jaiwithani
commited on
Commit
•
57e92c4
1
Parent(s):
2d05274
Upload folder using huggingface_hub
Browse files- iter_4/config.json +62 -0
- iter_4/model/config.json +27 -0
- iter_4/model/generation_config.json +6 -0
- iter_4/model/model.safetensors +3 -0
- iter_4/opt.pt +3 -0
- iter_4/run_context.json +7 -0
- iter_4/training_state.json +9 -0
iter_4/config.json
ADDED
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"model_config": {
|
3 |
+
"model_type": "llama2",
|
4 |
+
"mamba": null,
|
5 |
+
"llama2": {
|
6 |
+
"attention_bias": false,
|
7 |
+
"attention_dropout": 0.0,
|
8 |
+
"bos_token_id": -1,
|
9 |
+
"eos_token_id": -2,
|
10 |
+
"hidden_act": "silu",
|
11 |
+
"hidden_size": 48,
|
12 |
+
"initializer_range": 0.02,
|
13 |
+
"intermediate_size": 48,
|
14 |
+
"max_position_embeddings": 513,
|
15 |
+
"num_attention_heads": 2,
|
16 |
+
"num_hidden_layers": 2,
|
17 |
+
"num_key_value_heads": 2,
|
18 |
+
"pretraining_tp": 1,
|
19 |
+
"rms_norm_eps": 1e-06,
|
20 |
+
"rope_scaling": null,
|
21 |
+
"rope_theta": 10000.0,
|
22 |
+
"tie_word_embeddings": false,
|
23 |
+
"use_cache": true,
|
24 |
+
"vocab_size": 4096
|
25 |
+
}
|
26 |
+
},
|
27 |
+
"run_name": "debug__2024_03_19_00_59_57",
|
28 |
+
"output_dir": "/Users/jaidhyani/Library/Application Support/delphi/debug__2024_03_19_00_59_57",
|
29 |
+
"huggingface": {
|
30 |
+
"repo_id": "jaiwithani/testing",
|
31 |
+
"push_checkpoints_to_hub": true
|
32 |
+
},
|
33 |
+
"device": "auto",
|
34 |
+
"eval_interval": 1,
|
35 |
+
"log_interval": 1,
|
36 |
+
"eval_iters": 1,
|
37 |
+
"eval_only": false,
|
38 |
+
"always_save_checkpoint": true,
|
39 |
+
"init_from": "scratch",
|
40 |
+
"wandb_config": {
|
41 |
+
"log": true,
|
42 |
+
"project": "delphi",
|
43 |
+
"entity": "jaiwithani"
|
44 |
+
},
|
45 |
+
"batch_size": 64,
|
46 |
+
"max_seq_len": 512,
|
47 |
+
"max_epochs": 1,
|
48 |
+
"grad_clip": 1.0,
|
49 |
+
"optimizer": {
|
50 |
+
"gradient_accumulation_steps": 4,
|
51 |
+
"learning_rate": 0.0005,
|
52 |
+
"weight_decay": 0.1,
|
53 |
+
"beta1": 0.9,
|
54 |
+
"beta2": 0.95,
|
55 |
+
"grad_clip": 1.0,
|
56 |
+
"decay_lr": true,
|
57 |
+
"warmup_iters": 1000,
|
58 |
+
"min_lr": 0.0
|
59 |
+
},
|
60 |
+
"train_sample_limit": 1024,
|
61 |
+
"val_sample_limit": -1
|
62 |
+
}
|
iter_4/model/config.json
ADDED
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"architectures": [
|
3 |
+
"LlamaForCausalLM"
|
4 |
+
],
|
5 |
+
"attention_bias": false,
|
6 |
+
"attention_dropout": 0.0,
|
7 |
+
"bos_token_id": -1,
|
8 |
+
"eos_token_id": -2,
|
9 |
+
"hidden_act": "silu",
|
10 |
+
"hidden_size": 48,
|
11 |
+
"initializer_range": 0.02,
|
12 |
+
"intermediate_size": 48,
|
13 |
+
"max_position_embeddings": 513,
|
14 |
+
"model_type": "llama",
|
15 |
+
"num_attention_heads": 2,
|
16 |
+
"num_hidden_layers": 2,
|
17 |
+
"num_key_value_heads": 2,
|
18 |
+
"pretraining_tp": 1,
|
19 |
+
"rms_norm_eps": 1e-06,
|
20 |
+
"rope_scaling": null,
|
21 |
+
"rope_theta": 10000.0,
|
22 |
+
"tie_word_embeddings": false,
|
23 |
+
"torch_dtype": "float32",
|
24 |
+
"transformers_version": "4.39.0.dev0",
|
25 |
+
"use_cache": true,
|
26 |
+
"vocab_size": 4096
|
27 |
+
}
|
iter_4/model/generation_config.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"bos_token_id": -1,
|
4 |
+
"eos_token_id": -2,
|
5 |
+
"transformers_version": "4.39.0.dev0"
|
6 |
+
}
|
iter_4/model/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:edbfaf3965cab734e3e5b5a33b4b1703c39c7048434f266a3a657ccb7936909a
|
3 |
+
size 1705032
|
iter_4/opt.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4dae23d19e1bc53190b74f57318b32f2e390e4a054016524cc29c6c97acaaa20
|
3 |
+
size 3422425
|
iter_4/run_context.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"device": "mps",
|
3 |
+
"torch_version": "2.1.2",
|
4 |
+
"delphi_version": "0.1.1",
|
5 |
+
"transformers_version": "4.39.0.dev0",
|
6 |
+
"os": "Darwin Kernel Version 23.2.0: Wed Nov 15 21:53:34 PST 2023; root:xnu-10002.61.3~2/RELEASE_ARM64_T8103"
|
7 |
+
}
|
iter_4/training_state.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"iter_num": 4,
|
3 |
+
"local_iter_num": 4,
|
4 |
+
"best_val_loss": 8.323348045349121,
|
5 |
+
"running_mfu": -1.0,
|
6 |
+
"lr": 1.5e-06,
|
7 |
+
"epoch": 0,
|
8 |
+
"step": 3
|
9 |
+
}
|