Upload 8 files
Browse files- adapter_config.json +118 -0
- adapter_model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- trainer_state.json +454 -0
- training_args.bin +3 -0
adapter_config.json
ADDED
|
@@ -0,0 +1,118 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alora_invocation_tokens": null,
|
| 3 |
+
"alpha_pattern": {},
|
| 4 |
+
"arrow_config": null,
|
| 5 |
+
"auto_mapping": null,
|
| 6 |
+
"base_model_name_or_path": "Molmo2-4B",
|
| 7 |
+
"bias": "none",
|
| 8 |
+
"corda_config": null,
|
| 9 |
+
"ensure_weight_tying": false,
|
| 10 |
+
"eva_config": null,
|
| 11 |
+
"exclude_modules": null,
|
| 12 |
+
"fan_in_fan_out": false,
|
| 13 |
+
"inference_mode": true,
|
| 14 |
+
"init_lora_weights": true,
|
| 15 |
+
"layer_replication": null,
|
| 16 |
+
"layers_pattern": null,
|
| 17 |
+
"layers_to_transform": null,
|
| 18 |
+
"loftq_config": {},
|
| 19 |
+
"lora_alpha": 32,
|
| 20 |
+
"lora_bias": false,
|
| 21 |
+
"lora_dropout": 0.05,
|
| 22 |
+
"megatron_config": null,
|
| 23 |
+
"megatron_core": "megatron.core",
|
| 24 |
+
"modules_to_save": null,
|
| 25 |
+
"peft_type": "LORA",
|
| 26 |
+
"peft_version": "0.18.0",
|
| 27 |
+
"qalora_group_size": 16,
|
| 28 |
+
"r": 16,
|
| 29 |
+
"rank_pattern": {},
|
| 30 |
+
"revision": null,
|
| 31 |
+
"target_modules": [
|
| 32 |
+
"transformer.blocks.29.att_proj",
|
| 33 |
+
"transformer.blocks.24.att_proj",
|
| 34 |
+
"transformer.blocks.15.att_proj",
|
| 35 |
+
"transformer.blocks.7.att_proj",
|
| 36 |
+
"transformer.blocks.0.attn_out",
|
| 37 |
+
"transformer.blocks.8.att_proj",
|
| 38 |
+
"vision_backbone.image_pooling_2d.wv",
|
| 39 |
+
"transformer.blocks.24.attn_out",
|
| 40 |
+
"transformer.blocks.35.att_proj",
|
| 41 |
+
"transformer.blocks.22.attn_out",
|
| 42 |
+
"transformer.blocks.34.attn_out",
|
| 43 |
+
"transformer.blocks.3.attn_out",
|
| 44 |
+
"transformer.blocks.26.att_proj",
|
| 45 |
+
"transformer.blocks.6.att_proj",
|
| 46 |
+
"vision_backbone.image_projector.w1",
|
| 47 |
+
"transformer.blocks.10.att_proj",
|
| 48 |
+
"transformer.blocks.19.attn_out",
|
| 49 |
+
"transformer.blocks.13.att_proj",
|
| 50 |
+
"vision_backbone.image_pooling_2d.wk",
|
| 51 |
+
"transformer.blocks.25.att_proj",
|
| 52 |
+
"vision_backbone.image_projector.w2",
|
| 53 |
+
"transformer.blocks.21.att_proj",
|
| 54 |
+
"transformer.blocks.8.attn_out",
|
| 55 |
+
"transformer.blocks.27.attn_out",
|
| 56 |
+
"transformer.blocks.6.attn_out",
|
| 57 |
+
"transformer.blocks.35.attn_out",
|
| 58 |
+
"vision_backbone.image_pooling_2d.wq",
|
| 59 |
+
"transformer.blocks.9.attn_out",
|
| 60 |
+
"transformer.blocks.12.att_proj",
|
| 61 |
+
"transformer.blocks.22.att_proj",
|
| 62 |
+
"vision_backbone.image_pooling_2d.wo",
|
| 63 |
+
"transformer.blocks.10.attn_out",
|
| 64 |
+
"transformer.blocks.31.att_proj",
|
| 65 |
+
"transformer.blocks.25.attn_out",
|
| 66 |
+
"transformer.blocks.5.attn_out",
|
| 67 |
+
"transformer.blocks.7.attn_out",
|
| 68 |
+
"transformer.blocks.16.att_proj",
|
| 69 |
+
"transformer.blocks.14.attn_out",
|
| 70 |
+
"transformer.blocks.29.attn_out",
|
| 71 |
+
"transformer.blocks.34.att_proj",
|
| 72 |
+
"transformer.blocks.17.attn_out",
|
| 73 |
+
"transformer.blocks.5.att_proj",
|
| 74 |
+
"transformer.blocks.1.att_proj",
|
| 75 |
+
"transformer.blocks.20.att_proj",
|
| 76 |
+
"transformer.blocks.33.att_proj",
|
