Upload folder using huggingface_hub
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- README.md +357 -0
- adapter_config.json +21 -0
- adapter_model.bin +3 -0
- checkpoint-290/README.md +23 -0
- checkpoint-290/adapter_config.json +21 -0
- checkpoint-290/adapter_model.safetensors +3 -0
- checkpoint-290/optimizer.pt +3 -0
- checkpoint-290/rng_state.pth +3 -0
- checkpoint-290/scheduler.pt +3 -0
- checkpoint-290/trainer_state.json +891 -0
- checkpoint-290/training_args.bin +3 -0
- checkpoint-320/README.md +23 -0
- checkpoint-320/adapter_config.json +21 -0
- checkpoint-320/adapter_model.safetensors +3 -0
- checkpoint-320/optimizer.pt +3 -0
- checkpoint-320/rng_state.pth +3 -0
- checkpoint-320/scheduler.pt +3 -0
- checkpoint-320/trainer_state.json +981 -0
- checkpoint-320/training_args.bin +3 -0
- checkpoint-330/README.md +23 -0
- checkpoint-330/adapter_config.json +21 -0
- checkpoint-330/adapter_model.safetensors +3 -0
- checkpoint-330/optimizer.pt +3 -0
- checkpoint-330/rng_state.pth +3 -0
- checkpoint-330/scheduler.pt +3 -0
- checkpoint-330/trainer_state.json +1011 -0
- checkpoint-330/training_args.bin +3 -0
- checkpoint-340/README.md +23 -0
- checkpoint-340/adapter_config.json +21 -0
- checkpoint-340/adapter_model.safetensors +3 -0
- checkpoint-340/optimizer.pt +3 -0
- checkpoint-340/rng_state.pth +3 -0
- checkpoint-340/scheduler.pt +3 -0
- checkpoint-340/trainer_state.json +1041 -0
- checkpoint-340/training_args.bin +3 -0
- checkpoint-350/README.md +23 -0
- checkpoint-350/adapter_config.json +21 -0
- checkpoint-350/adapter_model.safetensors +3 -0
- checkpoint-350/optimizer.pt +3 -0
- checkpoint-350/rng_state.pth +3 -0
- checkpoint-350/scheduler.pt +3 -0
- checkpoint-350/trainer_state.json +1071 -0
- checkpoint-350/training_args.bin +3 -0
- config.json +42 -0
- logs/events.out.tfevents.1699989718.node0370.palmetto.clemson.edu.1119956.2 +3 -0
- logs/events.out.tfevents.1699989813.node0370.palmetto.clemson.edu.1119956.4 +3 -0
- logs/events.out.tfevents.1699994287.node0370.palmetto.clemson.edu.1126403.1 +3 -0
- logs/events.out.tfevents.1699995040.node0370.palmetto.clemson.edu.1127435.1 +3 -0
- logs/events.out.tfevents.1700064888.node0277.palmetto.clemson.edu.1971495.1 +3 -0
- logs/events.out.tfevents.1700066157.node0277.palmetto.clemson.edu.1973537.1 +3 -0
README.md
ADDED
|
@@ -0,0 +1,357 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
|
| 7 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 8 |
+
- quant_method: bitsandbytes
|
| 9 |
+
- _load_in_8bit: False
|
| 10 |
+
- _load_in_4bit: False
|
| 11 |
+
- llm_int8_threshold: 6.0
|
| 12 |
+
- llm_int8_skip_modules: None
|
| 13 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 14 |
+
- llm_int8_has_fp16_weight: False
|
| 15 |
+
- bnb_4bit_quant_type: fp4
|
| 16 |
+
- bnb_4bit_use_double_quant: False
|
| 17 |
+
- bnb_4bit_compute_dtype: float32
|
| 18 |
+
- load_in_4bit: False
|
| 19 |
+
- load_in_8bit: False
|
| 20 |
+
|
| 21 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 22 |
+
- quant_method: bitsandbytes
|
| 23 |
+
- _load_in_8bit: False
|
| 24 |
+
- _load_in_4bit: False
|
| 25 |
+
- llm_int8_threshold: 6.0
|
| 26 |
+
- llm_int8_skip_modules: None
|
| 27 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 28 |
+
- llm_int8_has_fp16_weight: False
|
| 29 |
+
- bnb_4bit_quant_type: fp4
|
| 30 |
+
- bnb_4bit_use_double_quant: False
|
| 31 |
+
- bnb_4bit_compute_dtype: float32
|
| 32 |
+
- load_in_4bit: False
|
| 33 |
+
- load_in_8bit: False
|
| 34 |
+
|
| 35 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 36 |
+
- quant_method: bitsandbytes
|
| 37 |
+
- _load_in_8bit: False
|
| 38 |
+
- _load_in_4bit: False
|
| 39 |
+
- llm_int8_threshold: 6.0
|
| 40 |
+
- llm_int8_skip_modules: None
|
| 41 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 42 |
+
- llm_int8_has_fp16_weight: False
|
| 43 |
+
- bnb_4bit_quant_type: fp4
|
| 44 |
+
- bnb_4bit_use_double_quant: False
|
| 45 |
+
- bnb_4bit_compute_dtype: float32
|
| 46 |
+
- load_in_4bit: False
|
| 47 |
+
- load_in_8bit: False
|
| 48 |
+
|
| 49 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 50 |
+
- quant_method: bitsandbytes
|
| 51 |
+
- _load_in_8bit: False
|
| 52 |
+
- _load_in_4bit: False
|
| 53 |
+
- llm_int8_threshold: 6.0
|
| 54 |
+
- llm_int8_skip_modules: None
|
| 55 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 56 |
+
- llm_int8_has_fp16_weight: False
|
| 57 |
+
- bnb_4bit_quant_type: fp4
|
| 58 |
+
- bnb_4bit_use_double_quant: False
|
| 59 |
+
- bnb_4bit_compute_dtype: float32
|
| 60 |
+
- load_in_4bit: False
|
| 61 |
+
- load_in_8bit: False
|
| 62 |
+
|
| 63 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 64 |
+
- quant_method: bitsandbytes
|
| 65 |
+
- _load_in_8bit: False
|
| 66 |
+
- _load_in_4bit: False
|
| 67 |
+
- llm_int8_threshold: 6.0
|
| 68 |
+
- llm_int8_skip_modules: None
|
| 69 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 70 |
+
- llm_int8_has_fp16_weight: False
|
| 71 |
+
- bnb_4bit_quant_type: fp4
|
| 72 |
+
- bnb_4bit_use_double_quant: False
|
| 73 |
+
- bnb_4bit_compute_dtype: float32
|
| 74 |
+
- load_in_4bit: False
|
| 75 |
+
- load_in_8bit: False
|
| 76 |
+
|
| 77 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 78 |
+
- quant_method: bitsandbytes
|
| 79 |
+
- _load_in_8bit: False
|
| 80 |
+
- _load_in_4bit: False
|
| 81 |
+
- llm_int8_threshold: 6.0
|
| 82 |
+
- llm_int8_skip_modules: None
|
| 83 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 84 |
+
- llm_int8_has_fp16_weight: False
|
| 85 |
+
- bnb_4bit_quant_type: fp4
|
| 86 |
+
- bnb_4bit_use_double_quant: False
|
| 87 |
+
- bnb_4bit_compute_dtype: float32
|
| 88 |
+
- load_in_4bit: False
|
| 89 |
+
- load_in_8bit: False
|
| 90 |
+
|
| 91 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 92 |
+
- quant_method: bitsandbytes
|
| 93 |
+
- _load_in_8bit: False
|
| 94 |
+
- _load_in_4bit: False
|
| 95 |
+
- llm_int8_threshold: 6.0
|
| 96 |
+
- llm_int8_skip_modules: None
|
| 97 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 98 |
+
- llm_int8_has_fp16_weight: False
|
| 99 |
+
- bnb_4bit_quant_type: fp4
|
| 100 |
+
- bnb_4bit_use_double_quant: False
|
| 101 |
+
- bnb_4bit_compute_dtype: float32
|
| 102 |
+
- load_in_4bit: False
|
| 103 |
+
- load_in_8bit: False
|
| 104 |
+
|
| 105 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 106 |
+
- quant_method: bitsandbytes
|
| 107 |
+
- _load_in_8bit: False
|
| 108 |
+
- _load_in_4bit: False
|
| 109 |
+
- llm_int8_threshold: 6.0
|
| 110 |
+
- llm_int8_skip_modules: None
|
| 111 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 112 |
+
- llm_int8_has_fp16_weight: False
|
| 113 |
+
- bnb_4bit_quant_type: fp4
|
| 114 |
+
- bnb_4bit_use_double_quant: False
|
| 115 |
+
- bnb_4bit_compute_dtype: float32
|
| 116 |
+
- load_in_4bit: False
|
| 117 |
+
- load_in_8bit: False
|
| 118 |
+
|
| 119 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 120 |
+
- quant_method: bitsandbytes
|
| 121 |
+
- _load_in_8bit: False
|
| 122 |
+
- _load_in_4bit: False
|
| 123 |
+
- llm_int8_threshold: 6.0
|
| 124 |
+
- llm_int8_skip_modules: None
|
| 125 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 126 |
+
- llm_int8_has_fp16_weight: False
|
| 127 |
+
- bnb_4bit_quant_type: fp4
|
| 128 |
+
- bnb_4bit_use_double_quant: False
|
| 129 |
+
- bnb_4bit_compute_dtype: float32
|
| 130 |
+
- load_in_4bit: False
|
| 131 |
+
- load_in_8bit: False
|
| 132 |
+
|
| 133 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 134 |
+
- quant_method: bitsandbytes
|
| 135 |
+
- _load_in_8bit: False
|
| 136 |
+
- _load_in_4bit: False
|
| 137 |
+
- llm_int8_threshold: 6.0
|
| 138 |
+
- llm_int8_skip_modules: None
|
| 139 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 140 |
+
- llm_int8_has_fp16_weight: False
|
| 141 |
+
- bnb_4bit_quant_type: fp4
|
| 142 |
+
- bnb_4bit_use_double_quant: False
|
| 143 |
+
- bnb_4bit_compute_dtype: float32
|
| 144 |
+
- load_in_4bit: False
|
| 145 |
+
- load_in_8bit: False
|
| 146 |
+
|
| 147 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 148 |
+
- quant_method: bitsandbytes
|
| 149 |
+
- _load_in_8bit: False
|
| 150 |
+
- _load_in_4bit: False
|
| 151 |
+
- llm_int8_threshold: 6.0
|
| 152 |
+
- llm_int8_skip_modules: None
|
| 153 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 154 |
+
- llm_int8_has_fp16_weight: False
|
| 155 |
+
- bnb_4bit_quant_type: fp4
|
| 156 |
+
- bnb_4bit_use_double_quant: False
|
| 157 |
+
- bnb_4bit_compute_dtype: float32
|
| 158 |
+
- load_in_4bit: False
|
| 159 |
+
- load_in_8bit: False
|
| 160 |
+
|
| 161 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 162 |
+
- quant_method: bitsandbytes
|
| 163 |
+
- _load_in_8bit: False
|
| 164 |
+
- _load_in_4bit: False
|
| 165 |
+
- llm_int8_threshold: 6.0
|
| 166 |
+
- llm_int8_skip_modules: None
|
| 167 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 168 |
+
- llm_int8_has_fp16_weight: False
|
| 169 |
+
- bnb_4bit_quant_type: fp4
|
| 170 |
+
- bnb_4bit_use_double_quant: False
|
| 171 |
+
- bnb_4bit_compute_dtype: float32
|
| 172 |
+
- load_in_4bit: False
|
| 173 |
+
- load_in_8bit: False
|
| 174 |
+
|
| 175 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 176 |
+
- quant_method: bitsandbytes
|
| 177 |
+
- load_in_8bit: True
|
| 178 |
+
- load_in_4bit: False
|
| 179 |
+
- llm_int8_threshold: 6.0
|
| 180 |
+
- llm_int8_skip_modules: None
|
| 181 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 182 |
+
- llm_int8_has_fp16_weight: False
|
| 183 |
+
- bnb_4bit_quant_type: fp4
|
| 184 |
+
- bnb_4bit_use_double_quant: False
|
| 185 |
+
- bnb_4bit_compute_dtype: float32
|
| 186 |
+
|
| 187 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 188 |
+
- quant_method: bitsandbytes
|
| 189 |
+
- load_in_8bit: True
|
| 190 |
+
- load_in_4bit: False
|
| 191 |
+
- llm_int8_threshold: 6.0
|
| 192 |
+
- llm_int8_skip_modules: None
|
| 193 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 194 |
+
- llm_int8_has_fp16_weight: False
|
| 195 |
+
- bnb_4bit_quant_type: fp4
|
| 196 |
+
- bnb_4bit_use_double_quant: False
|
| 197 |
+
- bnb_4bit_compute_dtype: float32
|
| 198 |
+
|
| 199 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 200 |
+
- quant_method: bitsandbytes
|
| 201 |
+
- load_in_8bit: True
|
| 202 |
+
- load_in_4bit: False
|
| 203 |
+
- llm_int8_threshold: 6.0
|
| 204 |
+
- llm_int8_skip_modules: None
|
| 205 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 206 |
+
- llm_int8_has_fp16_weight: False
|
| 207 |
+
- bnb_4bit_quant_type: fp4
|
| 208 |
+
- bnb_4bit_use_double_quant: False
|
| 209 |
+
- bnb_4bit_compute_dtype: float32
|
| 210 |
+
|
| 211 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 212 |
+
- quant_method: bitsandbytes
|
| 213 |
+
- load_in_8bit: True
|
| 214 |
+
- load_in_4bit: False
|
| 215 |
+
- llm_int8_threshold: 6.0
|
| 216 |
+
- llm_int8_skip_modules: None
|
| 217 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 218 |
+
- llm_int8_has_fp16_weight: False
|
| 219 |
+
- bnb_4bit_quant_type: fp4
|
| 220 |
+
- bnb_4bit_use_double_quant: False
|
| 221 |
+
- bnb_4bit_compute_dtype: float32
|
| 222 |
+
|
| 223 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 224 |
+
- quant_method: bitsandbytes
|
| 225 |
+
- load_in_8bit: True
|
| 226 |
+
- load_in_4bit: False
|
| 227 |
+
- llm_int8_threshold: 6.0
|
| 228 |
+
- llm_int8_skip_modules: None
|
| 229 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 230 |
+
- llm_int8_has_fp16_weight: False
|
| 231 |
+
- bnb_4bit_quant_type: fp4
|
| 232 |
+
- bnb_4bit_use_double_quant: False
|
| 233 |
+
- bnb_4bit_compute_dtype: float32
|
| 234 |
+
|
| 235 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 236 |
+
- quant_method: bitsandbytes
|
| 237 |
+
- load_in_8bit: True
|
| 238 |
+
- load_in_4bit: False
|
| 239 |
+
- llm_int8_threshold: 6.0
|
| 240 |
+
- llm_int8_skip_modules: None
|
| 241 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 242 |
+
- llm_int8_has_fp16_weight: False
|
| 243 |
+
- bnb_4bit_quant_type: fp4
|
| 244 |
+
- bnb_4bit_use_double_quant: False
|
| 245 |
+
- bnb_4bit_compute_dtype: float32
|
| 246 |
+
|
| 247 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 248 |
+
- quant_method: bitsandbytes
|
| 249 |
+
- load_in_8bit: True
|
| 250 |
+
- load_in_4bit: False
|
| 251 |
+
- llm_int8_threshold: 6.0
|
| 252 |
+
- llm_int8_skip_modules: None
|
| 253 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 254 |
+
- llm_int8_has_fp16_weight: False
|
| 255 |
+
- bnb_4bit_quant_type: fp4
|
| 256 |
+
- bnb_4bit_use_double_quant: False
|
| 257 |
+
- bnb_4bit_compute_dtype: float32
|
| 258 |
+
|
| 259 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 260 |
+
- quant_method: bitsandbytes
|
| 261 |
+
- load_in_8bit: True
|
| 262 |
+
- load_in_4bit: False
|
| 263 |
+
- llm_int8_threshold: 6.0
|
| 264 |
+
- llm_int8_skip_modules: None
|
| 265 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 266 |
+
- llm_int8_has_fp16_weight: False
|
| 267 |
+
- bnb_4bit_quant_type: fp4
|
| 268 |
+
- bnb_4bit_use_double_quant: False
|
| 269 |
+
- bnb_4bit_compute_dtype: float32
|
| 270 |
+
|
| 271 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 272 |
+
- quant_method: bitsandbytes
|
| 273 |
+
- load_in_8bit: True
|
| 274 |
+
- load_in_4bit: False
|
| 275 |
+
- llm_int8_threshold: 6.0
|
| 276 |
+
- llm_int8_skip_modules: None
|
| 277 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 278 |
+
- llm_int8_has_fp16_weight: False
|
| 279 |
+
- bnb_4bit_quant_type: fp4
|
| 280 |
+
- bnb_4bit_use_double_quant: False
|
| 281 |
+
- bnb_4bit_compute_dtype: float32
|
| 282 |
+
|
| 283 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 284 |
+
- quant_method: bitsandbytes
|
| 285 |
+
- load_in_8bit: True
|
| 286 |
+
- load_in_4bit: False
|
| 287 |
+
- llm_int8_threshold: 6.0
|
| 288 |
+
- llm_int8_skip_modules: None
|
| 289 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 290 |
+
- llm_int8_has_fp16_weight: False
|
| 291 |
+
- bnb_4bit_quant_type: fp4
|
| 292 |
+
- bnb_4bit_use_double_quant: False
|
| 293 |
+
- bnb_4bit_compute_dtype: float32
|
| 294 |
+
|
| 295 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 296 |
+
- quant_method: bitsandbytes
|
| 297 |
+
- load_in_8bit: True
|
| 298 |
+
- load_in_4bit: False
|
| 299 |
+
- llm_int8_threshold: 6.0
|
| 300 |
+
- llm_int8_skip_modules: None
|
| 301 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 302 |
+
- llm_int8_has_fp16_weight: False
|
| 303 |
+
- bnb_4bit_quant_type: fp4
|
| 304 |
+
- bnb_4bit_use_double_quant: False
|
| 305 |
+
- bnb_4bit_compute_dtype: float32
|
| 306 |
+
|
| 307 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 308 |
+
- quant_method: bitsandbytes
|
| 309 |
+
- load_in_8bit: True
|
| 310 |
+
- load_in_4bit: False
|
| 311 |
+
- llm_int8_threshold: 6.0
|
| 312 |
+
- llm_int8_skip_modules: None
|
| 313 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 314 |
+
- llm_int8_has_fp16_weight: False
|
| 315 |
+
- bnb_4bit_quant_type: fp4
|
| 316 |
+
- bnb_4bit_use_double_quant: False
|
| 317 |
+
- bnb_4bit_compute_dtype: float32
|
| 318 |
+
|
| 319 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 320 |
+
- quant_method: bitsandbytes
|
| 321 |
+
- load_in_8bit: True
|
| 322 |
+
- load_in_4bit: False
|
| 323 |
+
- llm_int8_threshold: 6.0
|
| 324 |
+
- llm_int8_skip_modules: None
|
| 325 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 326 |
+
- llm_int8_has_fp16_weight: False
|
| 327 |
+
- bnb_4bit_quant_type: fp4
|
| 328 |
+
- bnb_4bit_use_double_quant: False
|
| 329 |
+
- bnb_4bit_compute_dtype: float32
|
| 330 |
+
### Framework versions
|
| 331 |
+
|
| 332 |
+
- PEFT 0.5.0
|
| 333 |
+
- PEFT 0.5.0
|
| 334 |
+
- PEFT 0.5.0
|
| 335 |
+
- PEFT 0.5.0
|
| 336 |
+
- PEFT 0.5.0
|
| 337 |
+
- PEFT 0.5.0
|
| 338 |
+
- PEFT 0.5.0
|
| 339 |
+
- PEFT 0.5.0
|
| 340 |
+
- PEFT 0.5.0
|
| 341 |
+
- PEFT 0.5.0
|
| 342 |
+
- PEFT 0.5.0
|
| 343 |
+
- PEFT 0.5.0
|
| 344 |
+
- PEFT 0.5.0
|
| 345 |
+
- PEFT 0.5.0
|
| 346 |
+
- PEFT 0.5.0
|
| 347 |
+
- PEFT 0.5.0
|
| 348 |
+
- PEFT 0.5.0
|
| 349 |
+
- PEFT 0.5.0
|
| 350 |
+
- PEFT 0.5.0
|
| 351 |
+
- PEFT 0.5.0
|
| 352 |
+
- PEFT 0.5.0
|
| 353 |
+
- PEFT 0.5.0
|
| 354 |
+
- PEFT 0.5.0
|
| 355 |
+
- PEFT 0.5.0
|
| 356 |
+
|
| 357 |
+
- PEFT 0.5.0
|
adapter_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "meta-llama/Llama-2-70b-hf",
|
| 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": 32,
|
| 11 |
+
"lora_dropout": 0.05,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 8,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"q_proj",
|
| 18 |
+
"v_proj"
|
| 19 |
+
],
|
| 20 |
+
"task_type": "CAUSAL_LM"
|
| 21 |
+
}
|
adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:120826b59370d1320fc855be8a66bc4ed0018b13167789cf4c4bdc57459bc50d
|
| 3 |
+
size 65652106
|
checkpoint-290/README.md
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
|
| 7 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 8 |
+
- quant_method: bitsandbytes
|
| 9 |
+
- _load_in_8bit: False
|
| 10 |
+
- _load_in_4bit: False
|
| 11 |
+
- llm_int8_threshold: 6.0
|
| 12 |
+
- llm_int8_skip_modules: None
|
| 13 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 14 |
+
- llm_int8_has_fp16_weight: False
|
| 15 |
+
- bnb_4bit_quant_type: fp4
|
| 16 |
+
- bnb_4bit_use_double_quant: False
|
| 17 |
+
- bnb_4bit_compute_dtype: float32
|
| 18 |
+
- load_in_4bit: False
|
| 19 |
+
- load_in_8bit: False
|
| 20 |
+
### Framework versions
|
| 21 |
+
|
| 22 |
+
|
| 23 |
+
- PEFT 0.5.0
|
checkpoint-290/adapter_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "meta-llama/Llama-2-70b-hf",
|
| 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": 32,
|
| 11 |
+
"lora_dropout": 0.05,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 8,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"q_proj",
|
| 18 |
+
"v_proj"
|
| 19 |
+
],
|
| 20 |
+
"task_type": "CAUSAL_LM"
|
| 21 |
+
}
|
checkpoint-290/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:33b52e7f27a554af95df693aebd9a2193c06c2b8b5b11c5550cfcb897cec6d90
|
| 3 |
+
size 65578776
|
checkpoint-290/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:63e609378f34d3612474ffc8b5abd122c6d145243a1294dd6cda2a1067d8e0ec
|
| 3 |
+
size 131345914
|
checkpoint-290/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dde7e0fcf5e900fab5c1d3d8900eb1c1683390e863c7d5842d9b61d44f86a207
|
| 3 |
+
size 14244
|
checkpoint-290/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:de03e6c88bd8caf4db5520e8793fef794ac2af1f927ce3195c6aac84b5dec25c
|
| 3 |
+
size 1064
|
checkpoint-290/trainer_state.json
ADDED
|
@@ -0,0 +1,891 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9927281737327576,
|
| 3 |
+
"best_model_checkpoint": "/scratch/kwamea/llama-output/checkpoint-290",
|
| 4 |
+
"epoch": 38.666666666666664,
|
| 5 |
+
"eval_steps": 5,
|
| 6 |
+
"global_step": 290,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.67,
|
| 13 |
+
"grad_norm": 0.243179589509964,
|
| 14 |
+
"learning_rate": 9.857142857142858e-05,
|
| 15 |
+
"loss": 1.9956,
|
| 16 |
+
"step": 5
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.67,
|
| 20 |
+
"eval_loss": 1.9701930284500122,
|
| 21 |
+
"eval_runtime": 17.115,
|
| 22 |
+
"eval_samples_per_second": 0.409,
|
| 23 |
+
"eval_steps_per_second": 0.058,
|
| 24 |
+
"step": 5
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 1.33,
|
| 28 |
+
"grad_norm": 0.34590908885002136,
|
| 29 |
+
"learning_rate": 9.714285714285715e-05,
|
| 30 |
+
"loss": 1.9758,
|
| 31 |
+
"step": 10
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 1.33,
|
| 35 |
+
"eval_loss": 1.8941271305084229,
|
| 36 |
+
"eval_runtime": 17.0912,
|
| 37 |
+
"eval_samples_per_second": 0.41,
|
| 38 |
+
"eval_steps_per_second": 0.059,
|
| 39 |
+
"step": 10
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 2.0,
|
| 43 |
+
"grad_norm": 0.31595832109451294,
|
| 44 |
+
"learning_rate": 9.571428571428573e-05,
|
| 45 |
+
"loss": 1.849,
|
| 46 |
+
"step": 15
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 2.0,
|
| 50 |
+
"eval_loss": 1.8046789169311523,
|
| 51 |
+
"eval_runtime": 17.098,
|
| 52 |
+
"eval_samples_per_second": 0.409,
|
| 53 |
+
"eval_steps_per_second": 0.058,
|
| 54 |
+
"step": 15
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 2.67,
|
| 58 |
+
"grad_norm": 0.3428090512752533,
|
| 59 |
+
"learning_rate": 9.428571428571429e-05,
|
| 60 |
+
"loss": 1.789,
|
| 61 |
+
"step": 20
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 2.67,
|
| 65 |
+
"eval_loss": 1.7658358812332153,
|
| 66 |
+
"eval_runtime": 17.0734,
|
| 67 |
+
"eval_samples_per_second": 0.41,
|
| 68 |
+
"eval_steps_per_second": 0.059,
|
| 69 |
+
"step": 20
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 3.33,
|
| 73 |
+
"grad_norm": 0.3102028965950012,
|
| 74 |
+
"learning_rate": 9.285714285714286e-05,
|
| 75 |
+
"loss": 1.7789,
|
| 76 |
+
"step": 25
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 3.33,
|
| 80 |
+
"eval_loss": 1.7225048542022705,
|
| 81 |
+
"eval_runtime": 17.0972,
|
| 82 |
+
"eval_samples_per_second": 0.409,
|
| 83 |
+
"eval_steps_per_second": 0.058,
|
| 84 |
+
"step": 25
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 4.0,
|
| 88 |
+
"grad_norm": 0.38602885603904724,
|
| 89 |
+
"learning_rate": 9.142857142857143e-05,
|
| 90 |
+
"loss": 1.7003,
|
| 91 |
+
"step": 30
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 4.0,
|
| 95 |
+
"eval_loss": 1.6749440431594849,
|
| 96 |
+
"eval_runtime": 17.1034,
|
| 97 |
+
"eval_samples_per_second": 0.409,
|
| 98 |
+
"eval_steps_per_second": 0.058,
|
| 99 |
+
"step": 30
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 4.67,
|
| 103 |
+
"grad_norm": 0.37120407819747925,
|
| 104 |
+
"learning_rate": 9e-05,
|
| 105 |
+
"loss": 1.6424,
|
| 106 |
+
"step": 35
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 4.67,
|
| 110 |
+
"eval_loss": 1.6231099367141724,
|
| 111 |
+
"eval_runtime": 17.1067,
|
| 112 |
+
"eval_samples_per_second": 0.409,
|
| 113 |
+
"eval_steps_per_second": 0.058,
|
| 114 |
+
"step": 35
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 5.33,
|
| 118 |
+
"grad_norm": 0.4633428454399109,
|
| 119 |
+
"learning_rate": 8.857142857142857e-05,
|
| 120 |
+
"loss": 1.6023,
|
| 121 |
+
"step": 40
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 5.33,
|
| 125 |
+
"eval_loss": 1.5727053880691528,
|
| 126 |
+
"eval_runtime": 17.1002,
|
| 127 |
+
"eval_samples_per_second": 0.409,
|
| 128 |
+
"eval_steps_per_second": 0.058,
|
| 129 |
+
"step": 40
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 6.0,
|
| 133 |
+
"grad_norm": 0.5034663081169128,
|
| 134 |
+
"learning_rate": 8.714285714285715e-05,
|
| 135 |
+
"loss": 1.5322,
|
| 136 |
+
"step": 45
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 6.0,
|
| 140 |
+
"eval_loss": 1.5312587022781372,
|
| 141 |
+
"eval_runtime": 17.1159,
|
| 142 |
+
"eval_samples_per_second": 0.409,
|
| 143 |
+
"eval_steps_per_second": 0.058,
|
| 144 |
+
"step": 45
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 6.67,
|
| 148 |
+
"grad_norm": 0.5549929141998291,
|
| 149 |
+
"learning_rate": 8.571428571428571e-05,
|
| 150 |
+
"loss": 1.4788,
|
| 151 |
+
"step": 50
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 6.67,
|
| 155 |
+
"eval_loss": 1.492464303970337,
|
| 156 |
+
"eval_runtime": 17.0823,
|
| 157 |
+
"eval_samples_per_second": 0.41,
|
| 158 |
+
"eval_steps_per_second": 0.059,
|
| 159 |
+
"step": 50
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 7.33,
|
| 163 |
+
"grad_norm": 0.49194690585136414,
|
| 164 |
+
"learning_rate": 8.428571428571429e-05,
|
| 165 |
+
"loss": 1.4632,
|
| 166 |
+
"step": 55
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 7.33,
|
| 170 |
+
"eval_loss": 1.4622489213943481,
|
| 171 |
+
"eval_runtime": 17.1022,
|
| 172 |
+
"eval_samples_per_second": 0.409,
|
| 173 |
+
"eval_steps_per_second": 0.058,
|
| 174 |
+
"step": 55
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 8.0,
|
| 178 |
+
"grad_norm": 0.5866131782531738,
|
| 179 |
+
"learning_rate": 8.285714285714287e-05,
|
| 180 |
+
"loss": 1.3951,
|
| 181 |
+
"step": 60
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 8.0,
|
| 185 |
+
"eval_loss": 1.435951828956604,
|
| 186 |
+
"eval_runtime": 17.1087,
|
| 187 |
+
"eval_samples_per_second": 0.409,
|
| 188 |
+
"eval_steps_per_second": 0.058,
|
| 189 |
+
"step": 60
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 8.67,
|
| 193 |
+
"grad_norm": 0.6252542734146118,
|
| 194 |
+
"learning_rate": 8.142857142857143e-05,
|
| 195 |
+
"loss": 1.3796,
|
| 196 |
+
"step": 65
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 8.67,
|
| 200 |
+
"eval_loss": 1.413227915763855,
|
| 201 |
+
"eval_runtime": 17.0914,
|
| 202 |
+
"eval_samples_per_second": 0.41,
|
| 203 |
+
"eval_steps_per_second": 0.059,
|
| 204 |
+
"step": 65
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 9.33,
|
| 208 |
+
"grad_norm": 0.6751863360404968,
|
| 209 |
+
"learning_rate": 8e-05,
|
| 210 |
+
"loss": 1.3257,
|
| 211 |
+
"step": 70
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 9.33,
|
| 215 |
+
"eval_loss": 1.395649790763855,
|
| 216 |
+
"eval_runtime": 17.0885,
|
| 217 |
+
"eval_samples_per_second": 0.41,
|
| 218 |
+
"eval_steps_per_second": 0.059,
|
| 219 |
+
"step": 70
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 10.0,
|
| 223 |
+
"grad_norm": 0.8878222703933716,
|
| 224 |
+
"learning_rate": 7.857142857142858e-05,
|
| 225 |
+
"loss": 1.2795,
|
| 226 |
+
"step": 75
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 10.0,
|
| 230 |
+
"eval_loss": 1.3699487447738647,
|
| 231 |
+
"eval_runtime": 17.1031,
|
| 232 |
+
"eval_samples_per_second": 0.409,
|
| 233 |
+
"eval_steps_per_second": 0.058,
|
| 234 |
+
"step": 75
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 10.67,
|
| 238 |
+
"grad_norm": 0.8470121026039124,
|
| 239 |
+
"learning_rate": 7.714285714285715e-05,
|
| 240 |
+
"loss": 1.2449,
|
| 241 |
+
"step": 80
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 10.67,
|
| 245 |
+
"eval_loss": 1.347831130027771,
|
| 246 |
+
"eval_runtime": 17.0985,
|
| 247 |
+
"eval_samples_per_second": 0.409,
|
| 248 |
+
"eval_steps_per_second": 0.058,
|
| 249 |
+
"step": 80
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 11.33,
|
| 253 |
+
"grad_norm": 1.0655425786972046,
|
| 254 |
+
"learning_rate": 7.571428571428571e-05,
|
| 255 |
+
"loss": 1.1983,
|
| 256 |
+
"step": 85
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 11.33,
|
| 260 |
+
"eval_loss": 1.3311971426010132,
|
| 261 |
+
"eval_runtime": 17.0784,
|
| 262 |
+
"eval_samples_per_second": 0.41,
|
| 263 |
+
"eval_steps_per_second": 0.059,
|
| 264 |
+
"step": 85
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 12.0,
|
| 268 |
+
"grad_norm": 1.2651888132095337,
|
| 269 |
+
"learning_rate": 7.428571428571429e-05,
|
| 270 |
+
"loss": 1.1467,
|
| 271 |
+
"step": 90
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 12.0,
|
| 275 |
+
"eval_loss": 1.3095277547836304,
|
| 276 |
+
"eval_runtime": 17.0903,
|
| 277 |
+
"eval_samples_per_second": 0.41,
|
| 278 |
+
"eval_steps_per_second": 0.059,
|
| 279 |
+
"step": 90
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 12.67,
|
| 283 |
+
"grad_norm": 1.248926043510437,
|
| 284 |
+
"learning_rate": 7.285714285714286e-05,
|
| 285 |
+
"loss": 1.0922,
|
| 286 |
+
"step": 95
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 12.67,
|
| 290 |
+
"eval_loss": 1.2942878007888794,
|
| 291 |
+
"eval_runtime": 17.0947,
|
| 292 |
+
"eval_samples_per_second": 0.409,
|
| 293 |
+
"eval_steps_per_second": 0.058,
|
| 294 |
+
"step": 95
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 13.33,
|
| 298 |
+
"grad_norm": 1.896952509880066,
|
| 299 |
+
"learning_rate": 7.142857142857143e-05,
|
| 300 |
+
"loss": 1.0403,
|
| 301 |
+
"step": 100
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 13.33,
|
| 305 |
+
"eval_loss": 1.2803159952163696,
|
| 306 |
+
"eval_runtime": 17.0819,
|
| 307 |
+
"eval_samples_per_second": 0.41,
|
| 308 |
+
"eval_steps_per_second": 0.059,
|
| 309 |
+
"step": 100
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 14.0,
|
| 313 |
+
"grad_norm": 1.862244725227356,
|
| 314 |
+
"learning_rate": 7e-05,
|
| 315 |
+
"loss": 1.0049,
|
| 316 |
+
"step": 105
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 14.0,
|
| 320 |
+
"eval_loss": 1.2643567323684692,
|
| 321 |
+
"eval_runtime": 17.0849,
|
| 322 |
+
"eval_samples_per_second": 0.41,
|
| 323 |
+
"eval_steps_per_second": 0.059,
|
| 324 |
+
"step": 105
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 14.67,
|
| 328 |
+
"grad_norm": 1.7487821578979492,
|
| 329 |
+
"learning_rate": 6.857142857142858e-05,
|
| 330 |
+
"loss": 0.9262,
|
| 331 |
+
"step": 110
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 14.67,
|
| 335 |
+
"eval_loss": 1.2471646070480347,
|
| 336 |
+
"eval_runtime": 17.1278,
|
| 337 |
+
"eval_samples_per_second": 0.409,
|
| 338 |
+
"eval_steps_per_second": 0.058,
|
| 339 |
+
"step": 110
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 15.33,
|
| 343 |
+
"grad_norm": 1.838605284690857,
|
| 344 |
+
"learning_rate": 6.714285714285714e-05,
|
| 345 |
+
"loss": 0.8965,
|
| 346 |
+
"step": 115
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 15.33,
|
| 350 |
+
"eval_loss": 1.2377034425735474,
|
| 351 |
+
"eval_runtime": 17.0731,
|
| 352 |
+
"eval_samples_per_second": 0.41,
|
| 353 |
+
"eval_steps_per_second": 0.059,
|
| 354 |
+
"step": 115
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 16.0,
|
| 358 |
+
"grad_norm": 3.117398977279663,
|
| 359 |
+
"learning_rate": 6.571428571428571e-05,
|
| 360 |
+
"loss": 0.8581,
|
| 361 |
+
"step": 120
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 16.0,
|
| 365 |
+
"eval_loss": 1.2083133459091187,
|
| 366 |
+
"eval_runtime": 17.1304,
|
| 367 |
+
"eval_samples_per_second": 0.409,
|
| 368 |
+
"eval_steps_per_second": 0.058,
|
| 369 |
+
"step": 120
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 16.67,
|
| 373 |
+
"grad_norm": 2.5655250549316406,
|
| 374 |
+
"learning_rate": 6.428571428571429e-05,
|
| 375 |
+
"loss": 0.7929,
|
| 376 |
+
"step": 125
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 16.67,
|
| 380 |
+
"eval_loss": 1.1945828199386597,
|
| 381 |
+
"eval_runtime": 17.104,
|
| 382 |
+
"eval_samples_per_second": 0.409,
|
| 383 |
+
"eval_steps_per_second": 0.058,
|
| 384 |
+
"step": 125
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 17.33,
|
| 388 |
+
"grad_norm": 2.168546199798584,
|
| 389 |
+
"learning_rate": 6.285714285714286e-05,
|
| 390 |
+
"loss": 0.7543,
|
| 391 |
+
"step": 130
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 17.33,
|
| 395 |
+
"eval_loss": 1.1876276731491089,
|
| 396 |
+
"eval_runtime": 17.1046,
|
| 397 |
+
"eval_samples_per_second": 0.409,
|
| 398 |
+
"eval_steps_per_second": 0.058,
|
| 399 |
+
"step": 130
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 18.0,
|
| 403 |
+
"grad_norm": 2.5984208583831787,
|
| 404 |
+
"learning_rate": 6.142857142857143e-05,
|
| 405 |
+
"loss": 0.716,
|
| 406 |
+
"step": 135
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 18.0,
|
| 410 |
+
"eval_loss": 1.1714750528335571,
|
| 411 |
+
"eval_runtime": 17.0807,
|
| 412 |
+
"eval_samples_per_second": 0.41,
|
| 413 |
+
"eval_steps_per_second": 0.059,
|
| 414 |
+
"step": 135
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 18.67,
|
| 418 |
+
"grad_norm": 3.479024887084961,
|
| 419 |
+
"learning_rate": 6e-05,
|
| 420 |
+
"loss": 0.6681,
|
| 421 |
+
"step": 140
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 18.67,
|
| 425 |
+
"eval_loss": 1.169895052909851,
|
| 426 |
+
"eval_runtime": 17.0681,
|
| 427 |
+
"eval_samples_per_second": 0.41,
|
| 428 |
+
"eval_steps_per_second": 0.059,
|
| 429 |
+
"step": 140
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 19.33,
|
| 433 |
+
"grad_norm": 2.563386917114258,
|
| 434 |
+
"learning_rate": 5.8571428571428575e-05,
|
| 435 |
+
"loss": 0.6306,
|
| 436 |
+
"step": 145
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 19.33,
|
| 440 |
+
"eval_loss": 1.1741083860397339,
|
| 441 |
+
"eval_runtime": 17.0568,
|
| 442 |
+
"eval_samples_per_second": 0.41,
|
| 443 |
+
"eval_steps_per_second": 0.059,
|
| 444 |
+
"step": 145
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 20.0,
|
| 448 |
+
"grad_norm": 2.96592116355896,
|
| 449 |
+
"learning_rate": 5.714285714285714e-05,
|
| 450 |
+
"loss": 0.6183,
|
| 451 |
+
"step": 150
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 20.0,
|
| 455 |
+
"eval_loss": 1.1455965042114258,
|
| 456 |
+
"eval_runtime": 17.073,
|
| 457 |
+
"eval_samples_per_second": 0.41,
|
| 458 |
+
"eval_steps_per_second": 0.059,
|
| 459 |
+
"step": 150
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 20.67,
|
| 463 |
+
"grad_norm": 2.6751275062561035,
|
| 464 |
+
"learning_rate": 5.571428571428572e-05,
|
| 465 |
+
"loss": 0.5464,
|
| 466 |
+
"step": 155
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 20.67,
|
| 470 |
+
"eval_loss": 1.131102204322815,
|
| 471 |
+
"eval_runtime": 17.0578,
|
| 472 |
+
"eval_samples_per_second": 0.41,
|
| 473 |
+
"eval_steps_per_second": 0.059,
|
| 474 |
+
"step": 155
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 21.33,
|
| 478 |
+
"grad_norm": 2.3700051307678223,
|
| 479 |
+
"learning_rate": 5.428571428571428e-05,
|
| 480 |
+
"loss": 0.551,
|
| 481 |
+
"step": 160
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 21.33,
|
| 485 |
+
"eval_loss": 1.127384066581726,
|
| 486 |
+
"eval_runtime": 17.0546,
|
| 487 |
+
"eval_samples_per_second": 0.41,
|
| 488 |
+
"eval_steps_per_second": 0.059,
|
| 489 |
+
"step": 160
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 22.0,
|
| 493 |
+
"grad_norm": 3.3827567100524902,
|
| 494 |
+
"learning_rate": 5.285714285714286e-05,
|
| 495 |
+
"loss": 0.5179,
|
| 496 |
+
"step": 165
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 22.0,
|
| 500 |
+
"eval_loss": 1.111584186553955,
|
| 501 |
+
"eval_runtime": 17.0812,
|
| 502 |
+
"eval_samples_per_second": 0.41,
|
| 503 |
+
"eval_steps_per_second": 0.059,
|
| 504 |
+
"step": 165
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 22.67,
|
| 508 |
+
"grad_norm": 3.55790114402771,
|
| 509 |
+
"learning_rate": 5.142857142857143e-05,
|
| 510 |
+
"loss": 0.4831,
|
| 511 |
+
"step": 170
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 22.67,
|
| 515 |
+
"eval_loss": 1.0948525667190552,
|
| 516 |
+
"eval_runtime": 17.0547,
|
| 517 |
+
"eval_samples_per_second": 0.41,
|
| 518 |
+
"eval_steps_per_second": 0.059,
|
| 519 |
+
"step": 170
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 23.33,
|
| 523 |
+
"grad_norm": 3.0782699584960938,
|
| 524 |
+
"learning_rate": 5e-05,
|
| 525 |
+
"loss": 0.4587,
|
| 526 |
+
"step": 175
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 23.33,
|
| 530 |
+
"eval_loss": 1.0906586647033691,
|
| 531 |
+
"eval_runtime": 17.0666,
|
| 532 |
+
"eval_samples_per_second": 0.41,
|
| 533 |
+
"eval_steps_per_second": 0.059,
|
| 534 |
+
"step": 175
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 24.0,
|
| 538 |
+
"grad_norm": 3.3993167877197266,
|
| 539 |
+
"learning_rate": 4.8571428571428576e-05,
|
| 540 |
+
"loss": 0.4203,
|
| 541 |
+
"step": 180
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 24.0,
|
| 545 |
+
"eval_loss": 1.0688152313232422,
|
| 546 |
+
"eval_runtime": 17.0721,
|
| 547 |
+
"eval_samples_per_second": 0.41,
|
| 548 |
+
"eval_steps_per_second": 0.059,
|
| 549 |
+
"step": 180
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 24.67,
|
| 553 |
+
"grad_norm": 3.319303035736084,
|
| 554 |
+
"learning_rate": 4.714285714285714e-05,
|
| 555 |
+
"loss": 0.3975,
|
| 556 |
+
"step": 185
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 24.67,
|
| 560 |
+
"eval_loss": 1.0746583938598633,
|
| 561 |
+
"eval_runtime": 17.0709,
|
| 562 |
+
"eval_samples_per_second": 0.41,
|
| 563 |
+
"eval_steps_per_second": 0.059,
|
| 564 |
+
"step": 185
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 25.33,
|
| 568 |
+
"grad_norm": 2.4532127380371094,
|
| 569 |
+
"learning_rate": 4.5714285714285716e-05,
|
| 570 |
+
"loss": 0.3832,
|
| 571 |
+
"step": 190
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 25.33,
|
| 575 |
+
"eval_loss": 1.0772522687911987,
|
| 576 |
+
"eval_runtime": 17.0619,
|
| 577 |
+
"eval_samples_per_second": 0.41,
|
| 578 |
+
"eval_steps_per_second": 0.059,
|
| 579 |
+
"step": 190
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 26.0,
|
| 583 |
+
"grad_norm": 3.956822156906128,
|
| 584 |
+
"learning_rate": 4.428571428571428e-05,
|
| 585 |
+
"loss": 0.3725,
|
| 586 |
+
"step": 195
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 26.0,
|
| 590 |
+
"eval_loss": 1.0638784170150757,
|
| 591 |
+
"eval_runtime": 17.0807,
|
| 592 |
+
"eval_samples_per_second": 0.41,
|
| 593 |
+
"eval_steps_per_second": 0.059,
|
| 594 |
+
"step": 195
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 26.67,
|
| 598 |
+
"grad_norm": 2.76033353805542,
|
| 599 |
+
"learning_rate": 4.2857142857142856e-05,
|
| 600 |
+
"loss": 0.3473,
|
| 601 |
+
"step": 200
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 26.67,
|
| 605 |
+
"eval_loss": 1.04669988155365,
|
| 606 |
+
"eval_runtime": 17.0774,
|
| 607 |
+
"eval_samples_per_second": 0.41,
|
| 608 |
+
"eval_steps_per_second": 0.059,
|
| 609 |
+
"step": 200
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 27.33,
|
| 613 |
+
"grad_norm": 3.8683507442474365,
|
| 614 |
+
"learning_rate": 4.1428571428571437e-05,
|
| 615 |
+
"loss": 0.3243,
|
| 616 |
+
"step": 205
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 27.33,
|
| 620 |
+
"eval_loss": 1.0470303297042847,
|
| 621 |
+
"eval_runtime": 17.0718,
|
| 622 |
+
"eval_samples_per_second": 0.41,
|
| 623 |
+
"eval_steps_per_second": 0.059,
|
| 624 |
+
"step": 205
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 28.0,
|
| 628 |
+
"grad_norm": 4.535538196563721,
|
| 629 |
+
"learning_rate": 4e-05,
|
| 630 |
+
"loss": 0.3202,
|
| 631 |
+
"step": 210
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 28.0,
|
| 635 |
+
"eval_loss": 1.025539517402649,
|
| 636 |
+
"eval_runtime": 17.0604,
|
| 637 |
+
"eval_samples_per_second": 0.41,
|
| 638 |
+
"eval_steps_per_second": 0.059,
|
| 639 |
+
"step": 210
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 28.67,
|
| 643 |
+
"grad_norm": 2.6224355697631836,
|
| 644 |
+
"learning_rate": 3.857142857142858e-05,
|
| 645 |
+
"loss": 0.2958,
|
| 646 |
+
"step": 215
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 28.67,
|
| 650 |
+
"eval_loss": 1.0192126035690308,
|
| 651 |
+
"eval_runtime": 17.0657,
|
| 652 |
+
"eval_samples_per_second": 0.41,
|
| 653 |
+
"eval_steps_per_second": 0.059,
|
| 654 |
+
"step": 215
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 29.33,
|
| 658 |
+
"grad_norm": 2.5870041847229004,
|
| 659 |
+
"learning_rate": 3.7142857142857143e-05,
|
| 660 |
+
"loss": 0.2783,
|
| 661 |
+
"step": 220
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 29.33,
|
| 665 |
+
"eval_loss": 1.0211580991744995,
|
| 666 |
+
"eval_runtime": 17.0857,
|
| 667 |
+
"eval_samples_per_second": 0.41,
|
| 668 |
+
"eval_steps_per_second": 0.059,
|
| 669 |
+
"step": 220
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 30.0,
|
| 673 |
+
"grad_norm": 3.4565751552581787,
|
| 674 |
+
"learning_rate": 3.571428571428572e-05,
|
| 675 |
+
"loss": 0.2773,
|
| 676 |
+
"step": 225
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 30.0,
|
| 680 |
+
"eval_loss": 1.006419062614441,
|
| 681 |
+
"eval_runtime": 17.0807,
|
| 682 |
+
"eval_samples_per_second": 0.41,
|
| 683 |
+
"eval_steps_per_second": 0.059,
|
| 684 |
+
"step": 225
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 30.67,
|
| 688 |
+
"grad_norm": 2.4756500720977783,
|
| 689 |
+
"learning_rate": 3.428571428571429e-05,
|
| 690 |
+
"loss": 0.2482,
|
| 691 |
+
"step": 230
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 30.67,
|
| 695 |
+
"eval_loss": 1.0081219673156738,
|
| 696 |
+
"eval_runtime": 17.0576,
|
| 697 |
+
"eval_samples_per_second": 0.41,
|
| 698 |
+
"eval_steps_per_second": 0.059,
|
| 699 |
+
"step": 230
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 31.33,
|
| 703 |
+
"grad_norm": 2.38002610206604,
|
| 704 |
+
"learning_rate": 3.285714285714286e-05,
|
| 705 |
+
"loss": 0.2464,
|
| 706 |
+
"step": 235
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 31.33,
|
| 710 |
+
"eval_loss": 1.0151804685592651,
|
| 711 |
+
"eval_runtime": 17.0587,
|
| 712 |
+
"eval_samples_per_second": 0.41,
|
| 713 |
+
"eval_steps_per_second": 0.059,
|
| 714 |
+
"step": 235
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 32.0,
|
| 718 |
+
"grad_norm": 3.7081105709075928,
|
| 719 |
+
"learning_rate": 3.142857142857143e-05,
|
| 720 |
+
"loss": 0.2442,
|
| 721 |
+
"step": 240
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 32.0,
|
| 725 |
+
"eval_loss": 1.0032445192337036,
|
| 726 |
+
"eval_runtime": 17.1613,
|
| 727 |
+
"eval_samples_per_second": 0.408,
|
| 728 |
+
"eval_steps_per_second": 0.058,
|
| 729 |
+
"step": 240
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 32.67,
|
| 733 |
+
"grad_norm": 2.55924391746521,
|
| 734 |
+
"learning_rate": 3e-05,
|
| 735 |
+
"loss": 0.2193,
|
| 736 |
+
"step": 245
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 32.67,
|
| 740 |
+
"eval_loss": 0.9989615082740784,
|
| 741 |
+
"eval_runtime": 17.0447,
|
| 742 |
+
"eval_samples_per_second": 0.411,
|
| 743 |
+
"eval_steps_per_second": 0.059,
|
| 744 |
+
"step": 245
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 33.33,
|
| 748 |
+
"grad_norm": 1.9451407194137573,
|
| 749 |
+
"learning_rate": 2.857142857142857e-05,
|
| 750 |
+
"loss": 0.2101,
|
| 751 |
+
"step": 250
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 33.33,
|
| 755 |
+
"eval_loss": 1.0029457807540894,
|
| 756 |
+
"eval_runtime": 17.0816,
|
| 757 |
+
"eval_samples_per_second": 0.41,
|
| 758 |
+
"eval_steps_per_second": 0.059,
|
| 759 |
+
"step": 250
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 34.0,
|
| 763 |
+
"grad_norm": 2.713731527328491,
|
| 764 |
+
"learning_rate": 2.714285714285714e-05,
|
| 765 |
+
"loss": 0.2194,
|
| 766 |
+
"step": 255
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 34.0,
|
| 770 |
+
"eval_loss": 0.9959421753883362,
|
| 771 |
+
"eval_runtime": 17.0747,
|
| 772 |
+
"eval_samples_per_second": 0.41,
|
| 773 |
+
"eval_steps_per_second": 0.059,
|
| 774 |
+
"step": 255
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 34.67,
|
| 778 |
+
"grad_norm": 2.1633846759796143,
|
| 779 |
+
"learning_rate": 2.5714285714285714e-05,
|
| 780 |
+
"loss": 0.1958,
|
| 781 |
+
"step": 260
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 34.67,
|
| 785 |
+
"eval_loss": 0.9989770650863647,
|
| 786 |
+
"eval_runtime": 17.0821,
|
| 787 |
+
"eval_samples_per_second": 0.41,
|
| 788 |
+
"eval_steps_per_second": 0.059,
|
| 789 |
+
"step": 260
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 35.33,
|
| 793 |
+
"grad_norm": 3.9233529567718506,
|
| 794 |
+
"learning_rate": 2.4285714285714288e-05,
|
| 795 |
+
"loss": 0.1831,
|
| 796 |
+
"step": 265
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 35.33,
|
| 800 |
+
"eval_loss": 1.0072578191757202,
|
| 801 |
+
"eval_runtime": 17.0564,
|
| 802 |
+
"eval_samples_per_second": 0.41,
|
| 803 |
+
"eval_steps_per_second": 0.059,
|
| 804 |
+
"step": 265
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 36.0,
|
| 808 |
+
"grad_norm": 2.4143056869506836,
|
| 809 |
+
"learning_rate": 2.2857142857142858e-05,
|
| 810 |
+
"loss": 0.1753,
|
| 811 |
+
"step": 270
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 36.0,
|
| 815 |
+
"eval_loss": 0.9938892722129822,
|
| 816 |
+
"eval_runtime": 17.0668,
|
| 817 |
+
"eval_samples_per_second": 0.41,
|
| 818 |
+
"eval_steps_per_second": 0.059,
|
| 819 |
+
"step": 270
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 36.67,
|
| 823 |
+
"grad_norm": 2.706679582595825,
|
| 824 |
+
"learning_rate": 2.1428571428571428e-05,
|
| 825 |
+
"loss": 0.1698,
|
| 826 |
+
"step": 275
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 36.67,
|
| 830 |
+
"eval_loss": 0.9969200491905212,
|
| 831 |
+
"eval_runtime": 17.0643,
|
| 832 |
+
"eval_samples_per_second": 0.41,
|
| 833 |
+
"eval_steps_per_second": 0.059,
|
| 834 |
+
"step": 275
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 37.33,
|
| 838 |
+
"grad_norm": 1.872753620147705,
|
| 839 |
+
"learning_rate": 2e-05,
|
| 840 |
+
"loss": 0.16,
|
| 841 |
+
"step": 280
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 37.33,
|
| 845 |
+
"eval_loss": 0.9940390586853027,
|
| 846 |
+
"eval_runtime": 17.0728,
|
| 847 |
+
"eval_samples_per_second": 0.41,
|
| 848 |
+
"eval_steps_per_second": 0.059,
|
| 849 |
+
"step": 280
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 38.0,
|
| 853 |
+
"grad_norm": 2.7510581016540527,
|
| 854 |
+
"learning_rate": 1.8571428571428572e-05,
|
| 855 |
+
"loss": 0.1614,
|
| 856 |
+
"step": 285
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 38.0,
|
| 860 |
+
"eval_loss": 1.0066231489181519,
|
| 861 |
+
"eval_runtime": 17.072,
|
| 862 |
+
"eval_samples_per_second": 0.41,
|
| 863 |
+
"eval_steps_per_second": 0.059,
|
| 864 |
+
"step": 285
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 38.67,
|
| 868 |
+
"grad_norm": 1.8461092710494995,
|
| 869 |
+
"learning_rate": 1.7142857142857145e-05,
|
| 870 |
+
"loss": 0.1506,
|
| 871 |
+
"step": 290
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 38.67,
|
| 875 |
+
"eval_loss": 0.9927281737327576,
|
| 876 |
+
"eval_runtime": 17.0481,
|
| 877 |
+
"eval_samples_per_second": 0.411,
|
| 878 |
+
"eval_steps_per_second": 0.059,
|
| 879 |
+
"step": 290
|
| 880 |
+
}
|
| 881 |
+
],
|
| 882 |
+
"logging_steps": 5,
|
| 883 |
+
"max_steps": 350,
|
| 884 |
+
"num_input_tokens_seen": 0,
|
