lc8k_135m_2026-10-24/armA_hw step 300
Browse files- lc8k_135m_2026-10-24/armA_hw/step_300/blumodelcp_run1.meta +5 -0
- lc8k_135m_2026-10-24/armA_hw/step_300/rank_0__blumodelcp_run1_step_300.ckpt +3 -0
- lc8k_135m_2026-10-24/armA_hw/step_300/rank_0__blumodelcp_run1_step_300.zero_g1 +0 -0
- lc8k_135m_2026-10-24/armA_hw/step_300/rank_1__blumodelcp_run1_step_300.ckpt +3 -0
- lc8k_135m_2026-10-24/armA_hw/step_300/rank_1__blumodelcp_run1_step_300.zero_g1 +0 -0
lc8k_135m_2026-10-24/armA_hw/step_300/blumodelcp_run1.meta
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
rope_theta 50000
|
| 2 |
+
d_model 512
|
| 3 |
+
q_heads 4
|
| 4 |
+
kv_heads 2
|
| 5 |
+
head_dim 128
|
lc8k_135m_2026-10-24/armA_hw/step_300/rank_0__blumodelcp_run1_step_300.ckpt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5f527e870cda165f3f854fdd3217108521b2127697d79d14af1b827f6e1ec0c7
|
| 3 |
+
size 807359444
|
lc8k_135m_2026-10-24/armA_hw/step_300/rank_0__blumodelcp_run1_step_300.zero_g1
ADDED
|
Binary file (61.1 kB). View file
|
|
|
lc8k_135m_2026-10-24/armA_hw/step_300/rank_1__blumodelcp_run1_step_300.ckpt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c4cb91174555a8645ff99793bf6cb193f0519da9aa725f849e41cfa234465e41
|
| 3 |
+
size 805568094
|
lc8k_135m_2026-10-24/armA_hw/step_300/rank_1__blumodelcp_run1_step_300.zero_g1
ADDED
|
Binary file (57.9 kB). View file
|
|
|