1.5318353652954104
90th percentile: 1.8831948280334474
95th percentile: 2.0588745594024656
99th percentile: 2.1994183444976807
mean time: 1.501048755645752
Pipeline stage StressChecker completed in 8.16s
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-llam_2564_v10 status is now deployed due to DeploymentManager action
hastagaras-halu-8b-llam_2564_v10 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics