deqing commited on
Commit
ab4a6e2
·
verified ·
1 Parent(s): 4542e7c

Training in progress, step 13000

Browse files
Files changed (3) hide show
  1. adapter_model.safetensors +1 -1
  2. config.json +27 -21
  3. model.safetensors +2 -2
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bd3cce76dd9c80e45c8dc324cff572b8ec05355ee520419bce8cc29c2fe1de0f
3
  size 11301520
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0a15c640269b10d8e33856a5ca0dd6612cd37d02f1b45a934b5c542f341afafb
3
  size 11301520
config.json CHANGED
@@ -1,31 +1,37 @@
1
  {
2
- "_name_or_path": "fourier_cnt_pretrain",
3
  "architectures": [
4
- "Qwen2ForCausalLM"
5
  ],
 
6
  "attention_dropout": 0.0,
7
- "bos_token_id": 151643,
8
- "eos_token_id": 151643,
 
9
  "hidden_act": "silu",
10
- "hidden_size": 896,
11
  "initializer_range": 0.02,
12
- "intermediate_size": 4864,
13
- "max_position_embeddings": 32768,
14
- "max_window_layers": 24,
15
- "model_type": "qwen2",
16
- "num_attention_heads": 14,
17
- "num_hidden_layers": 24,
18
- "num_key_value_heads": 2,
19
- "pad_token_id": 151643,
20
- "rms_norm_eps": 1e-06,
21
- "rope_scaling": null,
22
- "rope_theta": 1000000.0,
23
- "sliding_window": null,
 
 
 
 
 
 
24
  "tie_word_embeddings": true,
25
- "torch_dtype": "bfloat16",
26
  "transformers_version": "4.48.2",
27
  "use_cache": true,
28
- "use_mrope": false,
29
- "use_sliding_window": false,
30
- "vocab_size": 151936
31
  }
 
1
  {
2
+ "_name_or_path": "llama_fourier_cnt_pretrain",
3
  "architectures": [
4
+ "LlamaForCausalLM"
5
  ],
6
+ "attention_bias": false,
7
  "attention_dropout": 0.0,
8
+ "bos_token_id": 128000,
9
+ "eos_token_id": 128001,
10
+ "head_dim": 64,
11
  "hidden_act": "silu",
12
+ "hidden_size": 2048,
13
  "initializer_range": 0.02,
14
+ "intermediate_size": 8192,
15
+ "max_position_embeddings": 131072,
16
+ "mlp_bias": false,
17
+ "model_type": "llama",
18
+ "num_attention_heads": 32,
19
+ "num_hidden_layers": 16,
20
+ "num_key_value_heads": 8,
21
+ "pad_token_id": 128001,
22
+ "pretraining_tp": 1,
23
+ "rms_norm_eps": 1e-05,
24
+ "rope_scaling": {
25
+ "factor": 32.0,
26
+ "high_freq_factor": 4.0,
27
+ "low_freq_factor": 1.0,
28
+ "original_max_position_embeddings": 8192,
29
+ "rope_type": "llama3"
30
+ },
31
+ "rope_theta": 500000.0,
32
  "tie_word_embeddings": true,
33
+ "torch_dtype": "float32",
34
  "transformers_version": "4.48.2",
35
  "use_cache": true,
36
+ "vocab_size": 128256
 
 
37
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:61d505a577901cd4ae53657ef4ba3967527a6be6e3cbe1d4c9c850433a007544
3
- size 988097824
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:626f785dc5e80339a672f2a321c71374d44b3f08630fa35f4cf923d0ba4f9a80
3
+ size 4943274328