pretrain update=10000 loss=3.3017
Browse files- README.md +9 -0
- config.json +56 -0
- model.safetensors +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +21 -0
README.md
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: other
|
| 3 |
+
library_name: pytorch
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
+
# TestGeniy 500M
|
| 7 |
+
|
| 8 |
+
1280d / 24L / GQA 10:2
|
| 9 |
+
3 local + 1 full NoPE, SwiGLU, QK-norm, tied emb, digit tokens.
|
config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"TestGeniyForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"model_type": "testgeniy",
|
| 6 |
+
"torch_dtype": "bfloat16",
|
| 7 |
+
"vocab_size": 49216,
|
| 8 |
+
"hidden_size": 1280,
|
| 9 |
+
"intermediate_size": 3584,
|
| 10 |
+
"num_hidden_layers": 24,
|
| 11 |
+
"num_attention_heads": 10,
|
| 12 |
+
"num_key_value_heads": 2,
|
| 13 |
+
"head_dim": 128,
|
| 14 |
+
"max_position_embeddings": 2048,
|
| 15 |
+
"rope_theta": 500000.0,
|
| 16 |
+
"local_block_size": 1024,
|
| 17 |
+
"tie_word_embeddings": true,
|
| 18 |
+
"layer_types": [
|
| 19 |
+
"block_local",
|
| 20 |
+
"block_local",
|
| 21 |
+
"block_local",
|
| 22 |
+
"full",
|
| 23 |
+
"block_local",
|
| 24 |
+
"block_local",
|
| 25 |
+
"block_local",
|
| 26 |
+
"full",
|
| 27 |
+
"block_local",
|
| 28 |
+
"block_local",
|
| 29 |
+
"block_local",
|
| 30 |
+
"full",
|
| 31 |
+
"block_local",
|
| 32 |
+
"block_local",
|
| 33 |
+
"block_local",
|
| 34 |
+
"full",
|
| 35 |
+
"block_local",
|
| 36 |
+
"block_local",
|
| 37 |
+
"block_local",
|
| 38 |
+
"full",
|
| 39 |
+
"block_local",
|
| 40 |
+
"block_local",
|
| 41 |
+
"block_local",
|
| 42 |
+
"full"
|
| 43 |
+
],
|
| 44 |
+
"digit_tokens": [
|
| 45 |
+
"<d0>",
|
| 46 |
+
"<d1>",
|
| 47 |
+
"<d2>",
|
| 48 |
+
"<d3>",
|
| 49 |
+
"<d4>",
|
| 50 |
+
"<d5>",
|
| 51 |
+
"<d6>",
|
| 52 |
+
"<d7>",
|
| 53 |
+
"<d8>",
|
| 54 |
+
"<d9>"
|
| 55 |
+
]
|
| 56 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:93086b7e0bd33b4861e41ecab2494f4986365d49e3fb8f699f324e782fcc5e11
|
| 3 |
+
size 1101624424
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<|begin|>",
|
| 5 |
+
"clean_up_tokenization_spaces": false,
|
| 6 |
+
"eos_token": "<|end|>",
|
| 7 |
+
"errors": "replace",
|
| 8 |
+
"extra_special_tokens": [
|
| 9 |
+
"<|doc|>",
|
| 10 |
+
"<|system|>",
|
| 11 |
+
"<|user|>",
|
| 12 |
+
"<|assistant|>"
|
| 13 |
+
],
|
| 14 |
+
"is_local": true,
|
| 15 |
+
"local_files_only": false,
|
| 16 |
+
"model_max_length": 1000000000000,
|
| 17 |
+
"pad_token": "<|pad|>",
|
| 18 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 19 |
+
"unk_token": "<|endoftext|>",
|
| 20 |
+
"vocab_size": 49152
|
| 21 |
+
}
|