Upload 2 files
Browse files- LilLM_300M_9_5B_Tokens.pt +3 -0
- config.py +16 -0
LilLM_300M_9_5B_Tokens.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d1fc8ffbf766c30da74939cc1629aaf12f424cbe905a5f4e44224b283f2eae96
|
3 |
+
size 3498930790
|
config.py
ADDED
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
from dataclasses import dataclass
|
2 |
+
|
3 |
+
@dataclass
|
4 |
+
class Config:
|
5 |
+
vocab_size: int = 2**13
|
6 |
+
d_model: int = 1024
|
7 |
+
n_layers: int = 24
|
8 |
+
max_seq_len: int = 512
|
9 |
+
q_heads: int = 16
|
10 |
+
kv_heads: int = 8
|
11 |
+
dropout: float = 0.0
|
12 |
+
max_batch_size: int = 32
|
13 |
+
hidden_dim: int = None
|
14 |
+
multiple_of: int = 128
|
15 |
+
eps: float = 1e-6
|
16 |
+
flash: bool = True
|