Uploading model files
Browse files- config.json +46 -0
- generation_config.json +6 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +848 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "distilgpt2",
|
| 3 |
+
"_num_labels": 1,
|
| 4 |
+
"activation_function": "gelu_new",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"GPT2LMHeadModel"
|
| 7 |
+
],
|
| 8 |
+
"attn_pdrop": 0.1,
|
| 9 |
+
"bos_token_id": 50256,
|
| 10 |
+
"embd_pdrop": 0.1,
|
| 11 |
+
"eos_token_id": 50256,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "LABEL_0"
|
| 14 |
+
},
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"label2id": {
|
| 17 |
+
"LABEL_0": 0
|
| 18 |
+
},
|
| 19 |
+
"layer_norm_epsilon": 1e-05,
|
| 20 |
+
"model_type": "gpt2",
|
| 21 |
+
"n_ctx": 1024,
|
| 22 |
+
"n_embd": 768,
|
| 23 |
+
"n_head": 12,
|
| 24 |
+
"n_inner": null,
|
| 25 |
+
"n_layer": 6,
|
| 26 |
+
"n_positions": 1024,
|
| 27 |
+
"reorder_and_upcast_attn": false,
|
| 28 |
+
"resid_pdrop": 0.1,
|
| 29 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 30 |
+
"scale_attn_weights": true,
|
| 31 |
+
"summary_activation": null,
|
| 32 |
+
"summary_first_dropout": 0.1,
|
| 33 |
+
"summary_proj_to_labels": true,
|
| 34 |
+
"summary_type": "cls_index",
|
| 35 |
+
"summary_use_proj": true,
|
| 36 |
+
"task_specific_params": {
|
| 37 |
+
"text-generation": {
|
| 38 |
+
"do_sample": true,
|
| 39 |
+
"max_length": 50
|
| 40 |
+
}
|
| 41 |
+
},
|
| 42 |
+
"torch_dtype": "float32",
|
| 43 |
+
"transformers_version": "4.38.2",
|
| 44 |
+
"use_cache": true,
|
| 45 |
+
"vocab_size": 50257
|
| 46 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 50256,
|
| 4 |
+
"eos_token_id": 50256,
|
| 5 |
+
"transformers_version": "4.38.2"
|
| 6 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6003283563edf92920968a45175d1f360227a7d4928a3c2241b97e5426899e9e
|
| 3 |
+
size 327657928
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c87093a3e5f2ffd8d01e35357728b9b06ec8f233262937b6f5bdfdff53ccfcb5
|
| 3 |
+
size 655364474
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c6abedc9b22385358e137f8bcd69ce1b9b6a63ef91d2c5e9da0c205e4afa508e
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:10ac3dbab3894fa91ac9181b6ac678310158f3bb546eda06cc32e36036f0592a
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,848 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 100.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 2500,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"eval_loss": 0.7225164175033569,
|
| 14 |
+
"eval_runtime": 2.4692,
|
| 15 |
+
"eval_samples_per_second": 80.998,
|
| 16 |
+
"eval_steps_per_second": 10.125,
|
| 17 |
+
"step": 25
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 2.0,
|
| 21 |
+
"eval_loss": 0.5052716732025146,
|
| 22 |
+
"eval_runtime": 2.4705,
|
| 23 |
+
"eval_samples_per_second": 80.955,
|
| 24 |
+
"eval_steps_per_second": 10.119,
|
| 25 |
+
"step": 50
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 3.0,
|
| 29 |
+
"eval_loss": 0.4475671350955963,
|
| 30 |
+
"eval_runtime": 2.4731,
|
| 31 |
+
"eval_samples_per_second": 80.871,
|
| 32 |
+
"eval_steps_per_second": 10.109,
|
| 33 |
+
"step": 75
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 4.0,
|
| 37 |
+
"eval_loss": 0.4104467034339905,
|
| 38 |
+
"eval_runtime": 2.4763,
|
| 39 |
+
"eval_samples_per_second": 80.765,
|
| 40 |
+
"eval_steps_per_second": 10.096,
|
| 41 |
+
"step": 100
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 5.0,
|
| 45 |
+
"eval_loss": 0.39656341075897217,
|
| 46 |
+
"eval_runtime": 2.4766,
|
| 47 |
+
"eval_samples_per_second": 80.755,
|
| 48 |
+
"eval_steps_per_second": 10.094,
|
| 49 |
+
"step": 125
