Commit
•
fa64459
1
Parent(s):
ea054ed
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 +1773 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "../MistralModels/models/Mixtral-8x7B-v0.1-small-4096",
|
| 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": 768,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 768,
|
| 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": "4.43.3",
|
| 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": "4.43.3"
|
| 6 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:81c550de27b0cfb958bd90295e8081ba41021975958f42ab55709ffed1088c28
|
| 3 |
+
size 276979168
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:37f947d6b54eeedc56f985f58c76b5403af723450de37052dd484bc469768e1b
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c97752dc999f5f01bcf49874b8735b2a56080af04fd7b12e23eeed58185e97d8
|
| 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 +1,1773 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 6.787229061126709,
|
| 3 |
+
"best_model_checkpoint": "./results/models/checkpoint-106530",
|
| 4 |
+
"epoch": 30.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 106530,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.14080540692762603,
|
| 13 |
+
"grad_norm": 0.2255859375,
|
| 14 |
+
"learning_rate": 0.00398873556744579,
|
| 15 |
+
"loss": 6.8932,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.28161081385525205,
|
| 20 |
+
"grad_norm": 0.9609375,
|
| 21 |
+
"learning_rate": 0.0039774711348915795,
|
| 22 |
+
"loss": 6.8292,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.42241622078287805,
|
| 27 |
+
"grad_norm": 1.6015625,
|
| 28 |
+
"learning_rate": 0.00396620670233737,
|
| 29 |
+
"loss": 6.8237,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.5632216277105041,
|
| 34 |
+
"grad_norm": 2.75,
|
| 35 |
+
"learning_rate": 0.00395494226978316,
|
| 36 |
+
"loss": 6.8763,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.7040270346381301,
|
| 41 |
+
"grad_norm": 1.375,
|
| 42 |
+
"learning_rate": 0.00394367783722895,
|
| 43 |
+
"loss": 6.9046,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.8448324415657561,
|
| 48 |
+
"grad_norm": 1.1171875,
|
| 49 |
+
"learning_rate": 0.003932413404674739,
|
| 50 |
+
"loss": 6.8564,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.9856378484933821,
|
| 55 |
+
"grad_norm": 2.25,
|
| 56 |
+
"learning_rate": 0.003921148972120529,
|
| 57 |
+
"loss": 6.8481,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 1.0,
|
| 62 |
+
"eval_loss": 6.840273380279541,
|
| 63 |
+
"eval_runtime": 197.295,
|
| 64 |
+
"eval_samples_per_second": 10.137,
|
| 65 |
+
"eval_steps_per_second": 1.267,
|
| 66 |
+
"step": 3551
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.1264432554210082,
|
| 70 |
+
"grad_norm": 6.875,
|
| 71 |
+
"learning_rate": 0.003909884539566319,
|
| 72 |
+
"loss": 6.8419,
|
| 73 |
+
"step": 4000
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.267248662348634,
|
| 77 |
+
"grad_norm": 0.349609375,
|
| 78 |
+
"learning_rate": 0.003898620107012109,
|
| 79 |
+
"loss": 6.9106,
|
| 80 |
+
"step": 4500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.4080540692762602,
|
| 84 |
+
"grad_norm": 1.96875,
|
| 85 |
+
"learning_rate": 0.003887355674457899,
|
| 86 |
+
"loss": 6.8742,
|
| 87 |
+
"step": 5000
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 1.5488594762038863,
|
| 91 |
+
"grad_norm": 1.734375,
|
| 92 |
+
"learning_rate": 0.0038760912419036893,
|
| 93 |
+
"loss": 6.8742,
|
| 94 |
+
"step": 5500
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 1.6896648831315122,
|
| 98 |
+
"grad_norm": 4.125,
|
| 99 |
+
"learning_rate": 0.003864826809349479,
|
| 100 |
+
"loss": 6.8928,
|
| 101 |
+
"step": 6000
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 1.8304702900591383,
|
| 105 |
+
"grad_norm": 2.125,
|
| 106 |
+
"learning_rate": 0.003853562376795269,
|
| 107 |
+
"loss": 6.8663,
|
| 108 |
+
"step": 6500
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.9712756969867642,
|
| 112 |
+
"grad_norm": 3.640625,
|
| 113 |
+
"learning_rate": 0.003842297944241059,
|
| 114 |
+
"loss": 6.8515,
|
| 115 |
+
"step": 7000
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 2.0,
|
| 119 |
+
"eval_loss": 6.85520601272583,
|
| 120 |
+
"eval_runtime": 192.2095,
|
| 121 |
+
"eval_samples_per_second": 10.405,
|
| 122 |
+
"eval_steps_per_second": 1.301,
|
| 123 |
+
"step": 7102
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 2.1120811039143903,
|
| 127 |
+
"grad_norm": 55.75,
|
| 128 |
+
"learning_rate": 0.0038310335116868485,
|
| 129 |
+
"loss": 6.8759,
|
| 130 |
+
"step": 7500
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 2.2528865108420164,
|
| 134 |
+
"grad_norm": 15.0,
|
| 135 |
+
"learning_rate": 0.0038197690791326386,
|
| 136 |
+
"loss": 6.9025,
|
| 137 |
+
"step": 8000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 2.3936919177696425,
|
| 141 |
+
"grad_norm": 0.259765625,
|
| 142 |
+
"learning_rate": 0.0038085046465784287,
|
| 143 |
+
"loss": 6.9002,
|
| 144 |
+
"step": 8500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 2.534497324697268,
|
| 148 |
+
"grad_norm": 0.42578125,
|
| 149 |
+
"learning_rate": 0.003797240214024219,
|
| 150 |
+
"loss": 6.8857,
|
| 151 |
+
"step": 9000
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 2.6753027316248943,
|
| 155 |
+
"grad_norm": 2.984375,
|
| 156 |
+
"learning_rate": 0.0037859757814700085,
|
| 157 |
+
"loss": 6.882,
|
| 158 |
+
"step": 9500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 2.8161081385525204,
|
| 162 |
+
"grad_norm": 1.953125,
|
| 163 |
+
"learning_rate": 0.0037747113489157986,
|
| 164 |
+
"loss": 6.8677,
|
| 165 |
+
"step": 10000
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 2.9569135454801465,
|
| 169 |
+
"grad_norm": 1.484375,
|
| 170 |
+
"learning_rate": 0.0037634469163615883,
|
| 171 |
+
"loss": 6.8696,
|
| 172 |
+
"step": 10500
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 3.0,
|
| 176 |
+
"eval_loss": 6.850285530090332,
|
| 177 |
+
"eval_runtime": 190.0725,
|
| 178 |
+
"eval_samples_per_second": 10.522,
|
| 179 |
+
"eval_steps_per_second": 1.315,
|
| 180 |
+
"step": 10653
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 3.0977189524077726,
|
| 184 |
+
"grad_norm": 0.94140625,
|
| 185 |
+
"learning_rate": 0.003752182483807378,
|
| 186 |
+
"loss": 6.8573,
|
| 187 |
+
"step": 11000
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 3.2385243593353983,
|
| 191 |
+
"grad_norm": 11.1875,
|
| 192 |
+
"learning_rate": 0.003740918051253168,
|
| 193 |
+
"loss": 6.8511,
|
| 194 |
+
"step": 11500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 3.3793297662630244,
|
| 198 |
+
"grad_norm": 2.3125,
|
| 199 |
+
"learning_rate": 0.0037296536186989583,
|
| 200 |
+
"loss": 6.8586,
|
| 201 |
+
"step": 12000
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 3.5201351731906505,
|
| 205 |
+
"grad_norm": 134.0,
|
| 206 |
+
"learning_rate": 0.003718389186144748,
|