| 885 |
+
"num_train_epochs": 50,
|
| 886 |
+
"save_steps": 10,
|
| 887 |
+
"total_flos": 1.9272976816637215e+18,
|
| 888 |
+
"train_batch_size": 2,
|
| 889 |
+
"trial_name": null,
|
| 890 |
+
"trial_params": null
|
| 891 |
+
}
|
checkpoint-290/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ed0ba6fb9a88dad56f61d9306f17b4e66e8767d898772faa97871a1388e82cf
|
| 3 |
+
size 4920
|
checkpoint-320/README.md
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
|
| 7 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 8 |
+
- quant_method: bitsandbytes
|
| 9 |
+
- _load_in_8bit: False
|
| 10 |
+
- _load_in_4bit: False
|
| 11 |
+
- llm_int8_threshold: 6.0
|
| 12 |
+
- llm_int8_skip_modules: None
|
| 13 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 14 |
+
- llm_int8_has_fp16_weight: False
|
| 15 |
+
- bnb_4bit_quant_type: fp4
|
| 16 |
+
- bnb_4bit_use_double_quant: False
|
| 17 |
+
- bnb_4bit_compute_dtype: float32
|
| 18 |
+
- load_in_4bit: False
|
| 19 |
+
- load_in_8bit: False
|
| 20 |
+
### Framework versions
|
| 21 |
+
|
| 22 |
+
|
| 23 |
+
- PEFT 0.5.0
|
checkpoint-320/adapter_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "meta-llama/Llama-2-70b-hf",
|
| 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": 32,
|
| 11 |
+
"lora_dropout": 0.05,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 8,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"q_proj",
|
| 18 |
+
"v_proj"
|
| 19 |
+
],
|
| 20 |
+
"task_type": "CAUSAL_LM"
|
| 21 |
+
}
|
checkpoint-320/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:503d1844ac09449a9022937fd6160c557504e89ac035c55bc1efe99d8474e6f2
|
| 3 |
+
size 65578776
|
checkpoint-320/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f932a7f6af7ecf22ed10ae5ef43e3515ce5ae8605087025bef91b87a52248bc5
|
| 3 |
+
size 131345914
|
checkpoint-320/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:33a70333a4fc39389b0589b00a370500b36754db76577b913a7a9f856a3ef8df
|
| 3 |
+
size 14244
|
checkpoint-320/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e41504e9cc891d7299640b1d039f4351835f3de1fa7153a1118aece45c9f2c20
|
| 3 |
+
size 1064
|
checkpoint-320/trainer_state.json
ADDED
|
@@ -0,0 +1,981 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9927281737327576,
|
| 3 |
+
"best_model_checkpoint": "/scratch/kwamea/llama-output/checkpoint-290",
|
| 4 |
+
"epoch": 42.666666666666664,
|
| 5 |
+
"eval_steps": 5,
|
| 6 |
+
"global_step": 320,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.67,
|
| 13 |
+
"grad_norm": 0.243179589509964,
|
| 14 |
+
"learning_rate": 9.857142857142858e-05,
|
| 15 |
+
"loss": 1.9956,
|
| 16 |
+
"step": 5
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.67,
|
| 20 |
+
"eval_loss": 1.9701930284500122,
|
| 21 |
+
"eval_runtime": 17.115,
|
| 22 |
+
"eval_samples_per_second": 0.409,
|
| 23 |
+
"eval_steps_per_second": 0.058,
|
| 24 |
+
"step": 5
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 1.33,
|
| 28 |
+
"grad_norm": 0.34590908885002136,
|
| 29 |
+
"learning_rate": 9.714285714285715e-05,
|
| 30 |
+
"loss": 1.9758,
|
| 31 |
+
"step": 10
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 1.33,
|
| 35 |
+
"eval_loss": 1.8941271305084229,
|
| 36 |
+
"eval_runtime": 17.0912,
|
| 37 |
+
"eval_samples_per_second": 0.41,
|
| 38 |
+
"eval_steps_per_second": 0.059,
|
| 39 |
+
"step": 10
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 2.0,
|
| 43 |
+
"grad_norm": 0.31595832109451294,
|
| 44 |
+
"learning_rate": 9.571428571428573e-05,
|
| 45 |
+
"loss": 1.849,
|
| 46 |
+
"step": 15
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 2.0,
|
| 50 |
+
"eval_loss": 1.8046789169311523,
|
| 51 |
+
"eval_runtime": 17.098,
|
| 52 |
+
"eval_samples_per_second": 0.409,
|
| 53 |
+
"eval_steps_per_second": 0.058,
|
| 54 |
+
"step": 15
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 2.67,
|
| 58 |
+
"grad_norm": 0.3428090512752533,
|
| 59 |
+
"learning_rate": 9.428571428571429e-05,
|
| 60 |
+
"loss": 1.789,
|
| 61 |
+
"step": 20
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 2.67,
|
| 65 |
+
"eval_loss": 1.7658358812332153,
|
| 66 |
+
"eval_runtime": 17.0734,
|
| 67 |
+
"eval_samples_per_second": 0.41,
|
| 68 |
+
"eval_steps_per_second": 0.059,
|
| 69 |
+
"step": 20
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 3.33,
|
| 73 |
+
"grad_norm": 0.3102028965950012,
|
| 74 |
+
"learning_rate": 9.285714285714286e-05,
|
| 75 |
+
"loss": 1.7789,
|
| 76 |
+
"step": 25
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 3.33,
|
| 80 |
+
"eval_loss": 1.7225048542022705,
|
| 81 |
+
"eval_runtime": 17.0972,
|
| 82 |
+
"eval_samples_per_second": 0.409,
|
| 83 |
+
"eval_steps_per_second": 0.058,
|
| 84 |
+
"step": 25
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 4.0,
|
| 88 |
+
"grad_norm": 0.38602885603904724,
|
| 89 |
+
"learning_rate": 9.142857142857143e-05,
|
| 90 |
+
"loss": 1.7003,
|
| 91 |
+
"step": 30
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 4.0,
|
| 95 |
+
"eval_loss": 1.6749440431594849,
|
| 96 |
+
"eval_runtime": 17.1034,
|
| 97 |
+
"eval_samples_per_second": 0.409,
|
| 98 |
+
"eval_steps_per_second": 0.058,
|
| 99 |
+
"step": 30
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 4.67,
|
| 103 |
+
"grad_norm": 0.37120407819747925,
|
| 104 |
+
"learning_rate": 9e-05,
|
| 105 |
+
"loss": 1.6424,
|
| 106 |
+
"step": 35
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 4.67,
|
| 110 |
+
"eval_loss": 1.6231099367141724,
|
| 111 |
+
"eval_runtime": 17.1067,
|
| 112 |
+
"eval_samples_per_second": 0.409,
|
| 113 |
+
"eval_steps_per_second": 0.058,
|
| 114 |
+
"step": 35
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 5.33,
|
| 118 |
+
"grad_norm": 0.4633428454399109,
|
| 119 |
+
"learning_rate": 8.857142857142857e-05,
|
| 120 |
+
"loss": 1.6023,
|
| 121 |
+
"step": 40
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 5.33,
|
| 125 |
+
"eval_loss": 1.5727053880691528,
|
| 126 |
+
"eval_runtime": 17.1002,
|
| 127 |
+
"eval_samples_per_second": 0.409,
|
| 128 |
+
"eval_steps_per_second": 0.058,
|
| 129 |
+
"step": 40
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 6.0,
|
| 133 |
+
"grad_norm": 0.5034663081169128,
|
| 134 |
+
"learning_rate": 8.714285714285715e-05,
|
| 135 |
+
"loss": 1.5322,
|
| 136 |
+
"step": 45
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 6.0,
|
| 140 |
+
"eval_loss": 1.5312587022781372,
|
| 141 |
+
"eval_runtime": 17.1159,
|
| 142 |
+
"eval_samples_per_second": 0.409,
|
| 143 |
+
"eval_steps_per_second": 0.058,
|
| 144 |
+
"step": 45
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 6.67,
|
| 148 |
+
"grad_norm": 0.5549929141998291,
|
| 149 |
+
"learning_rate": 8.571428571428571e-05,
|
| 150 |
+
"loss": 1.4788,
|
| 151 |
+
"step": 50
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 6.67,
|
| 155 |
+
"eval_loss": 1.492464303970337,
|
| 156 |
+
"eval_runtime": 17.0823,
|
| 157 |
+
"eval_samples_per_second": 0.41,
|
| 158 |
+
"eval_steps_per_second": 0.059,
|
| 159 |
+
"step": 50
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 7.33,
|
| 163 |
+
"grad_norm": 0.49194690585136414,
|
| 164 |
+
"learning_rate": 8.428571428571429e-05,
|
| 165 |
+
"loss": 1.4632,
|
| 166 |
+
"step": 55
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 7.33,
|
| 170 |
+
"eval_loss": 1.4622489213943481,
|
| 171 |
+
"eval_runtime": 17.1022,
|
| 172 |
+
"eval_samples_per_second": 0.409,
|
| 173 |
+
"eval_steps_per_second": 0.058,
|
| 174 |
+
"step": 55
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 8.0,
|
| 178 |
+
"grad_norm": 0.5866131782531738,
|
| 179 |
+
"learning_rate": 8.285714285714287e-05,
|
| 180 |
+
"loss": 1.3951,
|
| 181 |
+
"step": 60
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 8.0,
|
| 185 |
+
"eval_loss": 1.435951828956604,
|
| 186 |
+
"eval_runtime": 17.1087,
|
| 187 |
+
"eval_samples_per_second": 0.409,
|
| 188 |
+
"eval_steps_per_second": 0.058,
|
| 189 |
+
"step": 60
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 8.67,
|
| 193 |
+
"grad_norm": 0.6252542734146118,
|
| 194 |
+
"learning_rate": 8.142857142857143e-05,
|
| 195 |
+
"loss": 1.3796,
|
| 196 |
+
"step": 65
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 8.67,
|
| 200 |
+
"eval_loss": 1.413227915763855,
|
| 201 |
+
"eval_runtime": 17.0914,
|
| 202 |
+
"eval_samples_per_second": 0.41,
|
| 203 |
+
"eval_steps_per_second": 0.059,
|
| 204 |
+
"step": 65
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 9.33,
|
| 208 |
+
"grad_norm": 0.6751863360404968,
|
| 209 |
+
"learning_rate": 8e-05,
|
| 210 |
+
"loss": 1.3257,
|
| 211 |
+
"step": 70
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 9.33,
|
| 215 |
+
"eval_loss": 1.395649790763855,
|
| 216 |
+
"eval_runtime": 17.0885,
|
| 217 |
+
"eval_samples_per_second": 0.41,
|
| 218 |
+
"eval_steps_per_second": 0.059,
|
| 219 |
+
"step": 70
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 10.0,
|
| 223 |
+
"grad_norm": 0.8878222703933716,
|
| 224 |
+
"learning_rate": 7.857142857142858e-05,
|
| 225 |
+
"loss": 1.2795,
|
| 226 |
+
"step": 75
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 10.0,
|
| 230 |
+
"eval_loss": 1.3699487447738647,
|
| 231 |
+
"eval_runtime": 17.1031,
|
| 232 |
+
"eval_samples_per_second": 0.409,
|
| 233 |
+
"eval_steps_per_second": 0.058,
|
| 234 |
+
"step": 75
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 10.67,
|
| 238 |
+
"grad_norm": 0.8470121026039124,
|
| 239 |
+
"learning_rate": 7.714285714285715e-05,
|
| 240 |
+
"loss": 1.2449,
|
| 241 |
+
"step": 80
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 10.67,
|
| 245 |
+
"eval_loss": 1.347831130027771,
|
| 246 |
+
"eval_runtime": 17.0985,
|
| 247 |
+
"eval_samples_per_second": 0.409,
|
| 248 |
+
"eval_steps_per_second": 0.058,
|
| 249 |
+
"step": 80
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 11.33,
|
| 253 |
+
"grad_norm": 1.0655425786972046,
|
| 254 |
+
"learning_rate": 7.571428571428571e-05,
|
| 255 |
+
"loss": 1.1983,
|
| 256 |
+
"step": 85
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 11.33,
|
| 260 |
+
"eval_loss": 1.3311971426010132,
|
| 261 |
+
"eval_runtime": 17.0784,
|
| 262 |
+
"eval_samples_per_second": 0.41,
|
| 263 |
+
"eval_steps_per_second": 0.059,
|
| 264 |
+
"step": 85
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 12.0,
|
| 268 |
+
"grad_norm": 1.2651888132095337,
|
| 269 |
+
"learning_rate": 7.428571428571429e-05,
|
| 270 |
+
"loss": 1.1467,
|
| 271 |
+
"step": 90
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 12.0,
|
| 275 |
+
"eval_loss": 1.3095277547836304,
|
| 276 |
+
"eval_runtime": 17.0903,
|
| 277 |
+
"eval_samples_per_second": 0.41,
|
| 278 |
+
"eval_steps_per_second": 0.059,
|
| 279 |
+
"step": 90
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 12.67,
|
| 283 |
+
"grad_norm": 1.248926043510437,
|
| 284 |
+
"learning_rate": 7.285714285714286e-05,
|
| 285 |
+
"loss": 1.0922,
|
| 286 |
+
"step": 95
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 12.67,
|
| 290 |
+
"eval_loss": 1.2942878007888794,
|
| 291 |
+
"eval_runtime": 17.0947,
|
| 292 |
+
"eval_samples_per_second": 0.409,
|
| 293 |
+
"eval_steps_per_second": 0.058,
|
| 294 |
+
"step": 95
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 13.33,
|
| 298 |
+
"grad_norm": 1.896952509880066,
|
| 299 |
+
"learning_rate": 7.142857142857143e-05,
|
| 300 |
+
"loss": 1.0403,
|
| 301 |
+
"step": 100
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 13.33,
|
| 305 |
+
"eval_loss": 1.2803159952163696,
|
| 306 |
+
"eval_runtime": 17.0819,
|
| 307 |
+
"eval_samples_per_second": 0.41,
|
| 308 |
+
"eval_steps_per_second": 0.059,
|
| 309 |
+
"step": 100
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 14.0,
|
| 313 |
+
"grad_norm": 1.862244725227356,
|
| 314 |
+
"learning_rate": 7e-05,
|
| 315 |
+
"loss": 1.0049,
|
| 316 |
+
"step": 105
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 14.0,
|
| 320 |
+
"eval_loss": 1.2643567323684692,
|
| 321 |
+
"eval_runtime": 17.0849,
|
| 322 |
+
"eval_samples_per_second": 0.41,
|
| 323 |
+
"eval_steps_per_second": 0.059,
|
| 324 |
+
"step": 105
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 14.67,
|
| 328 |
+
"grad_norm": 1.7487821578979492,
|
| 329 |
+
"learning_rate": 6.857142857142858e-05,
|
| 330 |
+
"loss": 0.9262,
|
| 331 |
+
"step": 110
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 14.67,
|
| 335 |
+
"eval_loss": 1.2471646070480347,
|
| 336 |
+
"eval_runtime": 17.1278,
|
| 337 |
+
"eval_samples_per_second": 0.409,
|
| 338 |
+
"eval_steps_per_second": 0.058,
|
| 339 |
+
"step": 110
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 15.33,
|
| 343 |
+
"grad_norm": 1.838605284690857,
|
| 344 |
+
"learning_rate": 6.714285714285714e-05,
|
| 345 |
+
"loss": 0.8965,
|
| 346 |
+
"step": 115
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 15.33,
|
| 350 |
+
"eval_loss": 1.2377034425735474,
|
| 351 |
+
"eval_runtime": 17.0731,
|
| 352 |
+
"eval_samples_per_second": 0.41,
|
| 353 |
+
"eval_steps_per_second": 0.059,
|
| 354 |
+
"step": 115
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 16.0,
|
| 358 |
+
"grad_norm": 3.117398977279663,
|
| 359 |
+
"learning_rate": 6.571428571428571e-05,
|
| 360 |
+
"loss": 0.8581,
|
| 361 |
+
"step": 120
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 16.0,
|
| 365 |
+
"eval_loss": 1.2083133459091187,
|
| 366 |
+
"eval_runtime": 17.1304,
|
| 367 |
+
"eval_samples_per_second": 0.409,
|
| 368 |
+
"eval_steps_per_second": 0.058,
|
| 369 |
+
"step": 120
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 16.67,
|
| 373 |
+
"grad_norm": 2.5655250549316406,
|
| 374 |
+
"learning_rate": 6.428571428571429e-05,
|
| 375 |
+
"loss": 0.7929,
|
| 376 |
+
"step": 125
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 16.67,
|
| 380 |
+
"eval_loss": 1.1945828199386597,
|
| 381 |
+
"eval_runtime": 17.104,
|
| 382 |
+
"eval_samples_per_second": 0.409,
|
| 383 |
+
"eval_steps_per_second": 0.058,
|
| 384 |
+
"step": 125
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 17.33,
|
| 388 |
+
"grad_norm": 2.168546199798584,
|
| 389 |
+
"learning_rate": 6.285714285714286e-05,
|
| 390 |
+
"loss": 0.7543,
|
| 391 |
+
"step": 130
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 17.33,
|
| 395 |
+
"eval_loss": 1.1876276731491089,
|
| 396 |
+
"eval_runtime": 17.1046,
|
| 397 |
+
"eval_samples_per_second": 0.409,
|
| 398 |
+
"eval_steps_per_second": 0.058,
|
| 399 |
+
"step": 130
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 18.0,
|
| 403 |
+
"grad_norm": 2.5984208583831787,
|
| 404 |
+
"learning_rate": 6.142857142857143e-05,
|
| 405 |
+
"loss": 0.716,
|
| 406 |
+
"step": 135
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 18.0,
|
| 410 |
+
"eval_loss": 1.1714750528335571,
|
| 411 |
+
"eval_runtime": 17.0807,
|
| 412 |
+
"eval_samples_per_second": 0.41,
|
| 413 |
+
"eval_steps_per_second": 0.059,
|
| 414 |
+
"step": 135
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 18.67,
|
| 418 |
+
"grad_norm": 3.479024887084961,
|
| 419 |
+
"learning_rate": 6e-05,
|
| 420 |
+
"loss": 0.6681,
|
| 421 |
+
"step": 140
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 18.67,
|
| 425 |
+
"eval_loss": 1.169895052909851,
|
| 426 |
+
"eval_runtime": 17.0681,
|
| 427 |
+
"eval_samples_per_second": 0.41,
|
| 428 |
+
"eval_steps_per_second": 0.059,
|
| 429 |
+
"step": 140
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 19.33,
|
| 433 |
+
"grad_norm": 2.563386917114258,
|
| 434 |
+
"learning_rate": 5.8571428571428575e-05,
|
| 435 |
+
"loss": 0.6306,
|
| 436 |
+
"step": 145
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 19.33,
|
| 440 |
+
"eval_loss": 1.1741083860397339,
|
| 441 |
+
"eval_runtime": 17.0568,
|
| 442 |
+
"eval_samples_per_second": 0.41,
|
| 443 |
+
"eval_steps_per_second": 0.059,
|
| 444 |
+
"step": 145
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 20.0,
|
| 448 |
+
"grad_norm": 2.96592116355896,
|
| 449 |
+
"learning_rate": 5.714285714285714e-05,
|
| 450 |
+
"loss": 0.6183,
|
| 451 |
+
"step": 150
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 20.0,
|
| 455 |
+
"eval_loss": 1.1455965042114258,
|
| 456 |
+
"eval_runtime": 17.073,
|
| 457 |
+
"eval_samples_per_second": 0.41,
|
| 458 |
+
"eval_steps_per_second": 0.059,
|
| 459 |
+
"step": 150
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 20.67,
|
| 463 |
+
"grad_norm": 2.6751275062561035,
|
| 464 |
+
"learning_rate": 5.571428571428572e-05,
|
| 465 |
+
"loss": 0.5464,
|
| 466 |
+
"step": 155
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 20.67,
|
| 470 |
+
"eval_loss": 1.131102204322815,
|
| 471 |
+
"eval_runtime": 17.0578,
|
| 472 |
+
"eval_samples_per_second": 0.41,
|
| 473 |
+
"eval_steps_per_second": 0.059,
|
| 474 |
+
"step": 155
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 21.33,
|
| 478 |
+
"grad_norm": 2.3700051307678223,
|
| 479 |
+
"learning_rate": 5.428571428571428e-05,
|
| 480 |
+
"loss": 0.551,
|
| 481 |
+
"step": 160
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 21.33,
|
| 485 |
+
"eval_loss": 1.127384066581726,
|
| 486 |
+
"eval_runtime": 17.0546,
|
| 487 |
+
"eval_samples_per_second": 0.41,
|
| 488 |
+
"eval_steps_per_second": 0.059,
|
| 489 |
+
"step": 160
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 22.0,
|
| 493 |
+
"grad_norm": 3.3827567100524902,
|
| 494 |
+
"learning_rate": 5.285714285714286e-05,
|
| 495 |
+
"loss": 0.5179,
|
| 496 |
+
"step": 165
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 22.0,
|
| 500 |
+
"eval_loss": 1.111584186553955,
|
| 501 |
+
"eval_runtime": 17.0812,
|
| 502 |
+
"eval_samples_per_second": 0.41,
|
| 503 |
+
"eval_steps_per_second": 0.059,
|
| 504 |
+
"step": 165
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 22.67,
|
| 508 |
+
"grad_norm": 3.55790114402771,
|
| 509 |
+
"learning_rate": 5.142857142857143e-05,
|
| 510 |
+
"loss": 0.4831,
|
| 511 |
+
"step": 170
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 22.67,
|
| 515 |
+
"eval_loss": 1.0948525667190552,
|
| 516 |
+
"eval_runtime": 17.0547,
|
| 517 |
+
"eval_samples_per_second": 0.41,
|
| 518 |
+
"eval_steps_per_second": 0.059,
|
| 519 |
+
"step": 170
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 23.33,
|
| 523 |
+
"grad_norm": 3.0782699584960938,
|
| 524 |
+
"learning_rate": 5e-05,
|
| 525 |
+
"loss": 0.4587,
|
| 526 |
+
"step": 175
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 23.33,
|
| 530 |
+
"eval_loss": 1.0906586647033691,
|
| 531 |
+
"eval_runtime": 17.0666,
|
| 532 |
+
"eval_samples_per_second": 0.41,
|
| 533 |
+
"eval_steps_per_second": 0.059,
|
| 534 |
+
"step": 175
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 24.0,
|
| 538 |
+
"grad_norm": 3.3993167877197266,
|
| 539 |
+
"learning_rate": 4.8571428571428576e-05,
|
| 540 |
+
"loss": 0.4203,
|
| 541 |
+
"step": 180
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 24.0,
|
| 545 |
+
"eval_loss": 1.0688152313232422,
|
| 546 |
+
"eval_runtime": 17.0721,
|
| 547 |
+
"eval_samples_per_second": 0.41,
|
| 548 |
+
"eval_steps_per_second": 0.059,
|
| 549 |
+
"step": 180
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 24.67,
|
| 553 |
+
"grad_norm": 3.319303035736084,
|
| 554 |
+
"learning_rate": 4.714285714285714e-05,
|
| 555 |
+
"loss": 0.3975,
|
| 556 |
+
"step": 185
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 24.67,
|
| 560 |
+
"eval_loss": 1.0746583938598633,
|
| 561 |
+
"eval_runtime": 17.0709,
|
| 562 |
+
"eval_samples_per_second": 0.41,
|
| 563 |
+
"eval_steps_per_second": 0.059,
|
| 564 |
+
"step": 185
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 25.33,
|
| 568 |
+
"grad_norm": 2.4532127380371094,
|
| 569 |
+
"learning_rate": 4.5714285714285716e-05,
|
| 570 |
+
"loss": 0.3832,
|
| 571 |
+
"step": 190
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 25.33,
|
| 575 |
+
"eval_loss": 1.0772522687911987,
|
| 576 |
+
"eval_runtime": 17.0619,
|
| 577 |
+
"eval_samples_per_second": 0.41,
|
| 578 |
+
"eval_steps_per_second": 0.059,
|
| 579 |
+
"step": 190
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 26.0,
|
| 583 |
+
"grad_norm": 3.956822156906128,
|
| 584 |
+
"learning_rate": 4.428571428571428e-05,
|
| 585 |
+
"loss": 0.3725,
|
| 586 |
+
"step": 195
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 26.0,
|
| 590 |
+
"eval_loss": 1.0638784170150757,
|
| 591 |
+
"eval_runtime": 17.0807,
|
| 592 |
+
"eval_samples_per_second": 0.41,
|
| 593 |
+
"eval_steps_per_second": 0.059,
|
| 594 |
+
"step": 195
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 26.67,
|
| 598 |
+
"grad_norm": 2.76033353805542,
|
| 599 |
+
"learning_rate": 4.2857142857142856e-05,
|
| 600 |
+
"loss": 0.3473,
|
| 601 |
+
"step": 200
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 26.67,
|
| 605 |
+
"eval_loss": 1.04669988155365,
|
| 606 |
+
"eval_runtime": 17.0774,
|
| 607 |
+
"eval_samples_per_second": 0.41,
|
| 608 |
+
"eval_steps_per_second": 0.059,
|
| 609 |
+
"step": 200
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 27.33,
|
| 613 |
+
"grad_norm": 3.8683507442474365,
|
| 614 |
+
"learning_rate": 4.1428571428571437e-05,
|
| 615 |
+
"loss": 0.3243,
|
| 616 |
+
"step": 205
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 27.33,
|
| 620 |
+
"eval_loss": 1.0470303297042847,
|
| 621 |
+
"eval_runtime": 17.0718,
|
| 622 |
+
"eval_samples_per_second": 0.41,
|
| 623 |
+
"eval_steps_per_second": 0.059,
|
| 624 |
+
"step": 205
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 28.0,
|
| 628 |
+
"grad_norm": 4.535538196563721,
|
| 629 |
+
"learning_rate": 4e-05,
|
| 630 |
+
"loss": 0.3202,
|
| 631 |
+
"step": 210
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 28.0,
|
| 635 |
+
"eval_loss": 1.025539517402649,
|
| 636 |
+
"eval_runtime": 17.0604,
|
| 637 |
+
"eval_samples_per_second": 0.41,
|
| 638 |
+
"eval_steps_per_second": 0.059,
|
| 639 |
+
"step": 210
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 28.67,
|
| 643 |
+
"grad_norm": 2.6224355697631836,
|
| 644 |
+
"learning_rate": 3.857142857142858e-05,
|
| 645 |
+
"loss": 0.2958,
|
| 646 |
+
"step": 215
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 28.67,
|
| 650 |
+
"eval_loss": 1.0192126035690308,
|
| 651 |
+
"eval_runtime": 17.0657,
|
| 652 |
+
"eval_samples_per_second": 0.41,
|
| 653 |
+
"eval_steps_per_second": 0.059,
|
| 654 |
+
"step": 215
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 29.33,
|
| 658 |
+
"grad_norm": 2.5870041847229004,
|
| 659 |
+
"learning_rate": 3.7142857142857143e-05,
|
| 660 |
+
"loss": 0.2783,
|
| 661 |
+
"step": 220
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 29.33,
|
| 665 |
+
"eval_loss": 1.0211580991744995,
|
| 666 |
+
"eval_runtime": 17.0857,
|
| 667 |
+
"eval_samples_per_second": 0.41,
|
| 668 |
+
"eval_steps_per_second": 0.059,
|
| 669 |
+
"step": 220
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 30.0,
|
| 673 |
+
"grad_norm": 3.4565751552581787,
|
| 674 |
+
"learning_rate": 3.571428571428572e-05,
|
| 675 |
+
"loss": 0.2773,
|
| 676 |
+
"step": 225
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 30.0,
|
| 680 |
+
"eval_loss": 1.006419062614441,
|
| 681 |
+
"eval_runtime": 17.0807,
|
| 682 |
+
"eval_samples_per_second": 0.41,
|
| 683 |
+
"eval_steps_per_second": 0.059,
|
| 684 |
+
"step": 225
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 30.67,
|
| 688 |
+
"grad_norm": 2.4756500720977783,
|
| 689 |
+
"learning_rate": 3.428571428571429e-05,
|
| 690 |
+
"loss": 0.2482,
|
| 691 |
+
"step": 230
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 30.67,
|
| 695 |
+
"eval_loss": 1.0081219673156738,
|
| 696 |
+
"eval_runtime": 17.0576,
|
| 697 |
+
"eval_samples_per_second": 0.41,
|
| 698 |
+
"eval_steps_per_second": 0.059,
|
| 699 |
+
"step": 230
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 31.33,
|
| 703 |
+
"grad_norm": 2.38002610206604,
|
| 704 |
+
"learning_rate": 3.285714285714286e-05,
|
| 705 |
+
"loss": 0.2464,
|
| 706 |
+
"step": 235
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 31.33,
|
| 710 |
+
"eval_loss": 1.0151804685592651,
|
| 711 |
+
"eval_runtime": 17.0587,
|
| 712 |
+
"eval_samples_per_second": 0.41,
|
| 713 |
+
"eval_steps_per_second": 0.059,
|
| 714 |
+
"step": 235
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 32.0,
|
| 718 |
+
"grad_norm": 3.7081105709075928,
|
| 719 |
+
"learning_rate": 3.142857142857143e-05,
|
| 720 |
+
"loss": 0.2442,
|
| 721 |
+
"step": 240
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 32.0,
|
| 725 |
+
"eval_loss": 1.0032445192337036,
|
| 726 |
+
"eval_runtime": 17.1613,
|
| 727 |
+
"eval_samples_per_second": 0.408,
|
| 728 |
+
"eval_steps_per_second": 0.058,
|
| 729 |
+
"step": 240
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 32.67,
|
| 733 |
+
"grad_norm": 2.55924391746521,
|
| 734 |
+
"learning_rate": 3e-05,
|
| 735 |
+
"loss": 0.2193,
|
| 736 |
+
"step": 245
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 32.67,
|
| 740 |
+
"eval_loss": 0.9989615082740784,
|
| 741 |
+
"eval_runtime": 17.0447,
|
| 742 |
+
"eval_samples_per_second": 0.411,
|
| 743 |
+
"eval_steps_per_second": 0.059,
|
| 744 |
+
"step": 245
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 33.33,
|
| 748 |
+
"grad_norm": 1.9451407194137573,
|
| 749 |
+
"learning_rate": 2.857142857142857e-05,
|
| 750 |
+
"loss": 0.2101,
|
| 751 |
+
"step": 250
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 33.33,
|
| 755 |
+
"eval_loss": 1.0029457807540894,
|
| 756 |
+
"eval_runtime": 17.0816,
|
| 757 |
+
"eval_samples_per_second": 0.41,
|
| 758 |
+
"eval_steps_per_second": 0.059,
|
| 759 |
+
"step": 250
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 34.0,
|
| 763 |
+
"grad_norm": 2.713731527328491,
|
| 764 |
+
"learning_rate": 2.714285714285714e-05,
|
| 765 |
+
"loss": 0.2194,
|
| 766 |
+
"step": 255
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 34.0,
|
| 770 |
+
"eval_loss": 0.9959421753883362,
|
| 771 |
+
"eval_runtime": 17.0747,
|
| 772 |
+
"eval_samples_per_second": 0.41,
|
| 773 |
+
"eval_steps_per_second": 0.059,
|
| 774 |
+
"step": 255
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 34.67,
|
| 778 |
+
"grad_norm": 2.1633846759796143,
|
| 779 |
+
"learning_rate": 2.5714285714285714e-05,
|
| 780 |
+
"loss": 0.1958,
|
| 781 |
+
"step": 260
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 34.67,
|
| 785 |
+
"eval_loss": 0.9989770650863647,
|
| 786 |
+
"eval_runtime": 17.0821,
|
| 787 |
+
"eval_samples_per_second": 0.41,
|
| 788 |
+
"eval_steps_per_second": 0.059,
|
| 789 |
+
"step": 260
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 35.33,
|
| 793 |
+
"grad_norm": 3.9233529567718506,
|
| 794 |
+
"learning_rate": 2.4285714285714288e-05,
|
| 795 |
+
"loss": 0.1831,
|
| 796 |
+
"step": 265
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 35.33,
|
| 800 |
+
"eval_loss": 1.0072578191757202,
|
| 801 |
+
"eval_runtime": 17.0564,
|
| 802 |
+
"eval_samples_per_second": 0.41,
|
| 803 |
+
"eval_steps_per_second": 0.059,
|
| 804 |
+
"step": 265
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 36.0,
|
| 808 |
+
"grad_norm": 2.4143056869506836,
|
| 809 |
+
"learning_rate": 2.2857142857142858e-05,
|
| 810 |
+
"loss": 0.1753,
|
| 811 |
+
"step": 270
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 36.0,
|
| 815 |
+
"eval_loss": 0.9938892722129822,
|
| 816 |
+
"eval_runtime": 17.0668,
|
| 817 |
+
"eval_samples_per_second": 0.41,
|
| 818 |
+
"eval_steps_per_second": 0.059,
|
| 819 |
+
"step": 270
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 36.67,
|
| 823 |
+
"grad_norm": 2.706679582595825,
|
| 824 |
+
"learning_rate": 2.1428571428571428e-05,
|
| 825 |
+
"loss": 0.1698,
|
| 826 |
+
"step": 275
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 36.67,
|
| 830 |
+
"eval_loss": 0.9969200491905212,
|
| 831 |
+
"eval_runtime": 17.0643,
|
| 832 |
+
"eval_samples_per_second": 0.41,
|
| 833 |
+
"eval_steps_per_second": 0.059,
|
| 834 |
+
"step": 275
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 37.33,
|
| 838 |
+
"grad_norm": 1.872753620147705,
|
| 839 |
+
"learning_rate": 2e-05,
|
| 840 |
+
"loss": 0.16,
|
| 841 |
+
"step": 280
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 37.33,
|
| 845 |
+
"eval_loss": 0.9940390586853027,
|
| 846 |
+
"eval_runtime": 17.0728,
|
| 847 |
+
"eval_samples_per_second": 0.41,
|
| 848 |
+
"eval_steps_per_second": 0.059,
|
| 849 |
+
"step": 280
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 38.0,
|
| 853 |
+
"grad_norm": 2.7510581016540527,
|
| 854 |
+
"learning_rate": 1.8571428571428572e-05,
|
| 855 |
+
"loss": 0.1614,
|
| 856 |
+
"step": 285
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 38.0,
|
| 860 |
+
"eval_loss": 1.0066231489181519,
|
| 861 |
+
"eval_runtime": 17.072,
|
| 862 |
+
"eval_samples_per_second": 0.41,
|
| 863 |
+
"eval_steps_per_second": 0.059,
|
| 864 |
+
"step": 285
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 38.67,
|
| 868 |
+
"grad_norm": 1.8461092710494995,
|
| 869 |
+
"learning_rate": 1.7142857142857145e-05,
|
| 870 |
+
"loss": 0.1506,
|
| 871 |
+
"step": 290
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 38.67,
|
| 875 |
+
"eval_loss": 0.9927281737327576,
|
| 876 |
+
"eval_runtime": 17.0481,
|
| 877 |
+
"eval_samples_per_second": 0.411,
|
| 878 |
+
"eval_steps_per_second": 0.059,
|
| 879 |
+
"step": 290
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 39.33,
|
| 883 |
+
"grad_norm": 1.8425017595291138,
|
| 884 |
+
"learning_rate": 1.5714285714285715e-05,
|
| 885 |
+
"loss": 0.1419,
|
| 886 |
+
"step": 295
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 39.33,
|
| 890 |
+
"eval_loss": 1.0133570432662964,
|
| 891 |
+
"eval_runtime": 17.0642,
|
| 892 |
+
"eval_samples_per_second": 0.41,
|
| 893 |
+
"eval_steps_per_second": 0.059,
|
| 894 |
+
"step": 295
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 40.0,
|
| 898 |
+
"grad_norm": 2.0457987785339355,
|
| 899 |
+
"learning_rate": 1.4285714285714285e-05,
|
| 900 |
+
"loss": 0.1459,
|
| 901 |
+
"step": 300
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 40.0,
|
| 905 |
+
"eval_loss": 1.0127934217453003,
|
| 906 |
+
"eval_runtime": 17.0581,
|
| 907 |
+
"eval_samples_per_second": 0.41,
|
| 908 |
+
"eval_steps_per_second": 0.059,
|
| 909 |
+
"step": 300
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 40.67,
|
| 913 |
+
"grad_norm": 1.5630775690078735,
|
| 914 |
+
"learning_rate": 1.2857142857142857e-05,
|
| 915 |
+
"loss": 0.1225,
|
| 916 |
+
"step": 305
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 40.67,
|
| 920 |
+
"eval_loss": 1.0092624425888062,
|
| 921 |
+
"eval_runtime": 17.0483,
|
| 922 |
+
"eval_samples_per_second": 0.411,
|
| 923 |
+
"eval_steps_per_second": 0.059,
|
| 924 |
+
"step": 305
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 41.33,
|
| 928 |
+
"grad_norm": 1.37598717212677,
|
| 929 |
+
"learning_rate": 1.1428571428571429e-05,
|
| 930 |
+
"loss": 0.146,
|
| 931 |
+
"step": 310
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 41.33,
|
| 935 |
+
"eval_loss": 1.0083317756652832,
|
| 936 |
+
"eval_runtime": 17.0804,
|
| 937 |
+
"eval_samples_per_second": 0.41,
|
| 938 |
+
"eval_steps_per_second": 0.059,
|
| 939 |
+
"step": 310
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 42.0,
|
| 943 |
+
"grad_norm": 1.8867217302322388,
|
| 944 |
+
"learning_rate": 1e-05,
|
| 945 |
+
"loss": 0.13,
|
| 946 |
+
"step": 315
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 42.0,
|
| 950 |
+
"eval_loss": 1.0165104866027832,
|
| 951 |
+
"eval_runtime": 17.0621,
|
| 952 |
+
"eval_samples_per_second": 0.41,
|
| 953 |
+
"eval_steps_per_second": 0.059,
|
| 954 |
+
"step": 315
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 42.67,
|
| 958 |
+
"grad_norm": 1.4643555879592896,
|
| 959 |
+
"learning_rate": 8.571428571428573e-06,
|
| 960 |
+
"loss": 0.131,
|
| 961 |
+
"step": 320
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 42.67,
|
| 965 |
+
"eval_loss": 1.0264887809753418,
|
| 966 |
+
"eval_runtime": 17.0554,
|
| 967 |
+
"eval_samples_per_second": 0.41,
|
| 968 |
+
"eval_steps_per_second": 0.059,
|
| 969 |
+
"step": 320
|
| 970 |
+
}
|
| 971 |
+
],
|
| 972 |
+
"logging_steps": 5,
|
| 973 |
+
"max_steps": 350,
|
| 974 |
+
"num_input_tokens_seen": 0,
|
| 975 |
+
"num_train_epochs": 50,
|
| 976 |
+
"save_steps": 10,
|
| 977 |
+
"total_flos": 2.1266150580320993e+18,
|
| 978 |
+
"train_batch_size": 2,
|
| 979 |
+
"trial_name": null,
|
| 980 |
+
"trial_params": null
|
| 981 |
+
}
|
checkpoint-320/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ed0ba6fb9a88dad56f61d9306f17b4e66e8767d898772faa97871a1388e82cf
|
| 3 |
+
size 4920
|
checkpoint-330/README.md
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
|
| 7 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 8 |
+
- quant_method: bitsandbytes
|
| 9 |
+
- _load_in_8bit: False
|
| 10 |
+
- _load_in_4bit: False
|
| 11 |
+
- llm_int8_threshold: 6.0
|
| 12 |
+
- llm_int8_skip_modules: None
|
| 13 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 14 |
+
- llm_int8_has_fp16_weight: False
|
| 15 |
+
- bnb_4bit_quant_type: fp4
|
| 16 |
+
- bnb_4bit_use_double_quant: False
|
| 17 |
+
- bnb_4bit_compute_dtype: float32
|
| 18 |
+
- load_in_4bit: False
|
| 19 |
+
- load_in_8bit: False
|
| 20 |
+
### Framework versions
|
| 21 |
+
|
| 22 |
+
|
| 23 |
+
- PEFT 0.5.0
|
checkpoint-330/adapter_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "meta-llama/Llama-2-70b-hf",
|
| 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": 32,
|
| 11 |
+
"lora_dropout": 0.05,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 8,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"q_proj",
|
| 18 |
+
"v_proj"
|
| 19 |
+
],
|
| 20 |
+
"task_type": "CAUSAL_LM"
|
| 21 |
+
}
|
checkpoint-330/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5505e257fa18af21b6af22ca968df21297c4fb92614c82743718e81cf858cf9c
|
| 3 |
+
size 65578776
|
checkpoint-330/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8098efa09a64b1fcc8326dd62e171a71ab1c0f95e20015654d914ad8fbfb5880
|
| 3 |
+
size 131345914
|
checkpoint-330/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:42b24e02a21b68f4542793debee790ccdfacee8b8c9cad8a58a3c5770a53b720
|
| 3 |
+
size 14244
|
checkpoint-330/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aa7302be973ca42b6ea8fae42686b6439278ad5d7647a75f6503a767b9e0f09c
|
| 3 |
+
size 1064
|
checkpoint-330/trainer_state.json
ADDED
|
@@ -0,0 +1,1011 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9927281737327576,
|
| 3 |
+
"best_model_checkpoint": "/scratch/kwamea/llama-output/checkpoint-290",
|
| 4 |
+
"epoch": 44.0,
|
| 5 |
+
"eval_steps": 5,
|
| 6 |
+
"global_step": 330,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.67,
|
| 13 |
+
"grad_norm": 0.243179589509964,
|
| 14 |
+
"learning_rate": 9.857142857142858e-05,
|
| 15 |
+
"loss": 1.9956,
|
| 16 |
+
"step": 5
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.67,
|
| 20 |
+
"eval_loss": 1.9701930284500122,
|
| 21 |
+
"eval_runtime": 17.115,
|
| 22 |
+
"eval_samples_per_second": 0.409,
|
| 23 |
+
"eval_steps_per_second": 0.058,
|
| 24 |
+
"step": 5
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 1.33,
|
| 28 |
+
"grad_norm": 0.34590908885002136,
|
| 29 |
+
"learning_rate": 9.714285714285715e-05,
|
| 30 |
+
"loss": 1.9758,
|
| 31 |
+
"step": 10
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 1.33,
|
| 35 |
+
"eval_loss": 1.8941271305084229,
|
| 36 |
+
"eval_runtime": 17.0912,
|
| 37 |
+
"eval_samples_per_second": 0.41,
|
| 38 |
+
"eval_steps_per_second": 0.059,
|
| 39 |
+
"step": 10
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 2.0,
|
| 43 |
+
"grad_norm": 0.31595832109451294,
|
| 44 |
+
"learning_rate": 9.571428571428573e-05,
|
| 45 |
+
"loss": 1.849,
|
| 46 |
+
"step": 15
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 2.0,
|
| 50 |
+
"eval_loss": 1.8046789169311523,
|
| 51 |
+
"eval_runtime": 17.098,
|
| 52 |
+
"eval_samples_per_second": 0.409,
|
| 53 |
+
"eval_steps_per_second": 0.058,
|
| 54 |
+
"step": 15
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 2.67,
|
| 58 |
+
"grad_norm": 0.3428090512752533,
|
| 59 |
+
"learning_rate": 9.428571428571429e-05,
|
| 60 |
+
"loss": 1.789,
|
| 61 |
+
"step": 20
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 2.67,
|
| 65 |
+
"eval_loss": 1.7658358812332153,
|
| 66 |
+
"eval_runtime": 17.0734,
|
| 67 |
+
"eval_samples_per_second": 0.41,
|
| 68 |
+
"eval_steps_per_second": 0.059,
|
| 69 |
+
"step": 20
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 3.33,
|
| 73 |
+
"grad_norm": 0.3102028965950012,
|
| 74 |
+
"learning_rate": 9.285714285714286e-05,
|
| 75 |
+
"loss": 1.7789,
|
| 76 |
+
"step": 25
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 3.33,
|
| 80 |
+
"eval_loss": 1.7225048542022705,
|
| 81 |
+
"eval_runtime": 17.0972,
|
| 82 |
+
"eval_samples_per_second": 0.409,
|
| 83 |
+
"eval_steps_per_second": 0.058,
|
| 84 |
+
"step": 25
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 4.0,
|
| 88 |
+
"grad_norm": 0.38602885603904724,
|
| 89 |
+
"learning_rate": 9.142857142857143e-05,
|
| 90 |
+
"loss": 1.7003,
|
| 91 |
+
"step": 30
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 4.0,
|
| 95 |
+
"eval_loss": 1.6749440431594849,
|
| 96 |
+
"eval_runtime": 17.1034,
|
| 97 |
+
"eval_samples_per_second": 0.409,
|
| 98 |
+
"eval_steps_per_second": 0.058,
|
| 99 |
+
"step": 30
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 4.67,
|
| 103 |
+
"grad_norm": 0.37120407819747925,
|
| 104 |
+
"learning_rate": 9e-05,
|
| 105 |
+
"loss": 1.6424,
|
| 106 |
+
"step": 35
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 4.67,
|
| 110 |
+
"eval_loss": 1.6231099367141724,
|
| 111 |
+
"eval_runtime": 17.1067,
|
| 112 |
+
"eval_samples_per_second": 0.409,
|
| 113 |
+
"eval_steps_per_second": 0.058,
|
| 114 |
+
"step": 35
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 5.33,
|
| 118 |
+
"grad_norm": 0.4633428454399109,
|
| 119 |
+
"learning_rate": 8.857142857142857e-05,
|
| 120 |
+
"loss": 1.6023,
|
| 121 |
+
"step": 40
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 5.33,
|
| 125 |
+
"eval_loss": 1.5727053880691528,
|
| 126 |
+
"eval_runtime": 17.1002,
|
| 127 |
+
"eval_samples_per_second": 0.409,
|
| 128 |
+
"eval_steps_per_second": 0.058,
|
| 129 |
+
"step": 40
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 6.0,
|
| 133 |
+
"grad_norm": 0.5034663081169128,
|
| 134 |
+
"learning_rate": 8.714285714285715e-05,
|
| 135 |
+
"loss": 1.5322,
|
| 136 |
+
"step": 45
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 6.0,
|
| 140 |
+
"eval_loss": 1.5312587022781372,
|
| 141 |
+
"eval_runtime": 17.1159,
|
| 142 |
+
"eval_samples_per_second": 0.409,
|
| 143 |
+
"eval_steps_per_second": 0.058,
|
| 144 |
+
"step": 45
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 6.67,
|
| 148 |
+
"grad_norm": 0.5549929141998291,
|
| 149 |
+
"learning_rate": 8.571428571428571e-05,
|
| 150 |
+
"loss": 1.4788,
|
| 151 |
+
"step": 50
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 6.67,
|
| 155 |
+
"eval_loss": 1.492464303970337,
|
| 156 |
+
"eval_runtime": 17.0823,
|
| 157 |
+
"eval_samples_per_second": 0.41,
|
| 158 |
+
"eval_steps_per_second": 0.059,
|
| 159 |
+
"step": 50
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 7.33,
|
| 163 |
+
"grad_norm": 0.49194690585136414,
|
| 164 |
+
"learning_rate": 8.428571428571429e-05,
|
| 165 |
+
"loss": 1.4632,
|
| 166 |
+
"step": 55
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 7.33,
|
| 170 |
+
"eval_loss": 1.4622489213943481,
|
| 171 |
+
"eval_runtime": 17.1022,
|
| 172 |
+
"eval_samples_per_second": 0.409,
|
| 173 |
+
"eval_steps_per_second": 0.058,
|
| 174 |
+
"step": 55
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 8.0,
|
| 178 |
+
"grad_norm": 0.5866131782531738,
|
| 179 |
+
"learning_rate": 8.285714285714287e-05,
|
| 180 |
+
"loss": 1.3951,
|
| 181 |
+
"step": 60
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 8.0,
|
| 185 |
+
"eval_loss": 1.435951828956604,
|
| 186 |
+
"eval_runtime": 17.1087,
|
| 187 |
+
"eval_samples_per_second": 0.409,
|
| 188 |
+
"eval_steps_per_second": 0.058,
|
| 189 |
+
"step": 60
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 8.67,
|
| 193 |
+
"grad_norm": 0.6252542734146118,
|
| 194 |
+
"learning_rate": 8.142857142857143e-05,
|
| 195 |
+
"loss": 1.3796,
|
| 196 |
+
"step": 65
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 8.67,
|
| 200 |
+
"eval_loss": 1.413227915763855,
|
| 201 |
+
"eval_runtime": 17.0914,
|
| 202 |
+
"eval_samples_per_second": 0.41,
|
| 203 |
+
"eval_steps_per_second": 0.059,
|
| 204 |
+
"step": 65
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 9.33,
|
| 208 |
+
"grad_norm": 0.6751863360404968,
|
| 209 |
+
"learning_rate": 8e-05,
|
| 210 |
+
"loss": 1.3257,
|
| 211 |
+
"step": 70
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 9.33,
|
| 215 |
+
"eval_loss": 1.395649790763855,
|
| 216 |
+
"eval_runtime": 17.0885,
|
| 217 |
+
"eval_samples_per_second": 0.41,
|
| 218 |
+
"eval_steps_per_second": 0.059,
|
| 219 |
+
"step": 70
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 10.0,
|
| 223 |
+
"grad_norm": 0.8878222703933716,
|
| 224 |
+
"learning_rate": 7.857142857142858e-05,
|
| 225 |
+
"loss": 1.2795,
|
| 226 |
+
"step": 75
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 10.0,
|
| 230 |
+
"eval_loss": 1.3699487447738647,
|
| 231 |
+
"eval_runtime": 17.1031,
|
| 232 |
+
"eval_samples_per_second": 0.409,
|
| 233 |
+
"eval_steps_per_second": 0.058,
|
| 234 |
+
"step": 75
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 10.67,
|
| 238 |
+
"grad_norm": 0.8470121026039124,
|
| 239 |
+
"learning_rate": 7.714285714285715e-05,
|
| 240 |
+
"loss": 1.2449,
|
| 241 |
+
"step": 80
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 10.67,
|
| 245 |
+
"eval_loss": 1.347831130027771,
|
| 246 |
+
"eval_runtime": 17.0985,
|
| 247 |
+
"eval_samples_per_second": 0.409,
|
| 248 |
+
"eval_steps_per_second": 0.058,
|
| 249 |
+
"step": 80
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 11.33,
|
| 253 |
+
"grad_norm": 1.0655425786972046,
|
| 254 |
+
"learning_rate": 7.571428571428571e-05,
|
| 255 |
+
"loss": 1.1983,
|
| 256 |
+
"step": 85
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 11.33,
|
| 260 |
+
"eval_loss": 1.3311971426010132,
|
| 261 |
+
"eval_runtime": 17.0784,
|
| 262 |
+
"eval_samples_per_second": 0.41,
|
| 263 |
+
"eval_steps_per_second": 0.059,
|
| 264 |
+
"step": 85
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 12.0,
|
| 268 |
+
"grad_norm": 1.2651888132095337,
|
| 269 |
+
"learning_rate": 7.428571428571429e-05,
|
| 270 |
+
"loss": 1.1467,
|
| 271 |
+
"step": 90
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 12.0,
|
| 275 |
+
"eval_loss": 1.3095277547836304,
|
| 276 |
+
"eval_runtime": 17.0903,
|
| 277 |
+
"eval_samples_per_second": 0.41,
|
| 278 |
+
"eval_steps_per_second": 0.059,
|
| 279 |
+
"step": 90
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 12.67,
|
| 283 |
+
"grad_norm": 1.248926043510437,
|
| 284 |
+
"learning_rate": 7.285714285714286e-05,
|
| 285 |
+
"loss": 1.0922,
|
| 286 |
+
"step": 95
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 12.67,
|
| 290 |
+
"eval_loss": 1.2942878007888794,
|
| 291 |
+
"eval_runtime": 17.0947,
|
| 292 |
+
"eval_samples_per_second": 0.409,
|
| 293 |
+
"eval_steps_per_second": 0.058,
|
| 294 |
+
"step": 95
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 13.33,
|
| 298 |
+
"grad_norm": 1.896952509880066,
|
| 299 |
+
"learning_rate": 7.142857142857143e-05,
|
| 300 |
+
"loss": 1.0403,
|
| 301 |
+
"step": 100
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 13.33,
|
| 305 |
+
"eval_loss": 1.2803159952163696,
|
| 306 |
+
"eval_runtime": 17.0819,
|
| 307 |
+
"eval_samples_per_second": 0.41,
|
| 308 |
+
"eval_steps_per_second": 0.059,
|
| 309 |
+
"step": 100
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 14.0,
|
| 313 |
+
"grad_norm": 1.862244725227356,
|
| 314 |
+
"learning_rate": 7e-05,
|
| 315 |
+
"loss": 1.0049,
|
| 316 |
+
"step": 105
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 14.0,
|
| 320 |
+
"eval_loss": 1.2643567323684692,
|
| 321 |
+
"eval_runtime": 17.0849,
|
| 322 |
+
"eval_samples_per_second": 0.41,
|
| 323 |
+
"eval_steps_per_second": 0.059,
|
| 324 |
+
"step": 105
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 14.67,
|
| 328 |
+
"grad_norm": 1.7487821578979492,
|
| 329 |
+
"learning_rate": 6.857142857142858e-05,
|
| 330 |
+
"loss": 0.9262,
|
| 331 |
+
"step": 110
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 14.67,
|
| 335 |
+
"eval_loss": 1.2471646070480347,
|
| 336 |
+
"eval_runtime": 17.1278,
|
| 337 |
+
"eval_samples_per_second": 0.409,
|
| 338 |
+
"eval_steps_per_second": 0.058,
|
| 339 |
+
"step": 110
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 15.33,
|
| 343 |
+
"grad_norm": 1.838605284690857,
|
| 344 |
+
"learning_rate": 6.714285714285714e-05,
|
| 345 |
+
"loss": 0.8965,
|
| 346 |
+
"step": 115
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 15.33,
|
| 350 |
+
"eval_loss": 1.2377034425735474,
|
| 351 |
+
"eval_runtime": 17.0731,
|
| 352 |
+
"eval_samples_per_second": 0.41,
|
| 353 |
+
"eval_steps_per_second": 0.059,
|
| 354 |
+
"step": 115
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 16.0,
|
| 358 |
+
"grad_norm": 3.117398977279663,
|
| 359 |
+
"learning_rate": 6.571428571428571e-05,
|
| 360 |
+
"loss": 0.8581,
|
| 361 |
+
"step": 120
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 16.0,
|
| 365 |
+
"eval_loss": 1.2083133459091187,
|
| 366 |
+
"eval_runtime": 17.1304,
|
| 367 |
+
"eval_samples_per_second": 0.409,
|
| 368 |
+
"eval_steps_per_second": 0.058,
|
| 369 |
+
"step": 120
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 16.67,
|
| 373 |
+
"grad_norm": 2.5655250549316406,
|
| 374 |
+
"learning_rate": 6.428571428571429e-05,
|
| 375 |
+
"loss": 0.7929,
|
| 376 |
+
"step": 125
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 16.67,
|
| 380 |
+
"eval_loss": 1.1945828199386597,
|
| 381 |
+
"eval_runtime": 17.104,
|
| 382 |
+
"eval_samples_per_second": 0.409,
|
| 383 |
+
"eval_steps_per_second": 0.058,
|
| 384 |
+
"step": 125
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 17.33,
|
| 388 |
+
"grad_norm": 2.168546199798584,
|
| 389 |
+
"learning_rate": 6.285714285714286e-05,
|
| 390 |
+
"loss": 0.7543,
|
| 391 |
+
"step": 130
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 17.33,
|
| 395 |
+
"eval_loss": 1.1876276731491089,
|
| 396 |
+
"eval_runtime": 17.1046,
|
| 397 |
+
"eval_samples_per_second": 0.409,
|
| 398 |
+
"eval_steps_per_second": 0.058,
|
| 399 |
+
"step": 130
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 18.0,
|
| 403 |
+
"grad_norm": 2.5984208583831787,
|
| 404 |
+
"learning_rate": 6.142857142857143e-05,
|
| 405 |
+
"loss": 0.716,
|
| 406 |
+
"step": 135
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 18.0,
|
| 410 |
+
"eval_loss": 1.1714750528335571,
|
| 411 |
+
"eval_runtime": 17.0807,
|
| 412 |
+
"eval_samples_per_second": 0.41,
|
| 413 |
+
"eval_steps_per_second": 0.059,
|
| 414 |
+
"step": 135
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 18.67,
|
| 418 |
+
"grad_norm": 3.479024887084961,
|
| 419 |
+
"learning_rate": 6e-05,
|
| 420 |
+
"loss": 0.6681,
|
| 421 |
+
"step": 140
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 18.67,
|
| 425 |
+
"eval_loss": 1.169895052909851,
|
| 426 |
+
"eval_runtime": 17.0681,
|
| 427 |
+
"eval_samples_per_second": 0.41,
|
| 428 |
+
"eval_steps_per_second": 0.059,
|
| 429 |
+
"step": 140
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 19.33,
|
| 433 |
+
"grad_norm": 2.563386917114258,
|
| 434 |
+
"learning_rate": 5.8571428571428575e-05,
|
| 435 |
+
"loss": 0.6306,
|
| 436 |
+
"step": 145
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 19.33,
|
| 440 |
+
"eval_loss": 1.1741083860397339,
|
| 441 |
+
"eval_runtime": 17.0568,
|
| 442 |
+
"eval_samples_per_second": 0.41,
|
| 443 |
+
"eval_steps_per_second": 0.059,
|
| 444 |
+
"step": 145
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 20.0,
|
| 448 |
+
"grad_norm": 2.96592116355896,
|
| 449 |
+
"learning_rate": 5.714285714285714e-05,
|
| 450 |
+
"loss": 0.6183,
|
| 451 |
+
"step": 150
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 20.0,
|
| 455 |
+
"eval_loss": 1.1455965042114258,
|
| 456 |
+
"eval_runtime": 17.073,
|
| 457 |
+
"eval_samples_per_second": 0.41,
|
| 458 |
+
"eval_steps_per_second": 0.059,
|
| 459 |
+
"step": 150
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 20.67,
|
| 463 |
+
"grad_norm": 2.6751275062561035,
|
| 464 |
+
"learning_rate": 5.571428571428572e-05,
|
| 465 |
+
"loss": 0.5464,
|
| 466 |
+
"step": 155
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 20.67,
|
| 470 |
+
"eval_loss": 1.131102204322815,
|
| 471 |
+
"eval_runtime": 17.0578,
|
| 472 |
+
"eval_samples_per_second": 0.41,
|
| 473 |
+
"eval_steps_per_second": 0.059,
|
| 474 |
+
"step": 155
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 21.33,
|
| 478 |
+
"grad_norm": 2.3700051307678223,
|
| 479 |
+
"learning_rate": 5.428571428571428e-05,
|
| 480 |
+
"loss": 0.551,
|
| 481 |
+
"step": 160
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 21.33,
|
| 485 |
+
"eval_loss": 1.127384066581726,
|
| 486 |
+
"eval_runtime": 17.0546,
|
| 487 |
+
"eval_samples_per_second": 0.41,
|
| 488 |
+
"eval_steps_per_second": 0.059,
|
| 489 |
+
"step": 160
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 22.0,
|
| 493 |
+
"grad_norm": 3.3827567100524902,
|
| 494 |
+
"learning_rate": 5.285714285714286e-05,
|
| 495 |
+
"loss": 0.5179,
|
| 496 |
+
"step": 165
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 22.0,
|
| 500 |
+
"eval_loss": 1.111584186553955,
|
| 501 |
+
"eval_runtime": 17.0812,
|
| 502 |
+
"eval_samples_per_second": 0.41,
|
| 503 |
+
"eval_steps_per_second": 0.059,
|
| 504 |
+
"step": 165
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 22.67,
|
| 508 |
+
"grad_norm": 3.55790114402771,
|
| 509 |
+
"learning_rate": 5.142857142857143e-05,
|
| 510 |
+
"loss": 0.4831,
|
| 511 |
+
"step": 170
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 22.67,
|
| 515 |
+
"eval_loss": 1.0948525667190552,
|
| 516 |
+
"eval_runtime": 17.0547,
|
| 517 |
+
"eval_samples_per_second": 0.41,
|
| 518 |
+
"eval_steps_per_second": 0.059,
|
| 519 |
+
"step": 170
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 23.33,
|
| 523 |
+
"grad_norm": 3.0782699584960938,
|
| 524 |
+
"learning_rate": 5e-05,
|
| 525 |
+
"loss": 0.4587,
|
| 526 |
+
"step": 175
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 23.33,
|
| 530 |
+
"eval_loss": 1.0906586647033691,
|
| 531 |
+
"eval_runtime": 17.0666,
|
| 532 |
+
"eval_samples_per_second": 0.41,
|
| 533 |
+
"eval_steps_per_second": 0.059,
|
| 534 |
+
"step": 175
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 24.0,
|
| 538 |
+
"grad_norm": 3.3993167877197266,
|
| 539 |
+
"learning_rate": 4.8571428571428576e-05,
|
| 540 |
+
"loss": 0.4203,
|
| 541 |
+
"step": 180
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 24.0,
|
| 545 |
+
"eval_loss": 1.0688152313232422,
|
| 546 |
+
"eval_runtime": 17.0721,
|
| 547 |
+
"eval_samples_per_second": 0.41,
|
| 548 |
+
"eval_steps_per_second": 0.059,
|
| 549 |
+
"step": 180
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 24.67,
|
| 553 |
+
"grad_norm": 3.319303035736084,
|
| 554 |
+
"learning_rate": 4.714285714285714e-05,
|
| 555 |
+
"loss": 0.3975,
|
| 556 |
+
"step": 185
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 24.67,
|
| 560 |
+
"eval_loss": 1.0746583938598633,
|
| 561 |
+
"eval_runtime": 17.0709,
|
| 562 |
+
"eval_samples_per_second": 0.41,
|
| 563 |
+
"eval_steps_per_second": 0.059,
|
| 564 |
+
"step": 185
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 25.33,
|
| 568 |
+
"grad_norm": 2.4532127380371094,
|
| 569 |
+
"learning_rate": 4.5714285714285716e-05,
|
| 570 |
+
"loss": 0.3832,
|
| 571 |
+
"step": 190
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 25.33,
|
| 575 |
+
"eval_loss": 1.0772522687911987,
|
| 576 |
+
"eval_runtime": 17.0619,
|
| 577 |
+
"eval_samples_per_second": 0.41,
|
| 578 |
+
"eval_steps_per_second": 0.059,
|
| 579 |
+
"step": 190
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 26.0,
|
| 583 |
+
"grad_norm": 3.956822156906128,
|
| 584 |
+
"learning_rate": 4.428571428571428e-05,
|
| 585 |
+
"loss": 0.3725,
|
| 586 |
+
"step": 195
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 26.0,
|
| 590 |
+
"eval_loss": 1.0638784170150757,
|
| 591 |
+
"eval_runtime": 17.0807,
|
| 592 |
+
"eval_samples_per_second": 0.41,
|
| 593 |
+
"eval_steps_per_second": 0.059,
|
| 594 |
+
"step": 195
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 26.67,
|
| 598 |
+
"grad_norm": 2.76033353805542,
|
| 599 |
+
"learning_rate": 4.2857142857142856e-05,
|
| 600 |
+
"loss": 0.3473,
|
| 601 |
+
"step": 200
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 26.67,
|
| 605 |
+
"eval_loss": 1.04669988155365,
|
| 606 |
+
"eval_runtime": 17.0774,
|
| 607 |
+
"eval_samples_per_second": 0.41,
|
| 608 |
+
"eval_steps_per_second": 0.059,
|
| 609 |
+
"step": 200
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 27.33,
|
| 613 |
+
"grad_norm": 3.8683507442474365,
|
| 614 |
+
"learning_rate": 4.1428571428571437e-05,
|
| 615 |
+
"loss": 0.3243,
|
| 616 |
+
"step": 205
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 27.33,
|
| 620 |
+
"eval_loss": 1.0470303297042847,
|
| 621 |
+
"eval_runtime": 17.0718,
|
| 622 |
+
"eval_samples_per_second": 0.41,
|
| 623 |
+
"eval_steps_per_second": 0.059,
|
| 624 |
+
"step": 205
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 28.0,
|
| 628 |
+
"grad_norm": 4.535538196563721,
|
| 629 |
+
"learning_rate": 4e-05,
|
| 630 |
+
"loss": 0.3202,
|
| 631 |
+
"step": 210
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 28.0,
|
| 635 |
+
"eval_loss": 1.025539517402649,
|
| 636 |
+
"eval_runtime": 17.0604,
|
| 637 |
+
"eval_samples_per_second": 0.41,
|
| 638 |
+
"eval_steps_per_second": 0.059,
|
| 639 |
+
"step": 210
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 28.67,
|
| 643 |
+
"grad_norm": 2.6224355697631836,
|
| 644 |
+
"learning_rate": 3.857142857142858e-05,
|
| 645 |
+
"loss": 0.2958,
|
| 646 |
+
"step": 215
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 28.67,
|
| 650 |
+
"eval_loss": 1.0192126035690308,
|
| 651 |
+
"eval_runtime": 17.0657,
|
| 652 |
+
"eval_samples_per_second": 0.41,
|
| 653 |
+
"eval_steps_per_second": 0.059,
|
| 654 |
+
"step": 215
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 29.33,
|
| 658 |
+
"grad_norm": 2.5870041847229004,
|
| 659 |
+
"learning_rate": 3.7142857142857143e-05,
|
| 660 |
+
"loss": 0.2783,
|
| 661 |
+
"step": 220
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 29.33,
|
| 665 |
+
"eval_loss": 1.0211580991744995,
|
| 666 |
+
"eval_runtime": 17.0857,
|
| 667 |
+
"eval_samples_per_second": 0.41,
|
| 668 |
+
"eval_steps_per_second": 0.059,
|
| 669 |
+
"step": 220
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 30.0,
|
| 673 |
+
"grad_norm": 3.4565751552581787,
|
| 674 |
+
"learning_rate": 3.571428571428572e-05,
|
| 675 |
+
"loss": 0.2773,
|
| 676 |
+
"step": 225
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 30.0,
|
| 680 |
+
"eval_loss": 1.006419062614441,
|
| 681 |
+
"eval_runtime": 17.0807,
|
| 682 |
+
"eval_samples_per_second": 0.41,
|
| 683 |
+
"eval_steps_per_second": 0.059,
|
| 684 |
+
"step": 225
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 30.67,
|
| 688 |
+
"grad_norm": 2.4756500720977783,
|
| 689 |
+
"learning_rate": 3.428571428571429e-05,
|
| 690 |
+
"loss": 0.2482,
|
| 691 |
+
"step": 230
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 30.67,
|
| 695 |
+
"eval_loss": 1.0081219673156738,
|
| 696 |
+
"eval_runtime": 17.0576,
|
| 697 |
+
"eval_samples_per_second": 0.41,
|
| 698 |
+
"eval_steps_per_second": 0.059,
|
| 699 |
+
"step": 230
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 31.33,
|
| 703 |
+
"grad_norm": 2.38002610206604,
|
| 704 |
+
"learning_rate": 3.285714285714286e-05,
|
| 705 |
+
"loss": 0.2464,
|
| 706 |
+
"step": 235
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 31.33,
|
| 710 |
+
"eval_loss": 1.0151804685592651,
|
| 711 |
+
"eval_runtime": 17.0587,
|
| 712 |
+
"eval_samples_per_second": 0.41,
|
| 713 |
+
"eval_steps_per_second": 0.059,
|
| 714 |
+
"step": 235
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 32.0,
|
| 718 |
+
"grad_norm": 3.7081105709075928,
|
| 719 |
+
"learning_rate": 3.142857142857143e-05,
|
| 720 |
+
"loss": 0.2442,
|
| 721 |
+
"step": 240
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 32.0,
|
| 725 |
+
"eval_loss": 1.0032445192337036,
|
| 726 |
+
"eval_runtime": 17.1613,
|
| 727 |
+
"eval_samples_per_second": 0.408,
|
| 728 |
+
"eval_steps_per_second": 0.058,
|
| 729 |
+
"step": 240
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 32.67,
|
| 733 |
+
"grad_norm": 2.55924391746521,
|
| 734 |
+
"learning_rate": 3e-05,
|
| 735 |
+
"loss": 0.2193,
|
| 736 |
+
"step": 245
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 32.67,
|
| 740 |
+
"eval_loss": 0.9989615082740784,
|
| 741 |
+
"eval_runtime": 17.0447,
|
| 742 |
+
"eval_samples_per_second": 0.411,
|
| 743 |
+
"eval_steps_per_second": 0.059,
|
| 744 |
+
"step": 245
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 33.33,
|
| 748 |
+
"grad_norm": 1.9451407194137573,
|
| 749 |
+
"learning_rate": 2.857142857142857e-05,
|
| 750 |
+
"loss": 0.2101,
|
| 751 |
+
"step": 250
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 33.33,
|
| 755 |
+
"eval_loss": 1.0029457807540894,
|
| 756 |
+
"eval_runtime": 17.0816,
|
| 757 |
+
"eval_samples_per_second": 0.41,
|
| 758 |
+
"eval_steps_per_second": 0.059,
|
| 759 |
+
"step": 250
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 34.0,
|
| 763 |
+
"grad_norm": 2.713731527328491,
|
| 764 |
+
"learning_rate": 2.714285714285714e-05,
|
| 765 |
+
"loss": 0.2194,
|
| 766 |
+
"step": 255
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 34.0,
|
| 770 |
+
"eval_loss": 0.9959421753883362,
|
| 771 |
+
"eval_runtime": 17.0747,
|
| 772 |
+
"eval_samples_per_second": 0.41,
|
| 773 |
+
"eval_steps_per_second": 0.059,
|
| 774 |
+
"step": 255
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 34.67,
|
| 778 |
+
"grad_norm": 2.1633846759796143,
|
| 779 |
+
"learning_rate": 2.5714285714285714e-05,
|
| 780 |
+
"loss": 0.1958,
|
| 781 |
+
"step": 260
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 34.67,
|
| 785 |
+
"eval_loss": 0.9989770650863647,
|
| 786 |
+
"eval_runtime": 17.0821,
|
| 787 |
+
"eval_samples_per_second": 0.41,
|
| 788 |
+
"eval_steps_per_second": 0.059,
|
| 789 |
+
"step": 260
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 35.33,
|
| 793 |
+
"grad_norm": 3.9233529567718506,
|
| 794 |
+
"learning_rate": 2.4285714285714288e-05,
|
| 795 |
+
"loss": 0.1831,
|
| 796 |
+
"step": 265
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 35.33,
|
| 800 |
+
"eval_loss": 1.0072578191757202,
|
| 801 |
+
"eval_runtime": 17.0564,
|
| 802 |
+
"eval_samples_per_second": 0.41,
|
| 803 |
+
"eval_steps_per_second": 0.059,
|
| 804 |
+
"step": 265
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 36.0,
|
| 808 |
+
"grad_norm": 2.4143056869506836,
|
| 809 |
+
"learning_rate": 2.2857142857142858e-05,
|
| 810 |
+
"loss": 0.1753,
|
| 811 |
+
"step": 270
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 36.0,
|
| 815 |
+
"eval_loss": 0.9938892722129822,
|
| 816 |
+
"eval_runtime": 17.0668,
|
| 817 |
+
"eval_samples_per_second": 0.41,
|
| 818 |
+
"eval_steps_per_second": 0.059,
|
| 819 |
+
"step": 270
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 36.67,
|
| 823 |
+
"grad_norm": 2.706679582595825,
|
| 824 |
+
"learning_rate": 2.1428571428571428e-05,
|
| 825 |
+
"loss": 0.1698,
|
| 826 |
+
"step": 275
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 36.67,
|
| 830 |
+
"eval_loss": 0.9969200491905212,
|
| 831 |
+
"eval_runtime": 17.0643,
|
| 832 |
+
"eval_samples_per_second": 0.41,
|
| 833 |
+
"eval_steps_per_second": 0.059,
|
| 834 |
+
"step": 275
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 37.33,
|
| 838 |
+
"grad_norm": 1.872753620147705,
|
| 839 |
+
"learning_rate": 2e-05,
|
| 840 |
+
"loss": 0.16,
|
| 841 |
+
"step": 280
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 37.33,
|
| 845 |
+
"eval_loss": 0.9940390586853027,
|
| 846 |
+
"eval_runtime": 17.0728,
|
| 847 |
+
"eval_samples_per_second": 0.41,
|
| 848 |
+
"eval_steps_per_second": 0.059,
|
| 849 |
+
"step": 280
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 38.0,
|
| 853 |
+
"grad_norm": 2.7510581016540527,
|
| 854 |
+
"learning_rate": 1.8571428571428572e-05,
|
| 855 |
+
"loss": 0.1614,
|
| 856 |
+
"step": 285
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 38.0,
|
| 860 |
+
"eval_loss": 1.0066231489181519,
|
| 861 |
+
"eval_runtime": 17.072,
|
| 862 |
+
"eval_samples_per_second": 0.41,
|
| 863 |
+
"eval_steps_per_second": 0.059,
|
| 864 |
+
"step": 285
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 38.67,
|
| 868 |
+
"grad_norm": 1.8461092710494995,
|
| 869 |
+
"learning_rate": 1.7142857142857145e-05,
|
| 870 |
+
"loss": 0.1506,
|
| 871 |
+
"step": 290
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 38.67,
|
| 875 |
+
"eval_loss": 0.9927281737327576,
|
| 876 |
+
"eval_runtime": 17.0481,
|
| 877 |
+
"eval_samples_per_second": 0.411,
|
| 878 |
+
"eval_steps_per_second": 0.059,
|
| 879 |
+
"step": 290
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 39.33,
|
| 883 |
+
"grad_norm": 1.8425017595291138,
|
| 884 |
+
"learning_rate": 1.5714285714285715e-05,
|
| 885 |
+
"loss": 0.1419,
|
| 886 |
+
"step": 295
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 39.33,
|
| 890 |
+
"eval_loss": 1.0133570432662964,
|
| 891 |
+
"eval_runtime": 17.0642,
|
| 892 |
+
"eval_samples_per_second": 0.41,
|
| 893 |
+
"eval_steps_per_second": 0.059,
|
| 894 |
+
"step": 295
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 40.0,
|
| 898 |
+
"grad_norm": 2.0457987785339355,
|
| 899 |
+
"learning_rate": 1.4285714285714285e-05,
|
| 900 |
+
"loss": 0.1459,
|
| 901 |
+
"step": 300
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 40.0,
|
| 905 |
+
"eval_loss": 1.0127934217453003,
|
| 906 |
+
"eval_runtime": 17.0581,
|
| 907 |
+
"eval_samples_per_second": 0.41,
|
| 908 |
+
"eval_steps_per_second": 0.059,
|
| 909 |
+
"step": 300
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 40.67,
|
| 913 |
+
"grad_norm": 1.5630775690078735,
|
| 914 |
+
"learning_rate": 1.2857142857142857e-05,
|
| 915 |
+
"loss": 0.1225,
|
| 916 |
+
"step": 305
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 40.67,
|
| 920 |
+
"eval_loss": 1.0092624425888062,
|
| 921 |
+
"eval_runtime": 17.0483,
|
| 922 |
+
"eval_samples_per_second": 0.411,
|
| 923 |
+
"eval_steps_per_second": 0.059,
|
| 924 |
+
"step": 305
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 41.33,
|
| 928 |
+
"grad_norm": 1.37598717212677,
|
| 929 |
+
"learning_rate": 1.1428571428571429e-05,
|
| 930 |
+
"loss": 0.146,
|
| 931 |
+
"step": 310
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 41.33,
|
| 935 |
+
"eval_loss": 1.0083317756652832,
|
| 936 |
+
"eval_runtime": 17.0804,
|
| 937 |
+
"eval_samples_per_second": 0.41,
|
| 938 |
+
"eval_steps_per_second": 0.059,
|
| 939 |
+
"step": 310
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 42.0,
|
| 943 |
+
"grad_norm": 1.8867217302322388,
|
| 944 |
+
"learning_rate": 1e-05,
|
| 945 |
+
"loss": 0.13,
|
| 946 |
+
"step": 315
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 42.0,
|
| 950 |
+
"eval_loss": 1.0165104866027832,
|
| 951 |
+
"eval_runtime": 17.0621,
|
| 952 |
+
"eval_samples_per_second": 0.41,
|
| 953 |
+
"eval_steps_per_second": 0.059,
|
| 954 |
+
"step": 315
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 42.67,
|
| 958 |
+
"grad_norm": 1.4643555879592896,
|
| 959 |
+
"learning_rate": 8.571428571428573e-06,
|
| 960 |
+
"loss": 0.131,
|
| 961 |
+
"step": 320
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 42.67,
|
| 965 |
+
"eval_loss": 1.0264887809753418,
|
| 966 |
+
"eval_runtime": 17.0554,
|
| 967 |
+
"eval_samples_per_second": 0.41,
|
| 968 |
+
"eval_steps_per_second": 0.059,
|
| 969 |
+
"step": 320
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 43.33,
|
| 973 |
+
"grad_norm": 1.4163501262664795,
|
| 974 |
+
"learning_rate": 7.142857142857143e-06,
|
| 975 |
+
"loss": 0.1205,
|
| 976 |
+
"step": 325
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 43.33,
|
| 980 |
+
"eval_loss": 1.0175670385360718,
|
| 981 |
+
"eval_runtime": 17.0692,
|
| 982 |
+
"eval_samples_per_second": 0.41,
|
| 983 |
+
"eval_steps_per_second": 0.059,
|
| 984 |
+
"step": 325
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 44.0,
|
| 988 |
+
"grad_norm": 1.730556607246399,
|
| 989 |
+
"learning_rate": 5.7142857142857145e-06,
|
| 990 |
+
"loss": 0.1143,
|
| 991 |
+
"step": 330
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 44.0,
|
| 995 |
+
"eval_loss": 1.0272893905639648,
|
| 996 |
+
"eval_runtime": 17.0651,
|
| 997 |
+
"eval_samples_per_second": 0.41,
|
| 998 |
+
"eval_steps_per_second": 0.059,
|
| 999 |
+
"step": 330
|
| 1000 |
+
}
|
| 1001 |
+
],
|
| 1002 |
+
"logging_steps": 5,
|
| 1003 |
+
"max_steps": 350,
|
| 1004 |
+
"num_input_tokens_seen": 0,
|
| 1005 |
+
"num_train_epochs": 50,
|
| 1006 |
+
"save_steps": 10,
|
| 1007 |
+
"total_flos": 2.1924911400521564e+18,
|
| 1008 |
+
"train_batch_size": 2,
|
| 1009 |
+
"trial_name": null,
|
| 1010 |
+
"trial_params": null
|
| 1011 |
+
}
|
checkpoint-330/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ed0ba6fb9a88dad56f61d9306f17b4e66e8767d898772faa97871a1388e82cf
|
| 3 |
+
size 4920
|
checkpoint-340/README.md
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
|
| 7 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 8 |
+
- quant_method: bitsandbytes
|
| 9 |
+
- _load_in_8bit: False
|
| 10 |
+
- _load_in_4bit: False
|
| 11 |
+
- llm_int8_threshold: 6.0
|
| 12 |
+
- llm_int8_skip_modules: None
|
| 13 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 14 |
+
- llm_int8_has_fp16_weight: False
|
| 15 |
+
- bnb_4bit_quant_type: fp4
|
| 16 |
+
- bnb_4bit_use_double_quant: False
|
| 17 |
+
- bnb_4bit_compute_dtype: float32
|
| 18 |
+
- load_in_4bit: False
|
| 19 |
+
- load_in_8bit: False
|
| 20 |
+
### Framework versions
|
| 21 |
+
|
| 22 |
+
|
| 23 |
+
- PEFT 0.5.0
|
checkpoint-340/adapter_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "meta-llama/Llama-2-70b-hf",
|
| 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": 32,
|
| 11 |
+
"lora_dropout": 0.05,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 8,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"q_proj",
|
| 18 |
+
"v_proj"
|
| 19 |
+
],
|
| 20 |
+
"task_type": "CAUSAL_LM"
|
| 21 |
+
}
|
checkpoint-340/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c897d7f97d7cfce98fed482198deb963e7ba374c7a773a5a3c57015ba333f55c
|
| 3 |
+
size 65578776
|
checkpoint-340/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d4b41aaa2ff189b2a834053ad096c135f5c88ccb616ee466390cb9b512c7a6cc
|
| 3 |
+
size 131345914
|
checkpoint-340/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b33e7788fe74a32e8025c78f1eaeeb839f4a42077f639cc8f0ccaaa0cc2a8f5e
|
| 3 |
+
size 14244
|
checkpoint-340/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6ace13ef9060efed9fdf80fb5c7e28a065010d0a88cfc05a35d53c0b45da76b6
|
| 3 |
+
size 1064
|
checkpoint-340/trainer_state.json
ADDED
|
@@ -0,0 +1,1041 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9927281737327576,
|
| 3 |
+
"best_model_checkpoint": "/scratch/kwamea/llama-output/checkpoint-290",
|
| 4 |
+
"epoch": 45.333333333333336,
|
| 5 |
+
"eval_steps": 5,
|
| 6 |
+
"global_step": 340,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.67,
|
| 13 |
+
"grad_norm": 0.243179589509964,
|
| 14 |
+
"learning_rate": 9.857142857142858e-05,
|
| 15 |
+
"loss": 1.9956,
|
| 16 |
+
"step": 5
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.67,
|
| 20 |
+
"eval_loss": 1.9701930284500122,
|
| 21 |
+
"eval_runtime": 17.115,
|
| 22 |
+
"eval_samples_per_second": 0.409,
|
| 23 |
+
"eval_steps_per_second": 0.058,
|
| 24 |
+
"step": 5
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 1.33,
|
| 28 |
+
"grad_norm": 0.34590908885002136,
|
| 29 |
+
"learning_rate": 9.714285714285715e-05,
|
| 30 |
+
"loss": 1.9758,
|
| 31 |
+
"step": 10
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 1.33,
|
| 35 |
+
"eval_loss": 1.8941271305084229,
|
| 36 |
+
"eval_runtime": 17.0912,
|
| 37 |
+
"eval_samples_per_second": 0.41,
|
| 38 |
+
"eval_steps_per_second": 0.059,
|
| 39 |
+
"step": 10
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 2.0,
|
| 43 |
+
"grad_norm": 0.31595832109451294,
|
| 44 |
+
"learning_rate": 9.571428571428573e-05,
|
| 45 |
+
"loss": 1.849,
|
| 46 |
+
"step": 15
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 2.0,
|
| 50 |
+
"eval_loss": 1.8046789169311523,
|
| 51 |
+
"eval_runtime": 17.098,
|
| 52 |
+
"eval_samples_per_second": 0.409,
|
| 53 |
+
"eval_steps_per_second": 0.058,
|
| 54 |
+
"step": 15
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 2.67,
|
| 58 |
+
"grad_norm": 0.3428090512752533,
|
| 59 |
+
"learning_rate": 9.428571428571429e-05,
|
| 60 |
+
"loss": 1.789,
|
| 61 |
+
"step": 20
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 2.67,
|
| 65 |
+
"eval_loss": 1.7658358812332153,
|
| 66 |
+
"eval_runtime": 17.0734,
|
| 67 |
+
"eval_samples_per_second": 0.41,
|
| 68 |
+
"eval_steps_per_second": 0.059,
|
| 69 |
+
"step": 20
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 3.33,
|
| 73 |
+
"grad_norm": 0.3102028965950012,
|
| 74 |
+
"learning_rate": 9.285714285714286e-05,
|
| 75 |
+
"loss": 1.7789,
|
| 76 |
+
"step": 25
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 3.33,
|
| 80 |
+
"eval_loss": 1.7225048542022705,
|
| 81 |
+
"eval_runtime": 17.0972,
|
| 82 |
+
"eval_samples_per_second": 0.409,
|
| 83 |
+
"eval_steps_per_second": 0.058,
|
| 84 |
+
"step": 25
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 4.0,
|
| 88 |
+
"grad_norm": 0.38602885603904724,
|
| 89 |
+
"learning_rate": 9.142857142857143e-05,
|
| 90 |
+
"loss": 1.7003,
|
| 91 |
+
"step": 30
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 4.0,
|
| 95 |
+
"eval_loss": 1.6749440431594849,
|
| 96 |
+
"eval_runtime": 17.1034,
|
| 97 |
+
"eval_samples_per_second": 0.409,
|
| 98 |
+
"eval_steps_per_second": 0.058,
|
| 99 |
+
"step": 30
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 4.67,
|
| 103 |
+
"grad_norm": 0.37120407819747925,
|
| 104 |
+
"learning_rate": 9e-05,
|
| 105 |
+
"loss": 1.6424,
|
| 106 |
+
"step": 35
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 4.67,
|
| 110 |
+
"eval_loss": 1.6231099367141724,
|
| 111 |
+
"eval_runtime": 17.1067,
|
| 112 |
+
"eval_samples_per_second": 0.409,
|
| 113 |
+
"eval_steps_per_second": 0.058,
|
| 114 |
+
"step": 35
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 5.33,
|
| 118 |
+
"grad_norm": 0.4633428454399109,
|
| 119 |
+
"learning_rate": 8.857142857142857e-05,
|
| 120 |
+
"loss": 1.6023,
|
| 121 |
+
"step": 40
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 5.33,
|
| 125 |
+
"eval_loss": 1.5727053880691528,
|
| 126 |
+
"eval_runtime": 17.1002,
|
| 127 |
+
"eval_samples_per_second": 0.409,
|
| 128 |
+
"eval_steps_per_second": 0.058,
|
| 129 |
+
"step": 40
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 6.0,
|
| 133 |
+
"grad_norm": 0.5034663081169128,
|
| 134 |
+
"learning_rate": 8.714285714285715e-05,
|
| 135 |
+
"loss": 1.5322,
|
| 136 |
+
"step": 45
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 6.0,
|
| 140 |
+
"eval_loss": 1.5312587022781372,
|
| 141 |
+
"eval_runtime": 17.1159,
|
| 142 |
+
"eval_samples_per_second": 0.409,
|
| 143 |
+
"eval_steps_per_second": 0.058,
|
| 144 |
+
"step": 45
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 6.67,
|
| 148 |
+
"grad_norm": 0.5549929141998291,
|
| 149 |
+
"learning_rate": 8.571428571428571e-05,
|
| 150 |
+
"loss": 1.4788,
|
| 151 |
+
"step": 50
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 6.67,
|
| 155 |
+
"eval_loss": 1.492464303970337,
|
| 156 |
+
"eval_runtime": 17.0823,
|
| 157 |
+
"eval_samples_per_second": 0.41,
|
| 158 |
+
"eval_steps_per_second": 0.059,
|
| 159 |
+
"step": 50
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 7.33,
|
| 163 |
+
"grad_norm": 0.49194690585136414,
|
| 164 |
+
"learning_rate": 8.428571428571429e-05,
|
| 165 |
+
"loss": 1.4632,
|
| 166 |
+
"step": 55
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 7.33,
|
| 170 |
+
"eval_loss": 1.4622489213943481,
|
| 171 |
+
"eval_runtime": 17.1022,
|
| 172 |
+
"eval_samples_per_second": 0.409,
|
| 173 |
+
"eval_steps_per_second": 0.058,
|
| 174 |
+
"step": 55
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 8.0,
|
| 178 |
+
"grad_norm": 0.5866131782531738,
|
| 179 |
+
"learning_rate": 8.285714285714287e-05,
|
| 180 |
+
"loss": 1.3951,
|
| 181 |
+
"step": 60
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 8.0,
|
| 185 |
+
"eval_loss": 1.435951828956604,
|
| 186 |
+
"eval_runtime": 17.1087,
|
| 187 |
+
"eval_samples_per_second": 0.409,
|
| 188 |
+
"eval_steps_per_second": 0.058,
|
| 189 |
+
"step": 60
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 8.67,
|
| 193 |
+
"grad_norm": 0.6252542734146118,
|
| 194 |
+
"learning_rate": 8.142857142857143e-05,
|
| 195 |
+
"loss": 1.3796,
|
| 196 |
+
"step": 65
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 8.67,
|
| 200 |
+
"eval_loss": 1.413227915763855,
|
| 201 |
+
"eval_runtime": 17.0914,
|
| 202 |
+
"eval_samples_per_second": 0.41,
|
| 203 |
+
"eval_steps_per_second": 0.059,
|
| 204 |
+
"step": 65
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 9.33,
|
| 208 |
+
"grad_norm": 0.6751863360404968,
|
| 209 |
+
"learning_rate": 8e-05,
|
| 210 |
+
"loss": 1.3257,
|
| 211 |
+
"step": 70
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 9.33,
|
| 215 |
+
"eval_loss": 1.395649790763855,
|
| 216 |
+
"eval_runtime": 17.0885,
|
| 217 |
+
"eval_samples_per_second": 0.41,
|
| 218 |
+
"eval_steps_per_second": 0.059,
|
| 219 |
+
"step": 70
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 10.0,
|
| 223 |
+
"grad_norm": 0.8878222703933716,
|
| 224 |
+
"learning_rate": 7.857142857142858e-05,
|
| 225 |
+
"loss": 1.2795,
|
| 226 |
+
"step": 75
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 10.0,
|
| 230 |
+
"eval_loss": 1.3699487447738647,
|
| 231 |
+
"eval_runtime": 17.1031,
|
| 232 |
+
"eval_samples_per_second": 0.409,
|
| 233 |
+
"eval_steps_per_second": 0.058,
|
| 234 |
+
"step": 75
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 10.67,
|
| 238 |
+
"grad_norm": 0.8470121026039124,
|
| 239 |
+
"learning_rate": 7.714285714285715e-05,
|
| 240 |
+
"loss": 1.2449,
|
| 241 |
+
"step": 80
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 10.67,
|
| 245 |
+
"eval_loss": 1.347831130027771,
|
| 246 |
+
"eval_runtime": 17.0985,
|
| 247 |
+
"eval_samples_per_second": 0.409,
|
| 248 |
+
"eval_steps_per_second": 0.058,
|
| 249 |
+
"step": 80
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 11.33,
|
| 253 |
+
"grad_norm": 1.0655425786972046,
|
| 254 |
+
"learning_rate": 7.571428571428571e-05,
|
| 255 |
+
"loss": 1.1983,
|
| 256 |
+
"step": 85
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 11.33,
|
| 260 |
+
"eval_loss": 1.3311971426010132,
|
| 261 |
+
"eval_runtime": 17.0784,
|
| 262 |
+
"eval_samples_per_second": 0.41,
|
| 263 |
+
"eval_steps_per_second": 0.059,
|
| 264 |
+
"step": 85
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 12.0,
|
| 268 |
+
"grad_norm": 1.2651888132095337,
|
| 269 |
+
"learning_rate": 7.428571428571429e-05,
|
| 270 |
+
"loss": 1.1467,
|
| 271 |
+
"step": 90
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 12.0,
|
| 275 |
+
"eval_loss": 1.3095277547836304,
|
| 276 |
+
"eval_runtime": 17.0903,
|
| 277 |
+
"eval_samples_per_second": 0.41,
|
| 278 |
+
"eval_steps_per_second": 0.059,
|
| 279 |
+
"step": 90
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 12.67,
|
| 283 |
+
"grad_norm": 1.248926043510437,
|
| 284 |
+
"learning_rate": 7.285714285714286e-05,
|
| 285 |
+
"loss": 1.0922,
|
| 286 |
+
"step": 95
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 12.67,
|
| 290 |
+
"eval_loss": 1.2942878007888794,
|
| 291 |
+
"eval_runtime": 17.0947,
|
| 292 |
+
"eval_samples_per_second": 0.409,
|
| 293 |
+
"eval_steps_per_second": 0.058,
|
| 294 |
+
"step": 95
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 13.33,
|
| 298 |
+
"grad_norm": 1.896952509880066,
|
| 299 |
+
"learning_rate": 7.142857142857143e-05,
|
| 300 |
+
"loss": 1.0403,
|
| 301 |
+
"step": 100
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 13.33,
|
| 305 |
+
"eval_loss": 1.2803159952163696,
|
| 306 |
+
"eval_runtime": 17.0819,
|
| 307 |
+
"eval_samples_per_second": 0.41,
|
| 308 |
+
"eval_steps_per_second": 0.059,
|
| 309 |
+
"step": 100
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 14.0,
|
| 313 |
+
"grad_norm": 1.862244725227356,
|
| 314 |
+
"learning_rate": 7e-05,
|
| 315 |
+
"loss": 1.0049,
|
| 316 |
+
"step": 105
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 14.0,
|
| 320 |
+
"eval_loss": 1.2643567323684692,
|
| 321 |
+
"eval_runtime": 17.0849,
|
| 322 |
+
"eval_samples_per_second": 0.41,
|
| 323 |
+
"eval_steps_per_second": 0.059,
|
| 324 |
+
"step": 105
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 14.67,
|
| 328 |
+
"grad_norm": 1.7487821578979492,
|
| 329 |
+
"learning_rate": 6.857142857142858e-05,
|
| 330 |
+
"loss": 0.9262,
|
| 331 |
+
"step": 110
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 14.67,
|
| 335 |
+
"eval_loss": 1.2471646070480347,
|
| 336 |
+
"eval_runtime": 17.1278,
|
| 337 |
+
"eval_samples_per_second": 0.409,
|
| 338 |
+
"eval_steps_per_second": 0.058,
|
| 339 |
+
"step": 110
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 15.33,
|
| 343 |
+
"grad_norm": 1.838605284690857,
|
| 344 |
+
"learning_rate": 6.714285714285714e-05,
|
| 345 |
+
"loss": 0.8965,
|
| 346 |
+
"step": 115
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 15.33,
|
| 350 |
+
"eval_loss": 1.2377034425735474,
|
| 351 |
+
"eval_runtime": 17.0731,
|
| 352 |
+
"eval_samples_per_second": 0.41,
|
| 353 |
+
"eval_steps_per_second": 0.059,
|
| 354 |
+
"step": 115
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 16.0,
|
| 358 |
+
"grad_norm": 3.117398977279663,
|
| 359 |
+
"learning_rate": 6.571428571428571e-05,
|
| 360 |
+
"loss": 0.8581,
|
| 361 |
+
"step": 120
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 16.0,
|
| 365 |
+
"eval_loss": 1.2083133459091187,
|
| 366 |
+
"eval_runtime": 17.1304,
|
| 367 |
+
"eval_samples_per_second": 0.409,
|
| 368 |
+
"eval_steps_per_second": 0.058,
|
| 369 |
+
"step": 120
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 16.67,
|
| 373 |
+
"grad_norm": 2.5655250549316406,
|
| 374 |
+
"learning_rate": 6.428571428571429e-05,
|
| 375 |
+
"loss": 0.7929,
|
| 376 |
+
"step": 125
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 16.67,
|
| 380 |
+
"eval_loss": 1.1945828199386597,
|
| 381 |
+
"eval_runtime": 17.104,
|
| 382 |
+
"eval_samples_per_second": 0.409,
|
| 383 |
+
"eval_steps_per_second": 0.058,
|
| 384 |
+
"step": 125
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 17.33,
|
| 388 |
+
"grad_norm": 2.168546199798584,
|
| 389 |
+
"learning_rate": 6.285714285714286e-05,
|
| 390 |
+
"loss": 0.7543,
|
| 391 |
+
"step": 130
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 17.33,
|
| 395 |
+
"eval_loss": 1.1876276731491089,
|
| 396 |
+
"eval_runtime": 17.1046,
|
| 397 |
+
"eval_samples_per_second": 0.409,
|
| 398 |
+
"eval_steps_per_second": 0.058,
|
| 399 |
+
"step": 130
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 18.0,
|
| 403 |
+
"grad_norm": 2.5984208583831787,
|
| 404 |
+
"learning_rate": 6.142857142857143e-05,
|
| 405 |
+
"loss": 0.716,
|
| 406 |
+
"step": 135
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 18.0,
|
| 410 |
+
"eval_loss": 1.1714750528335571,
|
| 411 |
+
"eval_runtime": 17.0807,
|
| 412 |
+
"eval_samples_per_second": 0.41,
|
| 413 |
+
"eval_steps_per_second": 0.059,
|
| 414 |
+
"step": 135
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 18.67,
|
| 418 |
+
"grad_norm": 3.479024887084961,
|
| 419 |
+
"learning_rate": 6e-05,
|
| 420 |
+
"loss": 0.6681,
|
| 421 |
+
"step": 140
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 18.67,
|
| 425 |
+
"eval_loss": 1.169895052909851,
|
| 426 |
+
"eval_runtime": 17.0681,
|
| 427 |
+
"eval_samples_per_second": 0.41,
|
| 428 |
+
"eval_steps_per_second": 0.059,
|
| 429 |
+
"step": 140
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 19.33,
|
| 433 |
+
"grad_norm": 2.563386917114258,
|
| 434 |
+
"learning_rate": 5.8571428571428575e-05,
|
| 435 |
+
"loss": 0.6306,
|
| 436 |
+
"step": 145
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 19.33,
|
| 440 |
+
"eval_loss": 1.1741083860397339,
|
| 441 |
+
"eval_runtime": 17.0568,
|
| 442 |
+
"eval_samples_per_second": 0.41,
|
| 443 |
+
"eval_steps_per_second": 0.059,
|
| 444 |
+
"step": 145
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 20.0,
|
| 448 |
+
"grad_norm": 2.96592116355896,
|
| 449 |
+
"learning_rate": 5.714285714285714e-05,
|
| 450 |
+
"loss": 0.6183,
|
| 451 |
+
"step": 150
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 20.0,
|
| 455 |
+
"eval_loss": 1.1455965042114258,
|
| 456 |
+
"eval_runtime": 17.073,
|
| 457 |
+
"eval_samples_per_second": 0.41,
|
| 458 |
+
"eval_steps_per_second": 0.059,
|
| 459 |
+
"step": 150
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 20.67,
|
| 463 |
+
"grad_norm": 2.6751275062561035,
|
| 464 |
+
"learning_rate": 5.571428571428572e-05,
|
| 465 |
+
"loss": 0.5464,
|
| 466 |
+
"step": 155
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 20.67,
|
| 470 |
+
"eval_loss": 1.131102204322815,
|
| 471 |
+
"eval_runtime": 17.0578,
|
| 472 |
+
"eval_samples_per_second": 0.41,
|
| 473 |
+
"eval_steps_per_second": 0.059,
|
| 474 |
+
"step": 155
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 21.33,
|
| 478 |
+
"grad_norm": 2.3700051307678223,
|
| 479 |
+
"learning_rate": 5.428571428571428e-05,
|
| 480 |
+
"loss": 0.551,
|
| 481 |
+
"step": 160
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 21.33,
|
| 485 |
+
"eval_loss": 1.127384066581726,
|
| 486 |
+
"eval_runtime": 17.0546,
|
| 487 |
+
"eval_samples_per_second": 0.41,
|
| 488 |
+
"eval_steps_per_second": 0.059,
|
| 489 |
+
"step": 160
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 22.0,
|
| 493 |
+
"grad_norm": 3.3827567100524902,
|
| 494 |
+
"learning_rate": 5.285714285714286e-05,
|
| 495 |
+
"loss": 0.5179,
|
| 496 |
+
"step": 165
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 22.0,
|
| 500 |
+
"eval_loss": 1.111584186553955,
|
| 501 |
+
"eval_runtime": 17.0812,
|
| 502 |
+
"eval_samples_per_second": 0.41,
|
| 503 |
+
"eval_steps_per_second": 0.059,
|
| 504 |
+
"step": 165
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 22.67,
|
| 508 |
+
"grad_norm": 3.55790114402771,
|
| 509 |
+
"learning_rate": 5.142857142857143e-05,
|
| 510 |
+
"loss": 0.4831,
|
| 511 |
+
"step": 170
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 22.67,
|
| 515 |
+
"eval_loss": 1.0948525667190552,
|
| 516 |
+
"eval_runtime": 17.0547,
|
| 517 |
+
"eval_samples_per_second": 0.41,
|
| 518 |
+
"eval_steps_per_second": 0.059,
|
| 519 |
+
"step": 170
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 23.33,
|
| 523 |
+
"grad_norm": 3.0782699584960938,
|
| 524 |
+
"learning_rate": 5e-05,
|
| 525 |
+
"loss": 0.4587,
|
| 526 |
+
"step": 175
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 23.33,
|
| 530 |
+
"eval_loss": 1.0906586647033691,
|
| 531 |
+
"eval_runtime": 17.0666,
|
| 532 |
+
"eval_samples_per_second": 0.41,
|
| 533 |
+
"eval_steps_per_second": 0.059,
|
| 534 |
+
"step": 175
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 24.0,
|
| 538 |
+
"grad_norm": 3.3993167877197266,
|
| 539 |
+
"learning_rate": 4.8571428571428576e-05,
|
| 540 |
+
"loss": 0.4203,
|
| 541 |
+
"step": 180
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 24.0,
|
| 545 |
+
"eval_loss": 1.0688152313232422,
|
| 546 |
+
"eval_runtime": 17.0721,
|
| 547 |
+
"eval_samples_per_second": 0.41,
|
| 548 |
+
"eval_steps_per_second": 0.059,
|
| 549 |
+
"step": 180
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 24.67,
|
| 553 |
+
"grad_norm": 3.319303035736084,
|
| 554 |
+
"learning_rate": 4.714285714285714e-05,
|
| 555 |
+
"loss": 0.3975,
|
| 556 |
+
"step": 185
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 24.67,
|
| 560 |
+
"eval_loss": 1.0746583938598633,
|
| 561 |
+
"eval_runtime": 17.0709,
|
| 562 |
+
"eval_samples_per_second": 0.41,
|
| 563 |
+
"eval_steps_per_second": 0.059,
|
| 564 |
+
"step": 185
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 25.33,
|
| 568 |
+
"grad_norm": 2.4532127380371094,
|
| 569 |
+
"learning_rate": 4.5714285714285716e-05,
|
| 570 |
+
"loss": 0.3832,
|
| 571 |
+
"step": 190
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 25.33,
|
| 575 |
+
"eval_loss": 1.0772522687911987,
|
| 576 |
+
"eval_runtime": 17.0619,
|
| 577 |
+
"eval_samples_per_second": 0.41,
|
| 578 |
+
"eval_steps_per_second": 0.059,
|
| 579 |
+
"step": 190
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 26.0,
|
| 583 |
+
"grad_norm": 3.956822156906128,
|
| 584 |
+
"learning_rate": 4.428571428571428e-05,
|
| 585 |
+
"loss": 0.3725,
|
| 586 |
+
"step": 195
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 26.0,
|
| 590 |
+
"eval_loss": 1.0638784170150757,
|
| 591 |
+
"eval_runtime": 17.0807,
|
| 592 |
+
"eval_samples_per_second": 0.41,
|
| 593 |
+
"eval_steps_per_second": 0.059,
|
| 594 |
+
"step": 195
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 26.67,
|
| 598 |
+
"grad_norm": 2.76033353805542,
|
| 599 |
+
"learning_rate": 4.2857142857142856e-05,
|
| 600 |
+
"loss": 0.3473,
|
| 601 |
+
"step": 200
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 26.67,
|
| 605 |
+
"eval_loss": 1.04669988155365,
|
| 606 |
+
"eval_runtime": 17.0774,
|
| 607 |
+
"eval_samples_per_second": 0.41,
|
| 608 |
+
"eval_steps_per_second": 0.059,
|
| 609 |
+
"step": 200
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 27.33,
|
| 613 |
+
"grad_norm": 3.8683507442474365,
|
| 614 |
+
"learning_rate": 4.1428571428571437e-05,
|
| 615 |
+
"loss": 0.3243,
|
| 616 |
+
"step": 205
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 27.33,
|
| 620 |
+
"eval_loss": 1.0470303297042847,
|
| 621 |
+
"eval_runtime": 17.0718,
|
| 622 |
+
"eval_samples_per_second": 0.41,
|
| 623 |
+
"eval_steps_per_second": 0.059,
|
| 624 |
+
"step": 205
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 28.0,
|
| 628 |
+
"grad_norm": 4.535538196563721,
|
| 629 |
+
"learning_rate": 4e-05,
|
| 630 |
+
"loss": 0.3202,
|
| 631 |
+
"step": 210
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 28.0,
|
| 635 |
+
"eval_loss": 1.025539517402649,
|
| 636 |
+
"eval_runtime": 17.0604,
|
| 637 |
+
"eval_samples_per_second": 0.41,
|
| 638 |
+
"eval_steps_per_second": 0.059,
|
| 639 |
+
"step": 210
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 28.67,
|
| 643 |
+
"grad_norm": 2.6224355697631836,
|
| 644 |
+
"learning_rate": 3.857142857142858e-05,
|
| 645 |
+
"loss": 0.2958,
|
| 646 |
+
"step": 215
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 28.67,
|
| 650 |
+
"eval_loss": 1.0192126035690308,
|
| 651 |
+
"eval_runtime": 17.0657,
|
| 652 |
+
"eval_samples_per_second": 0.41,
|
| 653 |
+
"eval_steps_per_second": 0.059,
|
| 654 |
+
"step": 215
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 29.33,
|
| 658 |
+
"grad_norm": 2.5870041847229004,
|
| 659 |
+
"learning_rate": 3.7142857142857143e-05,
|
| 660 |
+
"loss": 0.2783,
|
| 661 |
+
"step": 220
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 29.33,
|
| 665 |
+
"eval_loss": 1.0211580991744995,
|
| 666 |
+
"eval_runtime": 17.0857,
|
| 667 |
+
"eval_samples_per_second": 0.41,
|
| 668 |
+
"eval_steps_per_second": 0.059,
|
| 669 |
+
"step": 220
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 30.0,
|
| 673 |
+
"grad_norm": 3.4565751552581787,
|
| 674 |
+
"learning_rate": 3.571428571428572e-05,
|
| 675 |
+
"loss": 0.2773,
|
| 676 |
+
"step": 225
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 30.0,
|
| 680 |
+
"eval_loss": 1.006419062614441,
|
| 681 |
+
"eval_runtime": 17.0807,
|
| 682 |
+
"eval_samples_per_second": 0.41,
|
| 683 |
+
"eval_steps_per_second": 0.059,
|
| 684 |
+
"step": 225
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 30.67,
|
| 688 |
+
"grad_norm": 2.4756500720977783,
|
| 689 |
+
"learning_rate": 3.428571428571429e-05,
|
| 690 |
+
"loss": 0.2482,
|
| 691 |
+
"step": 230
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 30.67,
|
| 695 |
+
"eval_loss": 1.0081219673156738,
|
| 696 |
+
"eval_runtime": 17.0576,
|
| 697 |
+
"eval_samples_per_second": 0.41,
|
| 698 |
+
"eval_steps_per_second": 0.059,
|
| 699 |
+
"step": 230
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 31.33,
|
| 703 |
+
"grad_norm": 2.38002610206604,
|
| 704 |
+
"learning_rate": 3.285714285714286e-05,
|
| 705 |
+
"loss": 0.2464,
|
| 706 |
+
"step": 235
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 31.33,
|
| 710 |
+
"eval_loss": 1.0151804685592651,
|
| 711 |
+
"eval_runtime": 17.0587,
|
| 712 |
+
"eval_samples_per_second": 0.41,
|
| 713 |
+
"eval_steps_per_second": 0.059,
|
| 714 |
+
"step": 235
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 32.0,
|
| 718 |
+
"grad_norm": 3.7081105709075928,
|
| 719 |
+
"learning_rate": 3.142857142857143e-05,
|
| 720 |
+
"loss": 0.2442,
|
| 721 |
+
"step": 240
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 32.0,
|
| 725 |
+
"eval_loss": 1.0032445192337036,
|
| 726 |
+
"eval_runtime": 17.1613,
|
| 727 |
+
"eval_samples_per_second": 0.408,
|
| 728 |
+
"eval_steps_per_second": 0.058,
|
| 729 |
+
"step": 240
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 32.67,
|
| 733 |
+
"grad_norm": 2.55924391746521,
|
| 734 |
+
"learning_rate": 3e-05,
|
| 735 |
+
"loss": 0.2193,
|
| 736 |
+
"step": 245
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 32.67,
|
| 740 |
+
"eval_loss": 0.9989615082740784,
|
| 741 |
+
"eval_runtime": 17.0447,
|
| 742 |
+
"eval_samples_per_second": 0.411,
|
| 743 |
+
"eval_steps_per_second": 0.059,
|
| 744 |
+
"step": 245
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 33.33,
|
| 748 |
+
"grad_norm": 1.9451407194137573,
|
| 749 |
+
"learning_rate": 2.857142857142857e-05,
|
| 750 |
+
"loss": 0.2101,
|
| 751 |
+
"step": 250
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 33.33,
|
| 755 |
+
"eval_loss": 1.0029457807540894,
|
| 756 |
+
"eval_runtime": 17.0816,
|
| 757 |
+
"eval_samples_per_second": 0.41,
|
| 758 |
+
"eval_steps_per_second": 0.059,
|
| 759 |
+
"step": 250
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 34.0,
|
| 763 |
+
"grad_norm": 2.713731527328491,
|
| 764 |
+
"learning_rate": 2.714285714285714e-05,
|
| 765 |
+
"loss": 0.2194,
|
| 766 |
+
"step": 255
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 34.0,
|
| 770 |
+
"eval_loss": 0.9959421753883362,
|
| 771 |
+
"eval_runtime": 17.0747,
|
| 772 |
+
"eval_samples_per_second": 0.41,
|
| 773 |
+
"eval_steps_per_second": 0.059,
|
| 774 |
+
"step": 255
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 34.67,
|
| 778 |
+
"grad_norm": 2.1633846759796143,
|
| 779 |
+
"learning_rate": 2.5714285714285714e-05,
|
| 780 |
+
"loss": 0.1958,
|
| 781 |
+
"step": 260
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 34.67,
|
| 785 |
+
"eval_loss": 0.9989770650863647,
|
| 786 |
+
"eval_runtime": 17.0821,
|
| 787 |
+
"eval_samples_per_second": 0.41,
|
| 788 |
+
"eval_steps_per_second": 0.059,
|
| 789 |
+
"step": 260
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 35.33,
|
| 793 |
+
"grad_norm": 3.9233529567718506,
|
| 794 |
+
"learning_rate": 2.4285714285714288e-05,
|
| 795 |
+
"loss": 0.1831,
|
| 796 |
+
"step": 265
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 35.33,
|
| 800 |
+
"eval_loss": 1.0072578191757202,
|
| 801 |
+
"eval_runtime": 17.0564,
|
| 802 |
+
"eval_samples_per_second": 0.41,
|
| 803 |
+
"eval_steps_per_second": 0.059,
|
| 804 |
+
"step": 265
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 36.0,
|
| 808 |
+
"grad_norm": 2.4143056869506836,
|
| 809 |
+
"learning_rate": 2.2857142857142858e-05,
|
| 810 |
+
"loss": 0.1753,
|
| 811 |
+
"step": 270
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 36.0,
|
| 815 |
+
"eval_loss": 0.9938892722129822,
|
| 816 |
+
"eval_runtime": 17.0668,
|
| 817 |
+
"eval_samples_per_second": 0.41,
|
| 818 |
+
"eval_steps_per_second": 0.059,
|
| 819 |
+
"step": 270
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 36.67,
|
| 823 |
+
"grad_norm": 2.706679582595825,
|
| 824 |
+
"learning_rate": 2.1428571428571428e-05,
|
| 825 |
+
"loss": 0.1698,
|
| 826 |
+
"step": 275
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 36.67,
|
| 830 |
+
"eval_loss": 0.9969200491905212,
|
| 831 |
+
"eval_runtime": 17.0643,
|
| 832 |
+
"eval_samples_per_second": 0.41,
|
| 833 |
+
"eval_steps_per_second": 0.059,
|
| 834 |
+
"step": 275
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 37.33,
|
| 838 |
+
"grad_norm": 1.872753620147705,
|
| 839 |
+
"learning_rate": 2e-05,
|
| 840 |
+
"loss": 0.16,
|
| 841 |
+
"step": 280
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 37.33,
|
| 845 |
+
"eval_loss": 0.9940390586853027,
|
| 846 |
+
"eval_runtime": 17.0728,
|
| 847 |
+
"eval_samples_per_second": 0.41,
|
| 848 |
+
"eval_steps_per_second": 0.059,
|
| 849 |
+
"step": 280
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 38.0,
|
| 853 |
+
"grad_norm": 2.7510581016540527,
|
| 854 |
+
"learning_rate": 1.8571428571428572e-05,
|
| 855 |
+
"loss": 0.1614,
|
| 856 |
+
"step": 285
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 38.0,
|
| 860 |
+
"eval_loss": 1.0066231489181519,
|
| 861 |
+
"eval_runtime": 17.072,
|
| 862 |
+
"eval_samples_per_second": 0.41,
|
| 863 |
+
"eval_steps_per_second": 0.059,
|
| 864 |
+
"step": 285
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 38.67,
|
| 868 |
+
"grad_norm": 1.8461092710494995,
|
| 869 |
+
"learning_rate": 1.7142857142857145e-05,
|
| 870 |
+
"loss": 0.1506,
|
| 871 |
+
"step": 290
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 38.67,
|
| 875 |
+
"eval_loss": 0.9927281737327576,
|
| 876 |
+
"eval_runtime": 17.0481,
|
| 877 |
+
"eval_samples_per_second": 0.411,
|
| 878 |
+
"eval_steps_per_second": 0.059,
|
| 879 |
+
"step": 290
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 39.33,
|
| 883 |
+
"grad_norm": 1.8425017595291138,
|
| 884 |
+
"learning_rate": 1.5714285714285715e-05,
|
| 885 |
+
"loss": 0.1419,
|
| 886 |
+
"step": 295
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 39.33,
|
| 890 |
+
"eval_loss": 1.0133570432662964,
|
| 891 |
+
"eval_runtime": 17.0642,
|
| 892 |
+
"eval_samples_per_second": 0.41,
|
| 893 |
+
"eval_steps_per_second": 0.059,
|
| 894 |
+
"step": 295
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 40.0,
|
| 898 |
+
"grad_norm": 2.0457987785339355,
|
| 899 |
+
"learning_rate": 1.4285714285714285e-05,
|
| 900 |
+
"loss": 0.1459,
|
| 901 |
+
"step": 300
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 40.0,
|
| 905 |
+
"eval_loss": 1.0127934217453003,
|
| 906 |
+
"eval_runtime": 17.0581,
|
| 907 |
+
"eval_samples_per_second": 0.41,
|
| 908 |
+
"eval_steps_per_second": 0.059,
|
| 909 |
+
"step": 300
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 40.67,
|
| 913 |
+
"grad_norm": 1.5630775690078735,
|
| 914 |
+
"learning_rate": 1.2857142857142857e-05,
|
| 915 |
+
"loss": 0.1225,
|
| 916 |
+
"step": 305
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 40.67,
|
| 920 |
+
"eval_loss": 1.0092624425888062,
|
| 921 |
+
"eval_runtime": 17.0483,
|
| 922 |
+
"eval_samples_per_second": 0.411,
|
| 923 |
+
"eval_steps_per_second": 0.059,
|
| 924 |
+
"step": 305
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 41.33,
|
| 928 |
+
"grad_norm": 1.37598717212677,
|
| 929 |
+
"learning_rate": 1.1428571428571429e-05,
|
| 930 |
+
"loss": 0.146,
|
| 931 |
+
"step": 310
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 41.33,
|
| 935 |
+
"eval_loss": 1.0083317756652832,
|
| 936 |
+
"eval_runtime": 17.0804,
|
| 937 |
+
"eval_samples_per_second": 0.41,
|
| 938 |
+
"eval_steps_per_second": 0.059,
|
| 939 |
+
"step": 310
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 42.0,
|
| 943 |
+
"grad_norm": 1.8867217302322388,
|
| 944 |
+
"learning_rate": 1e-05,
|
| 945 |
+
"loss": 0.13,
|
| 946 |
+
"step": 315
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 42.0,
|
| 950 |
+
"eval_loss": 1.0165104866027832,
|
| 951 |
+
"eval_runtime": 17.0621,
|
| 952 |
+
"eval_samples_per_second": 0.41,
|
| 953 |
+
"eval_steps_per_second": 0.059,
|
| 954 |
+
"step": 315
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 42.67,
|
| 958 |
+
"grad_norm": 1.4643555879592896,
|
| 959 |
+
"learning_rate": 8.571428571428573e-06,
|
| 960 |
+
"loss": 0.131,
|
| 961 |
+
"step": 320
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 42.67,
|
| 965 |
+
"eval_loss": 1.0264887809753418,
|
| 966 |
+
"eval_runtime": 17.0554,
|
| 967 |
+
"eval_samples_per_second": 0.41,
|
| 968 |
+
"eval_steps_per_second": 0.059,
|
| 969 |
+
"step": 320
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 43.33,
|
| 973 |
+
"grad_norm": 1.4163501262664795,
|
| 974 |
+
"learning_rate": 7.142857142857143e-06,
|
| 975 |
+
"loss": 0.1205,
|
| 976 |
+
"step": 325
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 43.33,
|
| 980 |
+
"eval_loss": 1.0175670385360718,
|
| 981 |
+
"eval_runtime": 17.0692,
|
| 982 |
+
"eval_samples_per_second": 0.41,
|
| 983 |
+
"eval_steps_per_second": 0.059,
|
| 984 |
+
"step": 325
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 44.0,
|
| 988 |
+
"grad_norm": 1.730556607246399,
|
| 989 |
+
"learning_rate": 5.7142857142857145e-06,
|
| 990 |
+
"loss": 0.1143,
|
| 991 |
+
"step": 330
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 44.0,
|
| 995 |
+
"eval_loss": 1.0272893905639648,
|
| 996 |
+
"eval_runtime": 17.0651,
|
| 997 |
+
"eval_samples_per_second": 0.41,
|
| 998 |
+
"eval_steps_per_second": 0.059,
|
| 999 |
+
"step": 330
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 44.67,
|
| 1003 |
+
"grad_norm": 1.451904058456421,
|
| 1004 |
+
"learning_rate": 4.285714285714286e-06,
|
| 1005 |
+
"loss": 0.1087,
|
| 1006 |
+
"step": 335
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 44.67,
|
| 1010 |
+
"eval_loss": 1.0272446870803833,
|
| 1011 |
+
"eval_runtime": 17.0604,
|
| 1012 |
+
"eval_samples_per_second": 0.41,
|
| 1013 |
+
"eval_steps_per_second": 0.059,
|
| 1014 |
+
"step": 335
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 45.33,
|
| 1018 |
+
"grad_norm": 1.1883801221847534,
|
| 1019 |
+
"learning_rate": 2.8571428571428573e-06,
|
| 1020 |
+
"loss": 0.1173,
|
| 1021 |
+
"step": 340
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 45.33,
|
| 1025 |
+
"eval_loss": 1.0314446687698364,
|
| 1026 |
+
"eval_runtime": 17.0652,
|
| 1027 |
+
"eval_samples_per_second": 0.41,
|
| 1028 |
+
"eval_steps_per_second": 0.059,
|
| 1029 |
+
"step": 340
|
| 1030 |
+
}
|
| 1031 |
+
],
|
| 1032 |
+
"logging_steps": 5,
|
| 1033 |
+
"max_steps": 350,
|
| 1034 |
+
"num_input_tokens_seen": 0,
|
| 1035 |
+
"num_train_epochs": 50,
|
| 1036 |
+
"save_steps": 10,
|
| 1037 |
+
"total_flos": 2.2592117872263168e+18,
|
| 1038 |
+
"train_batch_size": 2,
|
| 1039 |
+
"trial_name": null,
|
| 1040 |
+
"trial_params": null
|
| 1041 |
+
}
|
checkpoint-340/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ed0ba6fb9a88dad56f61d9306f17b4e66e8767d898772faa97871a1388e82cf
|
| 3 |
+
size 4920
|
checkpoint-350/README.md
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
|
| 7 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 8 |
+
- quant_method: bitsandbytes
|
| 9 |
+
- _load_in_8bit: False
|
| 10 |
+
- _load_in_4bit: False
|
| 11 |
+
- llm_int8_threshold: 6.0
|
| 12 |
+
- llm_int8_skip_modules: None
|
| 13 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 14 |
+
- llm_int8_has_fp16_weight: False
|
| 15 |
+
- bnb_4bit_quant_type: fp4
|
| 16 |
+
- bnb_4bit_use_double_quant: False
|
| 17 |
+
- bnb_4bit_compute_dtype: float32
|
| 18 |
+
- load_in_4bit: False
|
| 19 |
+
- load_in_8bit: False
|
| 20 |
+
### Framework versions
|
| 21 |
+
|
| 22 |
+
|
| 23 |
+
- PEFT 0.5.0
|
checkpoint-350/adapter_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "meta-llama/Llama-2-70b-hf",
|
| 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": 32,
|
| 11 |
+
"lora_dropout": 0.05,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 8,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"q_proj",
|
| 18 |
+
"v_proj"
|
| 19 |
+
],
|
| 20 |
+
"task_type": "CAUSAL_LM"
|
| 21 |
+
}
|
checkpoint-350/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:67eccf78fe003c69b69f04b1f9b9a6e6a1978840cc4a849f3af0d407627c47d6
|
| 3 |
+
size 65578776
|
checkpoint-350/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4d76e77636c4237df069f3fc007ebc0115869843959146b38c8e52ae3e0864fa
|
| 3 |
+
size 131345914
|
checkpoint-350/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:72b00dace87f5d813f1a45a5136e238b8c42ac9391260f621c6e60ffca2a1a3b
|
| 3 |
+
size 14244
|
checkpoint-350/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:70732dec4f14211b78555ce1acfc091b59c12a05595924c42e5da10d3a62cd6d
|
| 3 |
+
size 1064
|
checkpoint-350/trainer_state.json
ADDED
|
@@ -0,0 +1,1071 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9927281737327576,
|
| 3 |
+
"best_model_checkpoint": "/scratch/kwamea/llama-output/checkpoint-290",
|
| 4 |
+
"epoch": 46.666666666666664,
|
| 5 |
+
"eval_steps": 5,
|
| 6 |
+
"global_step": 350,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.67,
|
| 13 |
+
"grad_norm": 0.243179589509964,
|
| 14 |
+
"learning_rate": 9.857142857142858e-05,
|
| 15 |
+
"loss": 1.9956,
|
| 16 |
+
"step": 5
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.67,
|
| 20 |
+
"eval_loss": 1.9701930284500122,
|
| 21 |
+
"eval_runtime": 17.115,
|
| 22 |
+
"eval_samples_per_second": 0.409,
|
| 23 |
+
"eval_steps_per_second": 0.058,
|
| 24 |
+
"step": 5
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 1.33,
|
| 28 |
+
"grad_norm": 0.34590908885002136,
|
| 29 |
+
"learning_rate": 9.714285714285715e-05,
|
| 30 |
+
"loss": 1.9758,
|
| 31 |
+
"step": 10
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 1.33,
|
| 35 |
+
"eval_loss": 1.8941271305084229,
|
| 36 |
+
"eval_runtime": 17.0912,
|
| 37 |
+
"eval_samples_per_second": 0.41,
|
| 38 |
+
"eval_steps_per_second": 0.059,
|
| 39 |
+
"step": 10
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 2.0,
|
| 43 |
+
"grad_norm": 0.31595832109451294,
|
| 44 |
+
"learning_rate": 9.571428571428573e-05,
|
| 45 |
+
"loss": 1.849,
|
| 46 |
+
"step": 15
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 2.0,
|
| 50 |
+
"eval_loss": 1.8046789169311523,
|
| 51 |
+
"eval_runtime": 17.098,
|
| 52 |
+
"eval_samples_per_second": 0.409,
|
| 53 |
+
"eval_steps_per_second": 0.058,
|
| 54 |
+
"step": 15
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 2.67,
|
| 58 |
+
"grad_norm": 0.3428090512752533,
|
| 59 |
+
"learning_rate": 9.428571428571429e-05,
|
| 60 |
+
"loss": 1.789,
|
| 61 |
+
"step": 20
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 2.67,
|
| 65 |
+
"eval_loss": 1.7658358812332153,
|
| 66 |
+
"eval_runtime": 17.0734,
|
| 67 |
+
"eval_samples_per_second": 0.41,
|
| 68 |
+
"eval_steps_per_second": 0.059,
|
| 69 |
+
"step": 20
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 3.33,
|
| 73 |
+
"grad_norm": 0.3102028965950012,
|
| 74 |
+
"learning_rate": 9.285714285714286e-05,
|
| 75 |
+
"loss": 1.7789,
|
| 76 |
+
"step": 25
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 3.33,
|
| 80 |
+
"eval_loss": 1.7225048542022705,
|
| 81 |
+
"eval_runtime": 17.0972,
|
| 82 |
+
"eval_samples_per_second": 0.409,
|
| 83 |
+
"eval_steps_per_second": 0.058,
|
| 84 |
+
"step": 25
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 4.0,
|
| 88 |
+
"grad_norm": 0.38602885603904724,
|
| 89 |
+
"learning_rate": 9.142857142857143e-05,
|
| 90 |
+
"loss": 1.7003,
|
| 91 |
+
"step": 30
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 4.0,
|
| 95 |
+
"eval_loss": 1.6749440431594849,
|
| 96 |
+
"eval_runtime": 17.1034,
|
| 97 |
+
"eval_samples_per_second": 0.409,
|
| 98 |
+
"eval_steps_per_second": 0.058,
|
| 99 |
+
"step": 30
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 4.67,
|
| 103 |
+
"grad_norm": 0.37120407819747925,
|
| 104 |
+
"learning_rate": 9e-05,
|
| 105 |
+
"loss": 1.6424,
|
| 106 |
+
"step": 35
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 4.67,
|
| 110 |
+
"eval_loss": 1.6231099367141724,
|
| 111 |
+
"eval_runtime": 17.1067,
|
| 112 |
+
"eval_samples_per_second": 0.409,
|
| 113 |
+
"eval_steps_per_second": 0.058,
|
| 114 |
+
"step": 35
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 5.33,
|
| 118 |
+
"grad_norm": 0.4633428454399109,
|
| 119 |
+
"learning_rate": 8.857142857142857e-05,
|
| 120 |
+
"loss": 1.6023,
|
| 121 |
+
"step": 40
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 5.33,
|
| 125 |
+
"eval_loss": 1.5727053880691528,
|
| 126 |
+
"eval_runtime": 17.1002,
|
| 127 |
+
"eval_samples_per_second": 0.409,
|
| 128 |
+
"eval_steps_per_second": 0.058,
|
| 129 |
+
"step": 40
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 6.0,
|
| 133 |
+
"grad_norm": 0.5034663081169128,
|
| 134 |
+
"learning_rate": 8.714285714285715e-05,
|
| 135 |
+
"loss": 1.5322,
|
| 136 |
+
"step": 45
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 6.0,
|
| 140 |
+
"eval_loss": 1.5312587022781372,
|
| 141 |
+
"eval_runtime": 17.1159,
|
| 142 |
+
"eval_samples_per_second": 0.409,
|
| 143 |
+
"eval_steps_per_second": 0.058,
|
| 144 |
+
"step": 45
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 6.67,
|
| 148 |
+
"grad_norm": 0.5549929141998291,
|
| 149 |
+
"learning_rate": 8.571428571428571e-05,
|
| 150 |
+
"loss": 1.4788,
|
| 151 |
+
"step": 50
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 6.67,
|
| 155 |
+
"eval_loss": 1.492464303970337,
|
| 156 |
+
"eval_runtime": 17.0823,
|
| 157 |
+
"eval_samples_per_second": 0.41,
|
| 158 |
+
"eval_steps_per_second": 0.059,
|
| 159 |
+
"step": 50
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 7.33,
|
| 163 |
+
"grad_norm": 0.49194690585136414,
|
| 164 |
+
"learning_rate": 8.428571428571429e-05,
|
| 165 |
+
"loss": 1.4632,
|
| 166 |
+
"step": 55
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 7.33,
|
| 170 |
+
"eval_loss": 1.4622489213943481,
|
| 171 |
+
"eval_runtime": 17.1022,
|
| 172 |
+
"eval_samples_per_second": 0.409,
|
| 173 |
+
"eval_steps_per_second": 0.058,
|
| 174 |
+
"step": 55
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 8.0,
|
| 178 |
+
"grad_norm": 0.5866131782531738,
|
| 179 |
+
"learning_rate": 8.285714285714287e-05,
|
| 180 |
+
"loss": 1.3951,
|
| 181 |
+
"step": 60
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 8.0,
|
| 185 |
+
"eval_loss": 1.435951828956604,
|
| 186 |
+
"eval_runtime": 17.1087,
|
| 187 |
+
"eval_samples_per_second": 0.409,
|
| 188 |
+
"eval_steps_per_second": 0.058,
|
| 189 |
+
"step": 60
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 8.67,
|
| 193 |
+
"grad_norm": 0.6252542734146118,
|
| 194 |
+
"learning_rate": 8.142857142857143e-05,
|
| 195 |
+
"loss": 1.3796,
|
| 196 |
+
"step": 65
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 8.67,
|
| 200 |
+
"eval_loss": 1.413227915763855,
|
| 201 |
+
"eval_runtime": 17.0914,
|
| 202 |
+
"eval_samples_per_second": 0.41,
|
| 203 |
+
"eval_steps_per_second": 0.059,
|
| 204 |
+
"step": 65
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 9.33,
|
| 208 |
+
"grad_norm": 0.6751863360404968,
|
| 209 |
+
"learning_rate": 8e-05,
|
| 210 |
+
"loss": 1.3257,
|
| 211 |
+
"step": 70
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 9.33,
|
| 215 |
+
"eval_loss": 1.395649790763855,
|
| 216 |
+
"eval_runtime": 17.0885,
|
| 217 |
+
"eval_samples_per_second": 0.41,
|
| 218 |
+
"eval_steps_per_second": 0.059,
|
| 219 |
+
"step": 70
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 10.0,
|
| 223 |
+
"grad_norm": 0.8878222703933716,
|
| 224 |
+
"learning_rate": 7.857142857142858e-05,
|
| 225 |
+
"loss": 1.2795,
|
| 226 |
+
"step": 75
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 10.0,
|
| 230 |
+
"eval_loss": 1.3699487447738647,
|
| 231 |
+
"eval_runtime": 17.1031,
|
| 232 |
+
"eval_samples_per_second": 0.409,
|
| 233 |
+
"eval_steps_per_second": 0.058,
|
| 234 |
+
"step": 75
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 10.67,
|
| 238 |
+
"grad_norm": 0.8470121026039124,
|
| 239 |
+
"learning_rate": 7.714285714285715e-05,
|
| 240 |
+
"loss": 1.2449,
|
| 241 |
+
"step": 80
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 10.67,
|
| 245 |
+
"eval_loss": 1.347831130027771,
|
| 246 |
+
"eval_runtime": 17.0985,
|
| 247 |
+
"eval_samples_per_second": 0.409,
|
| 248 |
+
"eval_steps_per_second": 0.058,
|
| 249 |
+
"step": 80
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 11.33,
|
| 253 |
+
"grad_norm": 1.0655425786972046,
|
| 254 |
+
"learning_rate": 7.571428571428571e-05,
|
| 255 |
+
"loss": 1.1983,
|
| 256 |
+
"step": 85
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 11.33,
|
| 260 |
+
"eval_loss": 1.3311971426010132,
|
| 261 |
+
"eval_runtime": 17.0784,
|
| 262 |
+
"eval_samples_per_second": 0.41,
|
| 263 |
+
"eval_steps_per_second": 0.059,
|
| 264 |
+
"step": 85
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 12.0,
|
| 268 |
+
"grad_norm": 1.2651888132095337,
|
| 269 |
+
"learning_rate": 7.428571428571429e-05,
|
| 270 |
+
"loss": 1.1467,
|
| 271 |
+
"step": 90
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 12.0,
|
| 275 |
+
"eval_loss": 1.3095277547836304,
|
| 276 |
+
"eval_runtime": 17.0903,
|
| 277 |
+
"eval_samples_per_second": 0.41,
|
| 278 |
+
"eval_steps_per_second": 0.059,
|
| 279 |
+
"step": 90
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 12.67,
|
| 283 |
+
"grad_norm": 1.248926043510437,
|
| 284 |
+
"learning_rate": 7.285714285714286e-05,
|
| 285 |
+
"loss": 1.0922,
|
| 286 |
+
"step": 95
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 12.67,
|
| 290 |
+
"eval_loss": 1.2942878007888794,
|
| 291 |
+
"eval_runtime": 17.0947,
|
| 292 |
+
"eval_samples_per_second": 0.409,
|
| 293 |
+
"eval_steps_per_second": 0.058,
|
| 294 |
+
"step": 95
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 13.33,
|
| 298 |
+
"grad_norm": 1.896952509880066,
|
| 299 |
+
"learning_rate": 7.142857142857143e-05,
|
| 300 |
+
"loss": 1.0403,
|
| 301 |
+
"step": 100
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 13.33,
|
| 305 |
+
"eval_loss": 1.2803159952163696,
|
| 306 |
+
"eval_runtime": 17.0819,
|
| 307 |
+
"eval_samples_per_second": 0.41,
|
| 308 |
+
"eval_steps_per_second": 0.059,
|
| 309 |
+
"step": 100
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 14.0,
|
| 313 |
+
"grad_norm": 1.862244725227356,
|
| 314 |
+
"learning_rate": 7e-05,
|
| 315 |
+
"loss": 1.0049,
|
| 316 |
+
"step": 105
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 14.0,
|
| 320 |
+
"eval_loss": 1.2643567323684692,
|
| 321 |
+
"eval_runtime": 17.0849,
|
| 322 |
+
"eval_samples_per_second": 0.41,
|
| 323 |
+
"eval_steps_per_second": 0.059,
|
| 324 |
+
"step": 105
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 14.67,
|
| 328 |
+
"grad_norm": 1.7487821578979492,
|
| 329 |
+
"learning_rate": 6.857142857142858e-05,
|
| 330 |
+
"loss": 0.9262,
|
| 331 |
+
"step": 110
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 14.67,
|
| 335 |
+
"eval_loss": 1.2471646070480347,
|
| 336 |
+
"eval_runtime": 17.1278,
|
| 337 |
+
"eval_samples_per_second": 0.409,
|
| 338 |
+
"eval_steps_per_second": 0.058,
|
| 339 |
+
"step": 110
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 15.33,
|
| 343 |
+
"grad_norm": 1.838605284690857,
|
| 344 |
+
"learning_rate": 6.714285714285714e-05,
|
| 345 |
+
"loss": 0.8965,
|
| 346 |
+
"step": 115
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 15.33,
|
| 350 |
+
"eval_loss": 1.2377034425735474,
|
| 351 |
+
"eval_runtime": 17.0731,
|
| 352 |
+
"eval_samples_per_second": 0.41,
|
| 353 |
+
"eval_steps_per_second": 0.059,
|
| 354 |
+
"step": 115
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 16.0,
|
| 358 |
+
"grad_norm": 3.117398977279663,
|
| 359 |
+
"learning_rate": 6.571428571428571e-05,
|
| 360 |
+
"loss": 0.8581,
|
| 361 |
+
"step": 120
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 16.0,
|
| 365 |
+
"eval_loss": 1.2083133459091187,
|
| 366 |
+
"eval_runtime": 17.1304,
|
| 367 |
+
"eval_samples_per_second": 0.409,
|
| 368 |
+
"eval_steps_per_second": 0.058,
|
| 369 |
+
"step": 120
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 16.67,
|
| 373 |
+
"grad_norm": 2.5655250549316406,
|
| 374 |
+
"learning_rate": 6.428571428571429e-05,
|
| 375 |
+
"loss": 0.7929,
|
| 376 |
+
"step": 125
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 16.67,
|
| 380 |
+
"eval_loss": 1.1945828199386597,
|
| 381 |
+
"eval_runtime": 17.104,
|
| 382 |
+
"eval_samples_per_second": 0.409,
|
| 383 |
+
"eval_steps_per_second": 0.058,
|
| 384 |
+
"step": 125
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 17.33,
|
| 388 |
+
"grad_norm": 2.168546199798584,
|
| 389 |
+
"learning_rate": 6.285714285714286e-05,
|
| 390 |
+
"loss": 0.7543,
|
| 391 |
+
"step": 130
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 17.33,
|
| 395 |
+
"eval_loss": 1.1876276731491089,
|
| 396 |
+
"eval_runtime": 17.1046,
|
| 397 |
+
"eval_samples_per_second": 0.409,
|
| 398 |
+
"eval_steps_per_second": 0.058,
|
| 399 |
+
"step": 130
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 18.0,
|
| 403 |
+
"grad_norm": 2.5984208583831787,
|
| 404 |
+
"learning_rate": 6.142857142857143e-05,
|
| 405 |
+
"loss": 0.716,
|
| 406 |
+
"step": 135
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 18.0,
|
| 410 |
+
"eval_loss": 1.1714750528335571,
|
| 411 |
+
"eval_runtime": 17.0807,
|
| 412 |
+
"eval_samples_per_second": 0.41,
|
| 413 |
+
"eval_steps_per_second": 0.059,
|
| 414 |
+
"step": 135
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 18.67,
|
| 418 |
+
"grad_norm": 3.479024887084961,
|
| 419 |
+
"learning_rate": 6e-05,
|
| 420 |
+
"loss": 0.6681,
|
| 421 |
+
"step": 140
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 18.67,
|
| 425 |
+
"eval_loss": 1.169895052909851,
|
| 426 |
+
"eval_runtime": 17.0681,
|
| 427 |
+
"eval_samples_per_second": 0.41,
|
| 428 |
+
"eval_steps_per_second": 0.059,
|
| 429 |
+
"step": 140
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 19.33,
|
| 433 |
+
"grad_norm": 2.563386917114258,
|
| 434 |
+
"learning_rate": 5.8571428571428575e-05,
|
| 435 |
+
"loss": 0.6306,
|
| 436 |
+
"step": 145
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 19.33,
|
| 440 |
+
"eval_loss": 1.1741083860397339,
|
| 441 |
+
"eval_runtime": 17.0568,
|
| 442 |
+
"eval_samples_per_second": 0.41,
|
| 443 |
+
"eval_steps_per_second": 0.059,
|
| 444 |
+
"step": 145
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 20.0,
|
| 448 |
+
"grad_norm": 2.96592116355896,
|
| 449 |
+
"learning_rate": 5.714285714285714e-05,
|
| 450 |
+
"loss": 0.6183,
|
| 451 |
+
"step": 150
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 20.0,
|
| 455 |
+
"eval_loss": 1.1455965042114258,
|
| 456 |
+
"eval_runtime": 17.073,
|
| 457 |
+
"eval_samples_per_second": 0.41,
|
| 458 |
+
"eval_steps_per_second": 0.059,
|
| 459 |
+
"step": 150
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 20.67,
|
| 463 |
+
"grad_norm": 2.6751275062561035,
|
| 464 |
+
"learning_rate": 5.571428571428572e-05,
|
| 465 |
+
"loss": 0.5464,
|
| 466 |
+
"step": 155
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 20.67,
|
| 470 |
+
"eval_loss": 1.131102204322815,
|
| 471 |
+
"eval_runtime": 17.0578,
|
| 472 |
+
"eval_samples_per_second": 0.41,
|
| 473 |
+
"eval_steps_per_second": 0.059,
|
| 474 |
+
"step": 155
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 21.33,
|
| 478 |
+
"grad_norm": 2.3700051307678223,
|
| 479 |
+
"learning_rate": 5.428571428571428e-05,
|
| 480 |
+
"loss": 0.551,
|
| 481 |
+
"step": 160
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 21.33,
|
| 485 |
+
"eval_loss": 1.127384066581726,
|
| 486 |
+
"eval_runtime": 17.0546,
|
| 487 |
+
"eval_samples_per_second": 0.41,
|
| 488 |
+
"eval_steps_per_second": 0.059,
|
| 489 |
+
"step": 160
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 22.0,
|
| 493 |
+
"grad_norm": 3.3827567100524902,
|
| 494 |
+
"learning_rate": 5.285714285714286e-05,
|
| 495 |
+
"loss": 0.5179,
|
| 496 |
+
"step": 165
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 22.0,
|
| 500 |
+
"eval_loss": 1.111584186553955,
|
| 501 |
+
"eval_runtime": 17.0812,
|
| 502 |
+
"eval_samples_per_second": 0.41,
|
| 503 |
+
"eval_steps_per_second": 0.059,
|
| 504 |
+
"step": 165
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 22.67,
|
| 508 |
+
"grad_norm": 3.55790114402771,
|
| 509 |
+
"learning_rate": 5.142857142857143e-05,
|
| 510 |
+
"loss": 0.4831,
|
| 511 |
+
"step": 170
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 22.67,
|
| 515 |
+
"eval_loss": 1.0948525667190552,
|
| 516 |
+
"eval_runtime": 17.0547,
|
| 517 |
+
"eval_samples_per_second": 0.41,
|
| 518 |
+
"eval_steps_per_second": 0.059,
|
| 519 |
+
"step": 170
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 23.33,
|
| 523 |
+
"grad_norm": 3.0782699584960938,
|
| 524 |
+
"learning_rate": 5e-05,
|
| 525 |
+
"loss": 0.4587,
|
| 526 |
+
"step": 175
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 23.33,
|
| 530 |
+
"eval_loss": 1.0906586647033691,
|
| 531 |
+
"eval_runtime": 17.0666,
|
| 532 |
+
"eval_samples_per_second": 0.41,
|
| 533 |
+
"eval_steps_per_second": 0.059,
|
| 534 |
+
"step": 175
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 24.0,
|
| 538 |
+
"grad_norm": 3.3993167877197266,
|
| 539 |
+
"learning_rate": 4.8571428571428576e-05,
|
| 540 |
+
"loss": 0.4203,
|
| 541 |
+
"step": 180
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 24.0,
|
| 545 |
+
"eval_loss": 1.0688152313232422,
|
| 546 |
+
"eval_runtime": 17.0721,
|
| 547 |
+
"eval_samples_per_second": 0.41,
|
| 548 |
+
"eval_steps_per_second": 0.059,
|
| 549 |
+
"step": 180
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 24.67,
|
| 553 |
+
"grad_norm": 3.319303035736084,
|
| 554 |
+
"learning_rate": 4.714285714285714e-05,
|
| 555 |
+
"loss": 0.3975,
|
| 556 |
+
"step": 185
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 24.67,
|
| 560 |
+
"eval_loss": 1.0746583938598633,
|
| 561 |
+
"eval_runtime": 17.0709,
|
| 562 |
+
"eval_samples_per_second": 0.41,
|
| 563 |
+
"eval_steps_per_second": 0.059,
|
| 564 |
+
"step": 185
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 25.33,
|
| 568 |
+
"grad_norm": 2.4532127380371094,
|
| 569 |
+
"learning_rate": 4.5714285714285716e-05,
|
| 570 |
+
"loss": 0.3832,
|
| 571 |
+
"step": 190
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 25.33,
|
| 575 |
+
"eval_loss": 1.0772522687911987,
|
| 576 |
+
"eval_runtime": 17.0619,
|
| 577 |
+
"eval_samples_per_second": 0.41,
|
| 578 |
+
"eval_steps_per_second": 0.059,
|
| 579 |
+
"step": 190
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 26.0,
|
| 583 |
+
"grad_norm": 3.956822156906128,
|
| 584 |
+
"learning_rate": 4.428571428571428e-05,
|
| 585 |
+
"loss": 0.3725,
|
| 586 |
+
"step": 195
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 26.0,
|
| 590 |
+
"eval_loss": 1.0638784170150757,
|
| 591 |
+
"eval_runtime": 17.0807,
|
| 592 |
+
"eval_samples_per_second": 0.41,
|
| 593 |
+
"eval_steps_per_second": 0.059,
|
| 594 |
+
"step": 195
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 26.67,
|
| 598 |
+
"grad_norm": 2.76033353805542,
|
| 599 |
+
"learning_rate": 4.2857142857142856e-05,
|
| 600 |
+
"loss": 0.3473,
|
| 601 |
+
"step": 200
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 26.67,
|
| 605 |
+
"eval_loss": 1.04669988155365,
|
| 606 |
+
"eval_runtime": 17.0774,
|
| 607 |
+
"eval_samples_per_second": 0.41,
|
| 608 |
+
"eval_steps_per_second": 0.059,
|
| 609 |
+
"step": 200
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 27.33,
|
| 613 |
+
"grad_norm": 3.8683507442474365,
|
| 614 |
+
"learning_rate": 4.1428571428571437e-05,
|
| 615 |
+
"loss": 0.3243,
|
| 616 |
+
"step": 205
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 27.33,
|
| 620 |
+
"eval_loss": 1.0470303297042847,
|
| 621 |
+
"eval_runtime": 17.0718,
|
| 622 |
+
"eval_samples_per_second": 0.41,
|
| 623 |
+
"eval_steps_per_second": 0.059,
|
| 624 |
+
"step": 205
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 28.0,
|
| 628 |
+
"grad_norm": 4.535538196563721,
|
| 629 |
+
"learning_rate": 4e-05,
|
| 630 |
+
"loss": 0.3202,
|
| 631 |
+
"step": 210
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 28.0,
|
| 635 |
+
"eval_loss": 1.025539517402649,
|
| 636 |
+
"eval_runtime": 17.0604,
|
| 637 |
+
"eval_samples_per_second": 0.41,
|
| 638 |
+
"eval_steps_per_second": 0.059,
|
| 639 |
+
"step": 210
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 28.67,
|
| 643 |
+
"grad_norm": 2.6224355697631836,
|
| 644 |
+
"learning_rate": 3.857142857142858e-05,
|
| 645 |
+
"loss": 0.2958,
|
| 646 |
+
"step": 215
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 28.67,
|
| 650 |
+
"eval_loss": 1.0192126035690308,
|
| 651 |
+
"eval_runtime": 17.0657,
|
| 652 |
+
"eval_samples_per_second": 0.41,
|
| 653 |
+
"eval_steps_per_second": 0.059,
|
| 654 |
+
"step": 215
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 29.33,
|
| 658 |
+
"grad_norm": 2.5870041847229004,
|
| 659 |
+
"learning_rate": 3.7142857142857143e-05,
|
| 660 |
+
"loss": 0.2783,
|
| 661 |
+
"step": 220
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 29.33,
|
| 665 |
+
"eval_loss": 1.0211580991744995,
|
| 666 |
+
"eval_runtime": 17.0857,
|
| 667 |
+
"eval_samples_per_second": 0.41,
|
| 668 |
+
"eval_steps_per_second": 0.059,
|
| 669 |
+
"step": 220
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 30.0,
|
| 673 |
+
"grad_norm": 3.4565751552581787,
|
| 674 |
+
"learning_rate": 3.571428571428572e-05,
|
| 675 |
+
"loss": 0.2773,
|
| 676 |
+
"step": 225
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 30.0,
|
| 680 |
+
"eval_loss": 1.006419062614441,
|
| 681 |
+
"eval_runtime": 17.0807,
|
| 682 |
+
"eval_samples_per_second": 0.41,
|
| 683 |
+
"eval_steps_per_second": 0.059,
|
| 684 |
+
"step": 225
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 30.67,
|
| 688 |
+
"grad_norm": 2.4756500720977783,
|
| 689 |
+
"learning_rate": 3.428571428571429e-05,
|
| 690 |
+
"loss": 0.2482,
|
| 691 |
+
"step": 230
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 30.67,
|
| 695 |
+
"eval_loss": 1.0081219673156738,
|
| 696 |
+
"eval_runtime": 17.0576,
|
| 697 |
+
"eval_samples_per_second": 0.41,
|
| 698 |
+
"eval_steps_per_second": 0.059,
|
| 699 |
+
"step": 230
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 31.33,
|
| 703 |
+
"grad_norm": 2.38002610206604,
|
| 704 |
+
"learning_rate": 3.285714285714286e-05,
|
| 705 |
+
"loss": 0.2464,
|
| 706 |
+
"step": 235
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 31.33,
|
| 710 |
+
"eval_loss": 1.0151804685592651,
|
| 711 |
+
"eval_runtime": 17.0587,
|
| 712 |
+
"eval_samples_per_second": 0.41,
|
| 713 |
+
"eval_steps_per_second": 0.059,
|
| 714 |
+
"step": 235
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 32.0,
|
| 718 |
+
"grad_norm": 3.7081105709075928,
|
| 719 |
+
"learning_rate": 3.142857142857143e-05,
|
| 720 |
+
"loss": 0.2442,
|
| 721 |
+
"step": 240
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 32.0,
|
| 725 |
+
"eval_loss": 1.0032445192337036,
|
| 726 |
+
"eval_runtime": 17.1613,
|
| 727 |
+
"eval_samples_per_second": 0.408,
|
| 728 |
+
"eval_steps_per_second": 0.058,
|
| 729 |
+
"step": 240
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 32.67,
|
| 733 |
+
"grad_norm": 2.55924391746521,
|
| 734 |
+
"learning_rate": 3e-05,
|
| 735 |
+
"loss": 0.2193,
|
| 736 |
+
"step": 245
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 32.67,
|
| 740 |
+
"eval_loss": 0.9989615082740784,
|
| 741 |
+
"eval_runtime": 17.0447,
|
| 742 |
+
"eval_samples_per_second": 0.411,
|
| 743 |
+
"eval_steps_per_second": 0.059,
|
| 744 |
+
"step": 245
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 33.33,
|
| 748 |
+
"grad_norm": 1.9451407194137573,
|
| 749 |
+
"learning_rate": 2.857142857142857e-05,
|
| 750 |
+
"loss": 0.2101,
|
| 751 |
+
"step": 250
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 33.33,
|
| 755 |
+
"eval_loss": 1.0029457807540894,
|
| 756 |
+
"eval_runtime": 17.0816,
|
| 757 |
+
"eval_samples_per_second": 0.41,
|
| 758 |
+
"eval_steps_per_second": 0.059,
|
| 759 |
+
"step": 250
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 34.0,
|
| 763 |
+
"grad_norm": 2.713731527328491,
|
| 764 |
+
"learning_rate": 2.714285714285714e-05,
|
| 765 |
+
"loss": 0.2194,
|
| 766 |
+
"step": 255
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 34.0,
|
| 770 |
+
"eval_loss": 0.9959421753883362,
|
| 771 |
+
"eval_runtime": 17.0747,
|
| 772 |
+
"eval_samples_per_second": 0.41,
|
| 773 |
+
"eval_steps_per_second": 0.059,
|
| 774 |
+
"step": 255
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 34.67,
|
| 778 |
+
"grad_norm": 2.1633846759796143,
|
| 779 |
+
"learning_rate": 2.5714285714285714e-05,
|
| 780 |
+
"loss": 0.1958,
|
| 781 |
+
"step": 260
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 34.67,
|
| 785 |
+
"eval_loss": 0.9989770650863647,
|
| 786 |
+
"eval_runtime": 17.0821,
|
| 787 |
+
"eval_samples_per_second": 0.41,
|
| 788 |
+
"eval_steps_per_second": 0.059,
|
| 789 |
+
"step": 260
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 35.33,
|
| 793 |
+
"grad_norm": 3.9233529567718506,
|
| 794 |
+
"learning_rate": 2.4285714285714288e-05,
|
| 795 |
+
"loss": 0.1831,
|
| 796 |
+
"step": 265
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 35.33,
|
| 800 |
+
"eval_loss": 1.0072578191757202,
|
| 801 |
+
"eval_runtime": 17.0564,
|
| 802 |
+
"eval_samples_per_second": 0.41,
|
| 803 |
+
"eval_steps_per_second": 0.059,
|
| 804 |
+
"step": 265
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 36.0,
|
| 808 |
+
"grad_norm": 2.4143056869506836,
|
| 809 |
+
"learning_rate": 2.2857142857142858e-05,
|
| 810 |
+
"loss": 0.1753,
|
| 811 |
+
"step": 270
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 36.0,
|
| 815 |
+
"eval_loss": 0.9938892722129822,
|
| 816 |
+
"eval_runtime": 17.0668,
|
| 817 |
+
"eval_samples_per_second": 0.41,
|
| 818 |
+
"eval_steps_per_second": 0.059,
|
| 819 |
+
"step": 270
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 36.67,
|
| 823 |
+
"grad_norm": 2.706679582595825,
|
| 824 |
+
"learning_rate": 2.1428571428571428e-05,
|
| 825 |
+
"loss": 0.1698,
|
| 826 |
+
"step": 275
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 36.67,
|
| 830 |
+
"eval_loss": 0.9969200491905212,
|
| 831 |
+
"eval_runtime": 17.0643,
|
| 832 |
+
"eval_samples_per_second": 0.41,
|
| 833 |
+
"eval_steps_per_second": 0.059,
|
| 834 |
+
"step": 275
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 37.33,
|
| 838 |
+
"grad_norm": 1.872753620147705,
|
| 839 |
+
"learning_rate": 2e-05,
|
| 840 |
+
"loss": 0.16,
|
| 841 |
+
"step": 280
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 37.33,
|
| 845 |
+
"eval_loss": 0.9940390586853027,
|
| 846 |
+
"eval_runtime": 17.0728,
|
| 847 |
+
"eval_samples_per_second": 0.41,
|
| 848 |
+
"eval_steps_per_second": 0.059,
|
| 849 |
+
"step": 280
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 38.0,
|
| 853 |
+
"grad_norm": 2.7510581016540527,
|
| 854 |
+
"learning_rate": 1.8571428571428572e-05,
|
| 855 |
+
"loss": 0.1614,
|
| 856 |
+
"step": 285
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 38.0,
|
| 860 |
+
"eval_loss": 1.0066231489181519,
|
| 861 |
+
"eval_runtime": 17.072,
|
| 862 |
+
"eval_samples_per_second": 0.41,
|
| 863 |
+
"eval_steps_per_second": 0.059,
|
| 864 |
+
"step": 285
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 38.67,
|
| 868 |
+
"grad_norm": 1.8461092710494995,
|
| 869 |
+
"learning_rate": 1.7142857142857145e-05,
|
| 870 |
+
"loss": 0.1506,
|
| 871 |
+
"step": 290
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 38.67,
|
| 875 |
+
"eval_loss": 0.9927281737327576,
|
| 876 |
+
"eval_runtime": 17.0481,
|
| 877 |
+
"eval_samples_per_second": 0.411,
|
| 878 |
+
"eval_steps_per_second": 0.059,
|
| 879 |
+
"step": 290
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 39.33,
|
| 883 |
+
"grad_norm": 1.8425017595291138,
|
| 884 |
+
"learning_rate": 1.5714285714285715e-05,
|
| 885 |
+
"loss": 0.1419,
|
| 886 |
+
"step": 295
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 39.33,
|
| 890 |
+
"eval_loss": 1.0133570432662964,
|
| 891 |
+
"eval_runtime": 17.0642,
|
| 892 |
+
"eval_samples_per_second": 0.41,
|
| 893 |
+
"eval_steps_per_second": 0.059,
|
| 894 |
+
"step": 295
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 40.0,
|
| 898 |
+
"grad_norm": 2.0457987785339355,
|
| 899 |
+
"learning_rate": 1.4285714285714285e-05,
|
| 900 |
+
"loss": 0.1459,
|
| 901 |
+
"step": 300
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 40.0,
|
| 905 |
+
"eval_loss": 1.0127934217453003,
|
| 906 |
+
"eval_runtime": 17.0581,
|
| 907 |
+
"eval_samples_per_second": 0.41,
|
| 908 |
+
"eval_steps_per_second": 0.059,
|
| 909 |
+
"step": 300
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 40.67,
|
| 913 |
+
"grad_norm": 1.5630775690078735,
|
| 914 |
+
"learning_rate": 1.2857142857142857e-05,
|
| 915 |
+
"loss": 0.1225,
|
| 916 |
+
"step": 305
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 40.67,
|
| 920 |
+
"eval_loss": 1.0092624425888062,
|
| 921 |
+
"eval_runtime": 17.0483,
|
| 922 |
+
"eval_samples_per_second": 0.411,
|
| 923 |
+
"eval_steps_per_second": 0.059,
|
| 924 |
+
"step": 305
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 41.33,
|
| 928 |
+
"grad_norm": 1.37598717212677,
|
| 929 |
+
"learning_rate": 1.1428571428571429e-05,
|
| 930 |
+
"loss": 0.146,
|
| 931 |
+
"step": 310
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 41.33,
|
| 935 |
+
"eval_loss": 1.0083317756652832,
|
| 936 |
+
"eval_runtime": 17.0804,
|
| 937 |
+
"eval_samples_per_second": 0.41,
|
| 938 |
+
"eval_steps_per_second": 0.059,
|
| 939 |
+
"step": 310
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 42.0,
|
| 943 |
+
"grad_norm": 1.8867217302322388,
|
| 944 |
+
"learning_rate": 1e-05,
|
| 945 |
+
"loss": 0.13,
|
| 946 |
+
"step": 315
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 42.0,
|
| 950 |
+
"eval_loss": 1.0165104866027832,
|
| 951 |
+
"eval_runtime": 17.0621,
|
| 952 |
+
"eval_samples_per_second": 0.41,
|
| 953 |
+
"eval_steps_per_second": 0.059,
|
| 954 |
+
"step": 315
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 42.67,
|
| 958 |
+
"grad_norm": 1.4643555879592896,
|
| 959 |
+
"learning_rate": 8.571428571428573e-06,
|
| 960 |
+
"loss": 0.131,
|
| 961 |
+
"step": 320
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 42.67,
|
| 965 |
+
"eval_loss": 1.0264887809753418,
|
| 966 |
+
"eval_runtime": 17.0554,
|
| 967 |
+
"eval_samples_per_second": 0.41,
|
| 968 |
+
"eval_steps_per_second": 0.059,
|
| 969 |
+
"step": 320
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 43.33,
|
| 973 |
+
"grad_norm": 1.4163501262664795,
|
| 974 |
+
"learning_rate": 7.142857142857143e-06,
|
| 975 |
+
"loss": 0.1205,
|
| 976 |
+
"step": 325
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 43.33,
|
| 980 |
+
"eval_loss": 1.0175670385360718,
|
| 981 |
+
"eval_runtime": 17.0692,
|
| 982 |
+
"eval_samples_per_second": 0.41,
|
| 983 |
+
"eval_steps_per_second": 0.059,
|
| 984 |
+
"step": 325
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 44.0,
|
| 988 |
+
"grad_norm": 1.730556607246399,
|
| 989 |
+
"learning_rate": 5.7142857142857145e-06,
|
| 990 |
+
"loss": 0.1143,
|
| 991 |
+
"step": 330
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 44.0,
|
| 995 |
+
"eval_loss": 1.0272893905639648,
|
| 996 |
+
"eval_runtime": 17.0651,
|
| 997 |
+
"eval_samples_per_second": 0.41,
|
| 998 |
+
"eval_steps_per_second": 0.059,
|
| 999 |
+
"step": 330
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 44.67,
|
| 1003 |
+
"grad_norm": 1.451904058456421,
|
| 1004 |
+
"learning_rate": 4.285714285714286e-06,
|
| 1005 |
+
"loss": 0.1087,
|
| 1006 |
+
"step": 335
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 44.67,
|
| 1010 |
+
"eval_loss": 1.0272446870803833,
|
| 1011 |
+
"eval_runtime": 17.0604,
|
| 1012 |
+
"eval_samples_per_second": 0.41,
|
| 1013 |
+
"eval_steps_per_second": 0.059,
|
| 1014 |
+
"step": 335
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 45.33,
|
| 1018 |
+
"grad_norm": 1.1883801221847534,
|
| 1019 |
+
"learning_rate": 2.8571428571428573e-06,
|
| 1020 |
+
"loss": 0.1173,
|
| 1021 |
+
"step": 340
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 45.33,
|
| 1025 |
+
"eval_loss": 1.0314446687698364,
|
| 1026 |
+
"eval_runtime": 17.0652,
|
| 1027 |
+
"eval_samples_per_second": 0.41,
|
| 1028 |
+
"eval_steps_per_second": 0.059,
|
| 1029 |
+
"step": 340
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 46.0,
|
| 1033 |
+
"grad_norm": 1.514420747756958,
|
| 1034 |
+
"learning_rate": 1.4285714285714286e-06,
|
| 1035 |
+
"loss": 0.1227,
|
| 1036 |
+
"step": 345
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 46.0,
|
| 1040 |
+
"eval_loss": 1.0313900709152222,
|
| 1041 |
+
"eval_runtime": 17.0795,
|
| 1042 |
+
"eval_samples_per_second": 0.41,
|
| 1043 |
+
"eval_steps_per_second": 0.059,
|
| 1044 |
+
"step": 345
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 46.67,
|
| 1048 |
+
"grad_norm": 1.1168299913406372,
|
| 1049 |
+
"learning_rate": 0.0,
|
| 1050 |
+
"loss": 0.1133,
|
| 1051 |
+
"step": 350
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 46.67,
|
| 1055 |
+
"eval_loss": 1.032753586769104,
|
| 1056 |
+
"eval_runtime": 17.0596,
|
| 1057 |
+
"eval_samples_per_second": 0.41,
|
| 1058 |
+
"eval_steps_per_second": 0.059,
|
| 1059 |
+
"step": 350
|
| 1060 |
+
}
|
| 1061 |
+
],
|
| 1062 |
+
"logging_steps": 5,
|
| 1063 |
+
"max_steps": 350,
|
| 1064 |
+
"num_input_tokens_seen": 0,
|
| 1065 |
+
"num_train_epochs": 50,
|
| 1066 |
+
"save_steps": 10,
|
| 1067 |
+
"total_flos": 2.325932434400477e+18,
|
| 1068 |
+
"train_batch_size": 2,
|
| 1069 |
+
"trial_name": null,
|
| 1070 |
+
"trial_params": null
|
| 1071 |
+
}
|
checkpoint-350/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ed0ba6fb9a88dad56f61d9306f17b4e66e8767d898772faa97871a1388e82cf
|
| 3 |
+
size 4920
|
config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "meta-llama/Llama-2-70b-hf",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 8192,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 28672,
|
| 14 |
+
"max_position_embeddings": 4096,
|
| 15 |
+
"model_type": "llama",
|
| 16 |
+
"num_attention_heads": 64,
|
| 17 |
+
"num_hidden_layers": 80,
|
| 18 |
+
"num_key_value_heads": 8,
|
| 19 |
+
"pretraining_tp": 1,
|
| 20 |
+
"quantization_config": {
|
| 21 |
+
"_load_in_4bit": false,
|
| 22 |
+
"_load_in_8bit": false,
|
| 23 |
+
"bnb_4bit_compute_dtype": "float32",
|
| 24 |
+
"bnb_4bit_quant_type": "fp4",
|
| 25 |
+
"bnb_4bit_use_double_quant": false,
|
| 26 |
+
"llm_int8_enable_fp32_cpu_offload": false,
|
| 27 |
+
"llm_int8_has_fp16_weight": false,
|
| 28 |
+
"llm_int8_skip_modules": null,
|
| 29 |
+
"llm_int8_threshold": 6.0,
|
| 30 |
+
"load_in_4bit": false,
|
| 31 |
+
"load_in_8bit": false,
|
| 32 |
+
"quant_method": "bitsandbytes"
|
| 33 |
+
},
|
| 34 |
+
"rms_norm_eps": 1e-05,
|
| 35 |
+
"rope_scaling": null,
|
| 36 |
+
"rope_theta": 10000.0,
|
| 37 |
+
"tie_word_embeddings": false,
|
| 38 |
+
"torch_dtype": "float16",
|
| 39 |
+
"transformers_version": "4.38.1",
|
| 40 |
+
"use_cache": true,
|
| 41 |
+
"vocab_size": 32000
|
| 42 |
+
}
|
logs/events.out.tfevents.1699989718.node0370.palmetto.clemson.edu.1119956.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a8338f7ab492bb36430b90fde21a7822be3a5f94276aa93f72ad83aeaf872e40
|
| 3 |
+
size 4868
|
logs/events.out.tfevents.1699989813.node0370.palmetto.clemson.edu.1119956.4
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4773b11c0e593bac1fe819e5c6324bb2a3542c475a7d375bb608e46408b2b009
|
| 3 |
+
size 4602
|
logs/events.out.tfevents.1699994287.node0370.palmetto.clemson.edu.1126403.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:edff87033b1a80ab70d0a0af49d7943790a698e659a0875f929d78a3299c699d
|
| 3 |
+
size 7962
|
logs/events.out.tfevents.1699995040.node0370.palmetto.clemson.edu.1127435.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f054998bd37e1977597fdb66e495fb58c70a3b0042fa2ef4428756252ee49a4c
|
| 3 |
+
size 33937
|
logs/events.out.tfevents.1700064888.node0277.palmetto.clemson.edu.1971495.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:587a2145d6c92f9cbc7c114d9b2d452e13239b0a925f3bdcaa5631c48bdca19a
|
| 3 |
+
size 9908
|
logs/events.out.tfevents.1700066157.node0277.palmetto.clemson.edu.1973537.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5fcffea7a0e86207d5c20daaabb71040aa71cb529a4bbddc4a309e37d9666f12
|
| 3 |
+
size 14528
|