Commit
•
a17c01a
1
Parent(s):
a95089b
Upload 9 files
Browse files- config.json +31 -0
- generation_config.json +6 -0
- model.safetensors +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +1 -0
- trainer_state.json +965 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "data/models/Mixtral-8x7B-v0.1-dna",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"MixtralForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 1,
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
+
"hidden_act": "silu",
|
| 10 |
+
"hidden_size": 256,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 256,
|
| 13 |
+
"max_position_embeddings": 512,
|
| 14 |
+
"model_type": "mixtral",
|
| 15 |
+
"num_attention_heads": 8,
|
| 16 |
+
"num_experts_per_tok": 1,
|
| 17 |
+
"num_hidden_layers": 8,
|
| 18 |
+
"num_key_value_heads": 8,
|
| 19 |
+
"num_local_experts": 8,
|
| 20 |
+
"output_router_logits": false,
|
| 21 |
+
"rms_norm_eps": 1e-05,
|
| 22 |
+
"rope_theta": 1000000.0,
|
| 23 |
+
"router_aux_loss_coef": 0.02,
|
| 24 |
+
"router_jitter_noise": 0.0,
|
| 25 |
+
"sliding_window": null,
|
| 26 |
+
"tie_word_embeddings": false,
|
| 27 |
+
"torch_dtype": "bfloat16",
|
| 28 |
+
"transformers_version": "4.41.1",
|
| 29 |
+
"use_cache": true,
|
| 30 |
+
"vocab_size": 4096
|
| 31 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"transformers_version": "4.41.1"
|
| 6 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1f78a71f85a3aafe5300898ee4efe396909c04368a96942ed4c644f5314fb284
|
| 3 |
+
size 33626152
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:21c2425b85b819f329f6809df4ca360e34a2b44b18e94e8e0472ab0b59df0b2a
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c95d699539c776f89b1e51df2cc923c3b744334b7554ff2cf6c8f0480172f30a
|
| 3 |
+
size 1064
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"tokenizer_class": "PreTrainedTokenizerFast", "unk_token": "[UNK]", "cls_token": "[CLS]", "sep_token": "[SEP]", "pad_token": "[PAD]", "mask_token": "[MASK]"}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,965 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 5.531345844268799,
|
| 3 |
+
"best_model_checkpoint": "./results/models/checkpoint-58916",
|
| 4 |
+
"epoch": 13.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 58916,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.11032656663724624,
|
| 13 |
+
"grad_norm": 0.07373046875,
|
| 14 |
+
"learning_rate": 0.00199558693733451,
|
| 15 |
+
"loss": 6.7817,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.22065313327449249,
|
| 20 |
+
"grad_norm": 0.10693359375,
|
| 21 |
+
"learning_rate": 0.0019911738746690205,
|
| 22 |
+
"loss": 6.1767,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.33097969991173876,
|
| 27 |
+
"grad_norm": 0.10595703125,
|
| 28 |
+
"learning_rate": 0.0019867608120035306,
|
| 29 |
+
"loss": 6.0418,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.44130626654898497,
|
| 34 |
+
"grad_norm": 0.08642578125,
|
| 35 |
+
"learning_rate": 0.0019823477493380406,
|
| 36 |
+
"loss": 5.9811,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.5516328331862312,
|
| 41 |
+
"grad_norm": 0.11572265625,
|
| 42 |
+
"learning_rate": 0.0019779346866725506,
|
| 43 |
+
"loss": 5.9448,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.6619593998234775,
|
| 48 |
+
"grad_norm": 0.07958984375,
|
| 49 |
+
"learning_rate": 0.001973521624007061,
|
| 50 |
+
"loss": 5.9146,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.7722859664607238,
|
| 55 |
+
"grad_norm": 0.10205078125,
|
| 56 |
+
"learning_rate": 0.001969108561341571,
|
| 57 |
+
"loss": 5.8869,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.8826125330979699,
|
| 62 |
+
"grad_norm": 0.09814453125,
|
| 63 |
+
"learning_rate": 0.001964695498676081,
|
| 64 |
+
"loss": 5.8551,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.9929390997352162,
|
| 69 |
+
"grad_norm": 0.1025390625,
|
| 70 |
+
"learning_rate": 0.001960282436010591,
|
| 71 |
+
"loss": 5.8427,
|
| 72 |
+
"step": 4500