| 207 |
+
"loss": 6.8683,
|
| 208 |
+
"step": 12500
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 3.6609405801182766,
|
| 212 |
+
"grad_norm": 3.65625,
|
| 213 |
+
"learning_rate": 0.003707124753590538,
|
| 214 |
+
"loss": 6.8525,
|
| 215 |
+
"step": 13000
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 3.8017459870459027,
|
| 219 |
+
"grad_norm": 1.78125,
|
| 220 |
+
"learning_rate": 0.0036958603210363278,
|
| 221 |
+
"loss": 6.8547,
|
| 222 |
+
"step": 13500
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 3.9425513939735284,
|
| 226 |
+
"grad_norm": 14.1875,
|
| 227 |
+
"learning_rate": 0.0036845958884821174,
|
| 228 |
+
"loss": 6.8456,
|
| 229 |
+
"step": 14000
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 4.0,
|
| 233 |
+
"eval_loss": 6.838648319244385,
|
| 234 |
+
"eval_runtime": 187.9085,
|
| 235 |
+
"eval_samples_per_second": 10.643,
|
| 236 |
+
"eval_steps_per_second": 1.33,
|
| 237 |
+
"step": 14204
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 4.0833568009011545,
|
| 241 |
+
"grad_norm": 1.5078125,
|
| 242 |
+
"learning_rate": 0.0036733314559279076,
|
| 243 |
+
"loss": 6.8422,
|
| 244 |
+
"step": 14500
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 4.224162207828781,
|
| 248 |
+
"grad_norm": 2.484375,
|
| 249 |
+
"learning_rate": 0.0036620670233736977,
|
| 250 |
+
"loss": 6.859,
|
| 251 |
+
"step": 15000
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 4.364967614756407,
|
| 255 |
+
"grad_norm": 6.65625,
|
| 256 |
+
"learning_rate": 0.003650802590819488,
|
| 257 |
+
"loss": 6.8541,
|
| 258 |
+
"step": 15500
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 4.505773021684033,
|
| 262 |
+
"grad_norm": 6.59375,
|
| 263 |
+
"learning_rate": 0.0036395381582652775,
|
| 264 |
+
"loss": 6.8385,
|
| 265 |
+
"step": 16000
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 4.646578428611659,
|
| 269 |
+
"grad_norm": 3.546875,
|
| 270 |
+
"learning_rate": 0.003628273725711067,
|
| 271 |
+
"loss": 6.8464,
|
| 272 |
+
"step": 16500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 4.787383835539285,
|
| 276 |
+
"grad_norm": 1.625,
|
| 277 |
+
"learning_rate": 0.0036170092931568573,
|
| 278 |
+
"loss": 6.8434,
|
| 279 |
+
"step": 17000
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 4.92818924246691,
|
| 283 |
+
"grad_norm": 16.25,
|
| 284 |
+
"learning_rate": 0.003605744860602647,
|
| 285 |
+
"loss": 6.8525,
|
| 286 |
+
"step": 17500
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 5.0,
|
| 290 |
+
"eval_loss": 6.835108280181885,
|
| 291 |
+
"eval_runtime": 207.3832,
|
| 292 |
+
"eval_samples_per_second": 9.644,
|
| 293 |
+
"eval_steps_per_second": 1.205,
|
| 294 |
+
"step": 17755
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 5.068994649394536,
|
| 298 |
+
"grad_norm": 2.5625,
|
| 299 |
+
"learning_rate": 0.003594480428048437,
|
| 300 |
+
"loss": 6.8404,
|
| 301 |
+
"step": 18000
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 5.2098000563221625,
|
| 305 |
+
"grad_norm": 4.8125,
|
| 306 |
+
"learning_rate": 0.0035832159954942273,
|
| 307 |
+
"loss": 6.8484,
|
| 308 |
+
"step": 18500
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 5.350605463249789,
|
| 312 |
+
"grad_norm": 3.8125,
|
| 313 |
+
"learning_rate": 0.003571951562940017,
|
| 314 |
+
"loss": 6.8502,
|
| 315 |
+
"step": 19000
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 5.491410870177415,
|
| 319 |
+
"grad_norm": 2.34375,
|
| 320 |
+
"learning_rate": 0.0035606871303858066,
|
| 321 |
+
"loss": 6.8384,
|
| 322 |
+
"step": 19500
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 5.632216277105041,
|
| 326 |
+
"grad_norm": 1.96875,
|
| 327 |
+
"learning_rate": 0.0035494226978315968,
|
| 328 |
+
"loss": 6.8478,
|
| 329 |
+
"step": 20000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 5.773021684032667,
|
| 333 |
+
"grad_norm": 3.375,
|
| 334 |
+
"learning_rate": 0.003538158265277387,
|
| 335 |
+
"loss": 6.8339,
|
| 336 |
+
"step": 20500
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 5.913827090960293,
|
| 340 |
+
"grad_norm": 4.65625,
|
| 341 |
+
"learning_rate": 0.0035268938327231766,
|
| 342 |
+
"loss": 6.843,
|
| 343 |
+
"step": 21000
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 6.0,
|
| 347 |
+
"eval_loss": 6.82509708404541,
|
| 348 |
+
"eval_runtime": 199.4174,
|
| 349 |
+
"eval_samples_per_second": 10.029,
|
| 350 |
+
"eval_steps_per_second": 1.254,
|
| 351 |
+
"step": 21306
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 6.054632497887919,
|
| 355 |
+
"grad_norm": 3.171875,
|
| 356 |
+
"learning_rate": 0.0035156294001689667,
|
| 357 |
+
"loss": 6.8405,
|
| 358 |
+
"step": 21500
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 6.195437904815545,
|
| 362 |
+
"grad_norm": 10.9375,
|
| 363 |
+
"learning_rate": 0.003504364967614757,
|
| 364 |
+
"loss": 6.8416,
|
| 365 |
+
"step": 22000
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 6.336243311743171,
|
| 369 |
+
"grad_norm": 7.3125,
|
| 370 |
+
"learning_rate": 0.003493100535060546,
|
| 371 |
+
"loss": 6.8379,
|
| 372 |
+
"step": 22500
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 6.477048718670797,
|
| 376 |
+
"grad_norm": 4.65625,
|
| 377 |
+
"learning_rate": 0.003481836102506336,
|
| 378 |
+
"loss": 6.8452,
|
| 379 |
+
"step": 23000
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 6.617854125598423,
|
| 383 |
+
"grad_norm": 1.7734375,
|
| 384 |
+
"learning_rate": 0.0034705716699521263,
|
| 385 |
+
"loss": 6.8463,
|
| 386 |
+
"step": 23500
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 6.758659532526049,
|
| 390 |
+
"grad_norm": 11.25,
|
| 391 |
+
"learning_rate": 0.003459307237397916,
|
| 392 |
+
"loss": 6.8358,
|
| 393 |
+
"step": 24000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 6.899464939453675,
|
| 397 |
+
"grad_norm": 1.7421875,
|
| 398 |
+
"learning_rate": 0.003448042804843706,
|
| 399 |
+
"loss": 6.8361,
|
| 400 |
+
"step": 24500
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 7.0,
|
| 404 |
+
"eval_loss": 6.824986457824707,
|
| 405 |
+
"eval_runtime": 181.2581,
|
| 406 |
+
"eval_samples_per_second": 11.034,
|
| 407 |
+
"eval_steps_per_second": 1.379,
|
| 408 |
+
"step": 24857
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 7.040270346381301,
|
| 412 |
+
"grad_norm": 3.84375,
|
| 413 |
+
"learning_rate": 0.0034367783722894962,
|
| 414 |
+
"loss": 6.8328,
|
| 415 |
+
"step": 25000
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 7.181075753308927,
|
| 419 |
+
"grad_norm": 1.4296875,
|
| 420 |
+
"learning_rate": 0.0034255139397352855,
|
| 421 |
+
"loss": 6.8272,
|
| 422 |
+
"step": 25500
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 7.321881160236553,
|
| 426 |
+
"grad_norm": 3.359375,
|
| 427 |
+
"learning_rate": 0.0034142495071810756,
|
| 428 |
+
"loss": 6.8164,
|
| 429 |
+
"step": 26000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 7.462686567164179,
|
| 433 |
+
"grad_norm": 8.875,
|
| 434 |
+
