Files changed (3) hide show
  1. README.md +12 -47
  2. adapter_config.json +4 -7
  3. adapter_model.bin +2 -2
README.md CHANGED
@@ -1,55 +1,20 @@
1
  ---
2
- license: other
3
- tags:
4
- - generated_from_trainer
5
- model-index:
6
- - name: empath-llama-7b
7
- results: []
8
  library_name: peft
9
  ---
10
-
11
- <!-- This model card has been generated automatically according to the information the Trainer had access to. You
12
- should probably proofread and complete it, then remove this comment. -->
13
-
14
- # empath-llama-7b
15
-
16
- This model is a fine-tuned version of [decapoda-research/llama-7b-hf](https://huggingface.co/decapoda-research/llama-7b-hf) on the None dataset.
17
-
18
- ## Model description
19
-
20
- More information needed
21
-
22
- ## Intended uses & limitations
23
-
24
- More information needed
25
-
26
- ## Training and evaluation data
27
-
28
- More information needed
29
-
30
  ## Training procedure
31
 
32
- ### Training hyperparameters
33
-
34
- The following hyperparameters were used during training:
35
- - learning_rate: 0.001
36
- - train_batch_size: 4
37
- - eval_batch_size: 8
38
- - seed: 42
39
- - gradient_accumulation_steps: 4
40
- - total_train_batch_size: 16
41
- - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
42
- - lr_scheduler_type: linear
43
- - lr_scheduler_warmup_steps: 100
44
- - training_steps: 100
45
-
46
- ### Training results
47
-
48
-
49
 
 
 
 
 
 
 
 
 
 
 
50
  ### Framework versions
51
 
52
- - Transformers 4.31.0.dev0
53
- - Pytorch 2.0.1+cu117
54
- - Datasets 2.12.0
55
- - Tokenizers 0.13.3
 
1
  ---
 
 
 
 
 
 
2
  library_name: peft
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
4
  ## Training procedure
5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
6
 
7
+ The following `bitsandbytes` quantization config was used during training:
8
+ - load_in_8bit: False
9
+ - load_in_4bit: True
10
+ - llm_int8_threshold: 6.0
11
+ - llm_int8_skip_modules: None
12
+ - llm_int8_enable_fp32_cpu_offload: False
13
+ - llm_int8_has_fp16_weight: False
14
+ - bnb_4bit_quant_type: nf4
15
+ - bnb_4bit_use_double_quant: True
16
+ - bnb_4bit_compute_dtype: bfloat16
17
  ### Framework versions
18
 
19
+
20
+ - PEFT 0.4.0.dev0
 
 
adapter_config.json CHANGED
@@ -1,22 +1,19 @@
1
  {
2
- "base_model_name_or_path": "decapoda-research/llama-7b-hf",
3
  "bias": "none",
4
  "fan_in_fan_out": false,
5
  "inference_mode": true,
6
  "init_lora_weights": true,
7
  "layers_pattern": null,
8
  "layers_to_transform": null,
9
- "lora_alpha": 16,
10
  "lora_dropout": 0.05,
11
  "modules_to_save": null,
12
  "peft_type": "LORA",
13
- "r": 8,
14
  "revision": null,
15
  "target_modules": [
16
- "q_proj",
17
- "k_proj",
18
- "v_proj",
19
- "o_proj"
20
  ],
21
  "task_type": "CAUSAL_LM"
22
  }
 
1
  {
2
+ "base_model_name_or_path": "vilsonrodrigues/falcon-7b-instruct-sharded",
3
  "bias": "none",
4
  "fan_in_fan_out": false,
5
  "inference_mode": true,
6
  "init_lora_weights": true,
7
  "layers_pattern": null,
8
  "layers_to_transform": null,
9
+ "lora_alpha": 32,
10
  "lora_dropout": 0.05,
11
  "modules_to_save": null,
12
  "peft_type": "LORA",
13
+ "r": 16,
14
  "revision": null,
15
  "target_modules": [
16
+ "query_key_value"
 
 
 
17
  ],
18
  "task_type": "CAUSAL_LM"
19
  }
adapter_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9c1aa6604e99b0dc7e58e02adbf00fb613be312e40a0dd79ce13a7f48f5ec96b
3
- size 33646413
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6d79e7b0463b1f92b20594a358fc3989e5f377e7905cc605dc9d95a739884cc1
3
+ size 18898161