Training in progress, step 68000, checkpoint
Browse files- checkpoint-68000/config.json +32 -0
- checkpoint-68000/generation_config.json +6 -0
- checkpoint-68000/model.safetensors +3 -0
- checkpoint-68000/optimizer.pt +3 -0
- checkpoint-68000/rng_state.pth +3 -0
- checkpoint-68000/scheduler.pt +3 -0
- checkpoint-68000/special_tokens_map.json +34 -0
- checkpoint-68000/tokenizer.json +0 -0
- checkpoint-68000/tokenizer_config.json +63 -0
- checkpoint-68000/trainer_state.json +424 -0
- checkpoint-68000/training_args.bin +3 -0
checkpoint-68000/config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"GPT2LMHeadModel"
|
| 5 |
+
],
|
| 6 |
+
"attn_pdrop": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"embd_pdrop": 0.1,
|
| 9 |
+
"eos_token_id": 1,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"layer_norm_epsilon": 1e-05,
|
| 12 |
+
"model_type": "gpt2",
|
| 13 |
+
"n_ctx": 512,
|
| 14 |
+
"n_embd": 256,
|
| 15 |
+
"n_head": 8,
|
| 16 |
+
"n_inner": 2048,
|
| 17 |
+
"n_layer": 8,
|
| 18 |
+
"n_positions": 512,
|
| 19 |
+
"reorder_and_upcast_attn": false,
|
| 20 |
+
"resid_pdrop": 0.1,
|
| 21 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 22 |
+
"scale_attn_weights": true,
|
| 23 |
+
"summary_activation": null,
|
| 24 |
+
"summary_first_dropout": 0.1,
|
| 25 |
+
"summary_proj_to_labels": true,
|
| 26 |
+
"summary_type": "cls_index",
|
| 27 |
+
"summary_use_proj": true,
|
| 28 |
+
"torch_dtype": "float32",
|
| 29 |
+
"transformers_version": "4.45.2",
|
| 30 |
+
"use_cache": true,
|
| 31 |
+
"vocab_size": 8192
|
| 32 |
+
}
|
checkpoint-68000/generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 0,
|
| 4 |
+
"eos_token_id": 1,
|
| 5 |
+
"transformers_version": "4.45.2"
|
| 6 |
+
}
|
checkpoint-68000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3c2c5d2b2ae10bfa3e36bdf42e9f68881818263ad4b11063609968275da1fd7b
|
| 3 |
+
size 51007160
|
checkpoint-68000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:31b4156b2aafea13fc1af62d5646e00ccd728e5eea1b133f06d118d967621d04
|
| 3 |
+
size 102078202
|
checkpoint-68000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c6ef44cfcce1a9cfde2d362b774fa2eacb22bf2355918ea2e8d5c8bf8b4ba6ef
|
| 3 |
+
size 14308
|
checkpoint-68000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:94a819f752f6d1142cfeeaf9129a460f11b7714ddab6a67b9336ea286b8cd5a9
|
| 3 |
+
size 1000
|
checkpoint-68000/special_tokens_map.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|endoftext|>"
|
| 4 |
+
],
|
| 5 |
+
"bos_token": {
|
| 6 |
+
"content": "<s>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"eos_token": {
|
| 13 |
+
"content": "</s>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": false,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false
|
| 18 |
+
},
|
| 19 |
+
"mask_token": {
|
| 20 |
+
"content": "<mask>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false
|
| 25 |
+
},
|
| 26 |
+
"pad_token": "</s>",
|
| 27 |
+
"unk_token": {
|
| 28 |
+
"content": "<unk>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false
|
| 33 |
+
}
|
| 34 |
+
}
|
checkpoint-68000/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-68000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<s>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "</s>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "<pad>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "<unk>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"4": {
|
| 36 |
+
"content": "<mask>",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
},
|
| 43 |
+
"5": {
|
| 44 |
+
"content": "<|endoftext|>",
|
| 45 |
+
"lstrip": false,
|
| 46 |
+
"normalized": false,
|
| 47 |
+
"rstrip": false,