"learning_rate": 0.0034029850746268657,
|
| 435 |
+
"loss": 6.8276,
|
| 436 |
+
"step": 26500
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 7.603491974091805,
|
| 440 |
+
"grad_norm": 2.078125,
|
| 441 |
+
"learning_rate": 0.003391720642072656,
|
| 442 |
+
"loss": 6.8248,
|
| 443 |
+
"step": 27000
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 7.744297381019431,
|
| 447 |
+
"grad_norm": 8.8125,
|
| 448 |
+
"learning_rate": 0.0033804562095184456,
|
| 449 |
+
"loss": 6.8188,
|
| 450 |
+
"step": 27500
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 7.885102787947057,
|
| 454 |
+
"grad_norm": 3.921875,
|
| 455 |
+
"learning_rate": 0.0033691917769642357,
|
| 456 |
+
"loss": 6.8232,
|
| 457 |
+
"step": 28000
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 8.0,
|
| 461 |
+
"eval_loss": 6.8248701095581055,
|
| 462 |
+
"eval_runtime": 170.7503,
|
| 463 |
+
"eval_samples_per_second": 11.713,
|
| 464 |
+
"eval_steps_per_second": 1.464,
|
| 465 |
+
"step": 28408
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 8.025908194874683,
|
| 469 |
+
"grad_norm": 5.9375,
|
| 470 |
+
"learning_rate": 0.0033579273444100254,
|
| 471 |
+
"loss": 6.8241,
|
| 472 |
+
"step": 28500
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 8.166713601802309,
|
| 476 |
+
"grad_norm": 5.8125,
|
| 477 |
+
"learning_rate": 0.003346662911855815,
|
| 478 |
+
"loss": 6.8311,
|
| 479 |
+
"step": 29000
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 8.307519008729935,
|
| 483 |
+
"grad_norm": 3.328125,
|
| 484 |
+
"learning_rate": 0.003335398479301605,
|
| 485 |
+
"loss": 6.8203,
|
| 486 |
+
"step": 29500
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 8.448324415657561,
|
| 490 |
+
"grad_norm": 7.46875,
|
| 491 |
+
"learning_rate": 0.0033241340467473953,
|
| 492 |
+
"loss": 6.8291,
|
| 493 |
+
"step": 30000
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 8.589129822585187,
|
| 497 |
+
"grad_norm": 4.59375,
|
| 498 |
+
"learning_rate": 0.003312869614193185,
|
| 499 |
+
"loss": 6.8261,
|
| 500 |
+
"step": 30500
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 8.729935229512813,
|
| 504 |
+
"grad_norm": 4.3125,
|
| 505 |
+
"learning_rate": 0.003301605181638975,
|
| 506 |
+
"loss": 6.8362,
|
| 507 |
+
"step": 31000
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 8.87074063644044,
|
| 511 |
+
"grad_norm": 11.25,
|
| 512 |
+
"learning_rate": 0.003290340749084765,
|
| 513 |
+
"loss": 6.8353,
|
| 514 |
+
"step": 31500
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 9.0,
|
| 518 |
+
"eval_loss": 6.821832656860352,
|
| 519 |
+
"eval_runtime": 210.9202,
|
| 520 |
+
"eval_samples_per_second": 9.482,
|
| 521 |
+
"eval_steps_per_second": 1.185,
|
| 522 |
+
"step": 31959
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 9.011546043368066,
|
| 526 |
+
"grad_norm": 4.46875,
|
| 527 |
+
"learning_rate": 0.003279076316530555,
|
| 528 |
+
"loss": 6.832,
|
| 529 |
+
"step": 32000
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 9.152351450295692,
|
| 533 |
+
"grad_norm": 4.0625,
|
| 534 |
+
"learning_rate": 0.0032678118839763446,
|
| 535 |
+
"loss": 6.8292,
|
| 536 |
+
"step": 32500
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 9.293156857223318,
|
| 540 |
+
"grad_norm": 1.984375,
|
| 541 |
+
"learning_rate": 0.0032565474514221347,
|
| 542 |
+
"loss": 6.8248,
|
| 543 |
+
"step": 33000
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 9.433962264150944,
|
| 547 |
+
"grad_norm": 43.25,
|
| 548 |
+
"learning_rate": 0.003245283018867925,
|
| 549 |
+
"loss": 6.8277,
|
| 550 |
+
"step": 33500
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 9.57476767107857,
|
| 554 |
+
"grad_norm": 1.7890625,
|
| 555 |
+
"learning_rate": 0.0032340185863137146,
|
| 556 |
+
"loss": 6.8229,
|
| 557 |
+
"step": 34000
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 9.715573078006196,
|
| 561 |
+
"grad_norm": 3.296875,
|
| 562 |
+
"learning_rate": 0.0032227541537595042,
|
| 563 |
+
"loss": 6.8194,
|
| 564 |
+
"step": 34500
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 9.85637848493382,
|
| 568 |
+
"grad_norm": 3.390625,
|
| 569 |
+
"learning_rate": 0.0032114897212052944,
|
| 570 |
+
"loss": 6.8225,
|
| 571 |
+
"step": 35000
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 9.997183891861447,
|
| 575 |
+
"grad_norm": 22.375,
|
| 576 |
+
"learning_rate": 0.003200225288651084,
|
| 577 |
+
"loss": 6.8291,
|
| 578 |
+
"step": 35500
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 10.0,
|
| 582 |
+
"eval_loss": 6.81277322769165,
|
| 583 |
+
"eval_runtime": 239.1719,
|
| 584 |
+
"eval_samples_per_second": 8.362,
|
| 585 |
+
"eval_steps_per_second": 1.045,
|
| 586 |
+
"step": 35510
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 10.137989298789073,
|
| 590 |
+
"grad_norm": 5.375,
|
| 591 |
+
"learning_rate": 0.003188960856096874,
|
| 592 |
+
"loss": 6.815,
|
| 593 |
+
"step": 36000
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 10.278794705716699,
|
| 597 |
+
"grad_norm": 5.25,
|
| 598 |
+
"learning_rate": 0.0031776964235426643,
|
| 599 |
+
"loss": 6.8173,
|
| 600 |
+
"step": 36500
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 10.419600112644325,
|
| 604 |
+
"grad_norm": 3.84375,
|
| 605 |
+
"learning_rate": 0.003166431990988454,
|
| 606 |
+
"loss": 6.8189,
|
| 607 |
+
"step": 37000
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 10.560405519571951,
|
| 611 |
+
"grad_norm": 1.1875,
|
| 612 |
+
"learning_rate": 0.0031551675584342437,
|
| 613 |
+
"loss": 6.828,
|
| 614 |
+
"step": 37500
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 10.701210926499577,
|
| 618 |
+
"grad_norm": 5.96875,
|
| 619 |
+
"learning_rate": 0.003143903125880034,
|
| 620 |
+
"loss": 6.8141,
|
| 621 |
+
"step": 38000
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 10.842016333427203,
|
| 625 |
+
"grad_norm": 117.0,
|
| 626 |
+
"learning_rate": 0.003132638693325824,
|
| 627 |
+
"loss": 6.8258,
|
| 628 |
+
"step": 38500
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 10.98282174035483,
|
| 632 |
+
"grad_norm": 3.40625,
|
| 633 |
+
"learning_rate": 0.0031213742607716136,
|
| 634 |
+
"loss": 6.8254,
|
| 635 |
+
"step": 39000
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 11.0,
|
| 639 |
+
"eval_loss": 6.814772129058838,
|
| 640 |
+
"eval_runtime": 228.9281,
|
| 641 |
+
"eval_samples_per_second": 8.736,
|
| 642 |
+
"eval_steps_per_second": 1.092,
|
| 643 |
+
"step": 39061
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 11.123627147282455,
|
| 647 |
+
"grad_norm": 1.578125,
|
| 648 |
+
"learning_rate": 0.0031101098282174037,
|
| 649 |
+
"loss": 6.8192,
|
| 650 |
+
"step": 39500
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 11.264432554210082,
|
| 654 |
+
"grad_norm": 2.6875,
|
| 655 |
+
"learning_rate": 0.003098845395663194,
|
| 656 |
+
"loss": 6.8217,
|
| 657 |
+
"step": 40000
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 11.405237961137708,
|
| 661 |
+
