Upload folder using huggingface_hub
Browse files- chat_template.jinja +6 -0
- config.json +40 -0
- generation_config.json +7 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +17 -0
- trainer_state.json +1994 -0
- training_args.bin +3 -0
chat_template.jinja
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system
|
| 2 |
+
You are a helpful AI assistant named SmolLM, trained by Hugging Face<|im_end|>
|
| 3 |
+
' }}{% endif %}{{'<|im_start|>' + message['role'] + '
|
| 4 |
+
' + message['content'] + '<|im_end|>' + '
|
| 5 |
+
'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant
|
| 6 |
+
' }}{% endif %}
|
config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"LlamaForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 1,
|
| 8 |
+
"dtype": "bfloat16",
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"head_dim": 64,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 576,
|
| 13 |
+
"initializer_range": 0.041666666666666664,
|
| 14 |
+
"intermediate_size": 1536,
|
| 15 |
+
"is_llama_config": true,
|
| 16 |
+
"max_position_embeddings": 8192,
|
| 17 |
+
"mlp_bias": false,
|
| 18 |
+
"model_type": "llama",
|
| 19 |
+
"num_attention_heads": 9,
|
| 20 |
+
"num_hidden_layers": 30,
|
| 21 |
+
"num_key_value_heads": 3,
|
| 22 |
+
"pad_token_id": 2,
|
| 23 |
+
"pretraining_tp": 1,
|
| 24 |
+
"rms_norm_eps": 1e-05,
|
| 25 |
+
"rope_interleaved": false,
|
| 26 |
+
"rope_parameters": {
|
| 27 |
+
"rope_theta": 100000,
|
| 28 |
+
"rope_type": "default"
|
| 29 |
+
},
|
| 30 |
+
"tie_word_embeddings": true,
|
| 31 |
+
"transformers.js_config": {
|
| 32 |
+
"kv_cache_dtype": {
|
| 33 |
+
"fp16": "float16",
|
| 34 |
+
"q4f16": "float16"
|
| 35 |
+
}
|
| 36 |
+
},
|
| 37 |
+
"transformers_version": "5.0.0",
|
| 38 |
+
"use_cache": false,
|
| 39 |
+
"vocab_size": 49152
|
| 40 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"pad_token_id": 2,
|
| 6 |
+
"transformers_version": "5.0.0"
|
| 7 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:82f447a455940a896048f8b2f115d2933867c50f0980b8429722b502e2ce9fde
|
| 3 |
+
size 269060552
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fddf1d3cb89f82e24f5e570a4864a45bb1f887d6974edc5c53c9ce3cca90ad8d
|
| 3 |
+
size 538293323
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f4a9f217e852f439efa6bd32fde98d6867f11aa6ea13ddc021ba10af6a0b0934
|
| 3 |
+
size 14645
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0ebf781c5a5513927f21974a6d69dc04390db4fd082cbb36d0f415e67d837ef7
|
| 3 |
+
size 1465
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<|im_start|>",
|
| 5 |
+
"clean_up_tokenization_spaces": false,
|
| 6 |
+
"eos_token": "<|im_end|>",
|
| 7 |
+
"extra_special_tokens": [
|
| 8 |
+
"<|im_start|>",
|
| 9 |
+
"<|im_end|>"
|
| 10 |
+
],
|
| 11 |
+
"is_local": false,
|
| 12 |
+
"model_max_length": 8192,
|
| 13 |
+
"pad_token": "<|im_end|>",
|
| 14 |
+
"tokenizer_class": "TokenizersBackend",
|
| 15 |
+
"unk_token": "<|endoftext|>",
|
| 16 |
+
"vocab_size": 49152
|
| 17 |
+
}
|
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": 3.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 2805,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.010703773080010704,
|
| 14 |
+
"grad_norm": 2.640625,
|
| 15 |
+
"learning_rate": 1.993582887700535e-05,
|
| 16 |
+
"loss": 1.5584056854248047,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.02140754616002141,
|
| 21 |
+
"grad_norm": 2.53125,
|
| 22 |
+
"learning_rate": 1.9864527629233515e-05,
|
| 23 |
+
"loss": 1.562470054626465,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.03211131924003211,
|
| 28 |
+
"grad_norm": 2.46875,
|
| 29 |
+
"learning_rate": 1.9793226381461677e-05,
|
| 30 |
+
"loss": 1.5967525482177733,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.04281509232004282,
|
| 35 |
+
"grad_norm": 2.546875,
|
| 36 |
+
"learning_rate": 1.972192513368984e-05,
|
| 37 |
+
"loss": 1.5231587409973144,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.05351886540005352,
|
| 42 |
+
"grad_norm": 2.015625,
|
| 43 |
+
"learning_rate": 1.9650623885918005e-05,
|
| 44 |
+
"loss": 1.4803240776062012,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.06422263848006422,
|
| 49 |
+
"grad_norm": 2.21875,
|
| 50 |
+
"learning_rate": 1.957932263814617e-05,
|
| 51 |
+
"loss": 1.5196543693542481,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.07492641156007493,
|
| 56 |
+
"grad_norm": 2.21875,
|
| 57 |
+
"learning_rate": 1.9508021390374332e-05,
|
| 58 |
+
"loss": 1.4593828201293946,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.08563018464008564,
|
| 63 |
+
"grad_norm": 2.28125,
|
| 64 |
+
"learning_rate": 1.9436720142602497e-05,
|
| 65 |
+
"loss": 1.4797739028930663,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.09633395772009633,
|
| 70 |
+
"grad_norm": 2.078125,
|
| 71 |
+
"learning_rate": 1.9365418894830663e-05,
|
| 72 |
+
"loss": 1.5016543388366699,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.10703773080010703,
|
| 77 |
+
"grad_norm": 2.6875,
|
| 78 |
+
"learning_rate": 1.9294117647058825e-05,
|
| 79 |
+
"loss": 1.4800514221191405,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.11774150388011774,
|
| 84 |
+
"grad_norm": 1.75,
|
| 85 |
+
"learning_rate": 1.922281639928699e-05,
|
| 86 |
+
"loss": 1.445749568939209,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.12844527696012845,
|
| 91 |
+
"grad_norm": 2.0625,
|
| 92 |
+
"learning_rate": 1.9151515151515152e-05,
|
| 93 |
+
"loss": 1.4519201278686524,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.13914905004013914,
|
| 98 |
+
"grad_norm": 2.03125,
|
| 99 |
+
"learning_rate": 1.9080213903743317e-05,
|
| 100 |
+
"loss": 1.3476288795471192,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.14985282312014986,
|
| 105 |
+
"grad_norm": 2.234375,
|
| 106 |
+
"learning_rate": 1.9008912655971482e-05,
|
| 107 |
+
"loss": 1.4554848670959473,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.16055659620016055,
|
| 112 |
+
"grad_norm": 1.9375,
|
| 113 |
+
"learning_rate": 1.8937611408199644e-05,
|
| 114 |
+
"loss": 1.4435239791870118,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.17126036928017127,
|
| 119 |
+
"grad_norm": 1.671875,
|
| 120 |
+
"learning_rate": 1.886631016042781e-05,
|
| 121 |
+
"loss": 1.3767672538757325,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.18196414236018196,
|
| 126 |
+
"grad_norm": 2.09375,
|
| 127 |
+
"learning_rate": 1.8795008912655972e-05,
|
| 128 |
+
"loss": 1.4352334022521973,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.19266791544019266,
|
| 133 |
+
"grad_norm": 2.0,
|
| 134 |
+
"learning_rate": 1.8723707664884137e-05,
|
| 135 |
+
"loss": 1.382822322845459,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.20337168852020338,
|
| 140 |
+
"grad_norm": 1.953125,
|
| 141 |
+
"learning_rate": 1.8652406417112302e-05,
|
| 142 |
+
"loss": 1.429026985168457,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.21407546160021407,
|
| 147 |
+
"grad_norm": 1.859375,
|
| 148 |
+
"learning_rate": 1.8581105169340464e-05,
|
| 149 |
+
"loss": 1.3564122200012207,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.2247792346802248,
|
| 154 |
+
"grad_norm": 1.7109375,
|
| 155 |
+
"learning_rate": 1.850980392156863e-05,
|
| 156 |
+
"loss": 1.457004451751709,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.23548300776023548,
|
| 161 |
+
"grad_norm": 1.9921875,
|
| 162 |
+
"learning_rate": 1.843850267379679e-05,
|
| 163 |
+
"loss": 1.3679749488830566,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.2461867808402462,
|
| 168 |
+
"grad_norm": 2.0625,
