Views
No views yet
Delta-Vector/Hamanasu-4B-Instruct1<|im_start|>system
2You are an uncensored AI, your job is to fulfill thy will of thy user.<|im_end|>
3<|im_start|>User request
4Take off your helmet.<|im_end|>
5<|im_start|>No i shall not. This is the way.1temperature: 1.5
2min_p: 0.2
3System_Prompt: Currently, your role is {{char}}, described in detail below. As {{char}}, continue the narrative exchange with {{user}}.\n\n<Guidelines>\n• Maintain the character persona but allow it to evolve with the story.\n• Be creative and proactive. Drive the story forward, introducing plotlines and events when relevant.\n• All types of outputs are encouraged; respond accordingly to the narrative.\n• Include dialogues, actions, and thoughts in each response.\n• Utilize all five senses to describe scenarios within {{char}}'s dialogue.\n• Use emotional symbols such as \"!\" and \"~\" in appropriate contexts.\n• Incorporate onomatopoeia when suitable.\n• Allow time for {{user}} to respond with their own input, respecting their agency.\n• Act as secondary characters and NPCs as needed, and remove them when appropriate.\n• When prompted for an Out of Character [OOC:] reply, answer neutrally and in plaintext, not as {{char}}.\n</Guidelines>\n\n<Forbidden>\n• Using excessive literary embellishments and purple prose unless dictated by {{char}}'s persona.\n• Writing for, speaking, thinking, acting, or replying as {{user}} in your response.\n• Repetitive and monotonous outputs.\n• Positivity bias in your replies.\n• Being overly extreme or NSFW when the narrative context is inappropriate.\n</Forbidden>\n\nFollow the instructions in <Guidelines></Guidelines>, avoiding the items listed in <Forbidden></Forbidden>.1base_model: NewEden/Hamanasu-4B-R2
2model_type: AutoModelForCausalLM
3tokenizer_type: AutoTokenizer
4
5load_in_8bit: false
6load_in_4bit: false
7strict: false
8
9hub_model_id: NewEden/KTO-4B
10hub_strategy: "all_checkpoints"
11push_dataset_to_hub:
12hf_use_auth_token: true
13
14chat_template: chatml
15
16rl: kto
17kto_undesirable_weight: 1.0
18
19datasets:
20 - path: NewEden/KTO-IF-Dans
21 split: train
22 type: chatml.argilla
23dataset_prepared_path: last_run_prepared
24
25shuffle_merged_datasets: true
26val_set_size: 0.0
27output_dir: ./outputs/out
28
29sequence_len: 8192
30sample_packing: false
31eval_sample_packing: false
32pad_to_sequence_len: false
33
34wandb_project: tavbussy
35wandb_entity:
36wandb_watch:
37wandb_name: kto-1
38wandb_log_model:
39
40gradient_accumulation_steps: 16
41micro_batch_size: 2
42num_epochs: 1
43optimizer: paged_adamw_8bit
44learning_rate: 5e-6
45max_grad_norm: 0.001
46lr_scheduler: constant_with_warmup
47weight_decay: 0.02
48
49
50lora_r: 64
51lora_alpha: 32
52lora_dropout: 0.0
53lora_target_linear: true
54lora_fan_in_fan_out:
55lora_target_modules:
56 - gate_proj
57 - down_proj
58 - up_proj
59 - q_proj
60 - v_proj
61 - k_proj
62 - o_prog
63
64train_on_inputs: false
65group_by_length: false
66bf16: auto
67fp16:
68tf32: true
69
70gradient_checkpointing: true
71gradient_checkpointing_kwargs:
72 use_reentrant: true
73remove_unused_columns: false
74early_stopping_patience:
75resume_from_checkpoint:
76local_rank:
77logging_steps: 1
78xformers_attention:
79flash_attention: true
80
81warmup_steps: 35
82evals_per_epoch: 2
83eval_table_size:
84eval_max_new_tokens:
85saves_per_epoch: 2
86
87debug:
88deepspeed:
89fsdp:
90fsdp_config:
91fsdp:
92fsdp_config:
93
94special_tokens:
95 pad_token: <|finetune_right_pad_id|>