"grad_norm": 35.0,
|
| 662 |
+
"learning_rate": 0.003087580963108983,
|
| 663 |
+
"loss": 6.8172,
|
| 664 |
+
"step": 40500
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 11.546043368065334,
|
| 668 |
+
"grad_norm": 9.5625,
|
| 669 |
+
"learning_rate": 0.0030763165305547732,
|
| 670 |
+
"loss": 6.8307,
|
| 671 |
+
"step": 41000
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 11.68684877499296,
|
| 675 |
+
"grad_norm": 9.3125,
|
| 676 |
+
"learning_rate": 0.0030650520980005634,
|
| 677 |
+
"loss": 6.8237,
|
| 678 |
+
"step": 41500
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 11.827654181920586,
|
| 682 |
+
"grad_norm": 5.3125,
|
| 683 |
+
"learning_rate": 0.003053787665446353,
|
| 684 |
+
"loss": 6.8306,
|
| 685 |
+
"step": 42000
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 11.968459588848212,
|
| 689 |
+
"grad_norm": 5.34375,
|
| 690 |
+
"learning_rate": 0.003042523232892143,
|
| 691 |
+
"loss": 6.8138,
|
| 692 |
+
"step": 42500
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 12.0,
|
| 696 |
+
"eval_loss": 6.806704044342041,
|
| 697 |
+
"eval_runtime": 228.3922,
|
| 698 |
+
"eval_samples_per_second": 8.757,
|
| 699 |
+
"eval_steps_per_second": 1.095,
|
| 700 |
+
"step": 42612
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 12.109264995775838,
|
| 704 |
+
"grad_norm": 1.9296875,
|
| 705 |
+
"learning_rate": 0.0030312588003379333,
|
| 706 |
+
"loss": 6.8157,
|
| 707 |
+
"step": 43000
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 12.250070402703464,
|
| 711 |
+
"grad_norm": 5.25,
|
| 712 |
+
"learning_rate": 0.003019994367783723,
|
| 713 |
+
"loss": 6.8189,
|
| 714 |
+
"step": 43500
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 12.39087580963109,
|
| 718 |
+
"grad_norm": 3.078125,
|
| 719 |
+
"learning_rate": 0.0030087299352295127,
|
| 720 |
+
"loss": 6.8276,
|
| 721 |
+
"step": 44000
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 12.531681216558717,
|
| 725 |
+
"grad_norm": 3.46875,
|
| 726 |
+
"learning_rate": 0.002997465502675303,
|
| 727 |
+
"loss": 6.8092,
|
| 728 |
+
"step": 44500
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 12.672486623486343,
|
| 732 |
+
"grad_norm": 1.2421875,
|
| 733 |
+
"learning_rate": 0.002986201070121093,
|
| 734 |
+
"loss": 6.8152,
|
| 735 |
+
"step": 45000
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 12.813292030413969,
|
| 739 |
+
"grad_norm": 3.078125,
|
| 740 |
+
"learning_rate": 0.0029749366375668826,
|
| 741 |
+
"loss": 6.8141,
|
| 742 |
+
"step": 45500
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 12.954097437341593,
|
| 746 |
+
"grad_norm": 2.90625,
|
| 747 |
+
"learning_rate": 0.0029636722050126727,
|
| 748 |
+
"loss": 6.8209,
|
| 749 |
+
"step": 46000
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 13.0,
|
| 753 |
+
"eval_loss": 6.805023670196533,
|
| 754 |
+
"eval_runtime": 230.3566,
|
| 755 |
+
"eval_samples_per_second": 8.682,
|
| 756 |
+
"eval_steps_per_second": 1.085,
|
| 757 |
+
"step": 46163
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 13.09490284426922,
|
| 761 |
+
"grad_norm": 2.90625,
|
| 762 |
+
"learning_rate": 0.0029524077724584624,
|
| 763 |
+
"loss": 6.8179,
|
| 764 |
+
"step": 46500
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 13.235708251196845,
|
| 768 |
+
"grad_norm": 6.0,
|
| 769 |
+
"learning_rate": 0.002941143339904252,
|
| 770 |
+
"loss": 6.8193,
|
| 771 |
+
"step": 47000
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 13.376513658124471,
|
| 775 |
+
"grad_norm": 4.03125,
|
| 776 |
+
"learning_rate": 0.0029298789073500422,
|
| 777 |
+
"loss": 6.8153,
|
| 778 |
+
"step": 47500
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 13.517319065052098,
|
| 782 |
+
"grad_norm": 3.65625,
|
| 783 |
+
"learning_rate": 0.0029186144747958324,
|
| 784 |
+
"loss": 6.8046,
|
| 785 |
+
"step": 48000
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 13.658124471979724,
|
| 789 |
+
"grad_norm": 52.0,
|
| 790 |
+
"learning_rate": 0.002907350042241622,
|
| 791 |
+
"loss": 6.8161,
|
| 792 |
+
"step": 48500
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 13.79892987890735,
|
| 796 |
+
"grad_norm": 16.625,
|
| 797 |
+
"learning_rate": 0.002896085609687412,
|
| 798 |
+
"loss": 6.8094,
|
| 799 |
+
"step": 49000
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 13.939735285834976,
|
| 803 |
+
"grad_norm": 1.6328125,
|
| 804 |
+
"learning_rate": 0.002884821177133202,
|
| 805 |
+
"loss": 6.8181,
|
| 806 |
+
"step": 49500
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 14.0,
|
| 810 |
+
"eval_loss": 6.8096513748168945,
|
| 811 |
+
"eval_runtime": 235.2835,
|
| 812 |
+
"eval_samples_per_second": 8.5,
|
| 813 |
+
"eval_steps_per_second": 1.063,
|
| 814 |
+
"step": 49714
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 14.080540692762602,
|
| 818 |
+
"grad_norm": 1.140625,
|
| 819 |
+
"learning_rate": 0.002873556744578992,
|
| 820 |
+
"loss": 6.8197,
|
| 821 |
+
"step": 50000
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 14.221346099690228,
|
| 825 |
+
"grad_norm": 1.265625,
|
| 826 |
+
"learning_rate": 0.0028622923120247817,
|
| 827 |
+
"loss": 6.8097,
|
| 828 |
+
"step": 50500
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 14.362151506617854,
|
| 832 |
+
"grad_norm": 4.75,
|
| 833 |
+
"learning_rate": 0.002851027879470572,
|
| 834 |
+
"loss": 6.82,
|
| 835 |
+
"step": 51000
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 14.50295691354548,
|
| 839 |
+
"grad_norm": 2.28125,
|
| 840 |
+
"learning_rate": 0.002839763446916362,
|
| 841 |
+
"loss": 6.8203,
|
| 842 |
+
"step": 51500
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 14.643762320473106,
|
| 846 |
+
"grad_norm": 5.125,
|
| 847 |
+
"learning_rate": 0.0028284990143621516,
|
| 848 |
+
"loss": 6.807,
|
| 849 |
+
"step": 52000
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 14.784567727400733,
|
| 853 |
+
"grad_norm": 2.046875,
|
| 854 |
+
"learning_rate": 0.0028172345818079413,
|
| 855 |
+
"loss": 6.8153,
|
| 856 |
+
"step": 52500
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 14.925373134328359,
|
| 860 |
+
"grad_norm": 8.0,
|
| 861 |
+
"learning_rate": 0.0028059701492537314,
|
| 862 |
+
"loss": 6.8102,
|
| 863 |
+
"step": 53000
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 15.0,
|
| 867 |
+
"eval_loss": 6.802714824676514,
|
| 868 |
+
"eval_runtime": 235.2095,
|
| 869 |
+
"eval_samples_per_second": 8.503,
|
| 870 |
+
"eval_steps_per_second": 1.063,
|
| 871 |
+
"step": 53265
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 15.066178541255985,
|
| 875 |
+
"grad_norm": 16.5,
|
| 876 |
+
"learning_rate": 0.002794705716699521,
|
| 877 |
+
"loss": 6.8108,
|
| 878 |
+
"step": 53500
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 15.20698394818361,
|
| 882 |
+
"grad_norm": 2.015625,
|
| 883 |
+
"learning_rate": 0.0027834412841453112,
|
| 884 |
+
"loss": 6.8123,
|
| 885 |
+
"step": 54000
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 15.347789355111237,