| 77 |
+
"transformer.blocks.1.attn_out",
|
| 78 |
+
"transformer.blocks.3.att_proj",
|
| 79 |
+
"transformer.blocks.17.att_proj",
|
| 80 |
+
"transformer.blocks.26.attn_out",
|
| 81 |
+
"transformer.blocks.33.attn_out",
|
| 82 |
+
"transformer.blocks.14.att_proj",
|
| 83 |
+
"transformer.blocks.31.attn_out",
|
| 84 |
+
"transformer.blocks.30.att_proj",
|
| 85 |
+
"transformer.blocks.32.att_proj",
|
| 86 |
+
"transformer.blocks.11.att_proj",
|
| 87 |
+
"transformer.blocks.32.attn_out",
|
| 88 |
+
"transformer.blocks.16.attn_out",
|
| 89 |
+
"transformer.blocks.4.att_proj",
|
| 90 |
+
"transformer.blocks.4.attn_out",
|
| 91 |
+
"transformer.blocks.0.att_proj",
|
| 92 |
+
"transformer.blocks.23.attn_out",
|
| 93 |
+
"transformer.blocks.20.attn_out",
|
| 94 |
+
"transformer.blocks.27.att_proj",
|
| 95 |
+
"transformer.blocks.9.att_proj",
|
| 96 |
+
"transformer.blocks.28.att_proj",
|
| 97 |
+
"transformer.blocks.19.att_proj",
|
| 98 |
+
"transformer.blocks.18.attn_out",
|
| 99 |
+
"transformer.blocks.2.att_proj",
|
| 100 |
+
"transformer.blocks.30.attn_out",
|
| 101 |
+
"transformer.blocks.23.att_proj",
|
| 102 |
+
"transformer.blocks.28.attn_out",
|
| 103 |
+
"transformer.blocks.12.attn_out",
|
| 104 |
+
"transformer.blocks.2.attn_out",
|
| 105 |
+
"transformer.blocks.21.attn_out",
|
| 106 |
+
"transformer.blocks.18.att_proj",
|
| 107 |
+
"transformer.blocks.15.attn_out",
|
| 108 |
+
"vision_backbone.image_projector.w3",
|
| 109 |
+
"transformer.blocks.13.attn_out",
|
| 110 |
+
"transformer.blocks.11.attn_out"
|
| 111 |
+
],
|
| 112 |
+
"target_parameters": null,
|
| 113 |
+
"task_type": "CAUSAL_LM",
|
| 114 |
+
"trainable_token_indices": null,
|
| 115 |
+
"use_dora": false,
|
| 116 |
+
"use_qalora": false,
|
| 117 |
+
"use_rslora": false
|
| 118 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4a7eff1a7f925fbb629d54f0c8281fb3d62cae0df9f32a2ebfad873eb0659483
|
| 3 |
+
size 2992080
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1fee794f339f36ef1996c4e56f6502eba59cdf8c2070eef674041c7e3dfd7a23
|
| 3 |
+
size 1538037
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:363c5df1543d2c82b2f13164f35bdd0367ceb32e7fa1b2f67c19df073a08b17b
|
| 3 |
+
size 14645
|
scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:419e9121be762e4368fecb31eef4f8ce6f2ce4d52aa3adc3c6e5a569f9a1f099
|
| 3 |
+
size 1383
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:09de904d0ccc8857cebd09674e133c4f831bdbf6f060f9b5a76febbc1b782279
|
| 3 |
+
size 1465
|
trainer_state.json
ADDED
|
@@ -0,0 +1,454 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 8.956651718983558,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 3000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.14947683109118087,
|
| 14 |
+
"grad_norm": 0.3962731659412384,
|
| 15 |
+
"learning_rate": 1.9706176961602673e-05,
|
| 16 |
+
"loss": 1.902,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.29895366218236175,
|
| 21 |
+
"grad_norm": 0.3334282338619232,
|
| 22 |
+
"learning_rate": 1.9372287145242073e-05,
|
| 23 |
+
"loss": 1.8032,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.4484304932735426,
|
| 28 |
+
"grad_norm": 0.8322853446006775,
|
| 29 |
+
"learning_rate": 1.903839732888147e-05,
|
| 30 |
+
"loss": 1.6039,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.5979073243647235,
|
| 35 |
+
"grad_norm": 2.1256463527679443,
