submission_id: hastagaras-llama3-rebaha_6335_v1
developer_uid: Hastagaras
status: inactive
model_repo: Hastagaras/Llama3-RebahanSthenoMaidBlackroot-8B
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 0.95, 'top_p': 1.0, 'min_p': 0.1, 'top_k': 60, '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: 2024-06-21T03:11:26+00:00
model_name: m
model_group: Hastagaras/Llama3-Rebaha
num_battles: 16663
num_wins: 8734
celo_rating: 1209.01
propriety_score: 0.7088928299974664
propriety_total_count: 7894.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: m
ineligible_reason: None
language_model: Hastagaras/Llama3-RebahanSthenoMaidBlackroot-8B
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-06-20
win_ratio: 0.5241553141691172
Resubmit model
Running pipeline stage MKMLizer
Starting job with name hastagaras-llama3-rebaha-6335-v1-mkmlizer
Waiting for job on hastagaras-llama3-rebaha-6335-v1-mkmlizer to finish
hastagaras-llama3-rebaha-6335-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
hastagaras-llama3-rebaha-6335-v1-mkmlizer: ║ _____ __ __ ║
hastagaras-llama3-rebaha-6335-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
hastagaras-llama3-rebaha-6335-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
hastagaras-llama3-rebaha-6335-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
hastagaras-llama3-rebaha-6335-v1-mkmlizer: ║ /___/ ║
hastagaras-llama3-rebaha-6335-v1-mkmlizer: ║ ║
hastagaras-llama3-rebaha-6335-v1-mkmlizer: ║ Version: 0.8.14 ║
hastagaras-llama3-rebaha-6335-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
hastagaras-llama3-rebaha-6335-v1-mkmlizer: ║ https://mk1.ai ║
hastagaras-llama3-rebaha-6335-v1-mkmlizer: ║ ║
hastagaras-llama3-rebaha-6335-v1-mkmlizer: ║ The license key for the current software has been verified as ║
hastagaras-llama3-rebaha-6335-v1-mkmlizer: ║ belonging to: ║
hastagaras-llama3-rebaha-6335-v1-mkmlizer: ║ ║
hastagaras-llama3-rebaha-6335-v1-mkmlizer: ║ Chai Research Corp. ║
hastagaras-llama3-rebaha-6335-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
hastagaras-llama3-rebaha-6335-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
hastagaras-llama3-rebaha-6335-v1-mkmlizer: ║ ║
hastagaras-llama3-rebaha-6335-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
hastagaras-llama3-rebaha-6335-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-llama3-rebaha-6335-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
hastagaras-llama3-rebaha-6335-v1-mkmlizer: Downloaded to shared memory in 26.742s
hastagaras-llama3-rebaha-6335-v1-mkmlizer: quantizing model to /dev/shm/model_cache
hastagaras-llama3-rebaha-6335-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
hastagaras-llama3-rebaha-6335-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 1%| | 2/291 [00:04<10:56, 2.27s/it] Loading 0: 5%|▌ | 15/291 [00:04<01:03, 4.38it/s] Loading 0: 11%|█▏ | 33/291 [00:04<00:22, 11.62it/s] Loading 0: 18%|█▊ | 51/291 [00:04<00:11, 20.91it/s] Loading 0: 22%|██▏ | 65/291 [00:05<00:09, 25.05it/s] Loading 0: 27%|██▋ | 78/291 [00:05<00:06, 33.59it/s] Loading 0: 33%|███▎ | 96/291 [00:05<00:04, 48.26it/s] Loading 0: 39%|███▉ | 113/291 [00:05<00:02, 63.82it/s] Loading 0: 45%|████▌ | 131/291 [00:05<00:01, 80.14it/s] Loading 0: 51%|█████ | 149/291 [00:05<00:01, 96.48it/s] Loading 0: 57%|█████▋ | 166/291 [00:06<00:01, 74.23it/s] Loading 0: 63%|██████▎ | 184/291 [00:06<00:01, 89.51it/s] Loading 0: 69%|██████▉ | 202/291 [00:06<00:00, 103.75it/s] Loading 0: 76%|███████▌ | 220/291 [00:06<00:00, 116.93it/s] Loading 0: 82%|████████▏ | 238/291 [00:06<00:00, 128.14it/s] Loading 0: 88%|████████▊ | 256/291 [00:06<00:00, 137.85it/s] Loading 0: 93%|█████████▎| 272/291 [00:07<00:00, 86.99it/s] Loading 0: 98%|█████████▊| 285/291 [00:07<00:00, 94.56it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
hastagaras-llama3-rebaha-6335-v1-mkmlizer: quantized model in 23.025s
hastagaras-llama3-rebaha-6335-v1-mkmlizer: Processed model Hastagaras/Llama3-RebahanSthenoMaidBlackroot-8B in 52.246s
hastagaras-llama3-rebaha-6335-v1-mkmlizer: creating bucket guanaco-mkml-models
hastagaras-llama3-rebaha-6335-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
hastagaras-llama3-rebaha-6335-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/hastagaras-llama3-rebaha-6335-v1
hastagaras-llama3-rebaha-6335-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/hastagaras-llama3-rebaha-6335-v1/config.json