|
| 889 |
+
"grad_norm": 2.90625,
|
| 890 |
+
"learning_rate": 0.0027721768515911013,
|
| 891 |
+
"loss": 6.8105,
|
| 892 |
+
"step": 54500
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 15.488594762038863,
|
| 896 |
+
"grad_norm": 8.375,
|
| 897 |
+
"learning_rate": 0.0027609124190368915,
|
| 898 |
+
"loss": 6.8163,
|
| 899 |
+
"step": 55000
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 15.629400168966487,
|
| 903 |
+
"grad_norm": 3.34375,
|
| 904 |
+
"learning_rate": 0.0027496479864826807,
|
| 905 |
+
"loss": 6.8062,
|
| 906 |
+
"step": 55500
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 15.770205575894114,
|
| 910 |
+
"grad_norm": 48.25,
|
| 911 |
+
"learning_rate": 0.002738383553928471,
|
| 912 |
+
"loss": 6.8065,
|
| 913 |
+
"step": 56000
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 15.91101098282174,
|
| 917 |
+
"grad_norm": 2.9375,
|
| 918 |
+
"learning_rate": 0.002727119121374261,
|
| 919 |
+
"loss": 6.8066,
|
| 920 |
+
"step": 56500
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 16.0,
|
| 924 |
+
"eval_loss": 6.799798011779785,
|
| 925 |
+
"eval_runtime": 204.1083,
|
| 926 |
+
"eval_samples_per_second": 9.799,
|
| 927 |
+
"eval_steps_per_second": 1.225,
|
| 928 |
+
"step": 56816
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 16.051816389749366,
|
| 932 |
+
"grad_norm": 136.0,
|
| 933 |
+
"learning_rate": 0.0027158546888200507,
|
| 934 |
+
"loss": 6.8064,
|
| 935 |
+
"step": 57000
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 16.192621796676992,
|
| 939 |
+
"grad_norm": 2.75,
|
| 940 |
+
"learning_rate": 0.002704590256265841,
|
| 941 |
+
"loss": 6.8083,
|
| 942 |
+
"step": 57500
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 16.333427203604618,
|
| 946 |
+
"grad_norm": 15.625,
|
| 947 |
+
"learning_rate": 0.002693325823711631,
|
| 948 |
+
"loss": 6.8038,
|
| 949 |
+
"step": 58000
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 16.474232610532244,
|
| 953 |
+
"grad_norm": 20.875,
|
| 954 |
+
"learning_rate": 0.00268206139115742,
|
| 955 |
+
"loss": 6.802,
|
| 956 |
+
"step": 58500
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 16.61503801745987,
|
| 960 |
+
"grad_norm": 15.3125,
|
| 961 |
+
"learning_rate": 0.0026707969586032103,
|
| 962 |
+
"loss": 6.8112,
|
| 963 |
+
"step": 59000
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 16.755843424387496,
|
| 967 |
+
"grad_norm": 8.1875,
|
| 968 |
+
"learning_rate": 0.0026595325260490004,
|
| 969 |
+
"loss": 6.8144,
|
| 970 |
+
"step": 59500
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 16.896648831315122,
|
| 974 |
+
"grad_norm": 5.03125,
|
| 975 |
+
"learning_rate": 0.0026482680934947905,
|
| 976 |
+
"loss": 6.8158,
|
| 977 |
+
"step": 60000
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 17.0,
|
| 981 |
+
"eval_loss": 6.8019022941589355,
|
| 982 |
+
"eval_runtime": 194.9738,
|
| 983 |
+
"eval_samples_per_second": 10.258,
|
| 984 |
+
"eval_steps_per_second": 1.282,
|
| 985 |
+
"step": 60367
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 17.03745423824275,
|
| 989 |
+
"grad_norm": 5.34375,
|
| 990 |
+
"learning_rate": 0.0026370036609405802,
|
| 991 |
+
"loss": 6.8102,
|
| 992 |
+
"step": 60500
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 17.178259645170375,
|
| 996 |
+
"grad_norm": 26.125,
|
| 997 |
+
"learning_rate": 0.0026257392283863703,
|
| 998 |
+
"loss": 6.8135,
|
| 999 |
+
"step": 61000
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 17.319065052098,
|
| 1003 |
+
"grad_norm": 6.71875,
|
| 1004 |
+
"learning_rate": 0.00261447479583216,
|
| 1005 |
+
"loss": 6.8102,
|
| 1006 |
+
"step": 61500
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 17.459870459025627,
|
| 1010 |
+
"grad_norm": 30.875,
|
| 1011 |
+
"learning_rate": 0.0026032103632779497,
|
| 1012 |
+
"loss": 6.8099,
|
| 1013 |
+
"step": 62000
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 17.600675865953253,
|
| 1017 |
+
"grad_norm": 78.0,
|
| 1018 |
+
"learning_rate": 0.00259194593072374,
|
| 1019 |
+
"loss": 6.8046,
|
| 1020 |
+
"step": 62500
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 17.74148127288088,
|
| 1024 |
+
"grad_norm": 14.625,
|
| 1025 |
+
"learning_rate": 0.00258068149816953,
|
| 1026 |
+
"loss": 6.809,
|
| 1027 |
+
"step": 63000
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 17.882286679808505,
|
| 1031 |
+
"grad_norm": 170.0,
|
| 1032 |
+
"learning_rate": 0.0025694170656153197,
|
| 1033 |
+
"loss": 6.8035,
|
| 1034 |
+
"step": 63500
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 18.0,
|
| 1038 |
+
"eval_loss": 6.801079273223877,
|
| 1039 |
+
"eval_runtime": 232.8698,
|
| 1040 |
+
"eval_samples_per_second": 8.588,
|
| 1041 |
+
"eval_steps_per_second": 1.074,
|
| 1042 |
+
"step": 63918
|
| 1043 |
+
},
|
| 1044 |
+
{
|
| 1045 |
+
"epoch": 18.02309208673613,
|
| 1046 |
+
"grad_norm": 2.296875,
|
| 1047 |
+
"learning_rate": 0.0025581526330611098,
|
| 1048 |
+
"loss": 6.8134,
|
| 1049 |
+
"step": 64000
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 18.163897493663757,
|
| 1053 |
+
"grad_norm": 87.0,
|
| 1054 |
+
"learning_rate": 0.0025468882005068995,
|
| 1055 |
+
"loss": 6.8024,
|
| 1056 |
+
"step": 64500
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 18.304702900591383,
|
| 1060 |
+
"grad_norm": 1.3828125,
|
| 1061 |
+
"learning_rate": 0.002535623767952689,
|
| 1062 |
+
"loss": 6.8068,
|
| 1063 |
+
"step": 65000
|
| 1064 |
+
},
|
| 1065 |
+
{
|
| 1066 |
+
"epoch": 18.44550830751901,
|
| 1067 |
+
"grad_norm": 22.5,
|
| 1068 |
+
"learning_rate": 0.0025243593353984793,
|
| 1069 |
+
"loss": 6.8083,
|
| 1070 |
+
"step": 65500
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 18.586313714446636,
|
| 1074 |
+
"grad_norm": 3.578125,
|
| 1075 |
+
"learning_rate": 0.0025130949028442694,
|
| 1076 |
+
"loss": 6.8071,
|
| 1077 |
+
"step": 66000
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 18.727119121374262,
|
| 1081 |
+
"grad_norm": 22.125,
|
| 1082 |
+
"learning_rate": 0.0025018304702900595,
|
| 1083 |
+
"loss": 6.8059,
|
| 1084 |
+
"step": 66500
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 18.867924528301888,
|
| 1088 |
+
"grad_norm": 16.75,
|
| 1089 |
+
"learning_rate": 0.002490566037735849,
|
| 1090 |
+
"loss": 6.8056,
|
| 1091 |
+
"step": 67000
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 19.0,
|
| 1095 |
+
"eval_loss": 6.797260761260986,
|
| 1096 |
+
"eval_runtime": 216.6541,
|
| 1097 |
+
"eval_samples_per_second": 9.231,
|
| 1098 |
+
"eval_steps_per_second": 1.154,
|
| 1099 |
+
"step": 67469
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"epoch": 19.008729935229514,
|
| 1103 |
+
"grad_norm": 4.625,
|
| 1104 |
+
"learning_rate": 0.002479301605181639,
|
| 1105 |
+
"loss": 6.8015,
|
| 1106 |
+
"step": 67500
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 19.14953534215714,
|
| 1110 |
+
"grad_norm": 2.734375,
|
| 1111 |
+
"learning_rate": 0.002468037172627429,
