/
stablelm-2-1_6b.yml
148 lines (127 loc) · 3.69 KB
/
stablelm-2-1_6b.yml
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
{
# parallelism settings
"pipe-parallel-size": 0,
"model-parallel-size": 1,
# model settings
"num-layers": 24,
"hidden-size": 2048,
"num-attention-heads": 32,
"seq-length": 4096,
"max-position-embeddings": 4096,
# architecture design
"attention_head_type": "multihead",
"norm": "layernorm",
"pos-emb": "rotary",
"rotary_pct": 0.25,
"rotary_interleaved": false, # GPT-NeoX style
"mlp_multiple_of": 256,
"mlp_type": "gated",
"activation": "silu",
"no-weight-tying": true,
"gpt_j_residual": false,
"gpt_j_tied": false,
"output_layer_parallelism": "column",
# init methods
"init_method": "normal",
"output_layer_init_method": "scaled_normal",
"init_method_std": 0.02,
# biases
"use_bias_in_norms": false,
"use_bias_in_attn_linear": false,
"use_bias_in_mlp": false,
# fused ops
"use_flash_cross_entropy": true,
"bias-gelu-fusion": false,
"scaled-upper-triang-masked-softmax-fusion": false,
"attention-config": [[["flash"], 24]],
# optimizer settings
"optimizer": {
"type": "Adam",
"params": {
"lr": 0.001,
"betas": [0.9, 0.95],
"eps": 1.0e-8,
}
},
"min_lr": 0.0001,
"train-iters": 540_000,
"lr-decay-iters": 540_000,
"lr-decay-style": "hybrid_cosine_inv_sqrt_2",
"warmup": 0.018,
"cooldown": 0.,
"reset_attention_mask": true,
"reset_position_ids": true,
# for all zero_optimization options, see https://www.deepspeed.ai/docs/config-json/#zero-optimizations-for-fp16-training
"zero_optimization": {
"stage": 1,
"allgather_partitions": true,
"allgather_bucket_size": 1260000000,
"overlap_comm": true,
"reduce_scatter": true,
"reduce_bucket_size": 1260000000,
"contiguous_gradients": true,
"cpu_offload": false,
},
# batch / data settings
"train_micro_batch_size_per_gpu": 2,
"gradient_accumulation_steps": 2,
"data-impl": "mmap",
"eval-interval": 500_000,
"eval-iters": 1,
"eval_batch_size": 1,
"eval_tasks": [],
# activation checkpointing
"checkpoint-activations": true,
"checkpoint-num-layers": 24,
"partition-activations": true,
"synchronize-each-layer": true,
# regularization
"gradient_clipping": 1,
"weight-decay": 0.1,
"hidden-dropout": 0.,
"attention-dropout": 0.,
# precision settings
"bf16": { "enabled": true },
"precision": "bfloat16",
"full_precision_lm_cross_entropy": true,
"fp32_allreduce": true,
# misc. training settings
"num-workers": 2,
"distributed-backend": "nccl",
# checkpoint settings
"checkpoint-factor": 2_000,
"s3_sync_interval": 20_000,
"extra-save-iters": [0],
"save": "",
"load": "",
"s3_path": "",
"train_data_paths": [],
"train_data_weights": [],
"valid-data-paths": ["minipile_validation_arcade100k_tokenized_text_document"],
"valid-data-weights": [1.0,],
"test-data-paths": ["minipile_validation_arcade100k_tokenized_text_document"],
"test-data-weights": [1.0,],
# tokenizer settings
"tokenizer-type": "TiktokenTokenizer",
"vocab-file": "arcade100k.tiktoken",
"log-interval": 10,
"steps_per_print": 10,
"wall_clock_breakdown": true,
"use_wandb": true,
"wandb_host": "https://stability.wandb.io",
"wandb_team": "stability-llm",
"wandb_project": "",
"wandb_group": "",
"wandb_name": "",
# "wandb_id": "",
# "wandb_resume": "must",
# MuP
"use-mup": false,
"save-base-shapes": false , # this only needs to be enabled once in order to generate the base-shapes-file on each rank
"base-shapes-file": "mup-base-shapes-small-fixed/base-shapes-small", # load base shapes from this file
"coord-check": false, # generate coord check plots to verify mup's implementation in neox
# multi-node launcher
"launcher": "slurm",
"deepspeed_slurm": true,
"seed": 1234
}