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 1.0,
|
| 76 |
+
"eval_loss": 5.741388320922852,
|
| 77 |
+
"eval_runtime": 7.2342,
|
| 78 |
+
"eval_samples_per_second": 69.116,
|
| 79 |
+
"eval_steps_per_second": 1.106,
|
| 80 |
+
"step": 4532
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.1032656663724625,
|
| 84 |
+
"grad_norm": 0.09912109375,
|
| 85 |
+
"learning_rate": 0.0019558693733451016,
|
| 86 |
+
"loss": 5.8385,
|
| 87 |
+
"step": 5000
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 1.2135922330097086,
|
| 91 |
+
"grad_norm": 0.10302734375,
|
| 92 |
+
"learning_rate": 0.0019514563106796117,
|
| 93 |
+
"loss": 5.8289,
|
| 94 |
+
"step": 5500
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 1.323918799646955,
|
| 98 |
+
"grad_norm": 0.11181640625,
|
| 99 |
+
"learning_rate": 0.001947043248014122,
|
| 100 |
+
"loss": 5.8157,
|
| 101 |
+
"step": 6000
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 1.4342453662842012,
|
| 105 |
+
"grad_norm": 0.11376953125,
|
| 106 |
+
"learning_rate": 0.001942630185348632,
|
| 107 |
+
"loss": 5.7893,
|
| 108 |
+
"step": 6500
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.5445719329214476,
|
| 112 |
+
"grad_norm": 0.1171875,
|
| 113 |
+
"learning_rate": 0.0019382171226831422,
|
| 114 |
+
"loss": 5.7912,
|
| 115 |
+
"step": 7000
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 1.6548984995586937,
|
| 119 |
+
"grad_norm": 0.119140625,
|
| 120 |
+
"learning_rate": 0.0019338040600176522,
|
| 121 |
+
"loss": 5.7802,
|
| 122 |
+
"step": 7500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 1.7652250661959399,
|
| 126 |
+
"grad_norm": 0.109375,
|
| 127 |
+
"learning_rate": 0.0019293909973521625,
|
| 128 |
+
"loss": 5.7805,
|
| 129 |
+
"step": 8000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.8755516328331863,
|
| 133 |
+
"grad_norm": 0.10693359375,
|
| 134 |
+
"learning_rate": 0.0019249779346866725,
|
| 135 |
+
"loss": 5.765,
|
| 136 |
+
"step": 8500
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 1.9858781994704324,
|
| 140 |
+
"grad_norm": 0.1064453125,
|
| 141 |
+
"learning_rate": 0.0019205648720211827,
|
| 142 |
+
"loss": 5.7674,
|
| 143 |
+
"step": 9000
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 2.0,
|
| 147 |
+
"eval_loss": 5.6664910316467285,
|
| 148 |
+
"eval_runtime": 6.8932,
|
| 149 |
+
"eval_samples_per_second": 72.535,
|
| 150 |
+
"eval_steps_per_second": 1.161,
|
| 151 |
+
"step": 9064
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 2.096204766107679,
|
| 155 |
+
"grad_norm": 0.11572265625,
|
| 156 |
+
"learning_rate": 0.001916151809355693,
|
| 157 |
+
"loss": 5.7568,
|
| 158 |
+
"step": 9500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 2.206531332744925,
|
| 162 |
+
"grad_norm": 0.109375,
|
| 163 |
+
"learning_rate": 0.001911738746690203,
|
| 164 |
+
"loss": 5.7619,
|
| 165 |
+
"step": 10000
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 2.316857899382171,
|
| 169 |
+
"grad_norm": 0.12060546875,
|
| 170 |
+
"learning_rate": 0.0019073256840247133,
|
| 171 |
+
"loss": 5.7423,
|
| 172 |
+
"step": 10500
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 2.4271844660194173,
|
| 176 |
+
"grad_norm": 0.1181640625,
|
| 177 |
+
"learning_rate": 0.0019029126213592233,
|
| 178 |
+
"loss": 5.7502,
|
| 179 |
+
"step": 11000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 2.537511032656664,
|
| 183 |
+
"grad_norm": 0.109375,
|
| 184 |
+
"learning_rate": 0.0018984995586937335,
|
| 185 |
+
"loss": 5.7371,
|
| 186 |
+
"step": 11500
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 2.64783759929391,
|
| 190 |
+
"grad_norm": 0.1142578125,
|
| 191 |
+
"learning_rate": 0.0018940864960282436,
|
| 192 |
+
"loss": 5.7316,
|
| 193 |
+
"step": 12000
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 2.758164165931156,
|
| 197 |
+
"grad_norm": 0.10595703125,
|
| 198 |
+
"learning_rate": 0.0018896734333627538,
|
| 199 |
+
"loss": 5.7315,
|
| 200 |
+
"step": 12500
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 2.8684907325684024,
|
| 204 |
+
"grad_norm": 0.142578125,
|
| 205 |
+
"learning_rate": 0.001885260370697264,
|
| 206 |
+
"loss": 5.7301,
|
| 207 |
+
"step": 13000
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 2.978817299205649,
|
| 211 |
+
"grad_norm": 0.140625,
|
| 212 |
+
"learning_rate": 0.001880847308031774,
|
| 213 |
+
"loss": 5.728,
|
| 214 |
+
"step": 13500
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 3.0,
|
| 218 |
+
"eval_loss": 5.630899429321289,
|
| 219 |
+
"eval_runtime": 7.3076,
|
| 220 |
+
"eval_samples_per_second": 68.422,
|
| 221 |
+
"eval_steps_per_second": 1.095,
|
| 222 |
+
"step": 13596
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 3.089143865842895,
|
| 226 |
+
"grad_norm": 0.11572265625,
|
| 227 |
+
"learning_rate": 0.001876434245366284,
|
| 228 |
+
"loss": 5.724,
|
| 229 |
+
"step": 14000
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 3.1994704324801413,
|
| 233 |
+
"grad_norm": 0.109375,
|
| 234 |
+
"learning_rate": 0.0018720211827007946,
|
| 235 |
+
"loss": 5.7181,
|
| 236 |
+
"step": 14500
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 3.3097969991173875,
|
| 240 |
+
"grad_norm": 0.11865234375,
|
| 241 |
+
"learning_rate": 0.0018676081200353046,
|
| 242 |
+
"loss": 5.7201,
|
| 243 |
+
"step": 15000
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 3.4201235657546336,
|
| 247 |
+
"grad_norm": 0.1240234375,
|
| 248 |
+
"learning_rate": 0.0018631950573698146,
|
| 249 |
+
"loss": 5.718,
|
| 250 |
+
"step": 15500
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 3.5304501323918798,
|
| 254 |
+
"grad_norm": 0.1328125,
|
| 255 |
+
"learning_rate": 0.0018587819947043247,
|
| 256 |
+
"loss": 5.6959,
|
| 257 |
+
"step": 16000
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 3.6407766990291264,
|
| 261 |
+
"grad_norm": 0.11181640625,
|
| 262 |
+
"learning_rate": 0.0018543689320388351,
|
| 263 |
+
"loss": 5.7112,
|
| 264 |
+
"step": 16500
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 3.7511032656663725,
|
| 268 |
+
"grad_norm": 0.1220703125,
|
| 269 |
+
"learning_rate": 0.0018499558693733451,
|
| 270 |
+
"loss": 5.6986,
|
| 271 |
+
"step": 17000
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 3.8614298323036187,
|
| 275 |
+
"grad_norm": 0.1318359375,
|
| 276 |
+
"learning_rate": 0.0018455428067078552,
|
| 277 |
+
"loss": 5.7103,
|
| 278 |
+
"step": 17500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 3.971756398940865,
|
| 282 |
+
"grad_norm": 0.12158203125,
|
| 283 |
+
"learning_rate": 0.0018411297440423656,
|
| 284 |
+
"loss": 5.7054,
|
| 285 |
+
"step": 18000
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 4.0,
|
| 289 |
+
"eval_loss": 5.601424694061279,
|
| 290 |
+
"eval_runtime": 6.7523,
|
| 291 |
+
"eval_samples_per_second": 74.049,
|
| 292 |
+
"eval_steps_per_second": 1.185,
|
| 293 |
+
"step": 18128
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 4.0820829655781115,
|
| 297 |
+
"grad_norm": 0.126953125,
|
| 298 |
+
"learning_rate": 0.0018367166813768757,
|
| 299 |
+
"loss": 5.6929,
|
| 300 |
+
"step": 18500
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 4.192409532215358,
|
| 304 |
+
"grad_norm": 0.1298828125,
|
| 305 |
+
"learning_rate": 0.0018323036187113857,
|
| 306 |
+
"loss": 5.6868,
|
| 307 |
+
"step": 19000
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 4.302736098852604,
|
| 311 |
+
"grad_norm": 0.1611328125,
|
| 312 |
+
"learning_rate": 0.0018278905560458957,
|
| 313 |
+
"loss": 5.6955,
|
| 314 |
+
"step": 19500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 4.41306266548985,
|
| 318 |
+
"grad_norm": 0.142578125,
|
| 319 |
+
"learning_rate": 0.0018234774933804062,
|
| 320 |
+
"loss": 5.6879,
|
| 321 |
+
"step": 20000
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 4.523389232127096,
|
| 325 |
+
"grad_norm": 0.16796875,
|
| 326 |
+
"learning_rate": 0.0018190644307149162,
|
| 327 |
+
"loss": 5.6919,
|
| 328 |
+
"step": 20500