|
| 48 |
+
"single_word": false,
|
| 49 |
+
"special": true
|
| 50 |
+
}
|
| 51 |
+
},
|
| 52 |
+
"additional_special_tokens": [
|
| 53 |
+
"<|endoftext|>"
|
| 54 |
+
],
|
| 55 |
+
"bos_token": "<s>",
|
| 56 |
+
"clean_up_tokenization_spaces": false,
|
| 57 |
+
"eos_token": "</s>",
|
| 58 |
+
"mask_token": "<mask>",
|
| 59 |
+
"model_max_length": 128,
|
| 60 |
+
"pad_token": "</s>",
|
| 61 |
+
"tokenizer_class": "PreTrainedTokenizerFast",
|
| 62 |
+
"unk_token": "<unk>"
|
| 63 |
+
}
|
checkpoint-68000/trainer_state.json
ADDED
|
@@ -0,0 +1,424 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 3.9619035720825195,
|
| 3 |
+
"best_model_checkpoint": "/home/p318482/babyLM_controlled/models_trained_last/de_clm/de_childes_30/checkpoint-44000",
|
| 4 |
+
"epoch": 51.08599324070597,
|
| 5 |
+
"eval_steps": 2000,
|
| 6 |
+
"global_step": 68000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.5020653398422832,
|
| 13 |
+
"eval_loss": 7.09099817276001,
|
| 14 |
+
"eval_runtime": 3.7115,
|
| 15 |
+
"eval_samples_per_second": 894.522,
|
| 16 |
+
"eval_steps_per_second": 56.042,
|
| 17 |
+
"step": 2000
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 3.0041306796845664,
|
| 21 |
+
"grad_norm": 1.3897417783737183,
|
| 22 |
+
"learning_rate": 1e-05,
|
| 23 |
+
"loss": 6.9957,
|
| 24 |
+
"step": 4000
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 3.0041306796845664,
|
| 28 |
+
"eval_loss": 5.826653003692627,
|
| 29 |
+
"eval_runtime": 3.7561,
|
| 30 |
+
"eval_samples_per_second": 883.906,
|
| 31 |
+
"eval_steps_per_second": 55.377,
|
| 32 |
+
"step": 4000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 4.50619601952685,
|
| 36 |
+
"eval_loss": 5.465098857879639,
|
| 37 |
+
"eval_runtime": 3.8583,
|
| 38 |
+
"eval_samples_per_second": 860.484,
|
| 39 |
+
"eval_steps_per_second": 53.91,
|
| 40 |
+
"step": 6000
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 6.008261359369133,
|
| 44 |
+
"grad_norm": 2.3630881309509277,
|
| 45 |
+
"learning_rate": 2e-05,
|
| 46 |
+
"loss": 5.213,
|
| 47 |
+
"step": 8000
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 6.008261359369133,
|
| 51 |
+
"eval_loss": 5.171680927276611,
|
| 52 |
+
"eval_runtime": 3.928,
|
| 53 |
+
"eval_samples_per_second": 845.218,
|
| 54 |
+
"eval_steps_per_second": 52.953,
|
| 55 |
+
"step": 8000
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 7.510326699211416,
|
| 59 |
+
"eval_loss": 4.959029197692871,
|
| 60 |
+
"eval_runtime": 3.8954,
|
| 61 |
+
"eval_samples_per_second": 852.288,
|
| 62 |
+
"eval_steps_per_second": 53.396,
|
| 63 |
+
"step": 10000
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 9.0123920390537,
|
| 67 |
+
"grad_norm": 2.578948736190796,
|
| 68 |
+
"learning_rate": 2.9995e-05,
|
| 69 |
+
"loss": 4.7279,
|
| 70 |
+
"step": 12000
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 9.0123920390537,
|
| 74 |
+
"eval_loss": 4.791839599609375,
|
| 75 |
+
"eval_runtime": 3.7027,
|
| 76 |
+
"eval_samples_per_second": 896.638,
|
| 77 |
+
"eval_steps_per_second": 56.175,
|
| 78 |
+
"step": 12000
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 10.514457378895981,
|
| 82 |
+
"eval_loss": 4.660107612609863,
|
| 83 |
+
"eval_runtime": 3.7788,
|
| 84 |
+
"eval_samples_per_second": 878.594,
|
| 85 |
+
"eval_steps_per_second": 55.044,
|
| 86 |
+
"step": 14000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 12.016522718738265,
|
| 90 |
+
"grad_norm": 2.3461837768554688,
|
| 91 |
+
"learning_rate": 3.999e-05,
|
| 92 |
+
"loss": 4.4214,
|
| 93 |
+
"step": 16000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 12.016522718738265,
|
| 97 |
+
"eval_loss": 4.552372932434082,
|
| 98 |
+
"eval_runtime": 3.7928,
|
| 99 |
+
"eval_samples_per_second": 875.332,
|
| 100 |
+
"eval_steps_per_second": 54.84,
|
| 101 |
+
"step": 16000
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 13.518588058580548,
|
| 105 |
+
"eval_loss": 4.458698272705078,
|
| 106 |
+
"eval_runtime": 3.8109,
|
| 107 |
+
"eval_samples_per_second": 871.192,
|
| 108 |
+
"eval_steps_per_second": 54.581,
|
| 109 |
+
"step": 18000
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 15.020653398422832,
|
| 113 |
+
"grad_norm": 2.1618399620056152,
|
| 114 |
+
"learning_rate": 4.9985e-05,
|
| 115 |
+
"loss": 4.195,
|
| 116 |
+
"step": 20000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 15.020653398422832,
|
| 120 |
+
"eval_loss": 4.3683576583862305,
|
| 121 |
+
"eval_runtime": 3.8444,
|
| 122 |
+
"eval_samples_per_second": 863.584,
|
| 123 |
+
"eval_steps_per_second": 54.104,
|
| 124 |
+
"step": 20000
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 16.522718738265116,
|
| 128 |
+
"eval_loss": 4.289999485015869,
|
| 129 |
+
"eval_runtime": 3.8372,
|
| 130 |
+
"eval_samples_per_second": 865.217,
|
| 131 |
+
"eval_steps_per_second": 54.206,
|
| 132 |
+
"step": 22000
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 18.0247840781074,
|
| 136 |
+
"grad_norm": 2.0583324432373047,
|
| 137 |
+
"learning_rate": 5.9980000000000005e-05,
|
| 138 |
+
"loss": 4.0109,
|
| 139 |
+
"step": 24000
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 18.0247840781074,
|
| 143 |
+
"eval_loss": 4.227661609649658,
|
| 144 |
+
"eval_runtime": 3.7524,
|
| 145 |
+
"eval_samples_per_second": 884.777,
|
| 146 |
+
"eval_steps_per_second": 55.432,
|
| 147 |
+
"step": 24000
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 19.52684941794968,
|
| 151 |
+
"eval_loss": 4.172571659088135,
|
| 152 |
+
"eval_runtime": 3.7915,
|
| 153 |
+
"eval_samples_per_second": 875.64,
|
| 154 |
+
"eval_steps_per_second": 54.859,
|
| 155 |
+
"step": 26000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 21.028914757791963,
|
| 159 |
+
"grad_norm": 2.0024306774139404,
|
| 160 |
+
"learning_rate": 6.997500000000001e-05,
|
| 161 |
+
"loss": 3.8596,
|
| 162 |
+
"step": 28000
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 21.028914757791963,
|
| 166 |
+
"eval_loss": 4.128843307495117,
|
| 167 |
+
"eval_runtime": 3.6349,
|
| 168 |
+
"eval_samples_per_second": 913.374,
|
| 169 |
+
"eval_steps_per_second": 57.223,
|
| 170 |
+
"step": 28000
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 22.53098009763425,
|
| 174 |
+
"eval_loss": 4.089205265045166,
|
| 175 |
+
"eval_runtime": 3.7809,
|
| 176 |
+
"eval_samples_per_second": 878.103,
|
| 177 |
+
"eval_steps_per_second": 55.014,
|
| 178 |
+
"step": 30000
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 24.03304543747653,
|
| 182 |
+
"grad_norm": 1.9181395769119263,
|
| 183 |
+
"learning_rate": 7.997e-05,
|
| 184 |
+
"loss": 3.7326,
|
| 185 |
+
"step": 32000
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 24.03304543747653,
|
| 189 |
+
"eval_loss": 4.058904647827148,
|
| 190 |
+
"eval_runtime": 3.7243,
|
| 191 |
+
"eval_samples_per_second": 891.451,
|
| 192 |
+
"eval_steps_per_second": 55.85,
|
| 193 |
+
"step": 32000
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 25.535110777318813,
|
| 197 |
+
"eval_loss": 4.027942657470703,
|
| 198 |
+
"eval_runtime": 3.7785,
|
| 199 |
+
"eval_samples_per_second": 878.652,
|
| 200 |
+
"eval_steps_per_second": 55.048,
|
| 201 |
+
"step": 34000
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 27.037176117161096,
|
| 205 |
+
"grad_norm": 1.8628945350646973,
|
| 206 |
+
"learning_rate": 8.9965e-05,
|
| 207 |
+
"loss": 3.6241,
|
| 208 |
+
"step": 36000
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 27.037176117161096,
|
| 212 |
+
"eval_loss": 4.002871513366699,
|
| 213 |
+
"eval_runtime": 3.8549,
|
| 214 |
+
"eval_samples_per_second": 861.239,
|
| 215 |
+
"eval_steps_per_second": 53.957,
|
| 216 |
+
"step": 36000
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 28.539241457003378,
|
| 220 |
+
"eval_loss": 3.990269660949707,
|
| 221 |
+
"eval_runtime": 3.9421,
|
| 222 |
+
"eval_samples_per_second": 842.197,
|
| 223 |
+
"eval_steps_per_second": 52.764,
|
| 224 |
+
"step": 38000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 30.041306796845664,
|
| 228 |
+
"grad_norm": 1.9048078060150146,
|
| 229 |
+
"learning_rate": 9.996000000000001e-05,
|
| 230 |
+
"loss": 3.5294,
|
| 231 |
+
"step": 40000
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 30.041306796845664,
|
| 235 |
+
"eval_loss": 3.9731788635253906,
|
| 236 |
+
"eval_runtime": 3.8486,
|
| 237 |
+
"eval_samples_per_second": 862.647,
|
| 238 |
+
"eval_steps_per_second": 54.045,
|
| 239 |
+
"step": 40000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 31.543372136687946,
|
| 243 |
+
"eval_loss": 3.969944715499878,
|
| 244 |
+
"eval_runtime": 3.8145,
|
| 245 |
+
"eval_samples_per_second": 870.363,
|
| 246 |
+
"eval_steps_per_second": 54.529,
|
| 247 |
+
"step": 42000
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 33.04543747653023,
|
| 251 |
+
"grad_norm": 1.8309521675109863,
|
| 252 |
+
"learning_rate": 9.336333333333334e-05,
|
| 253 |
+
"loss": 3.4361,
|
| 254 |
+
"step": 44000
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 33.04543747653023,
|
| 258 |
+
"eval_loss": 3.9619035720825195,
|
| 259 |
+
"eval_runtime": 3.8269,
|
| 260 |
+
"eval_samples_per_second": 867.55,
|
| 261 |
+
"eval_steps_per_second": 54.353,
|
| 262 |
+
"step": 44000
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 34.547502816372514,
|
| 266 |
+
"eval_loss": 3.9607603549957275,
|
| 267 |
+
"eval_runtime": 3.8521,
|
| 268 |
+
"eval_samples_per_second": 861.876,
|
| 269 |
+
"eval_steps_per_second": 53.997,
|
| 270 |
+
"step": 46000
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 36.0495681562148,
|
| 274 |
+
"grad_norm": 1.8812049627304077,
|
| 275 |
+
"learning_rate": 8.67e-05,
|
| 276 |
+
"loss": 3.3449,
|
| 277 |
+
"step": 48000
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 36.0495681562148,
|
| 281 |
+
"eval_loss": 3.9666340351104736,
|
| 282 |
+
"eval_runtime": 3.8283,
|
| 283 |
+
"eval_samples_per_second": 867.222,
|
| 284 |
+
"eval_steps_per_second": 54.332,
|
| 285 |
+
"step": 48000
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 37.55163349605708,
|
| 289 |
+
"eval_loss": 3.969193458557129,
|
| 290 |
+
"eval_runtime": 3.8919,
|
| 291 |
+
"eval_samples_per_second": 853.06,
|
| 292 |
+
"eval_steps_per_second": 53.445,
|
| 293 |
+
"step": 50000
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 39.05369883589936,
|
| 297 |
+
"grad_norm": 2.010619640350342,
|
| 298 |
+
"learning_rate": 8.003666666666667e-05,
|
| 299 |
+
"loss": 3.2664,
|
| 300 |
+
"step": 52000
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 39.05369883589936,
|
| 304 |
+
"eval_loss": 3.9745302200317383,
|
| 305 |
+
"eval_runtime": 3.8291,
|
| 306 |
+
"eval_samples_per_second": 867.037,
|
| 307 |
+
"eval_steps_per_second": 54.32,
|
| 308 |
+
"step": 52000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 40.55576417574164,
|
| 312 |
+
"eval_loss": 3.986243963241577,
|
| 313 |
+
"eval_runtime": 3.7036,
|
| 314 |
+
"eval_samples_per_second": 896.418,
|
| 315 |
+
"eval_steps_per_second": 56.161,
|
| 316 |
+
"step": 54000
