Commit
•
cbef7e0
1
Parent(s):
a1c4796
Upload folder using huggingface_hub
Browse files- README.md +9 -0
- adapter_config.json +26 -0
- adapter_model.bin +3 -0
- config.json +45 -0
- non_lora_trainables.bin +3 -0
- trainer_state.json +493 -0
README.md
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
### Framework versions
|
| 7 |
+
|
| 8 |
+
|
| 9 |
+
- PEFT 0.4.0
|
adapter_config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "./checkpoints/llava-v1.5-13b",
|
| 4 |
+
"bias": "none",
|
| 5 |
+
"fan_in_fan_out": false,
|
| 6 |
+
"inference_mode": true,
|
| 7 |
+
"init_lora_weights": true,
|
| 8 |
+
"layers_pattern": null,
|
| 9 |
+
"layers_to_transform": null,
|
| 10 |
+
"lora_alpha": 256,
|
| 11 |
+
"lora_dropout": 0.05,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 128,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"up_proj",
|
| 18 |
+
"q_proj",
|
| 19 |
+
"o_proj",
|
| 20 |
+
"down_proj",
|
| 21 |
+
"v_proj",
|
| 22 |
+
"k_proj",
|
| 23 |
+
"gate_proj"
|
| 24 |
+
],
|
| 25 |
+
"task_type": "CAUSAL_LM"
|
| 26 |
+
}
|
adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:578fd1ea49f26e2c27b0070f46a7f2b67d623e5580b8581b8ad1d9d1c3d3005f
|
| 3 |
+
size 1001584333
|
config.json
ADDED
|
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "./checkpoints/llava-v1.5-13b",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlavaLlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"bos_token_id": 1,
|
| 7 |
+
"eos_token_id": 2,
|
| 8 |
+
"freeze_mm_mlp_adapter": false,
|
| 9 |
+
"freeze_mm_vision_resampler": false,
|
| 10 |
+
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 5120,
|
| 12 |
+
"image_aspect_ratio": "pad",
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 13824,
|
| 15 |
+
"max_length": 4096,
|
| 16 |
+
"max_position_embeddings": 4096,
|
| 17 |
+
"mm_hidden_size": 1024,
|
| 18 |
+
"mm_projector_lr": 2e-05,
|
| 19 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 20 |
+
"mm_resampler_type": null,
|
| 21 |
+
"mm_use_im_patch_token": false,
|
| 22 |
+
"mm_use_im_start_end": false,
|
| 23 |
+
"mm_vision_select_feature": "patch",
|
| 24 |
+
"mm_vision_select_layer": -2,
|
| 25 |
+
"mm_vision_tower": "openai/clip-vit-large-patch14-336",
|
| 26 |
+
"model_type": "llava",
|
| 27 |
+
"num_attention_heads": 40,
|
| 28 |
+
"num_hidden_layers": 40,
|
| 29 |
+
"num_key_value_heads": 40,
|
| 30 |
+
"pad_token_id": 0,
|
| 31 |
+
"pretraining_tp": 1,
|
| 32 |
+
"rms_norm_eps": 1e-05,
|
| 33 |
+
"rope_scaling": null,
|
| 34 |
+
"tie_word_embeddings": false,
|
| 35 |
+
"tokenizer_model_max_length": 2048,
|
| 36 |
+
"tokenizer_padding_side": "right",
|
| 37 |
+
"torch_dtype": "float16",
|
| 38 |
+
"transformers_version": "4.31.0",
|
| 39 |
+
"tune_mm_mlp_adapter": false,
|
| 40 |
+
"tune_mm_vision_resampler": false,
|
| 41 |
+
"unfreeze_mm_vision_tower": false,
|
| 42 |
+
"use_cache": true,
|
| 43 |
+
"use_mm_proj": true,
|
| 44 |
+
"vocab_size": 32000
|
| 45 |
+
}
|
non_lora_trainables.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c2d67d2a1da533bf002c4744e434d52e8f72174af2c4144f9549ba3807899b67
|
| 3 |
+
size 62936807
|
trainer_state.json
ADDED
|
@@ -0,0 +1,493 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.9433962264150946,
|
| 5 |
+
"global_step": 78,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.04,
|
| 12 |
+
"learning_rate": 0.00013333333333333334,
|
| 13 |
+
"loss": 1.5444,
|
| 14 |
+
"step": 1
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.08,
|
| 18 |
+
"learning_rate": 0.0002666666666666667,
|
| 19 |
+
"loss": 1.543,
|
| 20 |
+
"step": 2
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.11,
|
| 24 |
+
"learning_rate": 0.0004,
|
| 25 |
+
"loss": 1.3151,
|
| 26 |
+
"step": 3
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.15,
|
| 30 |
+
"learning_rate": 0.0003998245660197717,
|
| 31 |
+
"loss": 1.2922,
|
| 32 |
+
"step": 4
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.19,
|
| 36 |
+
"learning_rate": 0.0003992985718499009,
|
| 37 |
+
"loss": 1.1978,
|
| 38 |
+
"step": 5
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.23,
|
| 42 |
