Modelin V6.2 checkpoint-2800 versiyonu yüklendi.
Browse files- config.json +34 -0
- generation_config.json +10 -0
- model.safetensors +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +11 -0
- trainer_state.json +1994 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Phi3ForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_dropout": 0.0,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"dtype": "float32",
|
| 8 |
+
"embd_pdrop": 0.0,
|
| 9 |
+
"eos_token_id": 0,
|
| 10 |
+
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 1024,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"max_position_embeddings": 1024,
|
| 15 |
+
"model_type": "phi3",
|
| 16 |
+
"num_attention_heads": 8,
|
| 17 |
+
"num_hidden_layers": 24,
|
| 18 |
+
"num_key_value_heads": 2,
|
| 19 |
+
"original_max_position_embeddings": 4096,
|
| 20 |
+
"pad_token_id": 4,
|
| 21 |
+
"partial_rotary_factor": 1.0,
|
| 22 |
+
"resid_pdrop": 0.0,
|
| 23 |
+
"rms_norm_eps": 1e-05,
|
| 24 |
+
"rope_parameters": {
|
| 25 |
+
"partial_rotary_factor": 1.0,
|
| 26 |
+
"rope_theta": 10000.0,
|
| 27 |
+
"rope_type": "default"
|
| 28 |
+
},
|
| 29 |
+
"sliding_window": null,
|
| 30 |
+
"tie_word_embeddings": false,
|
| 31 |
+
"transformers_version": "5.7.0",
|
| 32 |
+
"use_cache": false,
|
| 33 |
+
"vocab_size": 50000
|
| 34 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 0,
|
| 4 |
+
"eos_token_id": 0,
|
| 5 |
+
"output_attentions": false,
|
| 6 |
+
"output_hidden_states": false,
|
| 7 |
+
"pad_token_id": 4,
|
| 8 |
+
"transformers_version": "5.7.0",
|
| 9 |
+
"use_cache": false
|
| 10 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9083048a5056ff226dcc651d383b0985ef7fa66b68f78be63bf7da5cac2b80a2
|
| 3 |
+
size 1567445472
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"bos_token": "<|endoftext|>",
|
| 4 |
+
"eos_token": "<|endoftext|>",
|
| 5 |
+
"is_local": false,
|
| 6 |
+
"local_files_only": false,
|
| 7 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 8 |
+
"pad_token": "<|endoftext|>",
|
| 9 |
+
"tokenizer_class": "TokenizersBackend",
|
| 10 |
+
"unk_token": "<|endoftext|>"
|
| 11 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1994 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0223,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 2800,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.001,
|
| 14 |
+
"grad_norm": 0.3872889280319214,
|
| 15 |
+
"learning_rate": 3.8817000000000005e-06,
|
| 16 |
+
"loss": 1.5489984512329102,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.002,
|
| 21 |
+
"grad_norm": 0.3785003423690796,
|
| 22 |
+
"learning_rate": 8.1947e-06,
|
| 23 |
+
"loss": 1.554966926574707,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.003,
|
| 28 |
+
"grad_norm": 0.38793623447418213,
|
| 29 |
+
"learning_rate": 1.25077e-05,
|
| 30 |
+
"loss": 1.546360397338867,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.004,
|
| 35 |
+
"grad_norm": 0.4206889569759369,
|
| 36 |
+
"learning_rate": 1.68207e-05,
|
| 37 |
+
"loss": 1.606406021118164,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.005,
|
| 42 |
+
"grad_norm": 0.38157036900520325,
|
| 43 |
+
"learning_rate": 2.11337e-05,
|
| 44 |
+
"loss": 1.5404697418212892,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.006,
|
| 49 |
+
"grad_norm": 0.39413636922836304,
|
| 50 |
+
"learning_rate": 2.54467e-05,
|
| 51 |
+
"loss": 1.5285799026489257,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.007,
|
| 56 |
+
"grad_norm": 0.40694475173950195,
|
| 57 |
+
"learning_rate": 2.97597e-05,
|
| 58 |
+
"loss": 1.504967498779297,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.008,
|
| 63 |
+
"grad_norm": 0.41990774869918823,
|
| 64 |
+
"learning_rate": 3.40727e-05,
|
| 65 |
+
"loss": 1.5729640960693358,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.009,
|
| 70 |
+
"grad_norm": 0.403705358505249,
|
| 71 |
+
"learning_rate": 3.83857e-05,
|
| 72 |
+
"loss": 1.5551136970520019,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.01,
|
| 77 |
+
"grad_norm": 0.3957897424697876,
|
| 78 |
+
"learning_rate": 4.2698700000000005e-05,
|
| 79 |
+
"loss": 1.5575121879577636,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.011,
|
| 84 |
+
"grad_norm": 0.40792497992515564,
|
| 85 |
+
"learning_rate": 4.70117e-05,
|
| 86 |
+
"loss": 1.518178939819336,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.012,
|
| 91 |
+
"grad_norm": 0.42794761061668396,
|
| 92 |
+
"learning_rate": 5.13247e-05,
|
| 93 |
+
"loss": 1.5593727111816407,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.013,
|
| 98 |
+
"grad_norm": 0.4005564749240875,
|
| 99 |
+
"learning_rate": 5.5637700000000004e-05,
|
| 100 |
+
"loss": 1.562605857849121,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.014,
|
| 105 |
+
"grad_norm": 0.3942921757698059,
|
| 106 |
+
"learning_rate": 5.99507e-05,
|
| 107 |
+
"loss": 1.5700313568115234,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.015,
|
| 112 |
+
"grad_norm": 0.399198055267334,
|
| 113 |
+
"learning_rate": 6.42637e-05,
|
| 114 |
+
"loss": 1.5733401298522949,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.016,
|
| 119 |
+
"grad_norm": 0.39514100551605225,
|
| 120 |
+
"learning_rate": 6.857670000000001e-05,
|
| 121 |
+
"loss": 1.5353406906127929,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.017,
|
| 126 |
+
"grad_norm": 0.3957034945487976,
|
| 127 |
+
"learning_rate": 7.288970000000001e-05,
|
| 128 |
+
"loss": 1.5085016250610352,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.018,
|
| 133 |
+
"grad_norm": 0.4351767897605896,
|
| 134 |
+
"learning_rate": 7.72027e-05,
|
| 135 |
+
"loss": 1.5530328750610352,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.019,
|
| 140 |
+
"grad_norm": 0.418708473443985,
|
| 141 |
+
"learning_rate": 8.15157e-05,
|
| 142 |
+
"loss": 1.5638229370117187,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.02,
|
| 147 |
+
"grad_norm": 0.40256965160369873,
|
| 148 |
+
"learning_rate": 8.58287e-05,
|
| 149 |
+
"loss": 1.5796945571899415,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.021,
|
| 154 |
+
"grad_norm": 0.397319495677948,
|
| 155 |
+
"learning_rate": 8.625982049284077e-05,
|
| 156 |
+
"loss": 1.5816753387451172,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.022,
|
| 161 |
+
"grad_norm": 0.41874557733535767,
|
| 162 |
+
"learning_rate": 8.625919997618398e-05,
|
| 163 |
+
"loss": 1.5102975845336915,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.023,
|
| 168 |
+
"grad_norm": 0.40078213810920715,
|
| 169 |
+
"learning_rate": 8.625813624026713e-05,
|
| 170 |
+
"loss": 1.571668529510498,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.024,
|
| 175 |
+
"grad_norm": 0.43663620948791504,
|
| 176 |
+
"learning_rate": 8.625662929602175e-05,
|
| 177 |
+
"loss": 1.5277430534362793,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.025,
|
| 182 |
+
"grad_norm": 0.4428839385509491,
|
| 183 |
+
"learning_rate": 8.625467915893403e-05,
|
| 184 |
+
"loss": 1.5955425262451173,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.026,
|
| 189 |
+
"grad_norm": 0.4413251578807831,
|
| 190 |
+
"learning_rate": 8.625228584904462e-05,
|
| 191 |
+
"loss": 1.5842302322387696,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.027,
|
| 196 |
+
"grad_norm": 0.42099958658218384,
|
| 197 |
+
"learning_rate": 8.62494493909485e-05,
