Hugging Face's logo

devoppro
/
FastLLM

Text Generation
Transformers
Safetensors
English
modern_llm
custom-architecture
rope
gqa
swiglu
rmsnorm
custom_code
Model card Files
xet
 Community
devoppro commited on
Commit
1113477
·
verified ·
1 Parent(s): f79ddd9

Training in progress, step 100, checkpoint

Browse files
Files changed (8) hide show
  1. last-checkpoint/config.json +21 -0
  2. last-checkpoint/model.safetensors +3 -0
  3. last-checkpoint/optimizer.pt +3 -0
  4. last-checkpoint/rng_state.pth +3 -0
  5. last-checkpoint/scaler.pt +3 -0
  6. last-checkpoint/scheduler.pt +3 -0
  7. last-checkpoint/trainer_state.json +104 -0
  8. last-checkpoint/training_args.bin +3 -0
last-checkpoint/config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "ModernLLMForCausalLM"
4
+ ],
5
+ "bos_token_id": 1,
6
+ "dtype": "float32",
7
+ "eos_token_id": 2,
8
+ "hidden_size": 768,
9
+ "intermediate_size": 2048,
10
+ "max_position_embeddings": 2048,
11
+ "model_type": "modern_llm",
12
+ "num_attention_heads": 12,
13
+ "num_hidden_layers": 12,
14
+ "num_key_value_heads": 4,
15
+ "pad_token_id": 0,
16
+ "rms_norm_eps": 1e-06,
17
+ "rope_theta": 1000000.0,
18
+ "transformers_version": "5.15.0",
19
+ "use_cache": false,
20
+ "vocab_size": 151936
21
+ }
last-checkpoint/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a746d1236d3fa2278997b4fe8b4862ddd7e6f6943a953a1e3a3dcb4e53feef47
3
+ size 1235573136
last-checkpoint/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f22e2f31d8b844f3b49fcf0dfcfd7d49a319f0f4f83070bfae39bda9c1efaa59
3
+ size 2471218763
last-checkpoint/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:61c19bab1174704a4a4441475683bf1270277af15d2e2c95e964789128e482c4
3
+ size 14645
last-checkpoint/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d15f80eff8e7aa3a49c0408f3d31fa2e6e9e75a2fbae063b0cc83bfa545cc3d7
3
+ size 1383
last-checkpoint/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:116b971bb63560610677655e38ba8be2afb415c6fd29c1d069804301bcde855e
3
+ size 1465
last-checkpoint/trainer_state.json ADDED
@@ -0,0 +1,104 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.002,
6
+ "eval_steps": 500,
7
+ "global_step": 100,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0002,
14
+ "grad_norm": 226.5673828125,
15
+ "learning_rate": 2.6999999999999996e-05,
16
+ "loss": 67.51285400390626,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.0004,
21
+ "grad_norm": 30.177888870239258,
22
+ "learning_rate": 5.6999999999999996e-05,
23
+ "loss": 29.44110107421875,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.0006,
28
+ "grad_norm": 7.557392120361328,
29
+ "learning_rate": 8.699999999999999e-05,
30
+ "loss": 16.708602905273438,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.0008,
35
+ "grad_norm": 4.4042744636535645,
36
+ "learning_rate": 0.000117,
37
+ "loss": 16.181797790527344,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.001,
42
+ "grad_norm": 2.3703675270080566,
43
+ "learning_rate": 0.000147,
44
+ "loss": 13.265884399414062,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.0012,
49
+ "grad_norm": 0.868837296962738,
50
+ "learning_rate": 0.00017699999999999997,
51
+ "loss": 11.693977355957031,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.0014,
56
+ "grad_norm": 1.1192706823349,
57
+ "learning_rate": 0.00020699999999999996,
58
+ "loss": 12.454998016357422,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.0016,
63
+ "grad_norm": 1.4453760385513306,
64
+ "learning_rate": 0.000237,
65
+ "loss": 13.03106231689453,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.0018,
70
+ "grad_norm": 1.499172568321228,
71
+ "learning_rate": 0.000267,
72
+ "loss": 10.518834686279297,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.002,
77
+ "grad_norm": 2.132214069366455,
78
+ "learning_rate": 0.00029699999999999996,
79
+ "loss": 12.466087341308594,
80
+ "step": 100
81
+ }
82
+ ],
83
+ "logging_steps": 10,
84
+ "max_steps": 50000,
85
+ "num_input_tokens_seen": 0,
86
+ "num_train_epochs": 9223372036854775807,
87
+ "save_steps": 100,
88
+ "stateful_callbacks": {
89
+ "TrainerControl": {
90
+ "args": {
91
+ "should_epoch_stop": false,
92
+ "should_evaluate": false,
93
+ "should_log": false,
94
+ "should_save": true,
95
+ "should_training_stop": false
96
+ },
97
+ "attributes": {}
98
+ }
99
+ },
100
+ "total_flos": 1889437483008000.0,
101
+ "train_batch_size": 2,
102
+ "trial_name": null,
103
+ "trial_params": null
104
+ }
last-checkpoint/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:edcf268bd411db3d29ef48993a8be91c48e0928ee9333782a1c880f5f6a50c85
3
+ size 5201