|
| 169 |
+
"learning_rate": 1.8367201426024957e-05,
|
| 170 |
+
"loss": 1.4186459541320802,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.2568905539202569,
|
| 175 |
+
"grad_norm": 1.8984375,
|
| 176 |
+
"learning_rate": 1.8295900178253122e-05,
|
| 177 |
+
"loss": 1.3645942687988282,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.2675943270002676,
|
| 182 |
+
"grad_norm": 1.796875,
|
| 183 |
+
"learning_rate": 1.8224598930481284e-05,
|
| 184 |
+
"loss": 1.3659990310668946,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.2782981000802783,
|
| 189 |
+
"grad_norm": 1.9375,
|
| 190 |
+
"learning_rate": 1.815329768270945e-05,
|
| 191 |
+
"loss": 1.3751505851745605,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.289001873160289,
|
| 196 |
+
"grad_norm": 1.96875,
|
| 197 |
+
"learning_rate": 1.808199643493761e-05,
|
| 198 |
+
"loss": 1.394303798675537,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.2997056462402997,
|
| 203 |
+
"grad_norm": 1.71875,
|
| 204 |
+
"learning_rate": 1.8010695187165777e-05,
|
| 205 |
+
"loss": 1.3266244888305665,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.3104094193203104,
|
| 210 |
+
"grad_norm": 1.6328125,
|
| 211 |
+
"learning_rate": 1.7939393939393942e-05,
|
| 212 |
+
"loss": 1.3767006874084473,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.3211131924003211,
|
| 217 |
+
"grad_norm": 1.8359375,
|
| 218 |
+
"learning_rate": 1.7868092691622104e-05,
|
| 219 |
+
"loss": 1.3508996963500977,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.3318169654803318,
|
| 224 |
+
"grad_norm": 1.8984375,
|
| 225 |
+
"learning_rate": 1.779679144385027e-05,
|
| 226 |
+
"loss": 1.299268627166748,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.34252073856034254,
|
| 231 |
+
"grad_norm": 1.6015625,
|
| 232 |
+
"learning_rate": 1.772549019607843e-05,
|
| 233 |
+
"loss": 1.335693073272705,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.35322451164035323,
|
| 238 |
+
"grad_norm": 1.6171875,
|
| 239 |
+
"learning_rate": 1.7654188948306597e-05,
|
| 240 |
+
"loss": 1.3631214141845702,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.3639282847203639,
|
| 245 |
+
"grad_norm": 1.7421875,
|
| 246 |
+
"learning_rate": 1.7582887700534762e-05,
|
| 247 |
+
"loss": 1.349259376525879,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.3746320578003746,
|
| 252 |
+
"grad_norm": 1.8515625,
|
| 253 |
+
"learning_rate": 1.7511586452762924e-05,
|
| 254 |
+
"loss": 1.3234673500061036,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.3853358308803853,
|
| 259 |
+
"grad_norm": 1.734375,
|
| 260 |
+
"learning_rate": 1.744028520499109e-05,
|
| 261 |
+
"loss": 1.34688138961792,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.39603960396039606,
|
| 266 |
+
"grad_norm": 1.796875,
|
| 267 |
+
"learning_rate": 1.736898395721925e-05,
|
| 268 |
+
"loss": 1.310294246673584,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.40674337704040675,
|
| 273 |
+
"grad_norm": 1.5703125,
|
| 274 |
+
"learning_rate": 1.7297682709447417e-05,
|
| 275 |
+
"loss": 1.3146047592163086,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.41744715012041744,
|
| 280 |
+
"grad_norm": 2.078125,
|
| 281 |
+
"learning_rate": 1.7226381461675582e-05,
|
| 282 |
+
"loss": 1.3516902923583984,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.42815092320042814,
|
| 287 |
+
"grad_norm": 1.6875,
|
| 288 |
+
"learning_rate": 1.7155080213903744e-05,
|
| 289 |
+
"loss": 1.3631095886230469,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.4388546962804388,
|
| 294 |
+
"grad_norm": 1.578125,
|
| 295 |
+
"learning_rate": 1.708377896613191e-05,
|
| 296 |
+
"loss": 1.3395885467529296,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.4495584693604496,
|
| 301 |
+