|
| 198 |
+
"loss": 1.5954334259033203,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.028,
|
| 203 |
+
"grad_norm": 0.43221530318260193,
|
| 204 |
+
"learning_rate": 8.624616981379467e-05,
|
| 205 |
+
"loss": 1.5589290618896485,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.029,
|
| 210 |
+
"grad_norm": 0.42665326595306396,
|
| 211 |
+
"learning_rate": 8.624244715128585e-05,
|
| 212 |
+
"loss": 1.5465147972106934,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.03,
|
| 217 |
+
"grad_norm": 0.42965951561927795,
|
| 218 |
+
"learning_rate": 8.623828144167817e-05,
|
| 219 |
+
"loss": 1.5350761413574219,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.031,
|
| 224 |
+
"grad_norm": 0.42243221402168274,
|
| 225 |
+
"learning_rate": 8.623367272778072e-05,
|
| 226 |
+
"loss": 1.5516321182250976,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.032,
|
| 231 |
+
"grad_norm": 0.4054717421531677,
|
| 232 |
+
"learning_rate": 8.622862105695517e-05,
|
| 233 |
+
"loss": 1.5238536834716796,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.033,
|
| 238 |
+
"grad_norm": 0.43124857544898987,
|
| 239 |
+
"learning_rate": 8.622312648111529e-05,
|
| 240 |
+
"loss": 1.5706765174865722,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.034,
|
| 245 |
+
"grad_norm": 0.4318324625492096,
|
| 246 |
+
"learning_rate": 8.621718905672629e-05,
|
| 247 |
+
"loss": 1.566463565826416,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.035,
|
| 252 |
+
"grad_norm": 0.43307042121887207,
|
| 253 |
+
"learning_rate": 8.621080884480445e-05,
|
| 254 |
+
"loss": 1.530984878540039,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.036,
|
| 259 |
+
"grad_norm": 0.43541616201400757,
|
| 260 |
+
"learning_rate": 8.620398591091628e-05,
|
| 261 |
+
"loss": 1.5489194869995118,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.037,
|
| 266 |
+
"grad_norm": 0.40485483407974243,
|
| 267 |
+
"learning_rate": 8.619672032517798e-05,
|
| 268 |
+
"loss": 1.576507568359375,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.038,
|
| 273 |
+
"grad_norm": 0.38804423809051514,
|
| 274 |
+
"learning_rate": 8.618901216225471e-05,
|
| 275 |
+
"loss": 1.5632071495056152,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.039,
|
| 280 |
+
"grad_norm": 0.44407907128334045,
|
| 281 |
+
"learning_rate": 8.618086150135976e-05,
|
| 282 |
+
"loss": 1.571460247039795,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.04,
|
| 287 |
+
"grad_norm": 0.4320124387741089,
|
| 288 |
+
"learning_rate": 8.617226842625377e-05,
|
| 289 |
+
"loss": 1.5724874496459962,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.041,
|
| 294 |
+
"grad_norm": 0.41030609607696533,
|
| 295 |
+
"learning_rate": 8.61632330252439e-05,
|
| 296 |
+
"loss": 1.552263069152832,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.042,
|
| 301 |
+
"grad_norm": 0.4280606210231781,
|
| 302 |
+
"learning_rate": 8.615375539118286e-05,
|
| 303 |
+
"loss": 1.5935471534729004,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.043,
|
| 308 |
+
"grad_norm": 0.425942063331604,
|
| 309 |
+
"learning_rate": 8.614383562146802e-05,
|
| 310 |
+
"loss": 1.618431854248047,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.044,
|
| 315 |
+
"grad_norm": 0.4112652540206909,
|
| 316 |
+
"learning_rate": 8.613347381804036e-05,
|
| 317 |
+
"loss": 1.6068254470825196,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.045,
|
| 322 |
+
"grad_norm": 0.43573281168937683,
|
| 323 |
+
"learning_rate": 8.612267008738342e-05,
|
| 324 |
+
"loss": 1.501287078857422,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.046,
|
| 329 |
+
"grad_norm": 0.43779465556144714,
|
| 330 |
+
"learning_rate": 8.611142454052226e-05,
|
| 331 |
+
"loss": 1.5798423767089844,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.047,
|
| 336 |
+
"grad_norm": 0.45562246441841125,
|
| 337 |
+
"learning_rate": 8.609973729302228e-05,
|
| 338 |
+
"loss": 1.5571372985839844,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.048,
|
| 343 |
+
"grad_norm": 0.4224706292152405,
|
| 344 |
+
"learning_rate": 8.608760846498805e-05,
|
| 345 |
+
"loss": 1.5518342971801757,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.049,
|
| 350 |
+
"grad_norm": 0.4402417242527008,
|
| 351 |
+
"learning_rate": 8.6075038181062e-05,
|
| 352 |
+
"loss": 1.5404040336608886,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.05,
|
| 357 |
+
"grad_norm": 0.42017725110054016,
|
| 358 |
+
"learning_rate": 8.606202657042327e-05,
|
| 359 |
+
"loss": 1.5324398040771485,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.051,
|
| 364 |
+
"grad_norm": 0.41633936762809753,
|
| 365 |
+
"learning_rate": 8.604857376678628e-05,
|
| 366 |
+
"loss": 1.6201072692871095,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.052,
|
| 371 |
+
"grad_norm": 0.4232465922832489,
|
| 372 |
+
"learning_rate": 8.603467990839943e-05,
|
| 373 |
+
"loss": 1.579312515258789,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.053,
|
| 378 |
+
"grad_norm": 0.4202162027359009,
|
| 379 |
+
"learning_rate": 8.602034513804359e-05,
|
| 380 |
+
"loss": 1.5241361618041993,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.054,
|
| 385 |
+
"grad_norm": 0.4149995446205139,
|
| 386 |
+
"learning_rate": 8.60055696030307e-05,
|
| 387 |
+
"loss": 1.5884086608886718,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.055,
|
| 392 |
+
"grad_norm": 0.4352942109107971,
|
| 393 |
+
"learning_rate": 8.599035345520225e-05,
|
| 394 |
+
"loss": 1.5795814514160156,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.056,
|
| 399 |
+
"grad_norm": 0.406890332698822,
|
| 400 |
+
"learning_rate": 8.59746968509277e-05,
|
| 401 |
+
"loss": 1.5849136352539062,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.057,
|
| 406 |
+
"grad_norm": 0.4194611608982086,
|
| 407 |
+
"learning_rate": 8.595859995110288e-05,
|
| 408 |
+
"loss": 1.5220673561096192,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.058,
|
| 413 |
+
"grad_norm": 0.42952999472618103,
|
| 414 |
+
"learning_rate": 8.594206292114837e-05,
|
| 415 |
+
"loss": 1.5716930389404298,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.059,
|
| 420 |
+
"grad_norm": 0.4449344575405121,
|
| 421 |
+
"learning_rate": 8.59250859310077e-05,
|
| 422 |
+
"loss": 1.5606157302856445,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.06,
|
| 427 |
+
"grad_norm": 0.4540117383003235,
|
| 428 |
+
"learning_rate": 8.590766915514571e-05,
|
| 429 |
+
"loss": 1.5590102195739746,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.061,
|
| 434 |
+
"grad_norm": 0.4081774652004242,
|
| 435 |
+
"learning_rate": 8.588981277254677e-05,
|
| 436 |
+
"loss": 1.513133430480957,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.062,
|
| 441 |
+
"grad_norm": 0.43133676052093506,
|
| 442 |
+
"learning_rate": 8.587151696671278e-05,
|
| 443 |
+
"loss": 1.5470260620117187,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.063,
|
| 448 |
+
"grad_norm": 0.42611169815063477,
|
| 449 |
+
"learning_rate": 8.585278192566148e-05,
|
| 450 |
+
"loss": 1.5632414817810059,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.064,
|
| 455 |
+
"grad_norm": 0.4463273286819458,
|
| 456 |
+
"learning_rate": 8.583360784192441e-05,
|
| 457 |
+
"loss": 1.6180675506591797,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.065,
|
| 462 |
+