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 6.0,
|
| 53 |
+
"eval_loss": 0.36905747652053833,
|
| 54 |
+
"eval_runtime": 2.4774,
|
| 55 |
+
"eval_samples_per_second": 80.731,
|
| 56 |
+
"eval_steps_per_second": 10.091,
|
| 57 |
+
"step": 150
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 7.0,
|
| 61 |
+
"eval_loss": 0.362547367811203,
|
| 62 |
+
"eval_runtime": 2.4792,
|
| 63 |
+
"eval_samples_per_second": 80.672,
|
| 64 |
+
"eval_steps_per_second": 10.084,
|
| 65 |
+
"step": 175
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 8.0,
|
| 69 |
+
"eval_loss": 0.3497239053249359,
|
| 70 |
+
"eval_runtime": 2.4872,
|
| 71 |
+
"eval_samples_per_second": 80.413,
|
| 72 |
+
"eval_steps_per_second": 10.052,
|
| 73 |
+
"step": 200
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 9.0,
|
| 77 |
+
"eval_loss": 0.3523653447628021,
|
| 78 |
+
"eval_runtime": 2.4878,
|
| 79 |
+
"eval_samples_per_second": 80.392,
|
| 80 |
+
"eval_steps_per_second": 10.049,
|
| 81 |
+
"step": 225
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 10.0,
|
| 85 |
+
"eval_loss": 0.33985191583633423,
|
| 86 |
+
"eval_runtime": 2.4877,
|
| 87 |
+
"eval_samples_per_second": 80.395,
|
| 88 |
+
"eval_steps_per_second": 10.049,
|
| 89 |
+
"step": 250
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 11.0,
|
| 93 |
+
"eval_loss": 0.3415805399417877,
|
| 94 |
+
"eval_runtime": 2.4896,
|
| 95 |
+
"eval_samples_per_second": 80.333,
|
| 96 |
+
"eval_steps_per_second": 10.042,
|
| 97 |
+
"step": 275
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 12.0,
|
| 101 |
+
"eval_loss": 0.33282220363616943,
|
| 102 |
+
"eval_runtime": 2.49,
|
| 103 |
+
"eval_samples_per_second": 80.321,
|
| 104 |
+
"eval_steps_per_second": 10.04,
|
| 105 |
+
"step": 300
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 13.0,
|
| 109 |
+
"eval_loss": 0.33800217509269714,
|
| 110 |
+
"eval_runtime": 2.491,
|
| 111 |
+
"eval_samples_per_second": 80.289,
|
| 112 |
+
"eval_steps_per_second": 10.036,
|
| 113 |
+
"step": 325
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 14.0,
|
| 117 |
+
"eval_loss": 0.3342490792274475,
|
| 118 |
+
"eval_runtime": 2.491,
|
| 119 |
+
"eval_samples_per_second": 80.288,
|
| 120 |
+
"eval_steps_per_second": 10.036,
|
| 121 |
+
"step": 350
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 15.0,
|
| 125 |
+
"eval_loss": 0.34117013216018677,
|
| 126 |
+
"eval_runtime": 2.4915,
|
| 127 |
+
"eval_samples_per_second": 80.273,
|
| 128 |
+
"eval_steps_per_second": 10.034,
|
| 129 |
+
"step": 375
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 16.0,
|
| 133 |
+
"eval_loss": 0.3388213813304901,
|
| 134 |
+
"eval_runtime": 2.4936,
|
| 135 |
+
"eval_samples_per_second": 80.205,
|
| 136 |
+
"eval_steps_per_second": 10.026,
|
| 137 |
+
"step": 400
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 17.0,
|
| 141 |
+
"eval_loss": 0.319783091545105,
|
| 142 |
+
"eval_runtime": 2.4927,
|
| 143 |
+
"eval_samples_per_second": 80.234,
|
| 144 |
+
"eval_steps_per_second": 10.029,
|
| 145 |
+
"step": 425
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 18.0,
|
| 149 |
+
"eval_loss": 0.3183771073818207,
|
| 150 |
+
"eval_runtime": 2.4921,
|
| 151 |
+
"eval_samples_per_second": 80.253,
|
| 152 |
+
"eval_steps_per_second": 10.032,
|
| 153 |
+
"step": 450
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 19.0,
|
| 157 |
+
"eval_loss": 0.31770122051239014,
|
| 158 |
+
"eval_runtime": 2.4935,
|
| 159 |
+
"eval_samples_per_second": 80.21,
|
| 160 |
+
"eval_steps_per_second": 10.026,
|
| 161 |
+
"step": 475
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 20.0,
|
| 165 |
+
"grad_norm": 0.8955293893814087,
|