|
| 1112 |
+
"loss": 6.8118,
|
| 1113 |
+
"step": 68000
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 19.290340749084766,
|
| 1117 |
+
"grad_norm": 45.0,
|
| 1118 |
+
"learning_rate": 0.0024567727400732187,
|
| 1119 |
+
"loss": 6.7995,
|
| 1120 |
+
"step": 68500
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 19.431146156012392,
|
| 1124 |
+
"grad_norm": 46.0,
|
| 1125 |
+
"learning_rate": 0.002445508307519009,
|
| 1126 |
+
"loss": 6.8098,
|
| 1127 |
+
"step": 69000
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 19.57195156294002,
|
| 1131 |
+
"grad_norm": 7.09375,
|
| 1132 |
+
"learning_rate": 0.002434243874964799,
|
| 1133 |
+
"loss": 6.7972,
|
| 1134 |
+
"step": 69500
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 19.712756969867645,
|
| 1138 |
+
"grad_norm": 91.5,
|
| 1139 |
+
"learning_rate": 0.002422979442410588,
|
| 1140 |
+
"loss": 6.8085,
|
| 1141 |
+
"step": 70000
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 19.853562376795267,
|
| 1145 |
+
"grad_norm": 1.7890625,
|
| 1146 |
+
"learning_rate": 0.0024117150098563783,
|
| 1147 |
+
"loss": 6.8034,
|
| 1148 |
+
"step": 70500
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 19.994367783722893,
|
| 1152 |
+
"grad_norm": 4.21875,
|
| 1153 |
+
"learning_rate": 0.0024004505773021685,
|
| 1154 |
+
"loss": 6.8024,
|
| 1155 |
+
"step": 71000
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 20.0,
|
| 1159 |
+
"eval_loss": 6.794472694396973,
|
| 1160 |
+
"eval_runtime": 212.8014,
|
| 1161 |
+
"eval_samples_per_second": 9.398,
|
| 1162 |
+
"eval_steps_per_second": 1.175,
|
| 1163 |
+
"step": 71020
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 20.13517319065052,
|
| 1167 |
+
"grad_norm": 4.21875,
|
| 1168 |
+
"learning_rate": 0.0023891861447479586,
|
| 1169 |
+
"loss": 6.7964,
|
| 1170 |
+
"step": 71500
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 20.275978597578145,
|
| 1174 |
+
"grad_norm": 27.5,
|
| 1175 |
+
"learning_rate": 0.0023779217121937483,
|
| 1176 |
+
"loss": 6.8033,
|
| 1177 |
+
"step": 72000
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 20.41678400450577,
|
| 1181 |
+
"grad_norm": 4.84375,
|
| 1182 |
+
"learning_rate": 0.0023666572796395384,
|
| 1183 |
+
"loss": 6.8066,
|
| 1184 |
+
"step": 72500
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 20.557589411433398,
|
| 1188 |
+
"grad_norm": 10.0625,
|
| 1189 |
+
"learning_rate": 0.0023553928470853285,
|
| 1190 |
+
"loss": 6.8076,
|
| 1191 |
+
"step": 73000
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 20.698394818361024,
|
| 1195 |
+
"grad_norm": 60.75,
|
| 1196 |
+
"learning_rate": 0.0023441284145311178,
|
| 1197 |
+
"loss": 6.8114,
|
| 1198 |
+
"step": 73500
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 20.83920022528865,
|
| 1202 |
+
"grad_norm": 47.0,
|
| 1203 |
+
"learning_rate": 0.002332863981976908,
|
| 1204 |
+
"loss": 6.8046,
|
| 1205 |
+
"step": 74000
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 20.980005632216276,
|
| 1209 |
+
"grad_norm": 4.5,
|
| 1210 |
+
"learning_rate": 0.002321599549422698,
|
| 1211 |
+
"loss": 6.8088,
|
| 1212 |
+
"step": 74500
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"epoch": 21.0,
|
| 1216 |
+
"eval_loss": 6.799040794372559,
|
| 1217 |
+
"eval_runtime": 211.5781,
|
| 1218 |
+
"eval_samples_per_second": 9.453,
|
| 1219 |
+
"eval_steps_per_second": 1.182,
|
| 1220 |
+
"step": 74571
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 21.120811039143902,
|
| 1224 |
+
"grad_norm": 1.7734375,
|
| 1225 |
+
"learning_rate": 0.0023103351168684877,
|
| 1226 |
+
"loss": 6.8154,
|
| 1227 |
+
"step": 75000
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 21.26161644607153,
|
| 1231 |
+
"grad_norm": 9.75,
|
| 1232 |
+
"learning_rate": 0.002299070684314278,
|
| 1233 |
+
"loss": 6.8074,
|
| 1234 |
+
"step": 75500
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 21.402421852999154,
|
| 1238 |
+
"grad_norm": 12.9375,
|
| 1239 |
+
"learning_rate": 0.0022878062517600675,
|
| 1240 |
+
"loss": 6.8097,
|
| 1241 |
+
"step": 76000
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 21.54322725992678,
|
| 1245 |
+
"grad_norm": 2.703125,
|
| 1246 |
+
"learning_rate": 0.002276541819205857,
|
| 1247 |
+
"loss": 6.8016,
|
| 1248 |
+
"step": 76500
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 21.684032666854407,
|
| 1252 |
+
"grad_norm": 2.875,
|
| 1253 |
+
"learning_rate": 0.0022652773866516473,
|
| 1254 |
+
"loss": 6.7996,
|
| 1255 |
+
"step": 77000
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 21.824838073782033,
|
| 1259 |
+
"grad_norm": 4.84375,
|
| 1260 |
+
"learning_rate": 0.0022540129540974375,
|
| 1261 |
+
"loss": 6.8058,
|
| 1262 |
+
"step": 77500
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 21.96564348070966,
|
| 1266 |
+
"grad_norm": 7.15625,
|
| 1267 |
+
"learning_rate": 0.0022427485215432276,
|
| 1268 |
+
"loss": 6.8024,
|
| 1269 |
+
"step": 78000
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 22.0,
|
| 1273 |
+
"eval_loss": 6.7944440841674805,
|
| 1274 |
+
"eval_runtime": 187.9546,
|
| 1275 |
+
"eval_samples_per_second": 10.641,
|
| 1276 |
+
"eval_steps_per_second": 1.33,
|
| 1277 |
+
"step": 78122
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 22.106448887637285,
|
| 1281 |
+
"grad_norm": 4.65625,
|
| 1282 |
+
"learning_rate": 0.0022314840889890173,
|
| 1283 |
+
"loss": 6.8025,
|
| 1284 |
+
"step": 78500
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 22.24725429456491,
|
| 1288 |
+
"grad_norm": 8.9375,
|
| 1289 |
+
"learning_rate": 0.002220219656434807,
|
| 1290 |
+
"loss": 6.8027,
|
| 1291 |
+
"step": 79000
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 22.388059701492537,
|
| 1295 |
+
"grad_norm": 8.75,
|
| 1296 |
+
"learning_rate": 0.002208955223880597,
|
| 1297 |
+
"loss": 6.8032,
|
| 1298 |
+
"step": 79500
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 22.528865108420163,
|
| 1302 |
+
"grad_norm": 10.625,
|
| 1303 |
+
"learning_rate": 0.0021976907913263868,
|
| 1304 |
+
"loss": 6.7998,
|
| 1305 |
+
"step": 80000
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 22.66967051534779,
|
| 1309 |
+
"grad_norm": 12.375,
|
| 1310 |
+
"learning_rate": 0.002186426358772177,
|
| 1311 |
+
"loss": 6.8102,
|
| 1312 |
+
"step": 80500
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 22.810475922275415,
|
| 1316 |
+
"grad_norm": 3.625,
|
| 1317 |
+
"learning_rate": 0.002175161926217967,
|
| 1318 |
+
"loss": 6.7944,
|
| 1319 |
+
"step": 81000
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 22.95128132920304,
|
| 1323 |
+
"grad_norm": 3.484375,
|
| 1324 |
+
"learning_rate": 0.0021638974936637567,
|
| 1325 |
+
"loss": 6.8079,
|
| 1326 |
+
"step": 81500
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 23.0,
|
| 1330 |
+
"eval_loss": 6.793390274047852,
|
| 1331 |
+
"eval_runtime": 214.7382,
|
| 1332 |
+
"eval_samples_per_second": 9.314,
|
| 1333 |
+
"eval_steps_per_second": 1.164,
|