"grad_norm": 1.7890625,
|
| 302 |
+
"learning_rate": 1.701247771836007e-05,
|
| 303 |
+
"loss": 1.322316837310791,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.46026224244046027,
|
| 308 |
+
"grad_norm": 2.03125,
|
| 309 |
+
"learning_rate": 1.6941176470588237e-05,
|
| 310 |
+
"loss": 1.3892762184143066,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.47096601552047096,
|
| 315 |
+
"grad_norm": 1.765625,
|
| 316 |
+
"learning_rate": 1.6869875222816402e-05,
|
| 317 |
+
"loss": 1.3081950187683105,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.48166978860048165,
|
| 322 |
+
"grad_norm": 1.75,
|
| 323 |
+
"learning_rate": 1.6798573975044564e-05,
|
| 324 |
+
"loss": 1.3405800819396974,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.4923735616804924,
|
| 329 |
+
"grad_norm": 1.6171875,
|
| 330 |
+
"learning_rate": 1.672727272727273e-05,
|
| 331 |
+
"loss": 1.3331517219543456,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.5030773347605031,
|
| 336 |
+
"grad_norm": 1.703125,
|
| 337 |
+
"learning_rate": 1.665597147950089e-05,
|
| 338 |
+
"loss": 1.3040351867675781,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.5137811078405138,
|
| 343 |
+
"grad_norm": 1.7265625,
|
| 344 |
+
"learning_rate": 1.6584670231729056e-05,
|
| 345 |
+
"loss": 1.319422149658203,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.5244848809205245,
|
| 350 |
+
"grad_norm": 1.8125,
|
| 351 |
+
"learning_rate": 1.6513368983957222e-05,
|
| 352 |
+
"loss": 1.3433240890502929,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.5351886540005352,
|
| 357 |
+
"grad_norm": 1.7265625,
|
| 358 |
+
"learning_rate": 1.6442067736185384e-05,
|
| 359 |
+
"loss": 1.3346479415893555,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.5458924270805459,
|
| 364 |
+
"grad_norm": 1.5625,
|
| 365 |
+
"learning_rate": 1.637076648841355e-05,
|
| 366 |
+
"loss": 1.3032867431640625,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.5565962001605566,
|
| 371 |
+
"grad_norm": 1.78125,
|
| 372 |
+
"learning_rate": 1.629946524064171e-05,
|
| 373 |
+
"loss": 1.3006314277648925,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.5672999732405672,
|
| 378 |
+
"grad_norm": 1.765625,
|
| 379 |
+
"learning_rate": 1.6228163992869876e-05,
|
| 380 |
+
"loss": 1.3416614532470703,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.578003746320578,
|
| 385 |
+
"grad_norm": 2.015625,
|
| 386 |
+
"learning_rate": 1.615686274509804e-05,
|
| 387 |
+
"loss": 1.303782081604004,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.5887075194005887,
|
| 392 |
+
"grad_norm": 1.578125,
|
| 393 |
+
"learning_rate": 1.6085561497326207e-05,
|
| 394 |
+
"loss": 1.2814931869506836,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.5994112924805994,
|
| 399 |
+
"grad_norm": 1.53125,
|
| 400 |
+
"learning_rate": 1.601426024955437e-05,
|
| 401 |
+
"loss": 1.3404861450195313,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.6101150655606101,
|
| 406 |
+
"grad_norm": 1.7734375,
|
| 407 |
+
"learning_rate": 1.594295900178253e-05,
|
| 408 |
+
"loss": 1.3594398498535156,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.6208188386406208,
|
| 413 |
+
"grad_norm": 1.609375,
|
| 414 |
+
"learning_rate": 1.5871657754010696e-05,
|
| 415 |
+
"loss": 1.2768223762512207,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.6315226117206315,
|
| 420 |
+
"grad_norm": 1.609375,
|
| 421 |
+
"learning_rate": 1.580035650623886e-05,
|
| 422 |
+
"loss": 1.3110815048217774,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.6422263848006422,
|
| 427 |
+
"grad_norm": 1.6640625,
|
| 428 |
+
"learning_rate": 1.5729055258467027e-05,
|
| 429 |
+
"loss": 1.2639217376708984,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.6529301578806529,
|
| 434 |
+
"grad_norm": 1.8203125,