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 4.633715798764342,
|
| 332 |
+
"grad_norm": 0.1416015625,
|
| 333 |
+
"learning_rate": 0.0018146513680494262,
|
| 334 |
+
"loss": 5.6906,
|
| 335 |
+
"step": 21000
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 4.744042365401588,
|
| 339 |
+
"grad_norm": 0.1533203125,
|
| 340 |
+
"learning_rate": 0.0018102383053839365,
|
| 341 |
+
"loss": 5.6761,
|
| 342 |
+
"step": 21500
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 4.854368932038835,
|
| 346 |
+
"grad_norm": 0.1298828125,
|
| 347 |
+
"learning_rate": 0.0018058252427184467,
|
| 348 |
+
"loss": 5.6859,
|
| 349 |
+
"step": 22000
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 4.964695498676081,
|
| 353 |
+
"grad_norm": 0.1611328125,
|
| 354 |
+
"learning_rate": 0.0018014121800529568,
|
| 355 |
+
"loss": 5.6948,
|
| 356 |
+
"step": 22500
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 5.0,
|
| 360 |
+
"eval_loss": 5.584114074707031,
|
| 361 |
+
"eval_runtime": 6.7803,
|
| 362 |
+
"eval_samples_per_second": 73.743,
|
| 363 |
+
"eval_steps_per_second": 1.18,
|
| 364 |
+
"step": 22660
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 5.075022065313328,
|
| 368 |
+
"grad_norm": 0.146484375,
|
| 369 |
+
"learning_rate": 0.001796999117387467,
|
| 370 |
+
"loss": 5.6773,
|
| 371 |
+
"step": 23000
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 5.185348631950574,
|
| 375 |
+
"grad_norm": 0.1494140625,
|
| 376 |
+
"learning_rate": 0.001792586054721977,
|
| 377 |
+
"loss": 5.6741,
|
| 378 |
+
"step": 23500
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 5.29567519858782,
|
| 382 |
+
"grad_norm": 0.138671875,
|
| 383 |
+
"learning_rate": 0.0017881729920564873,
|
| 384 |
+
"loss": 5.6694,
|
| 385 |
+
"step": 24000
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 5.406001765225066,
|
| 389 |
+
"grad_norm": 0.1474609375,
|
| 390 |
+
"learning_rate": 0.0017837599293909973,
|
| 391 |
+
"loss": 5.6803,
|
| 392 |
+
"step": 24500
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 5.516328331862312,
|
| 396 |
+
"grad_norm": 0.1611328125,
|
| 397 |
+
"learning_rate": 0.0017793468667255076,
|
| 398 |
+
"loss": 5.6683,
|
| 399 |
+
"step": 25000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 5.626654898499559,
|
| 403 |
+
"grad_norm": 0.1318359375,
|
| 404 |
+
"learning_rate": 0.0017749338040600176,
|
| 405 |
+
"loss": 5.6717,
|
| 406 |
+
"step": 25500
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 5.736981465136805,
|
| 410 |
+
"grad_norm": 0.126953125,
|
| 411 |
+
"learning_rate": 0.0017705207413945278,
|
| 412 |
+
"loss": 5.671,
|
| 413 |
+
"step": 26000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 5.847308031774051,
|
| 417 |
+
"grad_norm": 0.1474609375,
|
| 418 |
+
"learning_rate": 0.001766107678729038,
|
| 419 |
+
"loss": 5.6843,
|
| 420 |
+
"step": 26500
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 5.957634598411297,
|
| 424 |
+
"grad_norm": 0.1484375,
|
| 425 |
+
"learning_rate": 0.0017616946160635481,
|
| 426 |
+
"loss": 5.6726,
|
| 427 |
+
"step": 27000
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 6.0,
|
| 431 |
+
"eval_loss": 5.570712566375732,
|
| 432 |
+
"eval_runtime": 6.7662,
|
| 433 |
+
"eval_samples_per_second": 73.896,
|
| 434 |
+
"eval_steps_per_second": 1.182,
|
| 435 |
+
"step": 27192
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 6.067961165048544,
|
| 439 |
+
"grad_norm": 0.1455078125,
|
| 440 |
+
"learning_rate": 0.0017572815533980584,
|
| 441 |
+
"loss": 5.6475,
|
| 442 |
+
"step": 27500
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 6.17828773168579,
|
| 446 |
+
"grad_norm": 0.1328125,
|
| 447 |
+
"learning_rate": 0.0017528684907325684,
|
| 448 |
+
"loss": 5.6556,
|
| 449 |
+
"step": 28000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 6.288614298323036,
|
| 453 |
+
"grad_norm": 0.142578125,
|
| 454 |
+
"learning_rate": 0.0017484554280670786,
|
| 455 |
+
"loss": 5.6642,
|
| 456 |
+
"step": 28500
