Add model checkpoint
Browse files- config.json +3 -8
- model.safetensors +1 -1
config.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "/home/rllaskowski_a100/mot-training/experiments/mot_medium_32_8/checkpoint-
|
3 |
"activation_function": "gelu_new",
|
4 |
"architectures": [
|
5 |
"MoTModel"
|
@@ -18,20 +18,15 @@
|
|
18 |
"n_embd": 512,
|
19 |
"n_expert": 256,
|
20 |
"n_head": 8,
|
21 |
-
"n_inner":
|
22 |
"n_layer": 8,
|
23 |
"n_positions": 1024,
|
24 |
"reorder_and_upcast_attn": false,
|
25 |
"resid_pdrop": 0.1,
|
26 |
"scale_attn_by_inverse_layer_idx": false,
|
27 |
"scale_attn_weights": true,
|
28 |
-
"summary_activation": null,
|
29 |
-
"summary_first_dropout": 0.1,
|
30 |
-
"summary_proj_to_labels": true,
|
31 |
-
"summary_type": "cls_index",
|
32 |
-
"summary_use_proj": true,
|
33 |
"torch_dtype": "float32",
|
34 |
-
"transformers_version": "4.
|
35 |
"use_cache": true,
|
36 |
"use_discrete_routing": false,
|
37 |
"vocab_size": 50257
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "/home/rllaskowski_a100/mot-training/experiments/mot_medium_32_8/checkpoint-80000",
|
3 |
"activation_function": "gelu_new",
|
4 |
"architectures": [
|
5 |
"MoTModel"
|
|
|
18 |
"n_embd": 512,
|
19 |
"n_expert": 256,
|
20 |
"n_head": 8,
|
21 |
+
"n_inner": 65536,
|
22 |
"n_layer": 8,
|
23 |
"n_positions": 1024,
|
24 |
"reorder_and_upcast_attn": false,
|
25 |
"resid_pdrop": 0.1,
|
26 |
"scale_attn_by_inverse_layer_idx": false,
|
27 |
"scale_attn_weights": true,
|
|
|
|
|
|
|
|
|
|
|
28 |
"torch_dtype": "float32",
|
29 |
+
"transformers_version": "4.42.0.dev0",
|
30 |
"use_cache": true,
|
31 |
"use_discrete_routing": false,
|
32 |
"vocab_size": 50257
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 2290399320
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:16f5f35057f07a1f48e60e72ac84ffa132022ca1d9f53934d817a821682e2f93
|
3 |
size 2290399320
|