Upload folder using huggingface_hub
Browse files- config.json +36 -0
- generation_config.json +5 -0
- pytorch_model.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_use_flash_attention_2": true,
|
| 3 |
+
"add_linear_bias": false,
|
| 4 |
+
"apply_rotary_positional_embedding_kernel": false,
|
| 5 |
+
"architectures": [
|
| 6 |
+
"StFoundation2DForMaskedLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_probs_dropout_prob": 0.0,
|
| 9 |
+
"bin_alpha": 1.0,
|
| 10 |
+
"bin_num": 100,
|
| 11 |
+
"dropout": 0.0,
|
| 12 |
+
"hidden_act": "swiglu",
|
| 13 |
+
"hidden_dropout_prob": 0.0,
|
| 14 |
+
"hidden_size": 128,
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"intermediate_size": 320,
|
| 17 |
+
"layer_norm_eps": 1e-05,
|
| 18 |
+
"mask_token_id": -1,
|
| 19 |
+
"max_position_embeddings": 19264,
|
| 20 |
+
"model_type": "stfoundation2D",
|
| 21 |
+
"normalization_type": "LayerNorm",
|
| 22 |
+
"num_attention_heads": 2,
|
| 23 |
+
"num_hidden_layers": 4,
|
| 24 |
+
"pad_token_id": -2,
|
| 25 |
+
"position_embedding_type": "absolute",
|
| 26 |
+
"rotary_embedding_2d": true,
|
| 27 |
+
"rotary_percent": 1.0,
|
| 28 |
+
"self_attn_type": "core",
|
| 29 |
+
"seq_len_interpolation_factor": null,
|
| 30 |
+
"tie_word_embeddings": false,
|
| 31 |
+
"torch_dtype": "float32",
|
| 32 |
+
"transformers_version": "4.38.0.dev0",
|
| 33 |
+
"type_vocab_size": 2,
|
| 34 |
+
"use_cache": true,
|
| 35 |
+
"vocab_size": 16
|
| 36 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"pad_token_id": -2,
|
| 4 |
+
"transformers_version": "4.38.0.dev0"
|
| 5 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5a3bc3760fb5c453fadae6cd3e79e7b7ea42c676ff10c1c024495a38f792d033
|
| 3 |
+
size 13070510
|