| 166 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 167 |
+
"loss": 0.4631,
|
| 168 |
+
"step": 500
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 20.0,
|
| 172 |
+
"eval_loss": 0.31926316022872925,
|
| 173 |
+
"eval_runtime": 2.4537,
|
| 174 |
+
"eval_samples_per_second": 81.51,
|
| 175 |
+
"eval_steps_per_second": 10.189,
|
| 176 |
+
"step": 500
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 21.0,
|
| 180 |
+
"eval_loss": 0.3148637115955353,
|
| 181 |
+
"eval_runtime": 2.4822,
|
| 182 |
+
"eval_samples_per_second": 80.572,
|
| 183 |
+
"eval_steps_per_second": 10.072,
|
| 184 |
+
"step": 525
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 22.0,
|
| 188 |
+
"eval_loss": 0.31756889820098877,
|
| 189 |
+
"eval_runtime": 2.4835,
|
| 190 |
+
"eval_samples_per_second": 80.531,
|
| 191 |
+
"eval_steps_per_second": 10.066,
|
| 192 |
+
"step": 550
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 23.0,
|
| 196 |
+
"eval_loss": 0.3171720504760742,
|
| 197 |
+
"eval_runtime": 2.4904,
|
| 198 |
+
"eval_samples_per_second": 80.309,
|
| 199 |
+
"eval_steps_per_second": 10.039,
|
| 200 |
+
"step": 575
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 24.0,
|
| 204 |
+
"eval_loss": 0.3179776072502136,
|
| 205 |
+
"eval_runtime": 2.4915,
|
| 206 |
+
"eval_samples_per_second": 80.273,
|
| 207 |
+
"eval_steps_per_second": 10.034,
|
| 208 |
+
"step": 600
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 25.0,
|
| 212 |
+
"eval_loss": 0.3148040473461151,
|
| 213 |
+
"eval_runtime": 2.4928,
|
| 214 |
+
"eval_samples_per_second": 80.232,
|
| 215 |
+
"eval_steps_per_second": 10.029,
|
| 216 |
+
"step": 625
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 26.0,
|
| 220 |
+
"eval_loss": 0.3072579503059387,
|
| 221 |
+
"eval_runtime": 2.4926,
|
| 222 |
+
"eval_samples_per_second": 80.239,
|
| 223 |
+
"eval_steps_per_second": 10.03,
|
| 224 |
+
"step": 650
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 27.0,
|
| 228 |
+
"eval_loss": 0.3129171133041382,
|
| 229 |
+
"eval_runtime": 2.4936,
|
| 230 |
+
"eval_samples_per_second": 80.206,
|
| 231 |
+
"eval_steps_per_second": 10.026,
|
| 232 |
+
"step": 675
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 28.0,
|
| 236 |
+
"eval_loss": 0.3081643283367157,
|
| 237 |
+
"eval_runtime": 2.4941,
|
| 238 |
+
"eval_samples_per_second": 80.188,
|
| 239 |
+
"eval_steps_per_second": 10.024,
|
| 240 |
+
"step": 700
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 29.0,
|
| 244 |
+
"eval_loss": 0.3064418435096741,
|
| 245 |
+
"eval_runtime": 2.4964,
|
| 246 |
+
"eval_samples_per_second": 80.116,
|
| 247 |
+
"eval_steps_per_second": 10.014,
|
| 248 |
+
"step": 725
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 30.0,
|
| 252 |
+
"eval_loss": 0.30982914566993713,
|
| 253 |
+
"eval_runtime": 2.4963,
|
| 254 |
+
"eval_samples_per_second": 80.12,
|
| 255 |
+
"eval_steps_per_second": 10.015,
|
| 256 |
+
"step": 750
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 31.0,
|
| 260 |
+
"eval_loss": 0.3063754439353943,
|
| 261 |
+
"eval_runtime": 2.4956,
|
| 262 |
+
"eval_samples_per_second": 80.142,
|
| 263 |
+
"eval_steps_per_second": 10.018,
|
| 264 |
+
"step": 775
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 32.0,
|
| 268 |
+
"eval_loss": 0.3113013207912445,
|
| 269 |
+
"eval_runtime": 2.4972,
|
| 270 |
+
"eval_samples_per_second": 80.089,
|
| 271 |
+
"eval_steps_per_second": 10.011,
|
| 272 |
+
"step": 800
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 33.0,
|
| 276 |
+
"eval_loss": 0.30704861879348755,
|
| 277 |
+
"eval_runtime": 2.4962,
|
| 278 |
+
"eval_samples_per_second": 80.122,
|
| 279 |
+
"eval_steps_per_second": 10.015,
|
| 280 |
+
"step": 825
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 34.0,
|
| 284 |
+
"eval_loss": 0.2988375425338745,
|
| 285 |
+
"eval_runtime": 2.4974,
|
| 286 |
+
"eval_samples_per_second": 80.083,
|
| 287 |
+
"eval_steps_per_second": 10.01,
|
| 288 |
+
"step": 850
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 35.0,
|
| 292 |
+
"eval_loss": 0.3142584264278412,
|
| 293 |
+
"eval_runtime": 2.4958,
|
| 294 |
+
"eval_samples_per_second": 80.135,
|
| 295 |
+
"eval_steps_per_second": 10.017,
|
| 296 |
+
"step": 875
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 36.0,
|
| 300 |
+
"eval_loss": 0.3032761514186859,
|
| 301 |
+
"eval_runtime": 2.4976,
|
| 302 |
+
"eval_samples_per_second": 80.077,
|
| 303 |
+
"eval_steps_per_second": 10.01,
|
| 304 |
+
"step": 900
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 37.0,
|
| 308 |
+
"eval_loss": 0.30415403842926025,
|
| 309 |
+
"eval_runtime": 2.4973,
|
| 310 |
+
"eval_samples_per_second": 80.087,
|
| 311 |
+
"eval_steps_per_second": 10.011,
|
| 312 |
+
"step": 925
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 38.0,
|
| 316 |
+
"eval_loss": 0.30165913701057434,
|
| 317 |
+
"eval_runtime": 2.4982,
|
| 318 |
+
"eval_samples_per_second": 80.057,
|
| 319 |
+
"eval_steps_per_second": 10.007,
|
| 320 |
+
"step": 950
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 39.0,
|
| 324 |
+
"eval_loss": 0.3017444908618927,
|
| 325 |
+
"eval_runtime": 2.4977,
|
| 326 |
+
"eval_samples_per_second": 80.072,
|
| 327 |
+
"eval_steps_per_second": 10.009,
|
| 328 |
+
"step": 975
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 40.0,
|
| 332 |
+
"grad_norm": 2.4287023544311523,
|
| 333 |
+
"learning_rate": 1.2e-05,
|
| 334 |
+
"loss": 0.3457,
|
| 335 |
+
"step": 1000
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 40.0,
|
| 339 |
+
"eval_loss": 0.3025864064693451,
|
| 340 |
+
"eval_runtime": 2.4539,
|
| 341 |
+
"eval_samples_per_second": 81.504,
|
| 342 |
+
"eval_steps_per_second": 10.188,
|
| 343 |
+
"step": 1000
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 41.0,
|
| 347 |
+
"eval_loss": 0.30045461654663086,
|
| 348 |
+
"eval_runtime": 2.4819,
|
| 349 |
+
"eval_samples_per_second": 80.584,
|
| 350 |
+
"eval_steps_per_second": 10.073,
|
| 351 |
+
"step": 1025
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 42.0,
|
| 355 |
+
"eval_loss": 0.30064135789871216,
|
| 356 |
+
"eval_runtime": 2.4896,
|
| 357 |
+
"eval_samples_per_second": 80.334,
|
| 358 |
+
"eval_steps_per_second": 10.042,
|
| 359 |
+
"step": 1050
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 43.0,
|
| 363 |
+
"eval_loss": 0.29575350880622864,
|
| 364 |
+
"eval_runtime": 2.4903,
|
| 365 |
+
"eval_samples_per_second": 80.312,
|
| 366 |
+
"eval_steps_per_second": 10.039,
|
| 367 |
+
"step": 1075
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 44.0,
|
| 371 |
+
"eval_loss": 0.30160149931907654,
|
| 372 |
+
"eval_runtime": 2.4926,
|
| 373 |
+
"eval_samples_per_second": 80.239,
|
| 374 |
+
"eval_steps_per_second": 10.03,
|
| 375 |
+
"step": 1100
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 45.0,
|
| 379 |
+
"eval_loss": 0.30429255962371826,
|
| 380 |
+
"eval_runtime": 2.4921,
|
| 381 |
+
"eval_samples_per_second": 80.255,
|
| 382 |
+
"eval_steps_per_second": 10.032,
|
| 383 |
+
"step": 1125
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 46.0,
|
| 387 |
+
"eval_loss": 0.3015859127044678,
|
| 388 |
+
"eval_runtime": 2.4927,
|
| 389 |
+
"eval_samples_per_second": 80.236,
|
| 390 |
+
"eval_steps_per_second": 10.029,
|
| 391 |
+
"step": 1150
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 47.0,
|
| 395 |
+
"eval_loss": 0.29914650321006775,
|
| 396 |
+
"eval_runtime": 2.4947,
|
| 397 |
+
"eval_samples_per_second": 80.171,
|
| 398 |
+
"eval_steps_per_second": 10.021,
|
| 399 |
+
"step": 1175
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 48.0,
|
| 403 |
+
"eval_loss": 0.2971905469894409,
|
| 404 |
+
"eval_runtime": 2.4958,
|
| 405 |
+
"eval_samples_per_second": 80.134,
|
| 406 |
+
"eval_steps_per_second": 10.017,
|
| 407 |
+
"step": 1200
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 49.0,
|
| 411 |
+
"eval_loss": 0.29176658391952515,
|
| 412 |
+
"eval_runtime": 2.4963,
|
| 413 |
+
"eval_samples_per_second": 80.118,
|
| 414 |
+
"eval_steps_per_second": 10.015,
|
| 415 |
+
"step": 1225
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 50.0,
|
| 419 |
+
"eval_loss": 0.2934282720088959,
|
| 420 |
+
"eval_runtime": 2.4976,
|
| 421 |
+
"eval_samples_per_second": 80.076,
|
| 422 |
+
"eval_steps_per_second": 10.01,
|
| 423 |
+
"step": 1250
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 51.0,
|
| 427 |
+
"eval_loss": 0.2918751835823059,
|
| 428 |
+
"eval_runtime": 2.4964,
|
| 429 |
+
"eval_samples_per_second": 80.115,
|
| 430 |
+
"eval_steps_per_second": 10.014,
|
| 431 |
+
"step": 1275
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 52.0,
|
| 435 |
+
"eval_loss": 0.2914879620075226,
|
| 436 |
+
"eval_runtime": 2.4977,
|
| 437 |
+
"eval_samples_per_second": 80.075,
|
| 438 |
+
"eval_steps_per_second": 10.009,
|
| 439 |
+
"step": 1300
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 53.0,
|
| 443 |
+
"eval_loss": 0.2925909757614136,
|
| 444 |
+
"eval_runtime": 2.4975,
|
| 445 |
+
"eval_samples_per_second": 80.081,
|
| 446 |
+
"eval_steps_per_second": 10.01,
|
| 447 |
+
"step": 1325
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 54.0,
|
| 451 |
+
"eval_loss": 0.2940743565559387,
|
| 452 |
+
"eval_runtime": 2.4974,
|
| 453 |
+
"eval_samples_per_second": 80.085,
|
| 454 |
+
"eval_steps_per_second": 10.011,
|
| 455 |
+
"step": 1350
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 55.0,
|
| 459 |
+
"eval_loss": 0.2973780930042267,
|
| 460 |
+
"eval_runtime": 2.4989,
|
| 461 |
+
"eval_samples_per_second": 80.037,
|
| 462 |
+
"eval_steps_per_second": 10.005,
|
| 463 |
+
"step": 1375
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 56.0,
|
| 467 |
+
"eval_loss": 0.2954687178134918,
|
| 468 |
+
"eval_runtime": 2.498,
|
| 469 |
+
"eval_samples_per_second": 80.065,
|
| 470 |
+
"eval_steps_per_second": 10.008,
|
| 471 |
+
"step": 1400
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 57.0,
|
| 475 |
+
"eval_loss": 0.29051879048347473,
|
| 476 |
+
"eval_runtime": 2.4979,
|
| 477 |
+
"eval_samples_per_second": 80.069,
|
| 478 |
+
"eval_steps_per_second": 10.009,
|
| 479 |
+
"step": 1425
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 58.0,
|
| 483 |
+
"eval_loss": 0.29731473326683044,
|
| 484 |
+
"eval_runtime": 2.4993,
|
| 485 |
+
"eval_samples_per_second": 80.023,
|
| 486 |
+
"eval_steps_per_second": 10.003,
|
| 487 |
+
"step": 1450
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 59.0,
|
| 491 |
+
"eval_loss": 0.2933524250984192,
|
| 492 |
+
"eval_runtime": 2.4988,
|
| 493 |
+
"eval_samples_per_second": 80.04,
|
| 494 |
+
"eval_steps_per_second": 10.005,
|
| 495 |
+
"step": 1475
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 60.0,
|
| 499 |
+
"grad_norm": 1.120781421661377,
|
| 500 |
+
"learning_rate": 8.000000000000001e-06,
|
| 501 |
+
"loss": 0.3291,
|
| 502 |
+
"step": 1500
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 60.0,
|
| 506 |
+
"eval_loss": 0.2888854742050171,
|
| 507 |
+
"eval_runtime": 2.456,
|
| 508 |
+
"eval_samples_per_second": 81.434,
|
| 509 |
+
"eval_steps_per_second": 10.179,
|
| 510 |
+
"step": 1500
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 61.0,
|
| 514 |
+
"eval_loss": 0.2901514172554016,
|
| 515 |
+
"eval_runtime": 2.4865,
|
| 516 |
+
"eval_samples_per_second": 80.436,
|
| 517 |
+
"eval_steps_per_second": 10.054,
|
| 518 |
+
"step": 1525
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 62.0,
|
| 522 |
+
"eval_loss": 0.29295194149017334,
|
| 523 |
+
"eval_runtime": 2.4921,
|
| 524 |
+
"eval_samples_per_second": 80.253,
|
| 525 |
+
"eval_steps_per_second": 10.032,
|
| 526 |
+
"step": 1550
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 63.0,
|
| 530 |
+
"eval_loss": 0.29049646854400635,
|
| 531 |
+
"eval_runtime": 2.4924,
|
| 532 |
+
"eval_samples_per_second": 80.244,
|
| 533 |
+
"eval_steps_per_second": 10.031,
|
| 534 |
+
"step": 1575
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 64.0,
|
| 538 |
+
"eval_loss": 0.2913173735141754,
|
| 539 |
+
"eval_runtime": 2.4934,
|
| 540 |
+
"eval_samples_per_second": 80.212,
|
| 541 |
+
"eval_steps_per_second": 10.026,
|
| 542 |
+
"step": 1600
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 65.0,
|
| 546 |
+
"eval_loss": 0.28798389434814453,
|
| 547 |
+
"eval_runtime": 2.4945,
|
| 548 |
+
"eval_samples_per_second": 80.176,
|
| 549 |
+
"eval_steps_per_second": 10.022,
|
| 550 |
+
"step": 1625
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 66.0,
|
| 554 |
+
"eval_loss": 0.28929680585861206,
|
| 555 |
+
"eval_runtime": 2.4963,
|
| 556 |
+
"eval_samples_per_second": 80.118,
|
| 557 |
+
"eval_steps_per_second": 10.015,
|
| 558 |
+
"step": 1650
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 67.0,
|
| 562 |
+
"eval_loss": 0.2856563925743103,
|
| 563 |
+
"eval_runtime": 2.4957,
|
| 564 |
+
"eval_samples_per_second": 80.139,
|
| 565 |
+
"eval_steps_per_second": 10.017,
|
| 566 |
+
"step": 1675
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 68.0,
|
| 570 |
+
"eval_loss": 0.2869837284088135,
|
| 571 |
+
"eval_runtime": 2.496,
|
| 572 |
+
"eval_samples_per_second": 80.13,
|
| 573 |
+
"eval_steps_per_second": 10.016,
|
| 574 |
+
"step": 1700
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 69.0,
|
| 578 |
+
"eval_loss": 0.2902255356311798,
|
| 579 |
+
"eval_runtime": 2.4969,
|
| 580 |
+
"eval_samples_per_second": 80.098,
|
| 581 |
+
"eval_steps_per_second": 10.012,
|
| 582 |
+
"step": 1725
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 70.0,
|
| 586 |
+
"eval_loss": 0.28557059168815613,
|
| 587 |
+
"eval_runtime": 2.4957,
|
| 588 |
+
"eval_samples_per_second": 80.139,
|
| 589 |
+
"eval_steps_per_second": 10.017,
|
| 590 |
+
"step": 1750
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 71.0,
|
| 594 |
+
"eval_loss": 0.2883276343345642,
|
| 595 |
+
"eval_runtime": 2.4986,
|
| 596 |
+
"eval_samples_per_second": 80.045,
|
| 597 |
+
"eval_steps_per_second": 10.006,
|
| 598 |
+
"step": 1775
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 72.0,
|
| 602 |
+
"eval_loss": 0.286774605512619,
|
| 603 |
+
"eval_runtime": 2.4979,
|
| 604 |
+
"eval_samples_per_second": 80.068,
|
| 605 |
+
"eval_steps_per_second": 10.009,
|
| 606 |
+
"step": 1800
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 73.0,
|
| 610 |
+
"eval_loss": 0.28692272305488586,
|
| 611 |
+
"eval_runtime": 2.4977,
|
| 612 |
+
"eval_samples_per_second": 80.073,
|
| 613 |
+
"eval_steps_per_second": 10.009,
|
| 614 |
+
"step": 1825
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 74.0,
|
| 618 |
+
"eval_loss": 0.2842114567756653,
|
| 619 |
+
"eval_runtime": 2.4982,
|
| 620 |
+
"eval_samples_per_second": 80.058,
|
| 621 |
+
"eval_steps_per_second": 10.007,
|
| 622 |
+
"step": 1850
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 75.0,
|
| 626 |
+
"eval_loss": 0.2869337201118469,
|
| 627 |
+
"eval_runtime": 2.4984,
|
| 628 |
+
"eval_samples_per_second": 80.052,
|
| 629 |
+
"eval_steps_per_second": 10.007,
|
| 630 |
+
"step": 1875
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 76.0,
|
| 634 |
+
"eval_loss": 0.2843911647796631,
|
| 635 |
+
"eval_runtime": 2.498,
|
| 636 |
+
"eval_samples_per_second": 80.064,
|
| 637 |
+
"eval_steps_per_second": 10.008,
|
| 638 |
+
"step": 1900
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 77.0,
|
| 642 |
+
"eval_loss": 0.28588855266571045,
|
| 643 |
+
"eval_runtime": 2.4985,
|
| 644 |
+
"eval_samples_per_second": 80.047,
|
| 645 |
+
"eval_steps_per_second": 10.006,
|
| 646 |
+
"step": 1925
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 78.0,
|
| 650 |
+
"eval_loss": 0.2864097058773041,
|
| 651 |
+
"eval_runtime": 2.4994,
|
| 652 |
+
"eval_samples_per_second": 80.02,
|
| 653 |
+
"eval_steps_per_second": 10.002,
|
| 654 |
+
"step": 1950
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 79.0,
|
| 658 |
+
"eval_loss": 0.28731438517570496,
|
| 659 |
+
"eval_runtime": 2.4981,
|
| 660 |
+
"eval_samples_per_second": 80.061,
|
| 661 |
+
"eval_steps_per_second": 10.008,
|
| 662 |
+
"step": 1975
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 80.0,
|
| 666 |
+
"grad_norm": 1.4072085618972778,
|
| 667 |
+
"learning_rate": 4.000000000000001e-06,
|
| 668 |
+
"loss": 0.3199,
|
| 669 |
+
"step": 2000
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 80.0,
|
| 673 |
+
"eval_loss": 0.2888672947883606,
|
| 674 |
+
"eval_runtime": 2.4574,
|
| 675 |
+
"eval_samples_per_second": 81.386,
|
| 676 |
+
"eval_steps_per_second": 10.173,
|
| 677 |
+
"step": 2000
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 81.0,
|
| 681 |
+
"eval_loss": 0.28676241636276245,
|
| 682 |
+
"eval_runtime": 2.4848,
|
| 683 |
+
"eval_samples_per_second": 80.49,
|
| 684 |
+
"eval_steps_per_second": 10.061,
|
| 685 |
+
"step": 2025
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 82.0,
|
| 689 |
+
"eval_loss": 0.28567585349082947,
|
| 690 |
+
"eval_runtime": 2.4917,
|
| 691 |
+
"eval_samples_per_second": 80.268,
|
| 692 |
+
"eval_steps_per_second": 10.033,
|
| 693 |
+
"step": 2050
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 83.0,
|
| 697 |
+
"eval_loss": 0.2843726873397827,
|
| 698 |
+
"eval_runtime": 2.4926,
|
| 699 |
+
"eval_samples_per_second": 80.236,
|
| 700 |
+
"eval_steps_per_second": 10.03,
|
| 701 |
+
"step": 2075
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 84.0,
|
| 705 |
+
"eval_loss": 0.28754809498786926,
|
| 706 |
+
"eval_runtime": 2.4936,
|
| 707 |
+
"eval_samples_per_second": 80.205,
|
| 708 |
+
"eval_steps_per_second": 10.026,
|
| 709 |
+
"step": 2100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 85.0,
|
| 713 |
+
"eval_loss": 0.2854582667350769,
|
| 714 |
+
"eval_runtime": 2.4941,
|
| 715 |
+
"eval_samples_per_second": 80.188,
|
| 716 |
+
"eval_steps_per_second": 10.023,
|
| 717 |
+
"step": 2125
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 86.0,
|
| 721 |
+
"eval_loss": 0.2840147316455841,
|
| 722 |
+
"eval_runtime": 2.4944,
|
| 723 |
+
"eval_samples_per_second": 80.181,
|
| 724 |
+
"eval_steps_per_second": 10.023,
|
| 725 |
+
"step": 2150
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 87.0,
|
| 729 |
+
"eval_loss": 0.28520676493644714,
|
| 730 |
+
"eval_runtime": 2.4963,
|
| 731 |
+
"eval_samples_per_second": 80.119,
|
| 732 |
+
"eval_steps_per_second": 10.015,
|
| 733 |
+
"step": 2175
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 88.0,
|
| 737 |
+
"eval_loss": 0.28196609020233154,
|
| 738 |
+
