surya-narayanan commited on
Commit
5a78917
·
verified ·
1 Parent(s): 4142bf4

Training in progress, epoch 1

Browse files
adapter_config.json CHANGED
@@ -9,10 +9,11 @@
9
  "fan_in_fan_out": false,
10
  "inference_mode": true,
11
  "init_lora_weights": true,
 
12
  "layers_pattern": null,
13
  "layers_to_transform": null,
14
  "loftq_config": {},
15
- "lora_alpha": 1024,
16
  "lora_dropout": 0.1,
17
  "megatron_config": null,
18
  "megatron_core": "megatron.core",
@@ -20,14 +21,14 @@
20
  "ln_f"
21
  ],
22
  "peft_type": "LORA",
23
- "r": 512,
24
  "rank_pattern": {},
25
  "revision": null,
26
  "target_modules": [
27
- "o_proj",
28
  "q_proj",
 
29
  "k_proj",
30
- "v_proj"
31
  ],
32
  "task_type": null,
33
  "use_dora": false,
 
9
  "fan_in_fan_out": false,
10
  "inference_mode": true,
11
  "init_lora_weights": true,
12
+ "layer_replication": null,
13
  "layers_pattern": null,
14
  "layers_to_transform": null,
15
  "loftq_config": {},
16
+ "lora_alpha": 64,
17
  "lora_dropout": 0.1,
18
  "megatron_config": null,
19
  "megatron_core": "megatron.core",
 
21
  "ln_f"
22
  ],
23
  "peft_type": "LORA",
24
+ "r": 32,
25
  "rank_pattern": {},
26
  "revision": null,
27
  "target_modules": [
 
28
  "q_proj",
29
+ "v_proj",
30
  "k_proj",
31
+ "o_proj"
32
  ],
33
  "task_type": null,
34
  "use_dora": false,
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:718167fe18399b83a129f24f43ba96a4bf04fc7ebb0c4af364b1fc37c7d2ce85
3
- size 1744865936
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0557b1d6d039ad95afe7da85f1cdf882c0ef4b1c86c50e8d70cb2ef7ceda106e
3
+ size 109086416
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6a794173063075cf5abf159596877009b87216087bb460561e5f42f7f9d26ccc
3
- size 4920
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:354acc838f9e6454e14515171afbada8a746b1e31769d28ea20abd62f3daddfd
3
+ size 4984