"grad_norm": 0.4165816009044647,
|
| 463 |
+
"learning_rate": 8.581399491254493e-05,
|
| 464 |
+
"loss": 1.5370763778686523,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.066,
|
| 469 |
+
"grad_norm": 0.44205260276794434,
|
| 470 |
+
"learning_rate": 8.579394333907624e-05,
|
| 471 |
+
"loss": 1.558724594116211,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.067,
|
| 476 |
+
"grad_norm": 0.4382662773132324,
|
| 477 |
+
"learning_rate": 8.57734533275793e-05,
|
| 478 |
+
"loss": 1.5468637466430664,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.068,
|
| 483 |
+
"grad_norm": 0.40459638833999634,
|
| 484 |
+
"learning_rate": 8.575252508862066e-05,
|
| 485 |
+
"loss": 1.543410301208496,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.069,
|
| 490 |
+
"grad_norm": 0.42760446667671204,
|
| 491 |
+
"learning_rate": 8.573115883727036e-05,
|
| 492 |
+
"loss": 1.590751838684082,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.07,
|
| 497 |
+
"grad_norm": 0.42787182331085205,
|
| 498 |
+
"learning_rate": 8.570935479309969e-05,
|
| 499 |
+
"loss": 1.591897201538086,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.071,
|
| 504 |
+
"grad_norm": 0.4374030828475952,
|
| 505 |
+
"learning_rate": 8.568711318017893e-05,
|
| 506 |
+
"loss": 1.5722368240356446,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.072,
|
| 511 |
+
"grad_norm": 0.4523533582687378,
|
| 512 |
+
"learning_rate": 8.566443422707509e-05,
|
| 513 |
+
"loss": 1.5926239013671875,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.073,
|
| 518 |
+
"grad_norm": 0.44648537039756775,
|
| 519 |
+
"learning_rate": 8.564131816684949e-05,
|
| 520 |
+
"loss": 1.5328532218933106,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.074,
|
| 525 |
+
"grad_norm": 0.44178444147109985,
|
| 526 |
+
"learning_rate": 8.561776523705541e-05,
|
| 527 |
+
"loss": 1.5162950515747071,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.075,
|
| 532 |
+
"grad_norm": 0.459213525056839,
|
| 533 |
+
"learning_rate": 8.559377567973564e-05,
|
| 534 |
+
"loss": 1.627875518798828,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.076,
|
| 539 |
+
"grad_norm": 0.4523981511592865,
|
| 540 |
+
"learning_rate": 8.556934974142002e-05,
|
| 541 |
+
"loss": 1.5238592147827148,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.077,
|
| 546 |
+
"grad_norm": 0.4350197911262512,
|
| 547 |
+
"learning_rate": 8.554448767312285e-05,
|
| 548 |
+
"loss": 1.5533188819885253,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.078,
|
| 553 |
+
"grad_norm": 0.461459755897522,
|
| 554 |
+
"learning_rate": 8.551918973034032e-05,
|
| 555 |
+
"loss": 1.5818538665771484,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.079,
|
| 560 |
+
"grad_norm": 0.42179811000823975,
|
| 561 |
+
"learning_rate": 8.549345617304799e-05,
|
| 562 |
+
"loss": 1.5900559425354004,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.08,
|
| 567 |
+
"grad_norm": 0.4687558710575104,
|
| 568 |
+
"learning_rate": 8.546728726569792e-05,
|
| 569 |
+
"loss": 1.5136714935302735,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.081,
|
| 574 |
+
"grad_norm": 0.42299333214759827,
|
| 575 |
+
"learning_rate": 8.544068327721616e-05,
|
| 576 |
+
"loss": 1.6079557418823243,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.082,
|
| 581 |
+
"grad_norm": 0.45741724967956543,
|
| 582 |
+
"learning_rate": 8.541364448099983e-05,
|
| 583 |
+
"loss": 1.6250675201416016,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.083,
|
| 588 |
+
"grad_norm": 0.47489041090011597,
|
| 589 |
+
"learning_rate": 8.538617115491442e-05,
|
| 590 |
+
"loss": 1.5306575775146485,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.084,
|
| 595 |
+
"grad_norm": 0.43424174189567566,
|
| 596 |
+
"learning_rate": 8.535826358129085e-05,
|
| 597 |
+
"loss": 1.5837011337280273,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.085,
|
| 602 |
+
"grad_norm": 0.43130776286125183,
|
| 603 |
+
"learning_rate": 8.53299220469226e-05,
|
| 604 |
+
"loss": 1.560431957244873,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.086,
|
| 609 |
+
"grad_norm": 0.4119592607021332,
|
| 610 |
+
"learning_rate": 8.530114684306282e-05,
|
| 611 |
+
"loss": 1.572993278503418,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.087,
|
| 616 |
+
"grad_norm": 0.4472162425518036,
|
| 617 |
+
"learning_rate": 8.527193826542123e-05,
|
| 618 |
+
"loss": 1.5893125534057617,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.088,
|
| 623 |
+
"grad_norm": 0.43497446179389954,
|
| 624 |
+
"learning_rate": 8.524229661416117e-05,
|
| 625 |
+
"loss": 1.560661506652832,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.089,
|
| 630 |
+
"grad_norm": 0.43729737401008606,
|
| 631 |
+
"learning_rate": 8.521222219389646e-05,
|
| 632 |
+
"loss": 1.5510387420654297,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.09,
|
| 637 |
+
"grad_norm": 0.4537718594074249,
|
| 638 |
+
"learning_rate": 8.518171531368828e-05,
|
| 639 |
+
"loss": 1.5789440155029297,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.091,
|
| 644 |
+
"grad_norm": 0.4269010126590729,
|
| 645 |
+
"learning_rate": 8.515077628704208e-05,
|
| 646 |
+
"loss": 1.5246685028076172,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.092,
|
| 651 |
+
"grad_norm": 0.4279134273529053,
|
| 652 |
+
"learning_rate": 8.511940543190416e-05,
|
| 653 |
+
"loss": 1.5408878326416016,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.093,
|
| 658 |
+
"grad_norm": 0.4487226605415344,
|
| 659 |
+
"learning_rate": 8.508760307065864e-05,
|
| 660 |
+
"loss": 1.5889862060546875,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.094,
|
| 665 |
+
"grad_norm": 0.4346173107624054,
|
| 666 |
+
"learning_rate": 8.505536953012398e-05,
|
| 667 |
+
"loss": 1.4902814865112304,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.095,
|
| 672 |
+
"grad_norm": 0.431169718503952,
|
| 673 |
+
"learning_rate": 8.502270514154968e-05,
|
| 674 |
+
"loss": 1.5507546424865724,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.096,
|
| 679 |
+
"grad_norm": 0.46881142258644104,
|
| 680 |
+
"learning_rate": 8.498961024061287e-05,
|
| 681 |
+
"loss": 1.5701534271240234,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.097,
|
| 686 |
+
"grad_norm": 0.42858099937438965,
|
| 687 |
+
"learning_rate": 8.495608516741486e-05,
|
| 688 |
+
"loss": 1.4937341690063477,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.098,
|
| 693 |
+
"grad_norm": 0.4465695023536682,
|
| 694 |
+
"learning_rate": 8.492213026647764e-05,
|
| 695 |
+
"loss": 1.5935869216918945,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.099,
|
| 700 |
+
"grad_norm": 0.4498782157897949,
|
| 701 |
+
"learning_rate": 8.488774588674037e-05,
|
| 702 |
+
"loss": 1.5635653495788575,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.1,
|
| 707 |
+
"grad_norm": 0.4431057274341583,
|
| 708 |
+
"learning_rate": 8.485293238155573e-05,
|
| 709 |
+
"loss": 1.5423452377319335,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.101,
|
| 714 |
+
"grad_norm": 0.4575574994087219,
|
| 715 |
+
"learning_rate": 8.481769010868637e-05,
|
| 716 |
+
"loss": 1.5350657463073731,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.102,
|
| 721 |
+
"grad_norm": 0.4342168867588043,
|
| 722 |
+
"learning_rate": 8.478201943030117e-05,
|
| 723 |
+
"loss": 1.5452421188354493,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.103,
|
| 728 |
+
"grad_norm": 0.445075124502182,
|
| 729 |
+
"learning_rate": 8.474592071297152e-05,
|
| 730 |
+
"loss": 1.554254150390625,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.104,
|
| 735 |
+
"grad_norm": 0.4207341969013214,
|
| 736 |
+
"learning_rate": 8.470939432766762e-05,
|
| 737 |
+
"loss": 1.5895505905151368,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.105,
|
| 742 |
+
"grad_norm": 0.4459976851940155,
|
| 743 |
+
"learning_rate": 8.467244064975459e-05,
|
| 744 |
+
"loss": 1.546200942993164,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.106,
|
| 749 |
+
"grad_norm": 0.4187023341655731,
|
| 750 |
+
"learning_rate": 8.463506005898863e-05,
|
| 751 |
+
"loss": 1.5332408905029298,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.107,
|
| 756 |
+
"grad_norm": 0.4594968557357788,
|
| 757 |
+
"learning_rate": 8.459725293951314e-05,
|
| 758 |
+
"loss": 1.5771045684814453,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.108,
|
| 763 |
+
"grad_norm": 0.4230520725250244,
|
| 764 |
+
"learning_rate": 8.455901967985479e-05,
|
| 765 |
+
"loss": 1.5950983047485352,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.109,
|
| 770 |
+
"grad_norm": 0.43380364775657654,
|
| 771 |
+
"learning_rate": 8.452036067291948e-05,
|
| 772 |
+
"loss": 1.6060211181640625,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.11,
|
| 777 |
+
"grad_norm": 0.4305785596370697,
|
| 778 |
+
"learning_rate": 8.448127631598831e-05,
|
| 779 |
+
"loss": 1.5464977264404296,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.111,
|
| 784 |
+
"grad_norm": 0.46058452129364014,
|
| 785 |
+
"learning_rate": 8.44417670107135e-05,
|
| 786 |
+
"loss": 1.5613780975341798,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.112,
|
| 791 |
+
"grad_norm": 0.42492201924324036,
|
| 792 |
+
"learning_rate": 8.440183316311433e-05,
|
| 793 |
+
"loss": 1.5371337890625,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.113,
|
| 798 |
+
"grad_norm": 0.45703360438346863,
|
| 799 |
+
"learning_rate": 8.436147518357284e-05,
|
| 800 |
+
"loss": 1.5895401000976563,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.114,
|
| 805 |
+
"grad_norm": 0.4336182773113251,
|
| 806 |
+
"learning_rate": 8.432069348682973e-05,
|
| 807 |
+
"loss": 1.5472881317138671,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.115,
|
| 812 |
+
"grad_norm": 0.44153451919555664,
|
| 813 |
+
"learning_rate": 8.427948849198002e-05,
|
| 814 |
+
"loss": 1.5661738395690918,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.116,
|
| 819 |
+
"grad_norm": 0.43905603885650635,
|
| 820 |
+
"learning_rate": 8.42378606224688e-05,
|
| 821 |
+
"loss": 1.5574228286743164,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.117,
|
| 826 |
+
"grad_norm": 0.43091803789138794,
|
| 827 |
+
"learning_rate": 8.419581030608683e-05,
|
| 828 |
+
"loss": 1.5603097915649413,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.118,
|
| 833 |
+
"grad_norm": 0.4270052909851074,
|
| 834 |
+
"learning_rate": 8.415333797496616e-05,
|
| 835 |
+
"loss": 1.552748680114746,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.119,
|
| 840 |
+
"grad_norm": 0.4377681016921997,
|
| 841 |
+
"learning_rate": 8.411044406557571e-05,
|
| 842 |
+
"loss": 1.5209136962890626,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.12,
|
| 847 |
+
"grad_norm": 0.4468444585800171,
|
| 848 |
+
"learning_rate": 8.406712901871679e-05,
|
| 849 |
+
"loss": 1.5935579299926759,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.121,
|
| 854 |
+
"grad_norm": 0.40383124351501465,
|
| 855 |
+
"learning_rate": 8.402339327951851e-05,
|
| 856 |
+
"loss": 1.5492631912231445,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.122,
|
| 861 |
+
"grad_norm": 0.453901469707489,
|
| 862 |
+
"learning_rate": 8.397923729743324e-05,
|
| 863 |
+
"loss": 1.5286017417907716,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.123,
|
| 868 |
+
"grad_norm": 0.45355984568595886,
|
| 869 |
+
"learning_rate": 8.393466152623203e-05,
|
| 870 |
+
"loss": 1.534010887145996,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.124,
|
| 875 |
+
"grad_norm": 0.41849273443222046,
|
| 876 |
+
"learning_rate": 8.388966642399991e-05,
|
| 877 |
+
"loss": 1.5546277999877929,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.125,
|
| 882 |
+
"grad_norm": 0.42275470495224,
|
| 883 |
+
"learning_rate": 8.384425245313115e-05,
|
| 884 |
+
"loss": 1.5799736976623535,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.126,
|
| 889 |
+
"grad_norm": 0.44112563133239746,
|
| 890 |
+
"learning_rate": 8.379842008032457e-05,
|
| 891 |
+
"loss": 1.5725579261779785,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.127,
|
| 896 |
+
"grad_norm": 0.42606085538864136,
|
| 897 |
+
"learning_rate": 8.375216977657873e-05,
|
| 898 |
+
"loss": 1.534956932067871,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.128,
|
| 903 |
+
"grad_norm": 0.44213414192199707,
|
| 904 |
+
"learning_rate": 8.370550201718704e-05,
|
| 905 |
+
"loss": 1.496871566772461,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.129,
|
| 910 |
+
"grad_norm": 0.47176358103752136,
|
| 911 |
+
"learning_rate": 8.365841728173296e-05,
|
| 912 |
+
"loss": 1.5814940452575683,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.13,
|
| 917 |
+
"grad_norm": 0.4457390606403351,
|
| 918 |
+
"learning_rate": 8.361091605408497e-05,
|
| 919 |
+
"loss": 1.599702739715576,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.131,
|
| 924 |
+
"grad_norm": 0.43326595425605774,
|
| 925 |
+
"learning_rate": 8.356299882239168e-05,
|
| 926 |
+
"loss": 1.5610076904296875,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.132,
|
| 931 |
+
"grad_norm": 0.43634170293807983,
|
| 932 |
+
"learning_rate": 8.35146660790768e-05,
|
| 933 |
+
"loss": 1.5391773223876952,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.133,
|
| 938 |
+
"grad_norm": 0.4215056002140045,
|
| 939 |
+
"learning_rate": 8.346591832083405e-05,
|
| 940 |
+
"loss": 1.538853359222412,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.134,
|
| 945 |
+
"grad_norm": 0.4359005093574524,
|
| 946 |
+
"learning_rate": 8.341675604862204e-05,
|
| 947 |
+
"loss": 1.5893823623657226,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.135,
|
| 952 |
+
"grad_norm": 0.4620271623134613,
|
| 953 |
+
"learning_rate": 8.336717976765922e-05,
|
| 954 |
+
"loss": 1.544410514831543,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.136,
|
| 959 |
+
"grad_norm": 0.45624276995658875,
|
| 960 |
+
"learning_rate": 8.331718998741857e-05,
|
| 961 |
+
"loss": 1.5727739334106445,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.137,
|
| 966 |
+
"grad_norm": 0.47488826513290405,
|
| 967 |
+
"learning_rate": 8.32667872216224e-05,
|
| 968 |
+
"loss": 1.5734460830688477,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.138,
|
| 973 |
+
"grad_norm": 0.4481564462184906,
|
| 974 |
+
"learning_rate": 8.321597198823715e-05,
|
| 975 |
+
"loss": 1.5132453918457032,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.139,
|
| 980 |
+
"grad_norm": 0.44024789333343506,
|
| 981 |
+
"learning_rate": 8.316474480946796e-05,
|
| 982 |
+
"loss": 1.5181768417358399,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.14,
|
| 987 |
+
"grad_norm": 0.43633797764778137,
|
| 988 |
+
"learning_rate": 8.311310621175332e-05,
|
| 989 |
+
"loss": 1.5812137603759766,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.141,
|
| 994 |
+
"grad_norm": 0.4629945158958435,
|
| 995 |
+
"learning_rate": 8.306105672575975e-05,
|
| 996 |
+
"loss": 1.5248623847961427,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.142,
|
| 1001 |
+
"grad_norm": 0.44547826051712036,
|
| 1002 |
+
"learning_rate": 8.300859688637624e-05,
|
| 1003 |
+
"loss": 1.603738021850586,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.143,
|
| 1008 |
+
"grad_norm": 0.4610430896282196,
|
| 1009 |
+
"learning_rate": 8.295572723270881e-05,
|
| 1010 |
+
"loss": 1.5599722862243652,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.144,
|
| 1015 |
+
"grad_norm": 0.4239096939563751,
|
| 1016 |
+
"learning_rate": 8.290244830807493e-05,
|
| 1017 |
+
"loss": 1.5240245819091798,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.145,
|
| 1022 |
+
"grad_norm": 0.4612870216369629,
|
| 1023 |
+
"learning_rate": 8.284876065999803e-05,
|
| 1024 |
+
"loss": 1.5548505783081055,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.146,
|
| 1029 |
+
"grad_norm": 0.437900185585022,
|
| 1030 |
+
"learning_rate": 8.279466484020172e-05,
|
| 1031 |
+
"loss": 1.5739557266235351,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.147,
|
| 1036 |
+
"grad_norm": 0.459146648645401,
|
| 1037 |
+
"learning_rate": 8.274016140460427e-05,
|
| 1038 |
+
"loss": 1.5370497703552246,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.148,
|
| 1043 |
+
"grad_norm": 0.4295809268951416,
|
| 1044 |
+
"learning_rate": 8.268525091331279e-05,
|
| 1045 |
+
"loss": 1.5780242919921874,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.149,
|
| 1050 |
+
"grad_norm": 0.43193328380584717,
|
| 1051 |
+
"learning_rate": 8.262993393061756e-05,
|
| 1052 |
+
"loss": 1.5061514854431153,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.15,
|
| 1057 |
+
"grad_norm": 0.44761526584625244,
|
| 1058 |
+
"learning_rate": 8.257421102498613e-05,
|
| 1059 |
+
"loss": 1.5490300178527832,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.151,
|
| 1064 |
+
"grad_norm": 0.4909959137439728,
|
| 1065 |
+
"learning_rate": 8.251808276905761e-05,
|
| 1066 |
+
"loss": 1.5348563194274902,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.152,
|
| 1071 |
+
"grad_norm": 0.45388928055763245,
|
| 1072 |
+
"learning_rate": 8.246154973963663e-05,
|
| 1073 |
+
"loss": 1.5422224044799804,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.153,
|
| 1078 |
+
"grad_norm": 0.44226154685020447,
|
| 1079 |
+
"learning_rate": 8.240461251768759e-05,
|
| 1080 |
+
"loss": 1.5589332580566406,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.154,
|
| 1085 |
+
"grad_norm": 0.45566076040267944,
|
| 1086 |
+
"learning_rate": 8.23472716883285e-05,
|
| 1087 |
+
"loss": 1.55558500289917,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.155,
|
| 1092 |
+
"grad_norm": 0.45849281549453735,
|
| 1093 |
+
"learning_rate": 8.228952784082511e-05,
|
| 1094 |
+
"loss": 1.60096435546875,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.156,
|
| 1099 |
+
"grad_norm": 0.4728466272354126,
|
| 1100 |
+
"learning_rate": 8.22313815685848e-05,
|
| 1101 |
+
"loss": 1.594934844970703,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.157,
|
| 1106 |
+
"grad_norm": 0.4469038248062134,
|
| 1107 |
+
"learning_rate": 8.217283346915048e-05,
|
| 1108 |
+
"loss": 1.553289031982422,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.158,
|
| 1113 |
+
"grad_norm": 0.48276886343955994,
|
| 1114 |
+
"learning_rate": 8.211388414419445e-05,
|
| 1115 |
+
"loss": 1.503675937652588,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.159,
|
| 1120 |
+
"grad_norm": 0.4695449769496918,
|
| 1121 |
+
"learning_rate": 8.205453419951223e-05,
|
| 1122 |
+
"loss": 1.5649538040161133,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.16,
|
| 1127 |
+
"grad_norm": 0.43924853205680847,
|
| 1128 |
+
"learning_rate": 8.199478424501635e-05,
|
| 1129 |
+
"loss": 1.527259635925293,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.161,
|
| 1134 |
+
"grad_norm": 0.4189310073852539,
|
| 1135 |
+
"learning_rate": 8.193463489473004e-05,
|
| 1136 |
+
"loss": 1.55230655670166,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.162,
|
| 1141 |
+
"grad_norm": 0.44848498702049255,
|
| 1142 |
+
"learning_rate": 8.187408676678092e-05,
|
| 1143 |
+
"loss": 1.5449299812316895,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.163,
|
| 1148 |
+
"grad_norm": 0.4382420480251312,
|
| 1149 |
+
"learning_rate": 8.181314048339468e-05,
|
| 1150 |
+
"loss": 1.541883659362793,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.164,
|
| 1155 |
+
"grad_norm": 0.42705485224723816,
|
| 1156 |
+
"learning_rate": 8.175179667088869e-05,
|
| 1157 |
+
"loss": 1.5482391357421874,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.165,
|
| 1162 |
+
"grad_norm": 0.4264799654483795,
|
| 1163 |
+
"learning_rate": 8.169005595966554e-05,
|
| 1164 |
+
"loss": 1.562957191467285,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.166,
|
| 1169 |
+
"grad_norm": 0.46075376868247986,
|
| 1170 |
+
"learning_rate": 8.162791898420655e-05,
|
| 1171 |
+
"loss": 1.533614730834961,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.167,
|
| 1176 |
+
"grad_norm": 0.46955326199531555,
|
| 1177 |
+
"learning_rate": 8.156538638306531e-05,
|
| 1178 |
+
"loss": 1.5150230407714844,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.168,
|
| 1183 |
+
"grad_norm": 0.4578472077846527,
|
| 1184 |
+
"learning_rate": 8.150245879886102e-05,
|
| 1185 |
+
"loss": 1.579137420654297,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.169,
|
| 1190 |
+
"grad_norm": 0.42824026942253113,
|
| 1191 |
+
"learning_rate": 8.143913687827199e-05,
|
| 1192 |
+
"loss": 1.5843921661376954,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.17,
|
| 1197 |
+
"grad_norm": 0.43932676315307617,
|
| 1198 |
+
"learning_rate": 8.137542127202888e-05,
|
| 1199 |
+
"loss": 1.605410385131836,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.171,
|
| 1204 |
+
"grad_norm": 0.4417635500431061,
|
| 1205 |
+
"learning_rate": 8.131131263490822e-05,
|
| 1206 |
+
"loss": 1.5659252166748048,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.172,
|
| 1211 |
+
"grad_norm": 0.448393315076828,
|
| 1212 |
+
"learning_rate": 8.124681162572536e-05,
|
| 1213 |
+
"loss": 1.5661985397338867,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.173,
|
| 1218 |
+
"grad_norm": 0.4424435496330261,
|
| 1219 |
+
"learning_rate": 8.118191890732799e-05,
|
| 1220 |
+
"loss": 1.5341227531433106,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.174,
|
| 1225 |
+
"grad_norm": 0.4247022867202759,
|
| 1226 |
+
"learning_rate": 8.111663514658923e-05,
|
| 1227 |
+
"loss": 1.5798539161682128,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.175,
|
| 1232 |
+
"grad_norm": 0.4436150789260864,
|
| 1233 |
+
"learning_rate": 8.105096101440067e-05,
|
| 1234 |
+
"loss": 1.5780141830444336,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.176,
|
| 1239 |
+
"grad_norm": 0.4493919312953949,
|
| 1240 |
+
"learning_rate": 8.098489718566567e-05,
|
| 1241 |
+
"loss": 1.5496795654296875,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.177,
|
| 1246 |
+
"grad_norm": 0.44591131806373596,
|
| 1247 |
+
"learning_rate": 8.091844433929229e-05,
|
| 1248 |
+
"loss": 1.5309679985046387,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.178,
|
| 1253 |
+
"grad_norm": 0.44779840111732483,
|
| 1254 |
+
"learning_rate": 8.085160315818632e-05,
|
| 1255 |
+
"loss": 1.5156972885131836,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.179,
|
| 1260 |
+
"grad_norm": 0.44151243567466736,
|
| 1261 |
+
"learning_rate": 8.078437432924434e-05,
|
| 1262 |
+
"loss": 1.5683908462524414,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.18,
|
| 1267 |
+
"grad_norm": 0.43576961755752563,
|
| 1268 |
+
"learning_rate": 8.071675854334655e-05,
|
| 1269 |
+
"loss": 1.479705810546875,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.181,
|
| 1274 |
+
"grad_norm": 0.4513467848300934,
|
| 1275 |
+
"learning_rate": 8.06487564953498e-05,
|
| 1276 |
+
"loss": 1.553599739074707,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.182,
|
| 1281 |
+
"grad_norm": 0.4427294433116913,
|
| 1282 |
+
"learning_rate": 8.058036888408034e-05,
|
| 1283 |
+
"loss": 1.5128995895385742,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.183,
|
| 1288 |
+
"grad_norm": 0.458186537027359,
|
| 1289 |
+
"learning_rate": 8.051159641232666e-05,
|
| 1290 |
+
"loss": 1.5448317527770996,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.184,
|
| 1295 |
+
"grad_norm": 0.4836799204349518,
|
| 1296 |
+
"learning_rate": 8.044243978683234e-05,
|
| 1297 |
+
"loss": 1.5993682861328125,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.185,
|
| 1302 |
+
"grad_norm": 0.44361555576324463,
|
| 1303 |
+
"learning_rate": 8.03728997182887e-05,
|
| 1304 |
+
"loss": 1.5255595207214356,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.186,
|
| 1309 |
+
"grad_norm": 0.41420966386795044,
|
| 1310 |
+
"learning_rate": 8.030297692132756e-05,
|
| 1311 |
+
"loss": 1.582503890991211,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.187,
|
| 1316 |
+
"grad_norm": 0.44590333104133606,
|
| 1317 |
+
"learning_rate": 8.023267211451381e-05,
|
| 1318 |
+
"loss": 1.570404624938965,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.188,
|
| 1323 |
+
"grad_norm": 0.43962764739990234,
|
| 1324 |
+
"learning_rate": 8.016198602033813e-05,
|
| 1325 |
+
"loss": 1.5464605331420898,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.189,
|
| 1330 |
+
"grad_norm": 0.4597555994987488,
|
| 1331 |
+
"learning_rate": 8.009091936520949e-05,
|
| 1332 |
+
"loss": 1.4956825256347657,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.19,
|
| 1337 |
+
"grad_norm": 0.4748958349227905,
|
| 1338 |
+
"learning_rate": 8.001947287944775e-05,
|
| 1339 |
+
"loss": 1.5619684219360352,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.191,
|
| 1344 |
+
"grad_norm": 0.44536739587783813,
|
| 1345 |
+
"learning_rate": 7.99476472972761e-05,
|
| 1346 |
+
"loss": 1.5296592712402344,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.192,
|
| 1351 |
+
"grad_norm": 0.43877002596855164,
|
| 1352 |
+
"learning_rate": 7.98754433568135e-05,
|
| 1353 |
+
"loss": 1.531735324859619,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.193,
|
| 1358 |
+
"grad_norm": 0.4474166929721832,
|
| 1359 |
+
"learning_rate": 7.980286180006717e-05,
|
| 1360 |
+
"loss": 1.5319225311279296,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.194,
|
| 1365 |
+
"grad_norm": 0.4478398859500885,
|
| 1366 |
+
"learning_rate": 7.972990337292492e-05,
|
| 1367 |
+
"loss": 1.598667812347412,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.195,
|
| 1372 |
+
"grad_norm": 0.4543236196041107,
|
| 1373 |
+
"learning_rate": 7.965656882514742e-05,
|
| 1374 |
+
"loss": 1.5022565841674804,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.196,
|
| 1379 |
+
"grad_norm": 0.4551485776901245,
|
| 1380 |
+
"learning_rate": 7.958285891036068e-05,
|
| 1381 |
+
"loss": 1.4970108032226563,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.197,
|
| 1386 |
+
"grad_norm": 0.4381597638130188,
|
| 1387 |
+
"learning_rate": 7.950877438604807e-05,
|
| 1388 |
+
"loss": 1.5610149383544922,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.198,
|
| 1393 |
+
"grad_norm": 0.47578102350234985,
|
| 1394 |
+
"learning_rate": 7.943431601354271e-05,
|
| 1395 |
+
"loss": 1.4768410682678224,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.199,
|
| 1400 |
+
"grad_norm": 0.43383491039276123,
|
| 1401 |
+
"learning_rate": 7.93594845580196e-05,
|
| 1402 |
+
"loss": 1.615484619140625,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.2,
|
| 1407 |
+
"grad_norm": 0.4749927520751953,
|
| 1408 |
+
"learning_rate": 7.928428078848771e-05,
|
| 1409 |
+
"loss": 1.5539310455322266,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.201,
|
| 1414 |
+
"grad_norm": 0.4608052372932434,
|
| 1415 |
+
"learning_rate": 7.920870547778215e-05,
|
| 1416 |
+
"loss": 1.5318676948547363,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.202,
|
| 1421 |
+
"grad_norm": 0.45553329586982727,
|
| 1422 |
+
"learning_rate": 7.913275940255613e-05,
|
| 1423 |
+
"loss": 1.5542593002319336,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.203,
|
| 1428 |
+
"grad_norm": 0.4951193630695343,
|
| 1429 |
+
"learning_rate": 7.905644334327311e-05,
|
| 1430 |
+
"loss": 1.5068115234375,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.204,
|
| 1435 |
+
"grad_norm": 0.4648756980895996,
|
| 1436 |
+
"learning_rate": 7.897975808419868e-05,
|
| 1437 |
+
"loss": 1.5019064903259278,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.205,
|
| 1442 |
+
"grad_norm": 0.43299543857574463,
|
| 1443 |
+
"learning_rate": 7.890270441339251e-05,
|
| 1444 |
+
"loss": 1.536239242553711,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.206,
|
| 1449 |
+
"grad_norm": 0.4560064971446991,
|
| 1450 |
+
"learning_rate": 7.882528312270028e-05,
|
| 1451 |
+
"loss": 1.5249292373657226,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.207,
|
| 1456 |
+
"grad_norm": 0.4477216601371765,
|
| 1457 |
+
"learning_rate": 7.874749500774555e-05,
|
| 1458 |
+
"loss": 1.515114974975586,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.208,
|
| 1463 |
+
"grad_norm": 0.4781554341316223,
|
| 1464 |
+
"learning_rate": 7.866934086792156e-05,
|
| 1465 |
+
"loss": 1.5463275909423828,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.209,
|
| 1470 |
+
"grad_norm": 0.45784494280815125,
|
| 1471 |
+
"learning_rate": 7.859082150638303e-05,
|
| 1472 |
+
"loss": 1.5834733963012695,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.21,
|
| 1477 |
+
"grad_norm": 0.41224566102027893,
|
| 1478 |
+
"learning_rate": 7.851193773003785e-05,
|
| 1479 |
+
"loss": 1.5228580474853515,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.211,
|
| 1484 |
+
"grad_norm": 0.4573986530303955,
|
| 1485 |
+
"learning_rate": 7.84326903495389e-05,
|
| 1486 |
+
"loss": 1.4756481170654296,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.212,
|
| 1491 |
+
"grad_norm": 0.477107971906662,
|
| 1492 |
+
"learning_rate": 7.835308017927563e-05,
|
| 1493 |
+
"loss": 1.563486671447754,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.213,
|
| 1498 |
+
"grad_norm": 0.4521276652812958,
|
| 1499 |
+
"learning_rate": 7.82731080373657e-05,
|
| 1500 |
+
"loss": 1.5458997726440429,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.214,
|
| 1505 |
+
"grad_norm": 0.4387427568435669,
|
| 1506 |
+
"learning_rate": 7.819277474564661e-05,
|
| 1507 |
+
"loss": 1.5429878234863281,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.215,
|
| 1512 |
+
"grad_norm": 0.45496708154678345,
|
| 1513 |
+
"learning_rate": 7.811208112966722e-05,
|
| 1514 |
+
"loss": 1.5660066604614258,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.216,
|
| 1519 |
+
"grad_norm": 0.5154712796211243,
|
| 1520 |
+
"learning_rate": 7.803102801867928e-05,
|
| 1521 |
+
"loss": 1.5109113693237304,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.217,
|
| 1526 |
+
"grad_norm": 0.4356984496116638,
|
| 1527 |
+
"learning_rate": 7.794961624562892e-05,
|
| 1528 |
+
"loss": 1.5474672317504883,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.218,
|
| 1533 |
+
"grad_norm": 0.4773666560649872,
|
| 1534 |
+
"learning_rate": 7.786784664714808e-05,
|
| 1535 |
+
"loss": 1.52636775970459,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.219,
|
| 1540 |
+
"grad_norm": 0.4520673453807831,
|
| 1541 |
+
"learning_rate": 7.77857200635459e-05,
|
| 1542 |
+
"loss": 1.4999986648559571,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.22,
|
| 1547 |
+
"grad_norm": 0.45495954155921936,
|
| 1548 |
+
"learning_rate": 7.770323733880007e-05,
|
| 1549 |
+
"loss": 1.5395520210266114,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.221,
|
| 1554 |
+
"grad_norm": 0.46883487701416016,
|
| 1555 |
+
"learning_rate": 7.762039932054823e-05,
|
| 1556 |
+
"loss": 1.553999900817871,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.222,
|
| 1561 |
+
"grad_norm": 0.4731461703777313,
|
| 1562 |
+
"learning_rate": 7.753720686007916e-05,
|
| 1563 |
+
"loss": 1.5087655067443848,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.223,
|
| 1568 |
+
"grad_norm": 0.43786755204200745,
|
| 1569 |
+
"learning_rate": 7.745366081232414e-05,
|
| 1570 |
+
"loss": 1.5776697158813477,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.224,
|
| 1575 |
+
"grad_norm": 0.44203466176986694,
|
| 1576 |
+
"learning_rate": 7.736976203584805e-05,
|
| 1577 |
+
"loss": 1.5327131271362304,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.225,
|
| 1582 |
+
"grad_norm": 0.4511767625808716,
|
| 1583 |
+
"learning_rate": 7.728551139284063e-05,
|
| 1584 |
+
"loss": 1.5300714492797851,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.226,
|
| 1589 |
+
"grad_norm": 0.4910801649093628,
|
| 1590 |
+
"learning_rate": 7.720090974910761e-05,
|
| 1591 |
+
"loss": 1.4961822509765625,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.227,
|
| 1596 |
+
"grad_norm": 0.46052250266075134,
|
| 1597 |
+
"learning_rate": 7.711595797406177e-05,
|
| 1598 |
+
"loss": 1.5509103775024413,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.228,
|
| 1603 |
+
"grad_norm": 0.45214343070983887,
|
| 1604 |
+
"learning_rate": 7.703065694071401e-05,
|
| 1605 |
+
"loss": 1.528483772277832,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.229,
|
| 1610 |
+
"grad_norm": 0.46709343791007996,
|
| 1611 |
+
"learning_rate": 7.694500752566447e-05,
|
| 1612 |
+
"loss": 1.5288267135620117,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.23,
|
| 1617 |
+
"grad_norm": 0.45310071110725403,
|
| 1618 |
+
"learning_rate": 7.685901060909339e-05,
|
| 1619 |
+
"loss": 1.5159454345703125,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.231,
|
| 1624 |
+
"grad_norm": 0.4652255177497864,
|
| 1625 |
+
"learning_rate": 7.677266707475213e-05,
|
| 1626 |
+
"loss": 1.5301589012145995,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.232,
|
| 1631 |
+
"grad_norm": 0.45159515738487244,
|
| 1632 |
+
"learning_rate": 7.66859778099541e-05,
|
| 1633 |
+
"loss": 1.5472338676452637,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.233,
|
| 1638 |
+
"grad_norm": 0.4177556335926056,
|
| 1639 |
+
"learning_rate": 7.659894370556565e-05,
|
| 1640 |
+
"loss": 1.519093132019043,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.234,
|
| 1645 |
+
"grad_norm": 0.429817795753479,
|
| 1646 |
+
"learning_rate": 7.651156565599683e-05,
|
| 1647 |
+
"loss": 1.497427272796631,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.235,
|
| 1652 |
+
"grad_norm": 0.43000954389572144,
|
| 1653 |
+
"learning_rate": 7.642384455919233e-05,
|
| 1654 |
+
"loss": 1.4777793884277344,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.236,
|
| 1659 |
+
"grad_norm": 0.43708905577659607,
|
| 1660 |
+
"learning_rate": 7.63357813166221e-05,
|
| 1661 |
+
"loss": 1.5647817611694337,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.237,
|
| 1666 |
+
"grad_norm": 0.45687180757522583,
|
| 1667 |
+
"learning_rate": 7.624737683327225e-05,
|
| 1668 |
+
"loss": 1.4947479248046875,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.238,
|
| 1673 |
+
"grad_norm": 0.4629661440849304,
|
| 1674 |
+
"learning_rate": 7.615863201763559e-05,
|
| 1675 |
+
"loss": 1.5427149772644042,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.239,
|
| 1680 |
+
"grad_norm": 0.46899276971817017,
|
| 1681 |
+
"learning_rate": 7.606954778170244e-05,
|
| 1682 |
+
"loss": 1.5839698791503907,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.24,
|
| 1687 |
+
"grad_norm": 0.4606948494911194,
|
| 1688 |
+
"learning_rate": 7.598012504095115e-05,
|
| 1689 |
+
"loss": 1.541771697998047,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.241,
|
| 1694 |
+
"grad_norm": 0.45212459564208984,
|
| 1695 |
+
"learning_rate": 7.58903647143387e-05,
|
| 1696 |
+
"loss": 1.5472356796264648,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.242,
|
| 1701 |
+
"grad_norm": 0.45408064126968384,
|
| 1702 |
+
"learning_rate": 7.58002677242914e-05,
|
| 1703 |
+
"loss": 1.5932246208190919,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.243,
|
| 1708 |
+
"grad_norm": 0.48358407616615295,
|
| 1709 |
+
"learning_rate": 7.570983499669515e-05,
|
| 1710 |
+
"loss": 1.5044713973999024,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.244,
|
| 1715 |
+
"grad_norm": 0.44570690393447876,
|
| 1716 |
+
"learning_rate": 7.561906746088618e-05,
|
| 1717 |
+
"loss": 1.5048210144042968,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.245,
|
| 1722 |
+
"grad_norm": 0.4519187808036804,
|
| 1723 |
+
"learning_rate": 7.552796604964136e-05,
|
| 1724 |
+
"loss": 1.523080539703369,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.246,
|
| 1729 |
+
"grad_norm": 0.44542405009269714,
|
| 1730 |
+
"learning_rate": 7.543653169916864e-05,
|
| 1731 |
+
"loss": 1.5862494468688966,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.247,
|
| 1736 |
+
"grad_norm": 0.43162643909454346,
|
| 1737 |
+
"learning_rate": 7.534476534909743e-05,
|
| 1738 |
+
"loss": 1.553419303894043,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.248,
|
| 1743 |
+
"grad_norm": 0.4649028778076172,
|
| 1744 |
+
"learning_rate": 7.525266794246896e-05,
|
| 1745 |
+
"loss": 1.549135398864746,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.249,
|
| 1750 |
+
"grad_norm": 0.454800009727478,
|
| 1751 |
+
"learning_rate": 7.51602404257266e-05,
|
| 1752 |
+
"loss": 1.5687236785888672,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.25,
|
| 1757 |
+
"grad_norm": 0.4387585520744324,
|
| 1758 |
+
"learning_rate": 7.506748374870607e-05,
|
| 1759 |
+
"loss": 1.4585626602172852,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.251,
|
| 1764 |
+
"grad_norm": 0.44236883521080017,
|
| 1765 |
+
"learning_rate": 7.497439886462573e-05,
