Method
__init__
(
self,
vocab_size=160256,
hidden_size=4096,
n_layer=30,
n_head=32,
layer_norm_epsilon=1e-5,
initializer_range=0.02,
use_cache=True,
bos_token_id=1,
eos_token_id=2,
apply_residual_connection_post_layernorm=False,
hidden_dropout=0.0,
attention_dropout=0.0,
ffn_hidden_size=12288,
training_seqlen = 8192,
logn = True,
embed_layernorm = False,
**kwargs,
)
Source from the content-addressed store, hash-verified
| 48 | } |
| 49 | |
| 50 | def __init__( |
| 51 | self, |
| 52 | vocab_size=160256, |
| 53 | hidden_size=4096, |
| 54 | n_layer=30, |
| 55 | n_head=32, |
| 56 | layer_norm_epsilon=1e-5, |
| 57 | initializer_range=0.02, |
| 58 | use_cache=True, |
| 59 | bos_token_id=1, |
| 60 | eos_token_id=2, |
| 61 | apply_residual_connection_post_layernorm=False, |
| 62 | hidden_dropout=0.0, |
| 63 | attention_dropout=0.0, |
| 64 | ffn_hidden_size=12288, |
| 65 | training_seqlen = 8192, |
| 66 | logn = True, |
| 67 | embed_layernorm = False, |
| 68 | **kwargs, |
| 69 | ): |
| 70 | self.vocab_size = vocab_size |
| 71 | n_embed = kwargs.pop("n_embed", None) |
| 72 | self.hidden_size = hidden_size if n_embed is None else n_embed |
| 73 | self.n_layer = n_layer |
| 74 | self.n_head = n_head |
| 75 | self.layer_norm_epsilon = layer_norm_epsilon |
| 76 | self.initializer_range = initializer_range |
| 77 | self.use_cache = use_cache |
| 78 | self.apply_residual_connection_post_layernorm = apply_residual_connection_post_layernorm |
| 79 | self.hidden_dropout = hidden_dropout |
| 80 | self.attention_dropout = attention_dropout |
| 81 | self.bos_token_id = bos_token_id |
| 82 | self.eos_token_id = eos_token_id |
| 83 | self.logn = logn |
| 84 | self.ffn_hidden_size = ffn_hidden_size |
| 85 | self.training_seqlen = training_seqlen |
| 86 | self.embed_layernorm = embed_layernorm |
| 87 | |
| 88 | |
| 89 | super().__init__(bos_token_id=bos_token_id, eos_token_id=eos_token_id, **kwargs) |
| 90 | |
Callers
nothing calls this directly
Tested by
no test coverage detected