| 1334 |
+
"step": 81673
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 23.092086736130668,
|
| 1338 |
+
"grad_norm": 5.46875,
|
| 1339 |
+
"learning_rate": 0.0021526330611095464,
|
| 1340 |
+
"loss": 6.8065,
|
| 1341 |
+
"step": 82000
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"epoch": 23.232892143058294,
|
| 1345 |
+
"grad_norm": 3.125,
|
| 1346 |
+
"learning_rate": 0.0021413686285553365,
|
| 1347 |
+
"loss": 6.8017,
|
| 1348 |
+
"step": 82500
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"epoch": 23.37369754998592,
|
| 1352 |
+
"grad_norm": 77.5,
|
| 1353 |
+
"learning_rate": 0.0021301041960011266,
|
| 1354 |
+
"loss": 6.8055,
|
| 1355 |
+
"step": 83000
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 23.514502956913546,
|
| 1359 |
+
"grad_norm": 1.5390625,
|
| 1360 |
+
"learning_rate": 0.0021188397634469163,
|
| 1361 |
+
"loss": 6.8064,
|
| 1362 |
+
"step": 83500
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"epoch": 23.655308363841172,
|
| 1366 |
+
"grad_norm": 3.28125,
|
| 1367 |
+
"learning_rate": 0.0021075753308927065,
|
| 1368 |
+
"loss": 6.8029,
|
| 1369 |
+
"step": 84000
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 23.796113770768798,
|
| 1373 |
+
"grad_norm": 4.5,
|
| 1374 |
+
"learning_rate": 0.0020963108983384966,
|
| 1375 |
+
"loss": 6.7993,
|
| 1376 |
+
"step": 84500
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 23.936919177696424,
|
| 1380 |
+
"grad_norm": 3.078125,
|
| 1381 |
+
"learning_rate": 0.002085046465784286,
|
| 1382 |
+
"loss": 6.7938,
|
| 1383 |
+
"step": 85000
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 24.0,
|
| 1387 |
+
"eval_loss": 6.793313503265381,
|
| 1388 |
+
"eval_runtime": 192.9768,
|
| 1389 |
+
"eval_samples_per_second": 10.364,
|
| 1390 |
+
"eval_steps_per_second": 1.295,
|
| 1391 |
+
"step": 85224
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 24.07772458462405,
|
| 1395 |
+
"grad_norm": 16.875,
|
| 1396 |
+
"learning_rate": 0.002073782033230076,
|
| 1397 |
+
"loss": 6.8043,
|
| 1398 |
+
"step": 85500
|
| 1399 |
+
},
|
| 1400 |
+
{
|
| 1401 |
+
"epoch": 24.218529991551677,
|
| 1402 |
+
"grad_norm": 7.0,
|
| 1403 |
+
"learning_rate": 0.002062517600675866,
|
| 1404 |
+
"loss": 6.7964,
|
| 1405 |
+
"step": 86000
|
| 1406 |
+
},
|
| 1407 |
+
{
|
| 1408 |
+
"epoch": 24.359335398479303,
|
| 1409 |
+
"grad_norm": 4.46875,
|
| 1410 |
+
"learning_rate": 0.0020512531681216558,
|
| 1411 |
+
"loss": 6.8079,
|
| 1412 |
+
"step": 86500
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 24.50014080540693,
|
| 1416 |
+
"grad_norm": 3.1875,
|
| 1417 |
+
"learning_rate": 0.002039988735567446,
|
| 1418 |
+
"loss": 6.8029,
|
| 1419 |
+
"step": 87000
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 24.640946212334555,
|
| 1423 |
+
"grad_norm": 3.203125,
|
| 1424 |
+
"learning_rate": 0.002028724303013236,
|
| 1425 |
+
"loss": 6.7963,
|
| 1426 |
+
"step": 87500
|
| 1427 |
+
},
|
| 1428 |
+
{
|
| 1429 |
+
"epoch": 24.78175161926218,
|
| 1430 |
+
"grad_norm": 8.9375,
|
| 1431 |
+
"learning_rate": 0.0020174598704590253,
|
| 1432 |
+
"loss": 6.8056,
|
| 1433 |
+
"step": 88000
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 24.922557026189807,
|
| 1437 |
+
"grad_norm": 2.859375,
|
| 1438 |
+
"learning_rate": 0.0020061954379048154,
|
| 1439 |
+
"loss": 6.8061,
|
| 1440 |
+
"step": 88500
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 25.0,
|
| 1444 |
+
"eval_loss": 6.796795845031738,
|
| 1445 |
+
"eval_runtime": 181.8838,
|
| 1446 |
+
"eval_samples_per_second": 10.996,
|
| 1447 |
+
"eval_steps_per_second": 1.375,
|
| 1448 |
+
"step": 88775
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 25.063362433117433,
|
| 1452 |
+
"grad_norm": 6.09375,
|
| 1453 |
+
"learning_rate": 0.0019949310053506055,
|
| 1454 |
+
"loss": 6.7973,
|
| 1455 |
+
"step": 89000
|
| 1456 |
+
},
|
| 1457 |
+
{
|
| 1458 |
+
"epoch": 25.20416784004506,
|
| 1459 |
+
"grad_norm": 1.09375,
|
| 1460 |
+
"learning_rate": 0.001983666572796395,
|
| 1461 |
+
"loss": 6.8059,
|
| 1462 |
+
"step": 89500
|
| 1463 |
+
},
|
| 1464 |
+
{
|
| 1465 |
+
"epoch": 25.344973246972685,
|
| 1466 |
+
"grad_norm": 24.375,
|
| 1467 |
+
"learning_rate": 0.0019724021402421853,
|
| 1468 |
+
"loss": 6.8048,
|
| 1469 |
+
"step": 90000
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 25.48577865390031,
|
| 1473 |
+
"grad_norm": 3.171875,
|
| 1474 |
+
"learning_rate": 0.0019611377076879754,
|
| 1475 |
+
"loss": 6.7963,
|
| 1476 |
+
"step": 90500
|
| 1477 |
+
},
|
| 1478 |
+
{
|
| 1479 |
+
"epoch": 25.626584060827938,
|
| 1480 |
+
"grad_norm": 1.546875,
|
| 1481 |
+
"learning_rate": 0.0019498732751337651,
|
| 1482 |
+
"loss": 6.7984,
|
| 1483 |
+
"step": 91000
|
| 1484 |
+
},
|
| 1485 |
+
{
|
| 1486 |
+
"epoch": 25.76738946775556,
|
| 1487 |
+
"grad_norm": 24.375,
|
| 1488 |
+
"learning_rate": 0.0019386088425795553,
|
| 1489 |
+
"loss": 6.7948,
|
| 1490 |
+
"step": 91500
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 25.908194874683186,
|
| 1494 |
+
"grad_norm": 29.125,
|
| 1495 |
+
"learning_rate": 0.001927344410025345,
|
| 1496 |
+
"loss": 6.8014,
|
| 1497 |
+
"step": 92000
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 26.0,
|
| 1501 |
+
"eval_loss": 6.788823127746582,
|
| 1502 |
+
"eval_runtime": 205.9541,
|
| 1503 |
+
"eval_samples_per_second": 9.711,
|
| 1504 |
+
"eval_steps_per_second": 1.214,
|
| 1505 |
+
"step": 92326
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 26.049000281610812,
|
| 1509 |
+
"grad_norm": 3.046875,
|
| 1510 |
+
"learning_rate": 0.0019160799774711349,
|
| 1511 |
+
"loss": 6.794,
|
| 1512 |
+
"step": 92500
|
| 1513 |
+
},
|
| 1514 |
+
{
|
| 1515 |
+
"epoch": 26.18980568853844,
|
| 1516 |
+
"grad_norm": 2.1875,
|
| 1517 |
+
"learning_rate": 0.001904815544916925,
|
| 1518 |
+
"loss": 6.7969,
|
| 1519 |
+
"step": 93000
|
| 1520 |
+
},
|
| 1521 |
+
{
|
| 1522 |
+
"epoch": 26.330611095466065,
|
| 1523 |
+
"grad_norm": 3.78125,
|
| 1524 |
+
"learning_rate": 0.0018935511123627147,
|
| 1525 |
+
"loss": 6.7898,
|
| 1526 |
+
"step": 93500
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 26.47141650239369,
|
| 1530 |
+
"grad_norm": 20.875,
|
| 1531 |
+
"learning_rate": 0.0018822866798085048,
|
| 1532 |
+
"loss": 6.7979,
|
| 1533 |
+
"step": 94000
|
| 1534 |
+
},
|
| 1535 |
+
{
|
| 1536 |
+
"epoch": 26.612221909321317,
|
| 1537 |
+
"grad_norm": 3.21875,
|
| 1538 |
+
"learning_rate": 0.0018710222472542947,
|
| 1539 |
+
"loss": 6.7962,
|
| 1540 |
+
"step": 94500
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 26.753027316248943,
|
| 1544 |
+
"grad_norm": 11.375,
|
| 1545 |
+
"learning_rate": 0.0018597578147000844,
|
| 1546 |
+
"loss": 6.7987,
|
| 1547 |
+
"step": 95000
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 26.89383272317657,
|
| 1551 |
+
"grad_norm": 8.875,
|
| 1552 |
+
"learning_rate": 0.0018484933821458745,
|
| 1553 |
+
"loss": 6.801,
|
| 1554 |
+
"step": 95500
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 27.0,