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 42.057829515583926,
|
| 320 |
+
"grad_norm": 2.0288503170013428,
|
| 321 |
+
"learning_rate": 7.337333333333334e-05,
|
| 322 |
+
"loss": 3.1963,
|
| 323 |
+
"step": 56000
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 42.057829515583926,
|
| 327 |
+
"eval_loss": 3.9971587657928467,
|
| 328 |
+
"eval_runtime": 3.7762,
|
| 329 |
+
"eval_samples_per_second": 879.202,
|
| 330 |
+
"eval_steps_per_second": 55.083,
|
| 331 |
+
"step": 56000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 43.575666541494556,
|
| 335 |
+
"eval_loss": 4.007722854614258,
|
| 336 |
+
"eval_runtime": 3.9047,
|
| 337 |
+
"eval_samples_per_second": 850.256,
|
| 338 |
+
"eval_steps_per_second": 53.269,
|
| 339 |
+
"step": 58000
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 45.07773188133684,
|
| 343 |
+
"grad_norm": 2.1440820693969727,
|
| 344 |
+
"learning_rate": 6.670666666666668e-05,
|
| 345 |
+
"loss": 3.1346,
|
| 346 |
+
"step": 60000
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 45.07773188133684,
|
| 350 |
+
"eval_loss": 4.020359039306641,
|
| 351 |
+
"eval_runtime": 3.9205,
|
| 352 |
+
"eval_samples_per_second": 846.839,
|
| 353 |
+
"eval_steps_per_second": 53.055,
|
| 354 |
+
"step": 60000
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 46.57979722117912,
|
| 358 |
+
"eval_loss": 4.0261549949646,
|
| 359 |
+
"eval_runtime": 2.3464,
|
| 360 |
+
"eval_samples_per_second": 1414.93,
|
| 361 |
+
"eval_steps_per_second": 88.646,
|
| 362 |
+
"step": 62000
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 48.0818625610214,
|
| 366 |
+
"grad_norm": 2.2309792041778564,
|
| 367 |
+
"learning_rate": 6.004000000000001e-05,
|
| 368 |
+
"loss": 3.0792,
|
| 369 |
+
"step": 64000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 48.0818625610214,
|
| 373 |
+
"eval_loss": 4.040497303009033,
|
| 374 |
+
"eval_runtime": 2.2801,
|
| 375 |
+
"eval_samples_per_second": 1456.058,
|
| 376 |
+
"eval_steps_per_second": 91.223,
|
| 377 |
+
"step": 64000
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 49.583927900863685,
|
| 381 |
+
"eval_loss": 4.0522260665893555,
|
| 382 |
+
"eval_runtime": 2.3904,
|
| 383 |
+
"eval_samples_per_second": 1388.888,
|
| 384 |
+
"eval_steps_per_second": 87.015,
|
| 385 |
+
"step": 66000
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 51.08599324070597,
|
| 389 |
+
"grad_norm": 2.314417839050293,
|
| 390 |
+
"learning_rate": 5.337666666666667e-05,
|
| 391 |
+
"loss": 3.0286,
|
| 392 |
+
"step": 68000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 51.08599324070597,
|
| 396 |
+
"eval_loss": 4.06997537612915,
|
| 397 |
+
"eval_runtime": 3.8785,
|
| 398 |
+
"eval_samples_per_second": 855.993,
|
| 399 |
+
"eval_steps_per_second": 53.628,
|
| 400 |
+
"step": 68000
|
| 401 |
+
}
|
| 402 |
+
],
|
| 403 |
+
"logging_steps": 4000,
|
| 404 |
+
"max_steps": 100000,
|
| 405 |
+
"num_input_tokens_seen": 0,
|
| 406 |
+
"num_train_epochs": 76,
|
| 407 |
+
"save_steps": 4000,
|
| 408 |
+
"stateful_callbacks": {
|
| 409 |
+
"TrainerControl": {
|
| 410 |
+
"args": {
|
| 411 |
+
"should_epoch_stop": false,
|
| 412 |
+
"should_evaluate": false,
|
| 413 |
+
"should_log": false,
|
| 414 |
+
"should_save": true,
|
| 415 |
+
"should_training_stop": false
|
| 416 |
+
},
|
| 417 |
+
"attributes": {}
|
| 418 |
+
}
|
| 419 |
+
},
|
| 420 |
+
"total_flos": 1.7579208365973504e+16,
|
| 421 |
+
"train_batch_size": 16,
|
| 422 |
+
"trial_name": null,
|
| 423 |
+
"trial_params": null
|
| 424 |
+
}
|
checkpoint-68000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4e9ac166ba8168749d2e56e6bf1c4fed81d0a2370c265612a5c96db3fef84453
|
| 3 |
+
size 5496
|