"eval_runtime": 2.4972,
|
| 739 |
+
"eval_samples_per_second": 80.091,
|
| 740 |
+
"eval_steps_per_second": 10.011,
|
| 741 |
+
"step": 2200
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 89.0,
|
| 745 |
+
"eval_loss": 0.28386008739471436,
|
| 746 |
+
"eval_runtime": 2.4982,
|
| 747 |
+
"eval_samples_per_second": 80.057,
|
| 748 |
+
"eval_steps_per_second": 10.007,
|
| 749 |
+
"step": 2225
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 90.0,
|
| 753 |
+
"eval_loss": 0.2850269675254822,
|
| 754 |
+
"eval_runtime": 2.4975,
|
| 755 |
+
"eval_samples_per_second": 80.079,
|
| 756 |
+
"eval_steps_per_second": 10.01,
|
| 757 |
+
"step": 2250
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 91.0,
|
| 761 |
+
"eval_loss": 0.28362977504730225,
|
| 762 |
+
"eval_runtime": 2.4982,
|
| 763 |
+
"eval_samples_per_second": 80.057,
|
| 764 |
+
"eval_steps_per_second": 10.007,
|
| 765 |
+
"step": 2275
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 92.0,
|
| 769 |
+
"eval_loss": 0.2840833365917206,
|
| 770 |
+
"eval_runtime": 2.4978,
|
| 771 |
+
"eval_samples_per_second": 80.07,
|
| 772 |
+
"eval_steps_per_second": 10.009,
|
| 773 |
+
"step": 2300
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 93.0,
|
| 777 |
+
"eval_loss": 0.28477975726127625,
|
| 778 |
+
"eval_runtime": 2.4997,
|
| 779 |
+
"eval_samples_per_second": 80.01,
|
| 780 |
+
"eval_steps_per_second": 10.001,
|
| 781 |
+
"step": 2325
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 94.0,
|
| 785 |
+
"eval_loss": 0.2831202745437622,
|
| 786 |
+
"eval_runtime": 2.4995,
|
| 787 |
+
"eval_samples_per_second": 80.016,
|
| 788 |
+
"eval_steps_per_second": 10.002,
|
| 789 |
+
"step": 2350
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 95.0,
|
| 793 |
+
"eval_loss": 0.28298699855804443,
|
| 794 |
+
"eval_runtime": 2.5008,
|
| 795 |
+
"eval_samples_per_second": 79.975,
|
| 796 |
+
"eval_steps_per_second": 9.997,
|
| 797 |
+
"step": 2375
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 96.0,
|
| 801 |
+
"eval_loss": 0.2848021984100342,
|
| 802 |
+
"eval_runtime": 2.4987,
|
| 803 |
+
"eval_samples_per_second": 80.041,
|
| 804 |
+
"eval_steps_per_second": 10.005,
|
| 805 |
+
"step": 2400
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"epoch": 97.0,
|
| 809 |
+
"eval_loss": 0.2818942368030548,
|
| 810 |
+
"eval_runtime": 2.4999,
|
| 811 |
+
"eval_samples_per_second": 80.004,
|
| 812 |
+
"eval_steps_per_second": 10.0,
|
| 813 |
+
"step": 2425
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 98.0,
|
| 817 |
+
"eval_loss": 0.28425753116607666,
|
| 818 |
+
"eval_runtime": 2.5005,
|
| 819 |
+
"eval_samples_per_second": 79.982,
|
| 820 |
+
"eval_steps_per_second": 9.998,
|
| 821 |
+
"step": 2450
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 99.0,
|
| 825 |
+
"eval_loss": 0.28273478150367737,
|
| 826 |
+
"eval_runtime": 2.4987,
|
| 827 |
+
"eval_samples_per_second": 80.042,
|
| 828 |
+
"eval_steps_per_second": 10.005,
|
| 829 |
+
"step": 2475
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 100.0,
|
| 833 |
+
"grad_norm": 1.4653393030166626,
|
| 834 |
+
"learning_rate": 0.0,
|
| 835 |
+
"loss": 0.3157,
|
| 836 |
+
"step": 2500
|
| 837 |
+
}
|
| 838 |
+
],
|
| 839 |
+
"logging_steps": 500,
|
| 840 |
+
"max_steps": 2500,
|
| 841 |
+
"num_input_tokens_seen": 0,
|
| 842 |
+
"num_train_epochs": 100,
|
| 843 |
+
"save_steps": 500,
|
| 844 |
+
"total_flos": 1306483752960000.0,
|
| 845 |
+
"train_batch_size": 8,
|
| 846 |
+
"trial_name": null,
|
| 847 |
+
"trial_params": null
|
| 848 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5180705d15372e9f21caa77a1255bbc087918814aee5100bbab8263af8edeb38
|
| 3 |
+
size 4856
|