+
"learning_rate": 0.0003984229402628956,
|
| 43 |
+
"loss": 1.1409,
|
| 44 |
+
"step": 6
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.26,
|
| 48 |
+
"learning_rate": 0.000397199207414101,
|
| 49 |
+
"loss": 1.1181,
|
| 50 |
+
"step": 7
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.3,
|
| 54 |
+
"learning_rate": 0.00039562952014676116,
|
| 55 |
+
"loss": 1.1124,
|
| 56 |
+
"step": 8
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.34,
|
| 60 |
+
"learning_rate": 0.0003937166322257262,
|
| 61 |
+
"loss": 1.0952,
|
| 62 |
+
"step": 9
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.38,
|
| 66 |
+
"learning_rate": 0.00039146389950641346,
|
| 67 |
+
"loss": 1.0885,
|
| 68 |
+
"step": 10
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.42,
|
| 72 |
+
"learning_rate": 0.00038887527404749624,
|
| 73 |
+
"loss": 1.0787,
|
| 74 |
+
"step": 11
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.45,
|
| 78 |
+
"learning_rate": 0.00038595529717765027,
|
| 79 |
+
"loss": 1.0854,
|
| 80 |
+
"step": 12
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.49,
|
| 84 |
+
"learning_rate": 0.0003827090915285202,
|
| 85 |
+
"loss": 1.0807,
|
| 86 |
+
"step": 13
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.53,
|
| 90 |
+
"learning_rate": 0.0003791423520478826,
|
| 91 |
+
"loss": 1.055,
|
| 92 |
+
"step": 14
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.57,
|
| 96 |
+
"learning_rate": 0.0003752613360087727,
|
| 97 |
+
"loss": 1.056,
|
| 98 |
+
"step": 15
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.6,
|
| 102 |
+
"learning_rate": 0.00037107285203210135,
|
| 103 |
+
"loss": 1.0526,
|
| 104 |
+
"step": 16
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.64,
|
| 108 |
+
"learning_rate": 0.0003665842481420199,
|
| 109 |
+
"loss": 1.0693,
|
| 110 |
+
"step": 17
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.68,
|
| 114 |
+
"learning_rate": 0.0003618033988749895,
|
| 115 |
+
"loss": 1.0471,
|
| 116 |
+
"step": 18
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.72,
|
| 120 |
+
"learning_rate": 0.000356738691465168,
|
| 121 |
+
"loss": 1.0356,
|
| 122 |
+
"step": 19
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.75,
|
| 126 |
+
"learning_rate": 0.0003513990111303513,
|
| 127 |
+
"loss": 1.0433,
|
| 128 |
+
"step": 20
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.79,
|
| 132 |
+
"learning_rate": 0.00034579372548428235,
|
| 133 |
+
"loss": 1.0463,
|
| 134 |
+
"step": 21
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.83,
|
| 138 |
+
"learning_rate": 0.0003399326681026731,
|
| 139 |
+
"loss": 1.0564,
|
| 140 |
+
"step": 22
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.87,
|
| 144 |
+
"learning_rate": 0.00033382612127177166,
|
| 145 |
+
"loss": 1.0391,
|
| 146 |
+
"step": 23
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.91,
|
| 150 |
+
"learning_rate": 0.000327484797949738,
|
| 151 |
+
"loss": 1.035,
|
| 152 |
+
"step": 24
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.94,
|
| 156 |
+
"learning_rate": 0.000320919822972475,
|
| 157 |
+
"loss": 1.0371,
|
| 158 |
+
"step": 25
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.98,
|
| 162 |
+
"learning_rate": 0.0003141427135368864,
|
| 163 |
+
"loss": 1.0296,
|
| 164 |
+
"step": 26
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.02,
|
| 168 |
+
"learning_rate": 0.0003071653589957993,
|
| 169 |
+
"loss": 0.93,
|
| 170 |
+
"step": 27
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 1.06,
|
| 174 |
+
"learning_rate": 0.00030000000000000003,
|
| 175 |
+
"loss": 0.9064,
|
| 176 |
+
"step": 28
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.09,
|
| 180 |
+
"learning_rate": 0.00029265920702397236,
|
| 181 |
+
"loss": 0.904,
|
| 182 |
+
"step": 29
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 1.13,
|
| 186 |
+
"learning_rate": 0.00028515585831301456,
|
| 187 |
+
"loss": 0.893,
|
| 188 |
+
"step": 30
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 1.17,
|
| 192 |
+
"learning_rate": 0.0002775031172904206,
|
| 193 |
+
"loss": 0.8945,
|
| 194 |
+
"step": 31
