joelkoch commited on
Commit
21bcf31
·
verified ·
1 Parent(s): b997b0a

Upload folder using huggingface_hub

Browse files
Files changed (3) hide show
  1. config.json +38 -0
  2. generation_config.json +8 -0
  3. model.safetensors +3 -0
config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SmolLM3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 128000,
8
+ "dtype": "bfloat16",
9
+ "eos_token_id": 128012,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 64,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 64,
14
+ "layer_types": [
15
+ "full_attention"
16
+ ],
17
+ "max_position_embeddings": 64,
18
+ "max_window_layers": 28,
19
+ "mlp_bias": false,
20
+ "model_type": "smollm3",
21
+ "no_rope_layer_interval": 4,
22
+ "no_rope_layers": [
23
+ 1
24
+ ],
25
+ "num_attention_heads": 2,
26
+ "num_hidden_layers": 1,
27
+ "num_key_value_heads": 1,
28
+ "pad_token_id": 128004,
29
+ "pretraining_tp": 2,
30
+ "rms_norm_eps": 1e-06,
31
+ "rope_scaling": null,
32
+ "rope_theta": 5000000.0,
33
+ "sliding_window": null,
34
+ "transformers_version": "4.56.0",
35
+ "use_cache": false,
36
+ "use_sliding_window": false,
37
+ "vocab_size": 128256
38
+ }
generation_config.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 128000,
4
+ "eos_token_id": 128012,
5
+ "pad_token_id": 128004,
6
+ "transformers_version": "4.56.0",
7
+ "use_cache": false
8
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bf2a558fb3fdcc89f48c48fb86e18ee5de6f29fd0e304b328a063b11e5749953
3
+ size 16467496