|
| 36 |
+
"learning_rate": 1.870450751252087e-05,
|
| 37 |
+
"loss": 1.354,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.7473841554559043,
|
| 42 |
+
"grad_norm": 3.58968186378479,
|
| 43 |
+
"learning_rate": 1.837061769616027e-05,
|
| 44 |
+
"loss": 1.0853,
|
| 45 |
+
"step": 250
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.8968609865470852,
|
| 49 |
+
"grad_norm": 11.132929801940918,
|
| 50 |
+
"learning_rate": 1.8036727879799668e-05,
|
| 51 |
+
"loss": 0.8252,
|
| 52 |
+
"step": 300
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 1.0448430493273542,
|
| 56 |
+
"grad_norm": 4.746887683868408,
|
| 57 |
+
"learning_rate": 1.7702838063439068e-05,
|
| 58 |
+
"loss": 0.6159,
|
| 59 |
+
"step": 350
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.194319880418535,
|
| 63 |
+
"grad_norm": 7.573383808135986,
|
| 64 |
+
"learning_rate": 1.7368948247078464e-05,
|
| 65 |
+
"loss": 0.452,
|
| 66 |
+
"step": 400
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.343796711509716,
|
| 70 |
+
"grad_norm": 4.613552093505859,
|
| 71 |
+
"learning_rate": 1.7035058430717864e-05,
|
| 72 |
+
"loss": 0.3191,
|
| 73 |
+
"step": 450
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.493273542600897,
|
| 77 |
+
"grad_norm": 18.23060417175293,
|
| 78 |
+
"learning_rate": 1.6701168614357263e-05,
|
| 79 |
+
"loss": 0.2341,
|
| 80 |
+
"step": 500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.6427503736920777,
|
| 84 |
+
"grad_norm": 4.62631368637085,
|
| 85 |
+
"learning_rate": 1.6367278797996663e-05,
|
| 86 |
+
"loss": 0.1614,
|
| 87 |
+
"step": 550
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 1.7922272047832586,
|
| 91 |
+
"grad_norm": 1.3041956424713135,
|
| 92 |
+
"learning_rate": 1.6033388981636063e-05,
|
| 93 |
+
"loss": 0.1584,
|
| 94 |
+
"step": 600
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 1.9417040358744395,
|
| 98 |
+
"grad_norm": 2.2017955780029297,
|
| 99 |
+
"learning_rate": 1.569949916527546e-05,
|
| 100 |
+
"loss": 0.1428,
|
| 101 |
+
"step": 650
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 2.0896860986547083,
|
| 105 |
+
"grad_norm": 1.256799578666687,
|
| 106 |
+
"learning_rate": 1.536560934891486e-05,
|
| 107 |
+
"loss": 0.1266,
|
| 108 |
+
"step": 700
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 2.2391629297458895,
|
| 112 |
+
"grad_norm": 4.344259262084961,
|
| 113 |
+
"learning_rate": 1.5031719532554258e-05,
|
| 114 |
+
"loss": 0.1112,
|
| 115 |
+
"step": 750
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 2.38863976083707,
|
| 119 |
+
"grad_norm": 2.4069414138793945,
|
| 120 |
+
"learning_rate": 1.4697829716193656e-05,
|
| 121 |
+
"loss": 0.1158,
|
| 122 |
+
"step": 800
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 2.538116591928251,
|
| 126 |
+
"grad_norm": 1.6278817653656006,
|
| 127 |
+
"learning_rate": 1.4363939899833058e-05,
|
| 128 |
+
"loss": 0.0949,
|
| 129 |
+
"step": 850
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 2.687593423019432,
|
| 133 |
+
"grad_norm": 2.6387939453125,
|
| 134 |
+
"learning_rate": 1.4030050083472456e-05,
|
| 135 |
+
"loss": 0.095,
|
| 136 |
+
"step": 900
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 2.8370702541106128,
|
| 140 |
+
"grad_norm": 3.1809768676757812,
|
| 141 |
+
"learning_rate": 1.3696160267111853e-05,
|
| 142 |
+