hastagaras-llama3-rebaha-6335-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/hastagaras-llama3-rebaha-6335-v1/special_tokens_map.json
hastagaras-llama3-rebaha-6335-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/hastagaras-llama3-rebaha-6335-v1/tokenizer.json
hastagaras-llama3-rebaha-6335-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/hastagaras-llama3-rebaha-6335-v1/tokenizer_config.json
hastagaras-llama3-rebaha-6335-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/hastagaras-llama3-rebaha-6335-v1/flywheel_model.0.safetensors
hastagaras-llama3-rebaha-6335-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
hastagaras-llama3-rebaha-6335-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-llama3-rebaha-6335-v1-mkmlizer: warnings.warn(
hastagaras-llama3-rebaha-6335-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-llama3-rebaha-6335-v1-mkmlizer: warnings.warn(
hastagaras-llama3-rebaha-6335-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-llama3-rebaha-6335-v1-mkmlizer: warnings.warn(
hastagaras-llama3-rebaha-6335-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-llama3-rebaha-6335-v1-mkmlizer: return self.fget.__get__(instance, owner)()
hastagaras-llama3-rebaha-6335-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
hastagaras-llama3-rebaha-6335-v1-mkmlizer: Saving duration: 0.389s
hastagaras-llama3-rebaha-6335-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 5.406s
hastagaras-llama3-rebaha-6335-v1-mkmlizer: creating bucket guanaco-reward-models
hastagaras-llama3-rebaha-6335-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
hastagaras-llama3-rebaha-6335-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/hastagaras-llama3-rebaha-6335-v1_reward
hastagaras-llama3-rebaha-6335-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/hastagaras-llama3-rebaha-6335-v1_reward/config.json
hastagaras-llama3-rebaha-6335-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/hastagaras-llama3-rebaha-6335-v1_reward/special_tokens_map.json
hastagaras-llama3-rebaha-6335-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/hastagaras-llama3-rebaha-6335-v1_reward/tokenizer_config.json
hastagaras-llama3-rebaha-6335-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/hastagaras-llama3-rebaha-6335-v1_reward/vocab.json
hastagaras-llama3-rebaha-6335-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/hastagaras-llama3-rebaha-6335-v1_reward/tokenizer.json
hastagaras-llama3-rebaha-6335-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/hastagaras-llama3-rebaha-6335-v1_reward/merges.txt
hastagaras-llama3-rebaha-6335-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/hastagaras-llama3-rebaha-6335-v1_reward/reward.tensors
Job hastagaras-llama3-rebaha-6335-v1-mkmlizer completed after 83.79s with status: succeeded
Stopping job with name hastagaras-llama3-rebaha-6335-v1-mkmlizer
Pipeline stage MKMLizer completed in 84.18s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.13s
Running pipeline stage ISVCDeployer
Creating inference service hastagaras-llama3-rebaha-6335-v1
Waiting for inference service hastagaras-llama3-rebaha-6335-v1 to be ready
Inference service hastagaras-llama3-rebaha-6335-v1 ready after 40.21220111846924s
Pipeline stage ISVCDeployer completed in 45.88s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.1639022827148438s
Received healthy response to inference request in 1.3413653373718262s
Received healthy response to inference request in 1.3132927417755127s
Received healthy response to inference request in 1.2873764038085938s
Received healthy response to inference request in 1.4851489067077637s
5 requests
0 failed requests
5th percentile: 1.2925596714019776
10th percentile: 1.2977429389953614
20th percentile: 1.3081094741821289
30th percentile: 1.3189072608947754
40th percentile: 1.3301362991333008
50th percentile: 1.3413653373718262
60th percentile: 1.3988787651062011
70th percentile: 1.456392192840576
80th percentile: 1.6208995819091798
90th percentile: 1.892400932312012
95th percentile: 2.0281516075134274
99th percentile: 2.1367521476745606
mean time: 1.518217134475708
Pipeline stage StressChecker completed in 8.37s
Running pipeline stage DaemonicSafetyScorer
Pipeline stage DaemonicSafetyScorer completed in 0.04s
hastagaras-llama3-rebaha_6335_v1 status is now deployed due to DeploymentManager action
hastagaras-llama3-rebaha_6335_v1 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics