Epoch 0 - Val loss 4.2204
Browse files- config.json +1 -0
- model.safetensors +1 -1
config.json
CHANGED
|
@@ -24,6 +24,7 @@
|
|
| 24 |
"num_shared_experts": 2,
|
| 25 |
"rope_base": 100000,
|
| 26 |
"router_amp": true,
|
|
|
|
| 27 |
"self_att_type": "sqa",
|
| 28 |
"seq_len": 8192,
|
| 29 |
"shared_expert_dim": 384,
|
|
|
|
| 24 |
"num_shared_experts": 2,
|
| 25 |
"rope_base": 100000,
|
| 26 |
"router_amp": true,
|
| 27 |
+
"router_dtype": "bfloat16",
|
| 28 |
"self_att_type": "sqa",
|
| 29 |
"seq_len": 8192,
|
| 30 |
"shared_expert_dim": 384,
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5772284720
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:633ba3bae331e961159eabf91381c4a9f302f8d796c3764a7cc9181357334a3c
|
| 3 |
size 5772284720
|