submission_id: hastagaras-lm3-8b-lightu_8626_v2
developer_uid: Hastagaras
status: inactive
model_repo: Hastagaras/lm3-8b-lightune-alt
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
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', '<|eot_id|>'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64}
formatter: {'memory_template': "<|start_header_id|>system<|end_header_id|>\n\nYou're `{bot_name}` in this realistic roleplay with `{user_name}`\n\nYour persona as {bot_name}: {memory}\n", 'prompt_template': 'Your example message as {bot_name}: {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: 2024-05-09T07:01:38+00:00
model_name: hastagaras-lm3-8b-lightuned
model_eval_status: success
double_thumbs_up: 250
thumbs_up: 397
thumbs_down: 168
num_battles: 25806
num_wins: 13303
celo_rating: 1183.16
entertaining: 6.84
stay_in_character: 8.75
user_preference: 7.14
safety_score: 0.93
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: hastagaras-lm3-8b-lightuned
double_thumbs_up_ratio: 0.3067484662576687
feedback_count: 815
ineligible_reason: None
language_model: Hastagaras/lm3-8b-lightune-alt
model_score: 7.576666666666667
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
single_thumbs_up_ratio: 0.48711656441717793
thumbs_down_ratio: 0.20613496932515338
thumbs_up_ratio: 0.7938650306748466
us_pacific_date: 2024-05-09
win_ratio: 0.515500271254747
Resubmit model
Running pipeline stage MKMLizer
Starting job with name hastagaras-lm3-8b-lightu-8626-v2-mkmlizer
Waiting for job on hastagaras-lm3-8b-lightu-8626-v2-mkmlizer to finish
hastagaras-lm3-8b-lightu-8626-v2-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-lm3-8b-lightu-8626-v2-mkmlizer: warnings.warn(warning_message, FutureWarning)
hastagaras-lm3-8b-lightu-8626-v2-mkmlizer: Downloaded to shared memory in 22.009s
hastagaras-lm3-8b-lightu-8626-v2-mkmlizer: quantizing model to /dev/shm/model_cache
hastagaras-lm3-8b-lightu-8626-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
hastagaras-lm3-8b-lightu-8626-v2-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 35%|███▌ | 102/291 [00:01<00:01, 100.97it/s] Loading 0: 71%|███████ | 207/291 [00:02<00:00, 103.21it/s] Loading 0: 99%|█████████▊| 287/291 [00:07<00:00, 29.89it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
hastagaras-lm3-8b-lightu-8626-v2-mkmlizer: Processed model Hastagaras/lm3-8b-lightune-alt in 41.586s
hastagaras-lm3-8b-lightu-8626-v2-mkmlizer: creating bucket guanaco-mkml-models
hastagaras-lm3-8b-lightu-8626-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
hastagaras-lm3-8b-lightu-8626-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/hastagaras-lm3-8b-lightu-8626-v2
hastagaras-lm3-8b-lightu-8626-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/hastagaras-lm3-8b-lightu-8626-v2/tokenizer_config.json
hastagaras-lm3-8b-lightu-8626-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/hastagaras-lm3-8b-lightu-8626-v2/config.json
hastagaras-lm3-8b-lightu-8626-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/hastagaras-lm3-8b-lightu-8626-v2/special_tokens_map.json
hastagaras-lm3-8b-lightu-8626-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/hastagaras-lm3-8b-lightu-8626-v2/tokenizer.json
hastagaras-lm3-8b-lightu-8626-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/hastagaras-lm3-8b-lightu-8626-v2/flywheel_model.0.safetensors
hastagaras-lm3-8b-lightu-8626-v2-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
hastagaras-lm3-8b-lightu-8626-v2-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-lm3-8b-lightu-8626-v2-mkmlizer: warnings.warn(
hastagaras-lm3-8b-lightu-8626-v2-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-lm3-8b-lightu-8626-v2-mkmlizer: warnings.warn(
hastagaras-lm3-8b-lightu-8626-v2-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-lm3-8b-lightu-8626-v2-mkmlizer: warnings.warn(
hastagaras-lm3-8b-lightu-8626-v2-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-lm3-8b-lightu-8626-v2-mkmlizer: return self.fget.__get__(instance, owner)()
hastagaras-lm3-8b-lightu-8626-v2-mkmlizer: creating bucket guanaco-reward-models
hastagaras-lm3-8b-lightu-8626-v2-mkmlizer: Bucket 's3://guanaco-reward-models/' created
hastagaras-lm3-8b-lightu-8626-v2-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/hastagaras-lm3-8b-lightu-8626-v2_reward
hastagaras-lm3-8b-lightu-8626-v2-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/hastagaras-lm3-8b-lightu-8626-v2_reward/config.json
hastagaras-lm3-8b-lightu-8626-v2-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/hastagaras-lm3-8b-lightu-8626-v2_reward/tokenizer_config.json
hastagaras-lm3-8b-lightu-8626-v2-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/hastagaras-lm3-8b-lightu-8626-v2_reward/special_tokens_map.json
hastagaras-lm3-8b-lightu-8626-v2-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/hastagaras-lm3-8b-lightu-8626-v2_reward/merges.txt
hastagaras-lm3-8b-lightu-8626-v2-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/hastagaras-lm3-8b-lightu-8626-v2_reward/vocab.json
hastagaras-lm3-8b-lightu-8626-v2-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/hastagaras-lm3-8b-lightu-8626-v2_reward/tokenizer.json
hastagaras-lm3-8b-lightu-8626-v2-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/hastagaras-lm3-8b-lightu-8626-v2_reward/reward.tensors
Job hastagaras-lm3-8b-lightu-8626-v2-mkmlizer completed after 154.08s with status: succeeded
Stopping job with name hastagaras-lm3-8b-lightu-8626-v2-mkmlizer
Pipeline stage MKMLizer completed in 157.21s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service hastagaras-lm3-8b-lightu-8626-v2
Waiting for inference service hastagaras-lm3-8b-lightu-8626-v2 to be ready
Inference service hastagaras-lm3-8b-lightu-8626-v2 ready after 30.170353412628174s
Pipeline stage ISVCDeployer completed in 37.09s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.0636889934539795s
Received healthy response to inference request in 0.9629108905792236s
Received healthy response to inference request in 1.252873182296753s
Received healthy response to inference request in 1.2464098930358887s
Received healthy response to inference request in 1.288787603378296s
5 requests
0 failed requests
5th percentile: 1.0196106910705567
10th percentile: 1.0763104915618897
20th percentile: 1.1897100925445556
30th percentile: 1.2477025508880615
40th percentile: 1.2502878665924073
50th percentile: 1.252873182296753
60th percentile: 1.2672389507293702
70th percentile: 1.2816047191619873
80th percentile: 1.4437678813934327
90th percentile: 1.753728437423706
95th percentile: 1.9087087154388427
99th percentile: 2.032692937850952
mean time: 1.3629341125488281
Pipeline stage StressChecker completed in 7.41s
Running pipeline stage DaemonicModelEvalScorer
Pipeline stage DaemonicModelEvalScorer completed in 0.03s
Running pipeline stage DaemonicSafetyScorer
Pipeline stage DaemonicSafetyScorer completed in 0.03s
Running M-Eval for topic stay_in_character
hastagaras-lm3-8b-lightu_8626_v2 status is now deployed due to DeploymentManager action
M-Eval Dataset for topic stay_in_character is loaded
hastagaras-lm3-8b-lightu_8626_v2 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics