Upload training_log.json
Browse files- training_log.json +18 -0
training_log.json
ADDED
|
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"base_model_name": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
|
| 3 |
+
"base_model_class": "LlamaForCausalLM",
|
| 4 |
+
"base_loaded_in_4bit": false,
|
| 5 |
+
"base_loaded_in_8bit": true,
|
| 6 |
+
"projections": "q, k, v, o",
|
| 7 |
+
"loss": 1.4473,
|
| 8 |
+
"learning_rate": 2.364532019704433e-06,
|
| 9 |
+
"epoch": 0.25,
|
| 10 |
+
"current_steps": 2149,
|
| 11 |
+
"current_steps_adjusted": 2149,
|
| 12 |
+
"epoch_adjusted": 0.25,
|
| 13 |
+
"train_runtime": 6758.723,
|
| 14 |
+
"train_samples_per_second": 1.907,
|
| 15 |
+
"train_steps_per_second": 0.159,
|
| 16 |
+
"total_flos": 6.25660356722688e+16,
|
| 17 |
+
"train_loss": 1.498897494826206
|
| 18 |
+
}
|