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 6.398940864960283,
|
| 460 |
+
"grad_norm": 0.1474609375,
|
| 461 |
+
"learning_rate": 0.0017440423654015887,
|
| 462 |
+
"loss": 5.6667,
|
| 463 |
+
"step": 29000
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 6.509267431597529,
|
| 467 |
+
"grad_norm": 0.15625,
|
| 468 |
+
"learning_rate": 0.001739629302736099,
|
| 469 |
+
"loss": 5.6611,
|
| 470 |
+
"step": 29500
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 6.619593998234775,
|
| 474 |
+
"grad_norm": 0.1376953125,
|
| 475 |
+
"learning_rate": 0.0017352162400706092,
|
| 476 |
+
"loss": 5.6541,
|
| 477 |
+
"step": 30000
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 6.729920564872021,
|
| 481 |
+
"grad_norm": 0.146484375,
|
| 482 |
+
"learning_rate": 0.0017308031774051192,
|
| 483 |
+
"loss": 5.6624,
|
| 484 |
+
"step": 30500
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 6.840247131509267,
|
| 488 |
+
"grad_norm": 0.1484375,
|
| 489 |
+
"learning_rate": 0.0017263901147396292,
|
| 490 |
+
"loss": 5.6582,
|
| 491 |
+
"step": 31000
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 6.950573698146513,
|
| 495 |
+
"grad_norm": 0.150390625,
|
| 496 |
+
"learning_rate": 0.0017219770520741397,
|
| 497 |
+
"loss": 5.6548,
|
| 498 |
+
"step": 31500
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 7.0,
|
| 502 |
+
"eval_loss": 5.559887886047363,
|
| 503 |
+
"eval_runtime": 6.8808,
|
| 504 |
+
"eval_samples_per_second": 72.666,
|
| 505 |
+
"eval_steps_per_second": 1.163,
|
| 506 |
+
"step": 31724
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 7.0609002647837595,
|
| 510 |
+
"grad_norm": 0.154296875,
|
| 511 |
+
"learning_rate": 0.0017175639894086497,
|
| 512 |
+
"loss": 5.6465,
|
| 513 |
+
"step": 32000
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 7.171226831421007,
|
| 517 |
+
"grad_norm": 0.1298828125,
|
| 518 |
+
"learning_rate": 0.0017131509267431597,
|
| 519 |
+
"loss": 5.6571,
|
| 520 |
+
"step": 32500
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 7.281553398058253,
|
| 524 |
+
"grad_norm": 0.1708984375,
|
| 525 |
+
"learning_rate": 0.0017087378640776698,
|
| 526 |
+
"loss": 5.6475,
|
| 527 |
+
"step": 33000
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 7.391879964695499,
|
| 531 |
+
"grad_norm": 0.138671875,
|
| 532 |
+
"learning_rate": 0.0017043248014121802,
|
| 533 |
+
"loss": 5.6503,
|
| 534 |
+
"step": 33500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 7.502206531332745,
|
| 538 |
+
"grad_norm": 0.1572265625,
|
| 539 |
+
"learning_rate": 0.0016999117387466903,
|
| 540 |
+
"loss": 5.6645,
|
| 541 |
+
"step": 34000
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 7.612533097969991,
|
| 545 |
+
"grad_norm": 0.1396484375,
|
| 546 |
+
"learning_rate": 0.0016954986760812003,
|
| 547 |
+
"loss": 5.6403,
|
| 548 |
+
"step": 34500
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 7.722859664607237,
|
| 552 |
+
"grad_norm": 0.1259765625,
|
| 553 |
+
"learning_rate": 0.0016910856134157105,
|
| 554 |
+
"loss": 5.6528,
|
| 555 |
+
"step": 35000
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 7.8331862312444835,
|
| 559 |
+
"grad_norm": 0.138671875,
|
| 560 |
+
"learning_rate": 0.0016866725507502208,
|
| 561 |
+
"loss": 5.6419,
|
| 562 |
+
"step": 35500
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 7.94351279788173,
|
| 566 |
+
"grad_norm": 0.1396484375,
|
| 567 |
+
"learning_rate": 0.0016822594880847308,
|
| 568 |
+
"loss": 5.6511,
|
| 569 |
+
"step": 36000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 8.0,
|
| 573 |
+
"eval_loss": 5.553098678588867,
|
| 574 |
+
"eval_runtime": 8.0507,
|
| 575 |
+
"eval_samples_per_second": 62.106,
|
| 576 |
+
"eval_steps_per_second": 0.994,
|
| 577 |
+
"step": 36256
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 8.053839364518977,
|
| 581 |
+
"grad_norm": 0.1689453125,
|
| 582 |
+
"learning_rate": 0.0016778464254192408,
|
| 583 |
+
"loss": 5.6478,
|
| 584 |
+
"step": 36500
