Upload GptOssForCausalLM
Browse files- config.json +3 -2
- generation_config.json +1 -1
- model.safetensors +1 -1
config.json
CHANGED
|
@@ -4,6 +4,7 @@
|
|
| 4 |
],
|
| 5 |
"attention_bias": true,
|
| 6 |
"attention_dropout": 0.0,
|
|
|
|
| 7 |
"head_dim": 64,
|
| 8 |
"hidden_act": "silu",
|
| 9 |
"hidden_size": 8,
|
|
@@ -27,6 +28,7 @@
|
|
| 27 |
"beta_fast": 32.0,
|
| 28 |
"beta_slow": 1.0,
|
| 29 |
"factor": 32.0,
|
|
|
|
| 30 |
"rope_type": "yarn",
|
| 31 |
"truncate": false
|
| 32 |
},
|
|
@@ -34,8 +36,7 @@
|
|
| 34 |
"router_aux_loss_coef": 0.9,
|
| 35 |
"sliding_window": 128,
|
| 36 |
"tie_word_embeddings": false,
|
| 37 |
-
"
|
| 38 |
-
"transformers_version": "4.55.0.dev0",
|
| 39 |
"use_cache": true,
|
| 40 |
"vocab_size": 200019
|
| 41 |
}
|
|
|
|
| 4 |
],
|
| 5 |
"attention_bias": true,
|
| 6 |
"attention_dropout": 0.0,
|
| 7 |
+
"dtype": "float32",
|
| 8 |
"head_dim": 64,
|
| 9 |
"hidden_act": "silu",
|
| 10 |
"hidden_size": 8,
|
|
|
|
| 28 |
"beta_fast": 32.0,
|
| 29 |
"beta_slow": 1.0,
|
| 30 |
"factor": 32.0,
|
| 31 |
+
"original_max_position_embeddings": 4096,
|
| 32 |
"rope_type": "yarn",
|
| 33 |
"truncate": false
|
| 34 |
},
|
|
|
|
| 36 |
"router_aux_loss_coef": 0.9,
|
| 37 |
"sliding_window": 128,
|
| 38 |
"tie_word_embeddings": false,
|
| 39 |
+
"transformers_version": "4.57.0.dev0",
|
|
|
|
| 40 |
"use_cache": true,
|
| 41 |
"vocab_size": 200019
|
| 42 |
}
|
generation_config.json
CHANGED
|
@@ -1,4 +1,4 @@
|
|
| 1 |
{
|
| 2 |
"_from_model_config": true,
|
| 3 |
-
"transformers_version": "4.
|
| 4 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"_from_model_config": true,
|
| 3 |
+
"transformers_version": "4.57.0.dev0"
|
| 4 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 13728008
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:54fa84680fdca5d093b167896006d65ba8ce30da64db387f0f1a06d10849878b
|
| 3 |
size 13728008
|