|
| 435 |
+
"learning_rate": 1.565775401069519e-05,
|
| 436 |
+
"loss": 1.3356239318847656,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.6636339309606636,
|
| 441 |
+
"grad_norm": 1.8828125,
|
| 442 |
+
"learning_rate": 1.558645276292335e-05,
|
| 443 |
+
"loss": 1.3733593940734863,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.6743377040406744,
|
| 448 |
+
"grad_norm": 1.5703125,
|
| 449 |
+
"learning_rate": 1.5515151515151516e-05,
|
| 450 |
+
"loss": 1.2768065452575683,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.6850414771206851,
|
| 455 |
+
"grad_norm": 1.7578125,
|
| 456 |
+
"learning_rate": 1.544385026737968e-05,
|
| 457 |
+
"loss": 1.345008659362793,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.6957452502006958,
|
| 462 |
+
"grad_norm": 1.4921875,
|
| 463 |
+
"learning_rate": 1.5372549019607847e-05,
|
| 464 |
+
"loss": 1.2327005386352539,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.7064490232807065,
|
| 469 |
+
"grad_norm": 1.765625,
|
| 470 |
+
"learning_rate": 1.530124777183601e-05,
|
| 471 |
+
"loss": 1.327579879760742,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.7171527963607172,
|
| 476 |
+
"grad_norm": 1.6640625,
|
| 477 |
+
"learning_rate": 1.5229946524064172e-05,
|
| 478 |
+
"loss": 1.2693171501159668,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.7278565694407279,
|
| 483 |
+
"grad_norm": 1.640625,
|
| 484 |
+
"learning_rate": 1.5158645276292336e-05,
|
| 485 |
+
"loss": 1.3229084014892578,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.7385603425207385,
|
| 490 |
+
"grad_norm": 1.7265625,
|
| 491 |
+
"learning_rate": 1.5087344028520501e-05,
|
| 492 |
+
"loss": 1.3010024070739745,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.7492641156007492,
|
| 497 |
+
"grad_norm": 1.59375,
|
| 498 |
+
"learning_rate": 1.5016042780748665e-05,
|
| 499 |
+
"loss": 1.304527473449707,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.7599678886807599,
|
| 504 |
+
"grad_norm": 1.8671875,
|
| 505 |
+
"learning_rate": 1.4944741532976827e-05,
|
| 506 |
+
"loss": 1.2771072387695312,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.7706716617607706,
|
| 511 |
+
"grad_norm": 1.671875,
|
| 512 |
+
"learning_rate": 1.4873440285204992e-05,
|
| 513 |
+
"loss": 1.285037899017334,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.7813754348407814,
|
| 518 |
+
"grad_norm": 1.6171875,
|
| 519 |
+
"learning_rate": 1.4802139037433156e-05,
|
| 520 |
+
"loss": 1.2612761497497558,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.7920792079207921,
|
| 525 |
+
"grad_norm": 1.8125,
|
| 526 |
+
"learning_rate": 1.4730837789661321e-05,
|
| 527 |
+
"loss": 1.3110386848449707,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.8027829810008028,
|
| 532 |
+
"grad_norm": 1.671875,
|
| 533 |
+
"learning_rate": 1.4659536541889485e-05,
|
| 534 |
+
"loss": 1.3450962066650392,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.8134867540808135,
|
| 539 |
+
"grad_norm": 1.796875,
|
| 540 |
+
"learning_rate": 1.4588235294117647e-05,
|
| 541 |
+
"loss": 1.294900608062744,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.8241905271608242,
|
| 546 |
+
"grad_norm": 1.4765625,
|
| 547 |
+
"learning_rate": 1.4516934046345812e-05,
|
| 548 |
+
"loss": 1.3215585708618165,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.8348943002408349,
|
| 553 |
+
"grad_norm": 1.5859375,
|
| 554 |
+
"learning_rate": 1.4445632798573976e-05,
|
| 555 |
+
"loss": 1.3044111251831054,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.8455980733208456,
|
| 560 |
+
"grad_norm": 1.8671875,
|
| 561 |
+
"learning_rate": 1.4374331550802141e-05,
|
| 562 |
+
"loss": 1.3348912239074706,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.8563018464008563,
|
| 567 |
+
"grad_norm": 1.65625,
|
| 568 |
+
"learning_rate": 1.4303030303030305e-05,
|
| 569 |
+
"loss": 1.3434508323669434,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.867005619480867,
|
| 574 |
+
"grad_norm": 1.625,
|
| 575 |
+
"learning_rate": 1.4231729055258467e-05,
|
| 576 |
+
"loss": 1.291652297973633,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.8777093925608777,
|
| 581 |
+
"grad_norm": 1.4921875,
|
| 582 |
+
"learning_rate": 1.4160427807486632e-05,
|
| 583 |
+
"loss": 1.3067720413208008,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.8884131656408885,
|
| 588 |
+
"grad_norm": 1.9453125,
|
| 589 |
+
"learning_rate": 1.4089126559714796e-05,
|
| 590 |
+
"loss": 1.3196195602416991,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.8991169387208992,
|
| 595 |
+
"grad_norm": 1.5390625,
|
| 596 |
+
"learning_rate": 1.4017825311942961e-05,
|
| 597 |
+
"loss": 1.3129652976989745,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.9098207118009098,
|
| 602 |
+
"grad_norm": 1.5,
|
| 603 |
+
"learning_rate": 1.3946524064171123e-05,
|
| 604 |
+
"loss": 1.2702789306640625,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.9205244848809205,
|
| 609 |
+
"grad_norm": 1.9453125,
|
| 610 |
+
"learning_rate": 1.3875222816399288e-05,
|
| 611 |
+
"loss": 1.29964599609375,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.9312282579609312,
|
| 616 |
+
"grad_norm": 1.5703125,
|
| 617 |
+
"learning_rate": 1.3803921568627452e-05,
|
| 618 |
+
"loss": 1.301185131072998,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.9419320310409419,
|
| 623 |
+
"grad_norm": 1.4140625,
|
| 624 |
+
"learning_rate": 1.3732620320855616e-05,
|
| 625 |
+
"loss": 1.2731993675231934,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.9526358041209526,
|
| 630 |
+
"grad_norm": 1.421875,
|
| 631 |
+
"learning_rate": 1.3661319073083781e-05,
|
| 632 |
+
"loss": 1.2806821823120118,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.9633395772009633,
|
| 637 |
+
"grad_norm": 1.84375,
|
| 638 |
+
"learning_rate": 1.3590017825311943e-05,
|
| 639 |
+
"loss": 1.2375809669494628,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.974043350280974,
|
| 644 |
+
"grad_norm": 1.8046875,
|
| 645 |
+
"learning_rate": 1.3518716577540108e-05,
|
| 646 |
+
"loss": 1.2453808784484863,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.9847471233609848,
|
| 651 |
+
"grad_norm": 2.03125,
|
| 652 |
+
"learning_rate": 1.3447415329768272e-05,
|
| 653 |
+
"loss": 1.3074142456054687,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.9954508964409955,
|
| 658 |
+
"grad_norm": 1.484375,
|
| 659 |
+
"learning_rate": 1.3376114081996437e-05,
|
| 660 |
+
"loss": 1.2914584159851075,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.0053518865400053,
|
| 665 |
+
"grad_norm": 2.0625,
|
| 666 |
+
"learning_rate": 1.33048128342246e-05,
|
| 667 |
+
"loss": 1.3543176651000977,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.016055659620016,
|
| 672 |
+
"grad_norm": 1.765625,
|
| 673 |
+
"learning_rate": 1.3233511586452763e-05,
|
| 674 |
+
"loss": 1.3298683166503906,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 1.0267594327000267,
|
| 679 |
+
"grad_norm": 1.59375,
|
| 680 |
+
"learning_rate": 1.3162210338680928e-05,
|
| 681 |
+
"loss": 1.3020204544067382,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 1.0374632057800375,
|
| 686 |
+
"grad_norm": 1.6484375,
|
| 687 |
+
"learning_rate": 1.3090909090909092e-05,
|
| 688 |
+
"loss": 1.3046648025512695,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.048166978860048,
|
| 693 |
+
"grad_norm": 1.65625,
|
| 694 |
+
"learning_rate": 1.3019607843137257e-05,
|
| 695 |
+
"loss": 1.2308432579040527,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 1.0588707519400589,
|
| 700 |
