Publish merged standalone ClassOne model
Browse files- config.json +4 -4
- model-00001-of-00003.safetensors +2 -2
- model-00002-of-00003.safetensors +2 -2
- model-00003-of-00003.safetensors +2 -2
- tokenizer_config.json +1 -1
config.json
CHANGED
|
@@ -12,7 +12,7 @@
|
|
| 12 |
"attention_logit_cap": 50.0,
|
| 13 |
"chunk_size_feed_forward": 0,
|
| 14 |
"conv_kernel_size": 5,
|
| 15 |
-
"dtype": "
|
| 16 |
"gradient_clipping": 10000000000.0,
|
| 17 |
"hidden_act": "silu",
|
| 18 |
"hidden_size": 1024,
|
|
@@ -45,7 +45,7 @@
|
|
| 45 |
"audio_token_id": 258881,
|
| 46 |
"boa_token_id": 256000,
|
| 47 |
"boi_token_id": 255999,
|
| 48 |
-
"dtype": "
|
| 49 |
"eoa_token_id": 258883,
|
| 50 |
"eoa_token_index": 258883,
|
| 51 |
"eoi_token_id": 258882,
|
|
@@ -61,7 +61,7 @@
|
|
| 61 |
"attention_dropout": 0.0,
|
| 62 |
"attention_k_eq_v": false,
|
| 63 |
"bos_token_id": 2,
|
| 64 |
-
"dtype": "
|
| 65 |
"enable_moe_block": false,
|
| 66 |
"eos_token_id": 1,
|
| 67 |
"expert_intermediate_size": null,
|
|
@@ -172,7 +172,7 @@
|
|
| 172 |
"attention_dropout": 0.0,
|
| 173 |
"chunk_size_feed_forward": 0,
|
| 174 |
"default_output_length": 280,
|
| 175 |
-
"dtype": "
|
| 176 |
"global_head_dim": 64,
|
| 177 |
"head_dim": 64,
|
| 178 |
"hidden_activation": "gelu_pytorch_tanh",
|
|
|
|
| 12 |
"attention_logit_cap": 50.0,
|
| 13 |
"chunk_size_feed_forward": 0,
|
| 14 |
"conv_kernel_size": 5,
|
| 15 |
+
"dtype": "float16",
|
| 16 |
"gradient_clipping": 10000000000.0,
|
| 17 |
"hidden_act": "silu",
|
| 18 |
"hidden_size": 1024,
|
|
|
|
| 45 |
"audio_token_id": 258881,
|
| 46 |
"boa_token_id": 256000,
|
| 47 |
"boi_token_id": 255999,
|
| 48 |
+
"dtype": "float16",
|
| 49 |
"eoa_token_id": 258883,
|
| 50 |
"eoa_token_index": 258883,
|
| 51 |
"eoi_token_id": 258882,
|
|
|
|
| 61 |
"attention_dropout": 0.0,
|
| 62 |
"attention_k_eq_v": false,
|
| 63 |
"bos_token_id": 2,
|
| 64 |
+
"dtype": "float16",
|
| 65 |
"enable_moe_block": false,
|
| 66 |
"eos_token_id": 1,
|
| 67 |
"expert_intermediate_size": null,
|
|
|
|
| 172 |
"attention_dropout": 0.0,
|
| 173 |
"chunk_size_feed_forward": 0,
|
| 174 |
"default_output_length": 280,
|
| 175 |
+
"dtype": "float16",
|
| 176 |
"global_head_dim": 64,
|
| 177 |
"head_dim": 64,
|
| 178 |
"hidden_activation": "gelu_pytorch_tanh",
|
model-00001-of-00003.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1e56c4841074e28f20a9752500ff423f4e3a87a4936676f4896c8938f7fadcfc
|
| 3 |
+
size 1422125504
|
model-00002-of-00003.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ff90fa92f761fcd10a2372f917046fa52ec35a8dcca1183ed29431ab5d2c1f25
|
| 3 |
+
size 4987104842
|
model-00003-of-00003.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:60e7887b46aca1c22059e422078d12efb433f0efd48622a6f96ece40314f39b7
|
| 3 |
+
size 3799607828
|
tokenizer_config.json
CHANGED
|
@@ -28,7 +28,7 @@
|
|
| 28 |
"<|level_end|>"
|
| 29 |
],
|
| 30 |
"image_token": "<|image|>",
|
| 31 |
-
"is_local":
|
| 32 |
"local_files_only": false,
|
| 33 |
"mask_token": "<mask>",
|
| 34 |
"model_max_length": 1000000000000000019884624838656,
|
|
|
|
| 28 |
"<|level_end|>"
|
| 29 |
],
|
| 30 |
"image_token": "<|image|>",
|
| 31 |
+
"is_local": false,
|
| 32 |
"local_files_only": false,
|
| 33 |
"mask_token": "<mask>",
|
| 34 |
"model_max_length": 1000000000000000019884624838656,
|