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 8.164165931156223,
|
| 588 |
+
"grad_norm": 0.16796875,
|
| 589 |
+
"learning_rate": 0.0016734333627537513,
|
| 590 |
+
"loss": 5.6245,
|
| 591 |
+
"step": 37000
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 8.274492497793469,
|
| 595 |
+
"grad_norm": 0.1474609375,
|
| 596 |
+
"learning_rate": 0.0016690203000882613,
|
| 597 |
+
"loss": 5.6376,
|
| 598 |
+
"step": 37500
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 8.384819064430715,
|
| 602 |
+
"grad_norm": 0.2099609375,
|
| 603 |
+
"learning_rate": 0.0016646072374227714,
|
| 604 |
+
"loss": 5.6465,
|
| 605 |
+
"step": 38000
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 8.495145631067961,
|
| 609 |
+
"grad_norm": 0.1552734375,
|
| 610 |
+
"learning_rate": 0.0016601941747572816,
|
| 611 |
+
"loss": 5.6546,
|
| 612 |
+
"step": 38500
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 8.605472197705208,
|
| 616 |
+
"grad_norm": 0.15625,
|
| 617 |
+
"learning_rate": 0.0016557811120917918,
|
| 618 |
+
"loss": 5.6396,
|
| 619 |
+
"step": 39000
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 8.715798764342454,
|
| 623 |
+
"grad_norm": 0.1533203125,
|
| 624 |
+
"learning_rate": 0.0016513680494263019,
|
| 625 |
+
"loss": 5.6524,
|
| 626 |
+
"step": 39500
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 8.8261253309797,
|
| 630 |
+
"grad_norm": 0.15625,
|
| 631 |
+
"learning_rate": 0.0016469549867608121,
|
| 632 |
+
"loss": 5.6368,
|
| 633 |
+
"step": 40000
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 8.936451897616946,
|
| 637 |
+
"grad_norm": 0.1533203125,
|
| 638 |
+
"learning_rate": 0.0016425419240953221,
|
| 639 |
+
"loss": 5.6372,
|
| 640 |
+
"step": 40500
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 9.0,
|
| 644 |
+
"eval_loss": 5.547245979309082,
|
| 645 |
+
"eval_runtime": 6.7444,
|
| 646 |
+
"eval_samples_per_second": 74.136,
|
| 647 |
+
"eval_steps_per_second": 1.186,
|
| 648 |
+
"step": 40788
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 9.046778464254192,
|
| 652 |
+
"grad_norm": 0.1572265625,
|
| 653 |
+
"learning_rate": 0.0016381288614298324,
|
| 654 |
+
"loss": 5.6392,
|
| 655 |
+
"step": 41000
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 9.157105030891438,
|
| 659 |
+
"grad_norm": 0.1845703125,
|
| 660 |
+
"learning_rate": 0.0016337157987643424,
|
| 661 |
+
"loss": 5.6347,
|
| 662 |
+
"step": 41500
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 9.267431597528685,
|
| 666 |
+
"grad_norm": 0.2041015625,
|
| 667 |
+
"learning_rate": 0.0016293027360988527,
|
| 668 |
+
"loss": 5.6439,
|
| 669 |
+
"step": 42000
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 9.37775816416593,
|
| 673 |
+
"grad_norm": 0.15625,
|
| 674 |
+
"learning_rate": 0.0016248896734333627,
|
| 675 |
+
"loss": 5.6422,
|
| 676 |
+
"step": 42500
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 9.488084730803177,
|
| 680 |
+
"grad_norm": 0.17578125,
|
| 681 |
+
"learning_rate": 0.001620476610767873,
|
| 682 |
+
"loss": 5.6266,
|
| 683 |
+
"step": 43000
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 9.598411297440423,
|
| 687 |
+
"grad_norm": 0.140625,
|
| 688 |
+
"learning_rate": 0.0016160635481023832,
|
| 689 |
+
"loss": 5.6408,
|
| 690 |
+
"step": 43500
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 9.70873786407767,
|
| 694 |
+
"grad_norm": 0.1328125,
|
| 695 |
+
"learning_rate": 0.0016116504854368932,
|
| 696 |
+
"loss": 5.6357,
|
| 697 |
+
"step": 44000
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 9.819064430714917,
|
| 701 |
+
"grad_norm": 0.1474609375,
|
| 702 |
+
"learning_rate": 0.0016072374227714032,
|
| 703 |
+
"loss": 5.6387,
|
| 704 |
+
"step": 44500
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 9.929390997352161,
|
| 708 |
+
"grad_norm": 0.1435546875,
|
| 709 |
+
"learning_rate": 0.0016028243601059135,
|
| 710 |
+
"loss": 5.6347,
|
| 711 |
+
"step": 45000
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 10.0,