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 1.21,
|
| 198 |
+
"learning_rate": 0.0002697144094643631,
|
| 199 |
+
"loss": 0.8945,
|
| 200 |
+
"step": 32
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 1.25,
|
| 204 |
+
"learning_rate": 0.00026180339887498953,
|
| 205 |
+
"loss": 0.8792,
|
| 206 |
+
"step": 33
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.28,
|
| 210 |
+
"learning_rate": 0.00025378396412305314,
|
| 211 |
+
"loss": 0.8908,
|
| 212 |
+
"step": 34
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 1.32,
|
| 216 |
+
"learning_rate": 0.00024567017402213115,
|
| 217 |
+
"loss": 0.8892,
|
| 218 |
+
"step": 35
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 1.36,
|
| 222 |
+
"learning_rate": 0.00023747626291714498,
|
| 223 |
+
"loss": 0.8754,
|
| 224 |
+
"step": 36
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 1.4,
|
| 228 |
+
"learning_rate": 0.00022921660571248236,
|
| 229 |
+
"loss": 0.8414,
|
| 230 |
+
"step": 37
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 1.43,
|
| 234 |
+
"learning_rate": 0.00022090569265353072,
|
| 235 |
+
"loss": 0.8629,
|
| 236 |
+
"step": 38
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 1.47,
|
| 240 |
+
"learning_rate": 0.0002125581039058627,
|
| 241 |
+
"loss": 0.8653,
|
| 242 |
+
"step": 39
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 1.51,
|
| 246 |
+
"learning_rate": 0.0002041884839766714,
|
| 247 |
+
"loss": 0.8702,
|
| 248 |
+
"step": 40
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.55,
|
| 252 |
+
"learning_rate": 0.00019581151602332864,
|
| 253 |
+
"loss": 0.8799,
|
| 254 |
+
"step": 41
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 1.58,
|
| 258 |
+
"learning_rate": 0.00018744189609413734,
|
| 259 |
+
"loss": 0.8628,
|
| 260 |
+
"step": 42
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 1.62,
|
| 264 |
+
"learning_rate": 0.00017909430734646935,
|
| 265 |
+
"loss": 0.8751,
|
| 266 |
+
"step": 43
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 1.66,
|
| 270 |
+
"learning_rate": 0.0001707833942875177,
|
| 271 |
+
"loss": 0.8683,
|
| 272 |
+
"step": 44
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 1.7,
|
| 276 |
+
"learning_rate": 0.00016252373708285504,
|
| 277 |
+
"loss": 0.8522,
|
| 278 |
+
"step": 45
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 1.74,
|
| 282 |
+
"learning_rate": 0.00015432982597786884,
|
| 283 |
+
"loss": 0.8431,
|
| 284 |
+
"step": 46
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 1.77,
|
| 288 |
+
"learning_rate": 0.00014621603587694688,
|
| 289 |
+
"loss": 0.8575,
|
| 290 |
+
"step": 47
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.81,
|
| 294 |
+
"learning_rate": 0.00013819660112501054,
|
| 295 |
+
"loss": 0.8612,
|
| 296 |
+
"step": 48
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 1.85,
|
| 300 |
+
"learning_rate": 0.000130285590535637,
|
| 301 |
+
"loss": 0.8791,
|
| 302 |
+
"step": 49
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 1.89,
|
| 306 |
+
"learning_rate": 0.0001224968827095794,
|
| 307 |
+
"loss": 0.8104,
|
| 308 |
+
"step": 50
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 1.92,
|
| 312 |
+
"learning_rate": 0.00011484414168698547,
|
| 313 |
+
"loss": 0.8223,
|
| 314 |
+
"step": 51
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 1.96,
|
| 318 |
+
"learning_rate": 0.00010734079297602771,
|
| 319 |
+
"loss": 0.8642,
|
| 320 |
+
"step": 52
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 2.0,
|
| 324 |
+
"learning_rate": 0.00010000000000000005,
|
| 325 |
+
"loss": 0.7986,
|
| 326 |
+
"step": 53
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 2.04,
|
| 330 |
+
"learning_rate": 9.283464100420063e-05,
|
| 331 |
+
"loss": 0.7106,
|
| 332 |
+
"step": 54
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 2.08,
|
| 336 |
+
"learning_rate": 8.585728646311369e-05,
|
| 337 |
+
"loss": 0.7073,
|
| 338 |
+
"step": 55
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 2.11,
|
| 342 |
+
"learning_rate": 7.908017702752504e-05,
|
| 343 |
+
"loss": 0.7,
|
| 344 |
+
"step": 56
