Fischerboot/InternLM2-ToxicRP-QLORA-4Bit

Compute power from g4rg. Big Thanks.

See axolotl config

axolotl version: 0.4.0

mlflow_tracking_uri: http://127.0.0.1:2340
mlflow_experiment_name: Default

base_model: intervitens/internlm2-limarp-chat-20b
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer

load_in_8bit: false
load_in_4bit: true
strict: false

datasets:
  - path: ResplendentAI/Alpaca_NSFW_Shuffled
    type: alpaca
  - path: diffnamehard/toxic-dpo-v0.1-NoWarning-alpaca
    type: alpaca
dataset_prepared_path: last_run_prepared
val_set_size: 0.1
output_dir: ./outputs/qlora-out

adapter: qlora
lora_model_dir:

sequence_len: 8192
sample_packing: false
pad_to_sequence_len: true

lora_r: 32
lora_alpha: 16
lora_dropout: 0.05
lora_target_linear: true
lora_fan_in_fan_out:
lora_target_modules:
  - gate_proj
  - down_proj
  - up_proj
  - q_proj
  - v_proj
  - k_proj
  - o_proj

wandb_project:
wandb_entity:
wandb_watch:
wandb_name:
wandb_log_model:

gradient_accumulation_steps: 4
micro_batch_size: 2
num_epochs: 4
optimizer: adamw_bnb_8bit
lr_scheduler: cosine
learning_rate: 0.0002

train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32: false

gradient_checkpointing: true
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: true

loss_watchdog_threshold: 5.0
loss_watchdog_patience: 3

warmup_steps: 10
evals_per_epoch: 4
eval_table_size:
eval_max_new_tokens: 128
saves_per_epoch: 1
debug:
deepspeed:
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:

outputs/qlora-out

This model is a fine-tuned version of intervitens/internlm2-limarp-chat-20b on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.9896

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0002
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• distributed_type: multi-GPU
• num_devices: 7
• gradient_accumulation_steps: 4
• total_train_batch_size: 56
• total_eval_batch_size: 14
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 10
• num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss
1.4668	0.0476	1	1.4615
1.3541	0.2857	6	1.4253
1.2057	0.5714	12	1.2120
1.0818	0.8571	18	1.1259
1.0835	1.1429	24	1.0750
1.0503	1.4286	30	1.0451
1.0031	1.7143	36	1.0288
0.9728	2.0	42	1.0137
0.8879	2.2857	48	1.0082
0.8981	2.5714	54	0.9956
0.8613	2.8571	60	0.9926
0.8608	3.1429	66	0.9903
0.7841	3.4286	72	0.9903
0.9237	3.7143	78	0.9899
0.868	4.0	84	0.9896

Framework versions

• PEFT 0.10.0
• Transformers 4.40.2
• Pytorch 2.3.0
• Datasets 2.19.1
• Tokenizers 0.19.1