|
| 715 |
+
"eval_loss": 5.540464401245117,
|
| 716 |
+
"eval_runtime": 7.1327,
|
| 717 |
+
"eval_samples_per_second": 70.1,
|
| 718 |
+
"eval_steps_per_second": 1.122,
|
| 719 |
+
"step": 45320
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 10.03971756398941,
|
| 723 |
+
"grad_norm": 0.1513671875,
|
| 724 |
+
"learning_rate": 0.0015984112974404237,
|
| 725 |
+
"loss": 5.6329,
|
| 726 |
+
"step": 45500
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 10.150044130626656,
|
| 730 |
+
"grad_norm": 0.1728515625,
|
| 731 |
+
"learning_rate": 0.0015939982347749338,
|
| 732 |
+
"loss": 5.6368,
|
| 733 |
+
"step": 46000
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 10.260370697263902,
|
| 737 |
+
"grad_norm": 0.13671875,
|
| 738 |
+
"learning_rate": 0.001589585172109444,
|
| 739 |
+
"loss": 5.6292,
|
| 740 |
+
"step": 46500
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 10.370697263901148,
|
| 744 |
+
"grad_norm": 0.1396484375,
|
| 745 |
+
"learning_rate": 0.0015851721094439543,
|
| 746 |
+
"loss": 5.64,
|
| 747 |
+
"step": 47000
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 10.481023830538394,
|
| 751 |
+
"grad_norm": 0.1845703125,
|
| 752 |
+
"learning_rate": 0.0015807590467784643,
|
| 753 |
+
"loss": 5.63,
|
| 754 |
+
"step": 47500
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 10.59135039717564,
|
| 758 |
+
"grad_norm": 0.1826171875,
|
| 759 |
+
"learning_rate": 0.0015763459841129743,
|
| 760 |
+
"loss": 5.6312,
|
| 761 |
+
"step": 48000
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 10.701676963812886,
|
| 765 |
+
"grad_norm": 0.1708984375,
|
| 766 |
+
"learning_rate": 0.0015719329214474848,
|
| 767 |
+
"loss": 5.6319,
|
| 768 |
+
"step": 48500
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 10.812003530450133,
|
| 772 |
+
"grad_norm": 0.1533203125,
|
| 773 |
+
"learning_rate": 0.0015675198587819948,
|
| 774 |
+
"loss": 5.6298,
|
| 775 |
+
"step": 49000
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 10.922330097087379,
|
| 779 |
+
"grad_norm": 0.1904296875,
|
| 780 |
+
"learning_rate": 0.0015631067961165048,
|
| 781 |
+
"loss": 5.6281,
|
| 782 |
+
"step": 49500
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 11.0,
|
| 786 |
+
"eval_loss": 5.537391662597656,
|
| 787 |
+
"eval_runtime": 6.6373,
|
| 788 |
+
"eval_samples_per_second": 75.332,
|
| 789 |
+
"eval_steps_per_second": 1.205,
|
| 790 |
+
"step": 49852
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 11.032656663724625,
|
| 794 |
+
"grad_norm": 0.1513671875,
|
| 795 |
+
"learning_rate": 0.0015586937334510149,
|
| 796 |
+
"loss": 5.6283,
|
| 797 |
+
"step": 50000
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 11.142983230361871,
|
| 801 |
+
"grad_norm": 0.1435546875,
|
| 802 |
+
"learning_rate": 0.0015542806707855253,
|
| 803 |
+
"loss": 5.6227,
|
| 804 |
+
"step": 50500
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 11.253309796999117,
|
| 808 |
+
"grad_norm": 0.1767578125,
|
| 809 |
+
"learning_rate": 0.0015498676081200354,
|
| 810 |
+
"loss": 5.6277,
|
| 811 |
+
"step": 51000
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 11.363636363636363,
|
| 815 |
+
"grad_norm": 0.134765625,
|
| 816 |
+
"learning_rate": 0.0015454545454545454,
|
| 817 |
+
"loss": 5.6274,
|
| 818 |
+
"step": 51500
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 11.47396293027361,
|
| 822 |
+
"grad_norm": 0.140625,
|
| 823 |
+
"learning_rate": 0.0015410414827890556,
|
| 824 |
+
"loss": 5.6306,
|
| 825 |
+
"step": 52000
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 11.584289496910856,
|
| 829 |
+
"grad_norm": 0.150390625,
|
| 830 |
+
"learning_rate": 0.0015366284201235659,
|
| 831 |
+
"loss": 5.63,
|
| 832 |
+
"step": 52500
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 11.694616063548102,
|
| 836 |
+
"grad_norm": 0.1982421875,
|
| 837 |
+
"learning_rate": 0.001532215357458076,
|
| 838 |
+
"loss": 5.6296,
|
| 839 |
+
"step": 53000
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 11.804942630185348,