|
| 1766 |
+
"loss": 1.5203914642333984,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.252,
|
| 1771 |
+
"grad_norm": 0.4657570719718933,
|
| 1772 |
+
"learning_rate": 7.488098673007681e-05,
|
| 1773 |
+
"loss": 1.5481887817382813,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.253,
|
| 1778 |
+
"grad_norm": 0.4656415283679962,
|
| 1779 |
+
"learning_rate": 7.478724830501353e-05,
|
| 1780 |
+
"loss": 1.5086076736450196,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.254,
|
| 1785 |
+
"grad_norm": 0.4566405713558197,
|
| 1786 |
+
"learning_rate": 7.469318455274322e-05,
|
| 1787 |
+
"loss": 1.497421932220459,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.255,
|
| 1792 |
+
"grad_norm": 0.4292711019515991,
|
| 1793 |
+
"learning_rate": 7.459879643991647e-05,
|
| 1794 |
+
"loss": 1.5872754096984862,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.256,
|
| 1799 |
+
"grad_norm": 0.4766315519809723,
|
| 1800 |
+
"learning_rate": 7.450408493651717e-05,
|
| 1801 |
+
"loss": 1.4764188766479491,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.257,
|
| 1806 |
+
"grad_norm": 0.4654238820075989,
|
| 1807 |
+
"learning_rate": 7.440905101585258e-05,
|
| 1808 |
+
"loss": 1.5425111770629882,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 1.0003,
|
| 1813 |
+
"grad_norm": 0.42733854055404663,
|
| 1814 |
+
"learning_rate": 7.431369565454328e-05,
|
| 1815 |
+
"loss": 1.7178598403930665,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 1.0013,
|
| 1820 |
+
"grad_norm": 0.4512520730495453,
|
| 1821 |
+
"learning_rate": 7.421801983251313e-05,
|
| 1822 |
+
"loss": 1.434187126159668,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 1.0023,
|
| 1827 |
+
"grad_norm": 0.45792531967163086,
|
| 1828 |
+
"learning_rate": 7.412202453297922e-05,
|
| 1829 |
+
"loss": 1.5031817436218262,
|
| 1830 |
+
"step": 2600
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 1.0033,
|
| 1834 |
+
"grad_norm": 0.4429328739643097,
|
| 1835 |
+
"learning_rate": 7.402571074244179e-05,
|
| 1836 |
+
"loss": 1.4843154907226563,
|
| 1837 |
+
"step": 2610
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 1.0043,
|
| 1841 |
+
"grad_norm": 0.43615370988845825,
|
| 1842 |
+
"learning_rate": 7.392907945067407e-05,
|
| 1843 |
+
"loss": 1.4864605903625487,
|
| 1844 |
+
"step": 2620
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 1.0053,
|
| 1848 |
+
"grad_norm": 0.469474196434021,
|
| 1849 |
+
"learning_rate": 7.383213165071205e-05,
|
| 1850 |
+
"loss": 1.4709732055664062,
|
| 1851 |
+
"step": 2630
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 1.0063,
|
| 1855 |
+
"grad_norm": 0.45088157057762146,
|
| 1856 |
+
"learning_rate": 7.373486833884443e-05,
|
| 1857 |
+
"loss": 1.4104438781738282,
|
| 1858 |
+
"step": 2640
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 1.0073,
|
| 1862 |
+
"grad_norm": 0.458600252866745,
|
| 1863 |
+
"learning_rate": 7.363729051460219e-05,
|
| 1864 |
+
"loss": 1.4738828659057617,
|
| 1865 |
+
"step": 2650
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 1.0083,
|
| 1869 |
+
"grad_norm": 0.46691879630088806,
|
| 1870 |
+
"learning_rate": 7.353939918074846e-05,
|
| 1871 |
+
"loss": 1.4707263946533202,
|
| 1872 |
+
"step": 2660
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 1.0093,
|
| 1876 |
+
"grad_norm": 0.46489620208740234,
|
| 1877 |
+
"learning_rate": 7.344119534326815e-05,
|
| 1878 |
+
"loss": 1.3846729278564454,
|
| 1879 |
+
"step": 2670
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 1.0103,
|
| 1883 |
+
"grad_norm": 0.44752174615859985,
|
| 1884 |
+
"learning_rate": 7.334268001135764e-05,
|
| 1885 |
+
"loss": 1.4604171752929687,
|
| 1886 |
+
"step": 2680
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 1.0113,
|
| 1890 |
+
"grad_norm": 0.4594590365886688,
|
| 1891 |
+
"learning_rate": 7.324385419741438e-05,
|
| 1892 |
+
"loss": 1.4276948928833009,
|
| 1893 |
+
"step": 2690
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 1.0123,
|
| 1897 |
+
"grad_norm": 0.4594690501689911,
|
| 1898 |
+
"learning_rate": 7.314471891702651e-05,
|
| 1899 |
+
"loss": 1.3739115715026855,
|
| 1900 |
+
"step": 2700
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 1.0133,
|
| 1904 |
+
"grad_norm": 0.4549918472766876,
|
| 1905 |
+
"learning_rate": 7.304527518896245e-05,
|
| 1906 |
+
"loss": 1.4289373397827148,
|
| 1907 |
+
"step": 2710
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 1.0143,
|
| 1911 |
+
"grad_norm": 0.46599602699279785,
|
| 1912 |
+
"learning_rate": 7.294552403516034e-05,
|
| 1913 |
+
"loss": 1.447694969177246,
|
| 1914 |
+
"step": 2720
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 1.0153,
|
| 1918 |
+
"grad_norm": 0.4508320987224579,
|
| 1919 |
+
"learning_rate": 7.284546648071763e-05,
|
| 1920 |
+
"loss": 1.4010438919067383,
|
| 1921 |
+
"step": 2730
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 1.0163,
|
| 1925 |
+
"grad_norm": 0.4645399749279022,
|
| 1926 |
+
"learning_rate": 7.274510355388049e-05,
|
| 1927 |
+
"loss": 1.3724050521850586,
|
| 1928 |
+
"step": 2740
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 1.0173,
|
| 1932 |
+
"grad_norm": 0.5009335875511169,
|
| 1933 |
+
"learning_rate": 7.26444362860333e-05,
|
| 1934 |
+
"loss": 1.390232467651367,
|
| 1935 |
+
"step": 2750
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 1.0183,
|
| 1939 |
+
"grad_norm": 0.4698573648929596,
|
| 1940 |
+
"learning_rate": 7.254346571168798e-05,
|
| 1941 |
+
"loss": 1.3995256423950195,
|
| 1942 |
+
"step": 2760
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 1.0193,
|
| 1946 |
+
"grad_norm": 0.46238020062446594,
|
| 1947 |
+
"learning_rate": 7.24421928684734e-05,
|
| 1948 |
+
"loss": 1.3738612174987792,
|
| 1949 |
+
"step": 2770
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 1.0203,
|
| 1953 |
+
"grad_norm": 0.45451462268829346,
|
| 1954 |
+
"learning_rate": 7.234061879712472e-05,
|
| 1955 |
+
"loss": 1.3291207313537599,
|
| 1956 |
+
"step": 2780
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 1.0213,
|
| 1960 |
+
"grad_norm": 0.47129520773887634,
|
| 1961 |
+
"learning_rate": 7.223874454147269e-05,
|
| 1962 |
+
"loss": 1.356761646270752,
|
| 1963 |
+
"step": 2790
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 1.0223,
|
| 1967 |
+
"grad_norm": 0.4773348569869995,
|
| 1968 |
+
"learning_rate": 7.213657114843287e-05,
|
| 1969 |
+
"loss": 1.3456995010375976,
|
| 1970 |
+
"step": 2800
|
| 1971 |
+
}
|
| 1972 |
+
],
|
| 1973 |
+
"logging_steps": 10,
|
| 1974 |
+
"max_steps": 10000,
|
| 1975 |
+
"num_input_tokens_seen": 0,
|
| 1976 |
+
"num_train_epochs": 9223372036854775807,
|
| 1977 |
+
"save_steps": 100,
|
| 1978 |
+
"stateful_callbacks": {
|
| 1979 |
+
"TrainerControl": {
|
| 1980 |
+
"args": {
|
| 1981 |
+
"should_epoch_stop": false,
|
| 1982 |
+
"should_evaluate": false,
|
| 1983 |
+
"should_log": false,
|
| 1984 |
+
"should_save": true,
|
| 1985 |
+
"should_training_stop": false
|
| 1986 |
+
},
|
| 1987 |
+
"attributes": {}
|
| 1988 |
+
}
|
| 1989 |
+
},
|
| 1990 |
+
"total_flos": 3.7514887942478234e+17,
|
| 1991 |
+
"train_batch_size": 4,
|
| 1992 |
+
"trial_name": null,
|
| 1993 |
+
"trial_params": null
|
| 1994 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:652f1046a80d4fe70acd3bb17419b64c8e019931ca841ed54e28558b26e51397
|
| 3 |
+
size 5265
|