|
| 1558 |
+
"eval_loss": 6.789410591125488,
|
| 1559 |
+
"eval_runtime": 227.3793,
|
| 1560 |
+
"eval_samples_per_second": 8.796,
|
| 1561 |
+
"eval_steps_per_second": 1.099,
|
| 1562 |
+
"step": 95877
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 27.034638130104195,
|
| 1566 |
+
"grad_norm": 1.46875,
|
| 1567 |
+
"learning_rate": 0.0018372289495916644,
|
| 1568 |
+
"loss": 6.7975,
|
| 1569 |
+
"step": 96000
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 27.17544353703182,
|
| 1573 |
+
"grad_norm": 80.5,
|
| 1574 |
+
"learning_rate": 0.0018259645170374543,
|
| 1575 |
+
"loss": 6.7984,
|
| 1576 |
+
"step": 96500
|
| 1577 |
+
},
|
| 1578 |
+
{
|
| 1579 |
+
"epoch": 27.316248943959447,
|
| 1580 |
+
"grad_norm": 2.671875,
|
| 1581 |
+
"learning_rate": 0.0018147000844832442,
|
| 1582 |
+
"loss": 6.7878,
|
| 1583 |
+
"step": 97000
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"epoch": 27.457054350887073,
|
| 1587 |
+
"grad_norm": 9.0,
|
| 1588 |
+
"learning_rate": 0.0018034356519290341,
|
| 1589 |
+
"loss": 6.7973,
|
| 1590 |
+
"step": 97500
|
| 1591 |
+
},
|
| 1592 |
+
{
|
| 1593 |
+
"epoch": 27.5978597578147,
|
| 1594 |
+
"grad_norm": 2.1875,
|
| 1595 |
+
"learning_rate": 0.001792171219374824,
|
| 1596 |
+
"loss": 6.7993,
|
| 1597 |
+
"step": 98000
|
| 1598 |
+
},
|
| 1599 |
+
{
|
| 1600 |
+
"epoch": 27.738665164742326,
|
| 1601 |
+
"grad_norm": 5.65625,
|
| 1602 |
+
"learning_rate": 0.001780906786820614,
|
| 1603 |
+
"loss": 6.7979,
|
| 1604 |
+
"step": 98500
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 27.879470571669952,
|
| 1608 |
+
"grad_norm": 6.53125,
|
| 1609 |
+
"learning_rate": 0.001769642354266404,
|
| 1610 |
+
"loss": 6.7953,
|
| 1611 |
+
"step": 99000
|
| 1612 |
+
},
|
| 1613 |
+
{
|
| 1614 |
+
"epoch": 28.0,
|
| 1615 |
+
"eval_loss": 6.787979602813721,
|
| 1616 |
+
"eval_runtime": 193.1872,
|
| 1617 |
+
"eval_samples_per_second": 10.353,
|
| 1618 |
+
"eval_steps_per_second": 1.294,
|
| 1619 |
+
"step": 99428
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 28.020275978597578,
|
| 1623 |
+
"grad_norm": 8.5625,
|
| 1624 |
+
"learning_rate": 0.0017583779217121938,
|
| 1625 |
+
"loss": 6.7974,
|
| 1626 |
+
"step": 99500
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 28.161081385525204,
|
| 1630 |
+
"grad_norm": 7.25,
|
| 1631 |
+
"learning_rate": 0.0017471134891579837,
|
| 1632 |
+
"loss": 6.7964,
|
| 1633 |
+
"step": 100000
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 28.30188679245283,
|
| 1637 |
+
"grad_norm": 5.5625,
|
| 1638 |
+
"learning_rate": 0.0017358490566037738,
|
| 1639 |
+
"loss": 6.7967,
|
| 1640 |
+
"step": 100500
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 28.442692199380456,
|
| 1644 |
+
"grad_norm": 3.34375,
|
| 1645 |
+
"learning_rate": 0.0017245846240495635,
|
| 1646 |
+
"loss": 6.7968,
|
| 1647 |
+
"step": 101000
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 28.583497606308082,
|
| 1651 |
+
"grad_norm": 8.3125,
|
| 1652 |
+
"learning_rate": 0.0017133201914953536,
|
| 1653 |
+
"loss": 6.7985,
|
| 1654 |
+
"step": 101500
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 28.72430301323571,
|
| 1658 |
+
"grad_norm": 32.25,
|
| 1659 |
+
"learning_rate": 0.0017020557589411435,
|
| 1660 |
+
"loss": 6.7975,
|
| 1661 |
+
"step": 102000
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 28.865108420163335,
|
| 1665 |
+
"grad_norm": 5.25,
|
| 1666 |
+
"learning_rate": 0.0016907913263869332,
|
| 1667 |
+
"loss": 6.7887,
|
| 1668 |
+
"step": 102500
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 29.0,
|
| 1672 |
+
"eval_loss": 6.7877984046936035,
|
| 1673 |
+
"eval_runtime": 234.4697,
|
| 1674 |
+
"eval_samples_per_second": 8.53,
|
| 1675 |
+
"eval_steps_per_second": 1.066,
|
| 1676 |
+
"step": 102979
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 29.00591382709096,
|
| 1680 |
+
"grad_norm": 16.75,
|
| 1681 |
+
"learning_rate": 0.0016795268938327233,
|
| 1682 |
+
"loss": 6.7878,
|
| 1683 |
+
"step": 103000
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 29.146719234018587,
|
| 1687 |
+
"grad_norm": 4.65625,
|
| 1688 |
+
"learning_rate": 0.0016682624612785132,
|
| 1689 |
+
"loss": 6.7852,
|
| 1690 |
+
"step": 103500
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 29.287524640946213,
|
| 1694 |
+
"grad_norm": 8.0625,
|
| 1695 |
+
"learning_rate": 0.001656998028724303,
|
| 1696 |
+
"loss": 6.7974,
|
| 1697 |
+
"step": 104000
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 29.42833004787384,
|
| 1701 |
+
"grad_norm": 7.3125,
|
| 1702 |
+
"learning_rate": 0.001645733596170093,
|
| 1703 |
+
"loss": 6.8057,
|
| 1704 |
+
"step": 104500
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 29.569135454801465,
|
| 1708 |
+
"grad_norm": 1.0546875,
|
| 1709 |
+
"learning_rate": 0.001634469163615883,
|
| 1710 |
+
"loss": 6.7982,
|
| 1711 |
+
"step": 105000
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 29.70994086172909,
|
| 1715 |
+
"grad_norm": 11.125,
|
| 1716 |
+
"learning_rate": 0.0016232047310616728,
|
| 1717 |
+
"loss": 6.7948,
|
| 1718 |
+
"step": 105500
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 29.850746268656717,
|
| 1722 |
+
"grad_norm": 9.8125,
|
| 1723 |
+
"learning_rate": 0.0016119402985074627,
|
| 1724 |
+
"loss": 6.7843,
|
| 1725 |
+
"step": 106000
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 29.991551675584343,
|
| 1729 |
+
"grad_norm": 4.8125,
|
| 1730 |
+
"learning_rate": 0.0016006758659532524,
|
| 1731 |
+
"loss": 6.8003,
|
| 1732 |
+
"step": 106500
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 30.0,
|
| 1736 |
+
"eval_loss": 6.787229061126709,
|
| 1737 |
+
"eval_runtime": 216.4819,
|
| 1738 |
+
"eval_samples_per_second": 9.239,
|
| 1739 |
+
"eval_steps_per_second": 1.155,
|
| 1740 |
+
"step": 106530
|
| 1741 |
+
}
|
| 1742 |
+
],
|
| 1743 |
+
"logging_steps": 500,
|
| 1744 |
+
"max_steps": 177550,
|
| 1745 |
+
"num_input_tokens_seen": 0,
|
| 1746 |
+
"num_train_epochs": 50,
|
| 1747 |
+
"save_steps": 500,
|
| 1748 |
+
"stateful_callbacks": {
|
| 1749 |
+
"EarlyStoppingCallback": {
|
| 1750 |
+
"args": {
|
| 1751 |
+
"early_stopping_patience": 3,
|
| 1752 |
+
"early_stopping_threshold": 0.0
|
| 1753 |
+
},
|
| 1754 |
+
"attributes": {
|
| 1755 |
+
"early_stopping_patience_counter": 0
|
| 1756 |
+
}
|
| 1757 |
+
},
|
| 1758 |
+
"TrainerControl": {
|
| 1759 |
+
"args": {
|
| 1760 |
+
"should_epoch_stop": false,
|
| 1761 |
+
"should_evaluate": false,
|
| 1762 |
+
"should_log": false,
|
| 1763 |
+
"should_save": true,
|
| 1764 |
+
"should_training_stop": false
|
| 1765 |
+
},
|
| 1766 |
+
"attributes": {}
|
| 1767 |
+
}
|
| 1768 |
+
},
|
| 1769 |
+
"total_flos": 4.342179461275713e+18,
|
| 1770 |
+
"train_batch_size": 8,
|
| 1771 |
+
"trial_name": null,
|
| 1772 |
+
"trial_params": null
|
| 1773 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:03f7e64ba741ef0312bb7c236e2c187fd1f39c4a35b12a3a95d748b94a0d3a9c
|
| 3 |
+
size 5112
|