submission_id: nitral-ai-hathor-l3-8b-v-01_v6
developer_uid: Nitral-AI
status: inactive
model_repo: Nitral-AI/Hathor-L3-8B-v.01
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 0.95, 'top_p': 1.0, 'min_p': 0.08, 'top_k': 40, '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: 2024-06-11T21:41:49+00:00
model_name: nitral-ai-hathor-l3-8b-v-01_v1
model_eval_status: success
model_group: Nitral-AI/Hathor-L3-8B-v
num_battles: 15595
num_wins: 8545
celo_rating: 1224.56
propriety_score: 0.6697204233138525
propriety_total_count: 6331.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: nitral-ai-hathor-l3-8b-v-01_v1
ineligible_reason: None
language_model: Nitral-AI/Hathor-L3-8B-v.01
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-06-11
win_ratio: 0.5479320294966336
Resubmit model
Running pipeline stage MKMLizer
Starting job with name nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer
Waiting for job on nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer to finish
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: ║ _____ __ __ ║
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: ║ /___/ ║
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: ║ ║
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: ║ Version: 0.8.14 ║
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: ║ https://mk1.ai ║
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: ║ ║
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: ║ The license key for the current software has been verified as ║
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: ║ belonging to: ║
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: ║ ║
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: ║ Chai Research Corp. ║
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: ║ ║
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
nitral-ai-hathor-l3-8b-v-01-v6-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.
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: warnings.warn(warning_message, FutureWarning)
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: Downloaded to shared memory in 32.203s
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: quantizing model to /dev/shm/model_cache
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: Saving flywheel model at /dev/shm/model_cache
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 1%| | 2/291 [00:04<09:47, 2.03s/it] Loading 0: 8%|▊ | 23/291 [00:04<00:35, 7.57it/s] Loading 0: 15%|█▍ | 43/291 [00:04<00:15, 16.44it/s] Loading 0: 21%|██ | 60/291 [00:04<00:10, 22.29it/s] Loading 0: 29%|██▉ | 85/291 [00:04<00:05, 38.09it/s] Loading 0: 36%|███▌ | 105/291 [00:04<00:03, 52.36it/s] Loading 0: 45%|████▍ | 130/291 [00:04<00:02, 74.38it/s] Loading 0: 52%|█████▏ | 150/291 [00:05<00:01, 91.02it/s] Loading 0: 58%|█████▊ | 169/291 [00:05<00:01, 76.69it/s] Loading 0: 66%|██████▋ | 193/291 [00:05<00:00, 99.41it/s] Loading 0: 73%|███████▎ | 213/291 [00:05<00:00, 115.07it/s] Loading 0: 82%|████████▏ | 238/291 [00:05<00:00, 139.28it/s] Loading 0: 89%|████████▊ | 258/291 [00:05<00:00, 149.60it/s] Loading 0: 96%|█████████▌| 278/291 [00:06<00:00, 103.82it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: quantized model in 17.105s
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: Processed model Nitral-AI/Hathor-L3-8B-v.01 in 50.211s
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: creating bucket guanaco-mkml-models
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/nitral-ai-hathor-l3-8b-v-01-v6
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/nitral-ai-hathor-l3-8b-v-01-v6/config.json
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/nitral-ai-hathor-l3-8b-v-01-v6/special_tokens_map.json
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/nitral-ai-hathor-l3-8b-v-01-v6/tokenizer_config.json
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/nitral-ai-hathor-l3-8b-v-01-v6/tokenizer.json
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/nitral-ai-hathor-l3-8b-v-01-v6/flywheel_model.0.safetensors
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
nitral-ai-hathor-l3-8b-v-01-v6-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.
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: warnings.warn(
nitral-ai-hathor-l3-8b-v-01-v6-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.
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: warnings.warn(
nitral-ai-hathor-l3-8b-v-01-v6-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.
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: warnings.warn(
nitral-ai-hathor-l3-8b-v-01-v6-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()
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: return self.fget.__get__(instance, owner)()
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: Saving duration: 0.228s
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 6.964s
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: creating bucket guanaco-reward-models
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: Bucket 's3://guanaco-reward-models/' created
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/nitral-ai-hathor-l3-8b-v-01-v6_reward
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/nitral-ai-hathor-l3-8b-v-01-v6_reward/config.json
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/nitral-ai-hathor-l3-8b-v-01-v6_reward/tokenizer_config.json
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/nitral-ai-hathor-l3-8b-v-01-v6_reward/special_tokens_map.json
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/nitral-ai-hathor-l3-8b-v-01-v6_reward/vocab.json
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/nitral-ai-hathor-l3-8b-v-01-v6_reward/tokenizer.json
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/nitral-ai-hathor-l3-8b-v-01-v6_reward/merges.txt
nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/nitral-ai-hathor-l3-8b-v-01-v6_reward/reward.tensors
Job nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer completed after 73.23s with status: succeeded
Stopping job with name nitral-ai-hathor-l3-8b-v-01-v6-mkmlizer
Pipeline stage MKMLizer completed in 77.62s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service nitral-ai-hathor-l3-8b-v-01-v6
Waiting for inference service nitral-ai-hathor-l3-8b-v-01-v6 to be ready
Inference service nitral-ai-hathor-l3-8b-v-01-v6 ready after 30.210160732269287s
Pipeline stage ISVCDeployer completed in 37.55s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.076432943344116s
Received healthy response to inference request in 1.328660011291504s
Received healthy response to inference request in 1.310809850692749s
Received healthy response to inference request in 1.2620439529418945s
Received healthy response to inference request in 1.328148365020752s
5 requests
0 failed requests
5th percentile: 1.2717971324920654
10th percentile: 1.2815503120422362
20th percentile: 1.3010566711425782
30th percentile: 1.3142775535583495
40th percentile: 1.3212129592895507
50th percentile: 1.328148365020752
60th percentile: 1.3283530235290528
70th percentile: 1.3285576820373535
80th percentile: 1.4782145977020265
90th percentile: 1.7773237705230713
95th percentile: 1.9268783569335937
99th percentile: 2.0465220260620116
mean time: 1.461219024658203
Pipeline stage StressChecker completed in 7.93s
Running pipeline stage DaemonicModelEvalScorer
Pipeline stage DaemonicModelEvalScorer completed in 0.04s
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
nitral-ai-hathor-l3-8b-v-01_v6 status is now deployed due to DeploymentManager action
nitral-ai-hathor-l3-8b-v-01_v6 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics