Azazelle commited on
Commit
e661e28
1 Parent(s): 752b881

Upload 3 files

Browse files
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f69f9d9fc818285085540e80a28edfaf69d39220404026333582d00e3b4bf910
3
+ size 335750794
l3-8b_README.md ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ Llama 3 8B adapters
2
+
3
+ Changelog:
4
+ * 2024-06-21: v0.4 adapter update; LoRA this time
5
+ * 2024-06-13: v0.3 adapter update; attempt at DoRA adapter + some restructuring of dataset; reports of regresion from v0.2 version
6
+ * 2024-06-05: v0.2 adapter update; LoRA with updated dataset
7
+ * 2024-06-01: v0.1 adapter release
l3-8b_adapter_config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "",
5
+ "bias": "none",
6
+ "fan_in_fan_out": false,
7
+ "inference_mode": false,
8
+ "init_lora_weights": true,
9
+ "layer_replication": null,
10
+ "layers_pattern": null,
11
+ "layers_to_transform": null,
12
+ "loftq_config": {},
13
+ "lora_alpha": 16,
14
+ "lora_dropout": 0.05,
15
+ "megatron_config": null,
16
+ "megatron_core": "megatron.core",
17
+ "modules_to_save": [],
18
+ "peft_type": "LORA",
19
+ "r": 64,
20
+ "rank_pattern": {},
21
+ "revision": null,
22
+ "target_modules": [
23
+ "gate_proj",
24
+ "k_proj",
25
+ "down_proj",
26
+ "o_proj",
27
+ "v_proj",
28
+ "q_proj",
29
+ "up_proj"
30
+ ],
31
+ "task_type": "CAUSAL_LM",
32
+ "use_dora": false,
33
+ "use_rslora": false
34
+ }