"loss": 0.0918,
|
| 143 |
+
"step": 950
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 2.986547085201794,
|
| 147 |
+
"grad_norm": 2.9219062328338623,
|
| 148 |
+
"learning_rate": 1.3362270450751253e-05,
|
| 149 |
+
"loss": 0.087,
|
| 150 |
+
"step": 1000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 3.1345291479820627,
|
| 154 |
+
"grad_norm": 3.087089776992798,
|
| 155 |
+
"learning_rate": 1.3028380634390651e-05,
|
| 156 |
+
"loss": 0.0792,
|
| 157 |
+
"step": 1050
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 3.2840059790732434,
|
| 161 |
+
"grad_norm": 1.1930086612701416,
|
| 162 |
+
"learning_rate": 1.2694490818030052e-05,
|
| 163 |
+
"loss": 0.0769,
|
| 164 |
+
"step": 1100
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 3.4334828101644246,
|
| 168 |
+
"grad_norm": 3.4931893348693848,
|
| 169 |
+
"learning_rate": 1.236060100166945e-05,
|
| 170 |
+
"loss": 0.0848,
|
| 171 |
+
"step": 1150
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 3.5829596412556053,
|
| 175 |
+
"grad_norm": 2.244807004928589,
|
| 176 |
+
"learning_rate": 1.2026711185308848e-05,
|
| 177 |
+
"loss": 0.0765,
|
| 178 |
+
"step": 1200
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 3.7324364723467864,
|
| 182 |
+
"grad_norm": 2.385439872741699,
|
| 183 |
+
"learning_rate": 1.1692821368948248e-05,
|
| 184 |
+
"loss": 0.0689,
|
| 185 |
+
"step": 1250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 3.881913303437967,
|
| 189 |
+
"grad_norm": 4.636480808258057,
|
| 190 |
+
"learning_rate": 1.1358931552587646e-05,
|
| 191 |
+
"loss": 0.063,
|
| 192 |
+
"step": 1300
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 4.029895366218236,
|
| 196 |
+
"grad_norm": 1.3066422939300537,
|
| 197 |
+
"learning_rate": 1.1031719532554258e-05,
|
| 198 |
+
"loss": 0.074,
|
| 199 |
+
"step": 1350
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 4.179372197309417,
|
| 203 |
+
"grad_norm": 1.9610888957977295,
|
| 204 |
+
"learning_rate": 1.0697829716193657e-05,
|
| 205 |
+
"loss": 0.0638,
|
| 206 |
+
"step": 1400
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 4.328849028400598,
|
| 210 |
+
"grad_norm": 2.659454822540283,
|
| 211 |
+
"learning_rate": 1.0363939899833055e-05,
|
| 212 |
+
"loss": 0.0591,
|
| 213 |
+
"step": 1450
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 4.478325859491779,
|
| 217 |
+
"grad_norm": 1.8156203031539917,
|
| 218 |
+
"learning_rate": 1.0030050083472455e-05,
|
| 219 |
+
"loss": 0.0548,
|
| 220 |
+
"step": 1500
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 4.62780269058296,
|
| 224 |
+
"grad_norm": 3.0744662284851074,
|
| 225 |
+
"learning_rate": 9.696160267111854e-06,
|
| 226 |
+
"loss": 0.0586,
|
| 227 |
+
"step": 1550
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 4.77727952167414,
|
| 231 |
+
"grad_norm": 1.5125874280929565,
|
| 232 |
+
"learning_rate": 9.362270450751252e-06,
|
| 233 |
+
"loss": 0.0643,
|
| 234 |
+
"step": 1600
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 4.926756352765321,
|
| 238 |
+
"grad_norm": 1.3466798067092896,
|
| 239 |
+
"learning_rate": 9.028380634390652e-06,
|
| 240 |
+
"loss": 0.0614,
|
| 241 |
+
"step": 1650
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 5.074738415545591,
|
| 245 |
+
"grad_norm": 2.761390447616577,
|
| 246 |
+
"learning_rate": 8.694490818030052e-06,
|
| 247 |
+
"loss": 0.0568,
|
| 248 |
+
"step": 1700
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 5.2242152466367715,
|
| 252 |
+
"grad_norm": 1.61431086063385,
|
| 253 |
+
"learning_rate": 8.36060100166945e-06,
|
| 254 |
+
"loss": 0.0505,
|
| 255 |
+
"step": 1750
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 5.373692077727952,
|
| 259 |
+
"grad_norm": 0.6185945272445679,
|
| 260 |
+
"learning_rate": 8.02671118530885e-06,
|
| 261 |
+
"loss": 0.0539,
|
| 262 |
+
"step": 1800
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 5.523168908819133,
|
| 266 |
+
"grad_norm": 1.7833508253097534,
|
| 267 |
+
"learning_rate": 7.692821368948247e-06,
|
| 268 |
+
"loss": 0.049,
|
| 269 |
+
"step": 1850
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 5.672645739910314,
|
| 273 |
+
"grad_norm": 2.104482412338257,
|
| 274 |
+
"learning_rate": 7.358931552587647e-06,
|
| 275 |
+
"loss": 0.0486,
|
| 276 |
+
"step": 1900
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 5.822122571001495,
|
| 280 |
+
"grad_norm": 2.144746780395508,
|
| 281 |
+
"learning_rate": 7.025041736227045e-06,
|
| 282 |
+
"loss": 0.0487,
|
| 283 |
+
"step": 1950
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 5.971599402092676,
|
| 287 |
+
"grad_norm": 0.9488668441772461,
|
| 288 |
+
"learning_rate": 6.6911519198664446e-06,
|
| 289 |
+
"loss": 0.0533,
|
| 290 |
+
"step": 2000
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 6.119581464872945,
|
| 294 |
+
"grad_norm": 1.840523600578308,
|
| 295 |
+
"learning_rate": 6.357262103505843e-06,
|
| 296 |
+
"loss": 0.0426,
|
| 297 |
+
"step": 2050
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 6.2690582959641254,
|
| 301 |
+
"grad_norm": 3.228400707244873,
|
| 302 |
+
"learning_rate": 6.023372287145243e-06,
|
| 303 |
+
"loss": 0.0519,
|
| 304 |
+
"step": 2100
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 6.418535127055306,
|
| 308 |
+
"grad_norm": 3.9569454193115234,
|
| 309 |
+
"learning_rate": 5.689482470784642e-06,
|
| 310 |
+
"loss": 0.0426,
|
| 311 |
+
"step": 2150
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 6.568011958146487,
|
| 315 |
+
"grad_norm": 1.098770260810852,
|
| 316 |
+
"learning_rate": 5.35559265442404e-06,
|
| 317 |
+
"loss": 0.0402,
|
| 318 |
+
"step": 2200
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 6.7174887892376685,
|
| 322 |
+
"grad_norm": 1.297438621520996,
|
| 323 |
+
"learning_rate": 5.0217028380634394e-06,
|
| 324 |
+
"loss": 0.0461,
|
| 325 |
+
"step": 2250
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 6.866965620328849,
|
| 329 |
+
"grad_norm": 0.9027300477027893,
|
| 330 |
+
"learning_rate": 4.687813021702838e-06,
|
| 331 |
+
"loss": 0.0464,
|
| 332 |
+
"step": 2300
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 7.014947683109118,
|
| 336 |
+
"grad_norm": 1.7319854497909546,
|
| 337 |
+
"learning_rate": 4.353923205342238e-06,
|
| 338 |
+
"loss": 0.0669,
|
| 339 |
+
"step": 2350
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 7.164424514200299,
|
| 343 |
+
"grad_norm": 1.1381218433380127,
|
| 344 |
+
"learning_rate": 4.020033388981636e-06,
|
| 345 |
+
"loss": 0.0403,
|
| 346 |
+
"step": 2400
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 7.31390134529148,
|
| 350 |
+
"grad_norm": 1.5444351434707642,
|
| 351 |
+
"learning_rate": 3.6861435726210355e-06,
|
| 352 |
