Upload step 1000 checkpoint
Browse files- config.json +1 -0
- model.safetensors +1 -1
- train_config.json +2 -0
config.json
CHANGED
|
@@ -108,6 +108,7 @@
|
|
| 108 |
"r8c8": "<row_8_col_8>"
|
| 109 |
},
|
| 110 |
"vlm_load_backbone_weights": true,
|
|
|
|
| 111 |
"use_kv_bridge": false,
|
| 112 |
"transport_mode": "full",
|
| 113 |
"kv_bridge_mode": "affine",
|
|
|
|
| 108 |
"r8c8": "<row_8_col_8>"
|
| 109 |
},
|
| 110 |
"vlm_load_backbone_weights": true,
|
| 111 |
+
"vlm_checkpoint_path": "lusxvr/nanoVLM-230M-8k",
|
| 112 |
"use_kv_bridge": false,
|
| 113 |
"transport_mode": "full",
|
| 114 |
"kv_bridge_mode": "affine",
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 725313232
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cc71f1f836a1a5ad641f8953bc8788a234a1f2876d50d827a1e4d40efc139517
|
| 3 |
size 725313232
|
train_config.json
CHANGED
|
@@ -8,6 +8,7 @@
|
|
| 8 |
"gradient_accumulation_steps": 4,
|
| 9 |
"max_grad_norm": 1.0,
|
| 10 |
"max_training_steps": 10000,
|
|
|
|
| 11 |
"warmup_ratio": 0.03,
|
| 12 |
"stats_log_interval": 100,
|
| 13 |
"precision": "bf16",
|
|
@@ -55,5 +56,6 @@
|
|
| 55 |
"checkpoint_repo_pattern": "patrickamadeus/dt-base-full-{i}",
|
| 56 |
"hf_private": false,
|
| 57 |
"push_final_model_to_hub": true,
|
|
|
|
| 58 |
"resume_checkpoint_path": null
|
| 59 |
}
|
|
|
|
| 8 |
"gradient_accumulation_steps": 4,
|
| 9 |
"max_grad_norm": 1.0,
|
| 10 |
"max_training_steps": 10000,
|
| 11 |
+
"stop_after_step": 3100,
|
| 12 |
"warmup_ratio": 0.03,
|
| 13 |
"stats_log_interval": 100,
|
| 14 |
"precision": "bf16",
|
|
|
|
| 56 |
"checkpoint_repo_pattern": "patrickamadeus/dt-base-full-{i}",
|
| 57 |
"hf_private": false,
|
| 58 |
"push_final_model_to_hub": true,
|
| 59 |
+
"resume_from_vlm_checkpoint": true,
|
| 60 |
"resume_checkpoint_path": null
|
| 61 |
}
|