|
| 843 |
+
"grad_norm": 0.2490234375,
|
| 844 |
+
"learning_rate": 0.001527802294792586,
|
| 845 |
+
"loss": 5.6257,
|
| 846 |
+
"step": 53500
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 11.915269196822594,
|
| 850 |
+
"grad_norm": 0.1767578125,
|
| 851 |
+
"learning_rate": 0.0015233892321270964,
|
| 852 |
+
"loss": 5.6299,
|
| 853 |
+
"step": 54000
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 12.0,
|
| 857 |
+
"eval_loss": 5.5359296798706055,
|
| 858 |
+
"eval_runtime": 7.1034,
|
| 859 |
+
"eval_samples_per_second": 70.388,
|
| 860 |
+
"eval_steps_per_second": 1.126,
|
| 861 |
+
"step": 54384
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 12.02559576345984,
|
| 865 |
+
"grad_norm": 0.22265625,
|
| 866 |
+
"learning_rate": 0.0015189761694616064,
|
| 867 |
+
"loss": 5.6219,
|
| 868 |
+
"step": 54500
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 12.135922330097088,
|
| 872 |
+
"grad_norm": 0.162109375,
|
| 873 |
+
"learning_rate": 0.0015145631067961165,
|
| 874 |
+
"loss": 5.6342,
|
| 875 |
+
"step": 55000
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 12.246248896734334,
|
| 879 |
+
"grad_norm": 0.1669921875,
|
| 880 |
+
"learning_rate": 0.0015101500441306267,
|
| 881 |
+
"loss": 5.629,
|
| 882 |
+
"step": 55500
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 12.35657546337158,
|
| 886 |
+
"grad_norm": 0.1865234375,
|
| 887 |
+
"learning_rate": 0.001505736981465137,
|
| 888 |
+
"loss": 5.6198,
|
| 889 |
+
"step": 56000
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 12.466902030008827,
|
| 893 |
+
"grad_norm": 0.16015625,
|
| 894 |
+
"learning_rate": 0.001501323918799647,
|
| 895 |
+
"loss": 5.6308,
|
| 896 |
+
"step": 56500
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 12.577228596646073,
|
| 900 |
+
"grad_norm": 0.265625,
|
| 901 |
+
"learning_rate": 0.0014969108561341572,
|
| 902 |
+
"loss": 5.6205,
|
| 903 |
+
"step": 57000
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 12.687555163283319,
|
| 907 |
+
"grad_norm": 0.1416015625,
|
| 908 |
+
"learning_rate": 0.0014924977934686673,
|
| 909 |
+
"loss": 5.6075,
|
| 910 |
+
"step": 57500
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 12.797881729920565,
|
| 914 |
+
"grad_norm": 0.1708984375,
|
| 915 |
+
"learning_rate": 0.0014880847308031775,
|
| 916 |
+
"loss": 5.6294,
|
| 917 |
+
"step": 58000
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 12.908208296557811,
|
| 921 |
+
"grad_norm": 0.1484375,
|
| 922 |
+
"learning_rate": 0.0014836716681376875,
|
| 923 |
+
"loss": 5.6225,
|
| 924 |
+
"step": 58500
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 13.0,
|
| 928 |
+
"eval_loss": 5.531345844268799,
|
| 929 |
+
"eval_runtime": 6.7386,
|
| 930 |
+
"eval_samples_per_second": 74.2,
|
| 931 |
+
"eval_steps_per_second": 1.187,
|
| 932 |
+
"step": 58916
|
| 933 |
+
}
|
| 934 |
+
],
|
| 935 |
+
"logging_steps": 500,
|
| 936 |
+
"max_steps": 226600,
|
| 937 |
+
"num_input_tokens_seen": 0,
|
| 938 |
+
"num_train_epochs": 50,
|
| 939 |
+
"save_steps": 500,
|
| 940 |
+
"stateful_callbacks": {
|
| 941 |
+
"EarlyStoppingCallback": {
|
| 942 |
+
"args": {
|
| 943 |
+
"early_stopping_patience": 3,
|
| 944 |
+
"early_stopping_threshold": 0.0
|
| 945 |
+
},
|
| 946 |
+
"attributes": {
|
| 947 |
+
"early_stopping_patience_counter": 0
|
| 948 |
+
}
|
| 949 |
+
},
|
| 950 |
+
"TrainerControl": {
|
| 951 |
+
"args": {
|
| 952 |
+
"should_epoch_stop": false,
|
| 953 |
+
"should_evaluate": false,
|
| 954 |
+
"should_log": false,
|
| 955 |
+
"should_save": true,
|
| 956 |
+
"should_training_stop": false
|
| 957 |
+
},
|
| 958 |
+
"attributes": {}
|
| 959 |
+
}
|
| 960 |
+
},
|
| 961 |
+
"total_flos": 7.800284808610376e+17,
|
| 962 |
+
"train_batch_size": 64,
|
| 963 |
+
"trial_name": null,
|
| 964 |
+
"trial_params": null
|
| 965 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:32b62fbe21e999dc3647a99078ba6560a1f1a619d03fa179f34f90834502e558
|
| 3 |
+
size 5048
|