ml-ryanlee commited on
Commit
5c16c15
·
verified ·
1 Parent(s): 6e64891

Upload folder using huggingface_hub

Browse files
pre_decay/config.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "LoopLMForCausalLM"
4
+ ],
5
+ "context_length": 1024,
6
+ "d_ff": 2752,
7
+ "d_model": 1024,
8
+ "dtype": "float32",
9
+ "lb_loss_factor": 0.01,
10
+ "lz_loss_factor": 0.001,
11
+ "max_length": 1024,
12
+ "model_type": "loop-lm",
13
+ "model_variant": "moe",
14
+ "num_active": 2,
15
+ "num_experts": 8,
16
+ "num_heads": 16,
17
+ "num_layers": 16,
18
+ "num_layers_in_stack": 8,
19
+ "num_stacks": 2,
20
+ "rope_theta": 10000.0,
21
+ "transformers_version": "5.3.0",
22
+ "vocab_size": 50257,
23
+ "weight_tying": false,
24
+ "width_ratio": 8.0
25
+ }
pre_decay/generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "max_length": 1024,
4
+ "output_attentions": false,
5
+ "output_hidden_states": false,
6
+ "transformers_version": "5.3.0"
7
+ }
pre_decay/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc8e19044b33f237ec6fc510260adb77bc731a115519b0cda141699b4a72b622
3
+ size 2853330496