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 2.15,
|
| 348 |
+
"learning_rate": 7.251520205026205e-05,
|
| 349 |
+
"loss": 0.7022,
|
| 350 |
+
"step": 57
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 2.19,
|
| 354 |
+
"learning_rate": 6.617387872822842e-05,
|
| 355 |
+
"loss": 0.6812,
|
| 356 |
+
"step": 58
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 2.23,
|
| 360 |
+
"learning_rate": 6.0067331897326895e-05,
|
| 361 |
+
"loss": 0.6941,
|
| 362 |
+
"step": 59
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 2.26,
|
| 366 |
+
"learning_rate": 5.4206274515717736e-05,
|
| 367 |
+
"loss": 0.6749,
|
| 368 |
+
"step": 60
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 2.3,
|
| 372 |
+
"learning_rate": 4.8600988869648745e-05,
|
| 373 |
+
"loss": 0.6833,
|
| 374 |
+
"step": 61
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 2.34,
|
| 378 |
+
"learning_rate": 4.326130853483206e-05,
|
| 379 |
+
"loss": 0.6684,
|
| 380 |
+
"step": 62
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 2.38,
|
| 384 |
+
"learning_rate": 3.819660112501053e-05,
|
| 385 |
+
"loss": 0.6482,
|
| 386 |
+
"step": 63
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 2.42,
|
| 390 |
+
"learning_rate": 3.3415751857980115e-05,
|
| 391 |
+
"loss": 0.6638,
|
| 392 |
+
"step": 64
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 2.45,
|
| 396 |
+
"learning_rate": 2.8927147967898682e-05,
|
| 397 |
+
"loss": 0.6586,
|
| 398 |
+
"step": 65
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 2.49,
|
| 402 |
+
"learning_rate": 2.473866399122733e-05,
|
| 403 |
+
"loss": 0.6709,
|
| 404 |
+
"step": 66
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 2.53,
|
| 408 |
+
"learning_rate": 2.085764795211742e-05,
|
| 409 |
+
"loss": 0.6518,
|
| 410 |
+
"step": 67
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 2.57,
|
| 414 |
+
"learning_rate": 1.7290908471479805e-05,
|
| 415 |
+
"loss": 0.6567,
|
| 416 |
+
"step": 68
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 2.6,
|
| 420 |
+
"learning_rate": 1.4044702822349731e-05,
|
| 421 |
+
"loss": 0.678,
|
| 422 |
+
"step": 69
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 2.64,
|
| 426 |
+
"learning_rate": 1.1124725952503801e-05,
|
| 427 |
+
"loss": 0.6848,
|
| 428 |
+
"step": 70
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 2.68,
|
| 432 |
+
"learning_rate": 8.536100493586552e-06,
|
| 433 |
+
"loss": 0.6601,
|
| 434 |
+
"step": 71
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 2.72,
|
| 438 |
+
"learning_rate": 6.283367774273785e-06,
|
| 439 |
+
"loss": 0.6749,
|
| 440 |
+
"step": 72
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 2.75,
|
| 444 |
+
"learning_rate": 4.370479853238863e-06,
|
| 445 |
+
"loss": 0.6864,
|
| 446 |
+
"step": 73
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 2.79,
|
| 450 |
+
"learning_rate": 2.800792585899026e-06,
|
| 451 |
+
"loss": 0.6621,
|
| 452 |
+
"step": 74
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 2.83,
|
| 456 |
+
"learning_rate": 1.577059737104447e-06,
|
| 457 |
+
"loss": 0.6514,
|
| 458 |
+
"step": 75
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 2.87,
|
| 462 |
+
"learning_rate": 7.01428150099126e-07,
|
| 463 |
+
"loss": 0.6516,
|
| 464 |
+
"step": 76
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 2.91,
|
| 468 |
+
"learning_rate": 1.7543398022832336e-07,
|
| 469 |
+
"loss": 0.6504,
|
| 470 |
+
"step": 77
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 2.94,
|
| 474 |
+
"learning_rate": 0.0,
|
| 475 |
+
"loss": 0.6665,
|
| 476 |
+
"step": 78
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 2.94,
|
| 480 |
+
"step": 78,
|
| 481 |
+
"total_flos": 183654252937216.0,
|
| 482 |
+
"train_loss": 0.8923629743930621,
|
| 483 |
+
"train_runtime": 2340.4996,
|
| 484 |
+
"train_samples_per_second": 4.275,
|
| 485 |
+
"train_steps_per_second": 0.033
|
| 486 |
+
}
|
| 487 |
+
],
|
| 488 |
+
"max_steps": 78,
|
| 489 |
+
"num_train_epochs": 3,
|
| 490 |
+
"total_flos": 183654252937216.0,
|
| 491 |
+
"trial_name": null,
|
| 492 |
+
"trial_params": null
|
| 493 |
+
}
|