Upload 5 files
Browse files- adapter_config.json +26 -0
- adapter_model.safetensors +3 -0
- training_log.json +16 -0
- training_parameters.json +37 -0
- training_prompt.json +3 -0
adapter_config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "models/TheBloke_rocket-3B-GPTQ",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layers_pattern": null,
|
| 10 |
+
"layers_to_transform": null,
|
| 11 |
+
"loftq_config": {},
|
| 12 |
+
"lora_alpha": 512,
|
| 13 |
+
"lora_dropout": 0.05,
|
| 14 |
+
"megatron_config": null,
|
| 15 |
+
"megatron_core": "megatron.core",
|
| 16 |
+
"modules_to_save": null,
|
| 17 |
+
"peft_type": "LORA",
|
| 18 |
+
"r": 256,
|
| 19 |
+
"rank_pattern": {},
|
| 20 |
+
"revision": null,
|
| 21 |
+
"target_modules": [
|
| 22 |
+
"q_proj",
|
| 23 |
+
"v_proj"
|
| 24 |
+
],
|
| 25 |
+
"task_type": "CAUSAL_LM"
|
| 26 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6e9b16dedbf2d2d5d6a9245818ee5e5c64d3247e9aae3b089fa618c1875dc19d
|
| 3 |
+
size 335561896
|
training_log.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"base_model_name": "TheBloke_rocket-3B-GPTQ",
|
| 3 |
+
"base_model_class": "StableLMEpochForCausalLM",
|
| 4 |
+
"base_loaded_in_4bit": false,
|
| 5 |
+
"base_loaded_in_8bit": false,
|
| 6 |
+
"projections": "q, v",
|
| 7 |
+
"loss": 2.813,
|
| 8 |
+
"learning_rate": 6.464906595023967e-09,
|
| 9 |
+
"epoch": 2.96,
|
| 10 |
+
"current_steps": 959,
|
| 11 |
+
"train_runtime": 13914.3966,
|
| 12 |
+
"train_samples_per_second": 4.459,
|
| 13 |
+
"train_steps_per_second": 0.009,
|
| 14 |
+
"total_flos": 4.013119292571648e+16,
|
| 15 |
+
"train_loss": 2.8925536632537843
|
| 16 |
+
}
|
training_parameters.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"lora_name": "Yiff",
|
| 3 |
+
"always_override": true,
|
| 4 |
+
"q_proj_en": true,
|
| 5 |
+
"v_proj_en": true,
|
| 6 |
+
"k_proj_en": false,
|
| 7 |
+
"o_proj_en": false,
|
| 8 |
+
"gate_proj_en": false,
|
| 9 |
+
"down_proj_en": false,
|
| 10 |
+
"up_proj_en": false,
|
| 11 |
+
"save_steps": 0.0,
|
| 12 |
+
"micro_batch_size": 64,
|
| 13 |
+
"batch_size": 512,
|
| 14 |
+
"epochs": 3.0,
|
| 15 |
+
"learning_rate": "3e-5",
|
| 16 |
+
"lr_scheduler_type": "cosine",
|
| 17 |
+
"lora_rank": 256,
|
| 18 |
+
"lora_alpha": 512,
|
| 19 |
+
"lora_dropout": 0.05,
|
| 20 |
+
"cutoff_len": 512,
|
| 21 |
+
"dataset": "None",
|
| 22 |
+
"eval_dataset": "None",
|
| 23 |
+
"format": "None",
|
| 24 |
+
"eval_steps": 100.0,
|
| 25 |
+
"raw_text_file": "combined",
|
| 26 |
+
"overlap_len": 128,
|
| 27 |
+
"newline_favor_len": 128,
|
| 28 |
+
"higher_rank_limit": false,
|
| 29 |
+
"warmup_steps": 100.0,
|
| 30 |
+
"optimizer": "adamw_torch",
|
| 31 |
+
"hard_cut_string": "***",
|
| 32 |
+
"train_only_after": "",
|
| 33 |
+
"stop_at_loss": 0,
|
| 34 |
+
"add_eos_token": false,
|
| 35 |
+
"min_chars": 0.0,
|
| 36 |
+
"report_to": "None"
|
| 37 |
+
}
|
training_prompt.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"template_type": "raw_text"
|
| 3 |
+
}
|