Views
No views yet
1base_model: Heralax/demo-nursing-model-pretrain
2tokenizer_type: AutoTokenizer
3model_type: AutoModelForCausalLM
4load_in_8bit: false
5load_in_4bit: false
6strict: false
7datasets:
8- path: axolotl_correction_conversations_hidden-openstax-nursing.json
9 type: input_output
10- path: axolotl_rag_conversations_hidden-openstax-nursing.jsonl
11 type: input_output
12- path: pretraining_subset_1421673.jsonl
13 type: completion
14- path: factual_sft_completion/combined_all_0.jsonl
15 type: completion
16- path: factual_sft_completion/combined_all_2.jsonl
17 type: completion
18- path: factual_sft_completion/combined_all_3.jsonl
19 type: completion
20- path: factual_sft_completion/combined_all_1.jsonl
21 type: completion
22- path: generic_sft_completion/Augmentoolkit-Openthoughts-100mil-DifferentFormat_2012946.jsonl
23 type: completion
24- path: generic_sft_completion/Augmentoolkit-Augmentoolkit-Pippa-Thoughts_503236.jsonl
25 type: completion
26- path: generic_sft_completion/Augmentoolkit-Augmentoolkit-LMsys-800k-Thoughts_251557.jsonl
27 type: completion
28- path: generic_sft_completion/Augmentoolkit-Openthoughts-100mil-DifferentFormat_1006231.jsonl
29 type: completion
30- path: generic_sft_completion/Augmentoolkit-Augmentoolkit-Bluemoon-1mil-thoughts_503236.jsonl
31 type: completion
32- path: generic_sft_completion/Augmentoolkit-Augmentoolkit-Capybara-2point5mil-Thoughts_251557.jsonl
33 type: completion
34- path: generic_sft_completion/Augmentoolkit-Augmentoolkit-Generic-Grabbag-Thoughts_503115.jsonl
35 type: completion
36- path: generic_sft_completion/Augmentoolkit-Augmentoolkit-LMsys-800k-Thoughts_503236.jsonl
37 type: completion
38- path: generic_sft_completion/Augmentoolkit-Augmentoolkit-Pippa-Thoughts_251557.jsonl
39 type: completion
40- path: generic_sft_completion/Augmentoolkit-Augmentoolkit-Generic-Grabbag-Thoughts_1006473.jsonl
41 type: completion
42- path: generic_sft_completion/Augmentoolkit-Augmentoolkit-Capybara-2point5mil-Thoughts_503236.jsonl
43 type: completion
44- path: generic_sft_completion/Augmentoolkit-Augmentoolkit-Bluemoon-1mil-thoughts_251557.jsonl
45 type: completion
46dataset_prepared_path: last_finetune_prepared
47output_dir: ./finetune-model-output
48seed: 1337
49sequence_len: 5000
50sample_packing: true
51pad_to_sequence_len: false
52shuffle_merged_datasets: true
53gradient_accumulation_steps: 75
54micro_batch_size: 2
55eval_batch_size: 4
56num_epochs: 5
57optimizer: paged_adamw_8bit
58lr_scheduler: constant
59learning_rate: 2.0e-05
60noisy_embedding_alpha: 5
61weight_decay: 0
62train_on_inputs: false
63group_by_length: false
64bf16: true
65fp16: false
66tf32: false
67gradient_checkpointing: true
68logging_steps: 1
69xformers_attention: false
70flash_attention: true
71chat_template: chatml
72auto_resume_from_checkpoints: false
73warmup_ratio: 0.1
74evals_per_epoch: 1
75val_set_size: 0.04
76saves_per_epoch: 1
77eval_sample_packing: false
78save_total_limit: 2
79special_tokens:
80 pad_token: <unk>
81use_liger_kernel: true
82plugins:
83- axolotl.integrations.liger.LigerPlugin
84liger_rope: true
85liger_rms_norm: true
86liger_glu_activation: true
87liger_layer_norm: true
88liger_fused_linear_cross_entropy: true
89sequence_length: 10000
90wandb_project: test-project
91wandb_entity: ''
92wandb_watch: ''
93wandb_run_id: ''
94wandb_log_model: ''
95hub_model_id: Heralax/demo-nursing-model-sft-2
96hub_strategy: all_checkpoints
97Clinical-Nursing-Skills-WEB.pdf
Fundamentals_of_Nursing_-_WEB.pdfprompt.txt, template.txt, RAG dataset, and GGUF file are all inside this folder so that people can run this model themselves using Augmentoolkit's chat interface. Just download the things not in the checkpoint-xx/ folders (not the model.safetensors files), put them all in a folder, and configure the basic-server or rag-server config to point at the prompt, template, etc., (see the documentation pages for those utility pipelines) and bang, Augmentoolkit will run these models with the correct prompt template and configuration.