+
"loss": 0.043,
|
| 353 |
+
"step": 2450
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 7.463378176382661,
|
| 357 |
+
"grad_norm": 2.456782817840576,
|
| 358 |
+
"learning_rate": 3.3589315525876463e-06,
|
| 359 |
+
"loss": 0.0469,
|
| 360 |
+
"step": 2500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 7.612855007473842,
|
| 364 |
+
"grad_norm": 1.1411938667297363,
|
| 365 |
+
"learning_rate": 3.025041736227045e-06,
|
| 366 |
+
"loss": 0.054,
|
| 367 |
+
"step": 2550
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 7.762331838565022,
|
| 371 |
+
"grad_norm": 1.1636459827423096,
|
| 372 |
+
"learning_rate": 2.6911519198664443e-06,
|
| 373 |
+
"loss": 0.0442,
|
| 374 |
+
"step": 2600
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 7.911808669656203,
|
| 378 |
+
"grad_norm": 1.4487160444259644,
|
| 379 |
+
"learning_rate": 2.357262103505843e-06,
|
| 380 |
+
"loss": 0.0415,
|
| 381 |
+
"step": 2650
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 8.059790732436472,
|
| 385 |
+
"grad_norm": 1.9948341846466064,
|
| 386 |
+
"learning_rate": 2.023372287145242e-06,
|
| 387 |
+
"loss": 0.0468,
|
| 388 |
+
"step": 2700
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 8.209267563527654,
|
| 392 |
+
"grad_norm": 1.8844599723815918,
|
| 393 |
+
"learning_rate": 1.6894824707846414e-06,
|
| 394 |
+
"loss": 0.0407,
|
| 395 |
+
"step": 2750
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 8.358744394618833,
|
| 399 |
+
"grad_norm": 1.19595205783844,
|
| 400 |
+
"learning_rate": 1.3555926544240402e-06,
|
| 401 |
+
"loss": 0.0531,
|
| 402 |
+
"step": 2800
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 8.508221225710015,
|
| 406 |
+
"grad_norm": 2.4211935997009277,
|
| 407 |
+
"learning_rate": 1.0217028380634392e-06,
|
| 408 |
+
"loss": 0.0421,
|
| 409 |
+
"step": 2850
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 8.657698056801197,
|
| 413 |
+
"grad_norm": 1.1242051124572754,
|
| 414 |
+
"learning_rate": 6.878130217028381e-07,
|
| 415 |
+
"loss": 0.0472,
|
| 416 |
+
"step": 2900
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 8.807174887892376,
|
| 420 |
+
"grad_norm": 2.1767444610595703,
|
| 421 |
+
"learning_rate": 3.5392320534223706e-07,
|
| 422 |
+
"loss": 0.042,
|
| 423 |
+
"step": 2950
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 8.956651718983558,
|
| 427 |
+
"grad_norm": 3.1764280796051025,
|
| 428 |
+
"learning_rate": 2.003338898163606e-08,
|
| 429 |
+
"loss": 0.0422,
|
| 430 |
+
"step": 3000
|
| 431 |
+
}
|
| 432 |
+
],
|
| 433 |
+
"logging_steps": 50,
|
| 434 |
+
"max_steps": 3000,
|
| 435 |
+
"num_input_tokens_seen": 0,
|
| 436 |
+
"num_train_epochs": 9,
|
| 437 |
+
"save_steps": 600,
|
| 438 |
+
"stateful_callbacks": {
|
| 439 |
+
"TrainerControl": {
|
| 440 |
+
"args": {
|
| 441 |
+
"should_epoch_stop": false,
|
| 442 |
+
"should_evaluate": false,
|
| 443 |
+
"should_log": false,
|
| 444 |
+
"should_save": true,
|
| 445 |
+
"should_training_stop": true
|
| 446 |
+
},
|
| 447 |
+
"attributes": {}
|
| 448 |
+
}
|
| 449 |
+
},
|
| 450 |
+
"total_flos": 4.497547090227081e+17,
|
| 451 |
+
"train_batch_size": 1,
|
| 452 |
+
"trial_name": null,
|
| 453 |
+
"trial_params": null
|
| 454 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0afed9a30bc0f9f2754235eb9897d853f9c98584b20363945fc053d2b9c1af1a
|
| 3 |
+
size 5905
|