+
"grad_norm": 1.65625,
|
| 701 |
+
"learning_rate": 1.294830659536542e-05,
|
| 702 |
+
"loss": 1.2811461448669434,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 1.0695745250200697,
|
| 707 |
+
"grad_norm": 1.6640625,
|
| 708 |
+
"learning_rate": 1.2877005347593583e-05,
|
| 709 |
+
"loss": 1.3090335845947265,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 1.0802782981000802,
|
| 714 |
+
"grad_norm": 1.9921875,
|
| 715 |
+
"learning_rate": 1.2805704099821748e-05,
|
| 716 |
+
"loss": 1.2958572387695313,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 1.090982071180091,
|
| 721 |
+
"grad_norm": 1.6953125,
|
| 722 |
+
"learning_rate": 1.2734402852049912e-05,
|
| 723 |
+
"loss": 1.326209259033203,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 1.1016858442601016,
|
| 728 |
+
"grad_norm": 1.75,
|
| 729 |
+
"learning_rate": 1.2663101604278077e-05,
|
| 730 |
+
"loss": 1.2520675659179688,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 1.1123896173401124,
|
| 735 |
+
"grad_norm": 1.546875,
|
| 736 |
+
"learning_rate": 1.259180035650624e-05,
|
| 737 |
+
"loss": 1.3478898048400878,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 1.123093390420123,
|
| 742 |
+
"grad_norm": 1.4921875,
|
| 743 |
+
"learning_rate": 1.2520499108734403e-05,
|
| 744 |
+
"loss": 1.2806931495666505,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 1.1337971635001338,
|
| 749 |
+
"grad_norm": 1.75,
|
| 750 |
+
"learning_rate": 1.2449197860962568e-05,
|
| 751 |
+
"loss": 1.2603809356689453,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.1445009365801444,
|
| 756 |
+
"grad_norm": 1.484375,
|
| 757 |
+
"learning_rate": 1.2377896613190731e-05,
|
| 758 |
+
"loss": 1.2837313652038573,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 1.1552047096601552,
|
| 763 |
+
"grad_norm": 1.859375,
|
| 764 |
+
"learning_rate": 1.2306595365418897e-05,
|
| 765 |
+
"loss": 1.271355152130127,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 1.165908482740166,
|
| 770 |
+
"grad_norm": 1.6015625,
|
| 771 |
+
"learning_rate": 1.223529411764706e-05,
|
| 772 |
+
"loss": 1.2751256942749023,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 1.1766122558201766,
|
| 777 |
+
"grad_norm": 1.5390625,
|
| 778 |
+
"learning_rate": 1.2163992869875222e-05,
|
| 779 |
+
"loss": 1.2217981338500976,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 1.1873160289001874,
|
| 784 |
+
"grad_norm": 1.46875,
|
| 785 |
+
"learning_rate": 1.2092691622103388e-05,
|
| 786 |
+
"loss": 1.3460000038146973,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 1.198019801980198,
|
| 791 |
+
"grad_norm": 1.6328125,
|
| 792 |
+
"learning_rate": 1.2021390374331551e-05,
|
| 793 |
+
"loss": 1.3119497299194336,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 1.2087235750602088,
|
| 798 |
+
"grad_norm": 1.5,
|
| 799 |
+
"learning_rate": 1.1950089126559717e-05,
|
| 800 |
+
"loss": 1.326594066619873,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 1.2194273481402194,
|
| 805 |
+
"grad_norm": 1.8671875,
|
| 806 |
+
"learning_rate": 1.187878787878788e-05,
|
| 807 |
+
"loss": 1.313736343383789,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 1.2301311212202302,
|
| 812 |
+
"grad_norm": 1.625,
|
| 813 |
+
"learning_rate": 1.1807486631016042e-05,
|
| 814 |
+
"loss": 1.2580394744873047,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 1.2408348943002407,
|
| 819 |
+
"grad_norm": 1.6953125,
|
| 820 |
+
"learning_rate": 1.1736185383244208e-05,
|
| 821 |
+
"loss": 1.3472198486328124,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 1.2515386673802515,
|
| 826 |
+
"grad_norm": 1.6875,
|
| 827 |
+
"learning_rate": 1.1664884135472371e-05,
|
| 828 |
+
"loss": 1.3223270416259765,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 1.2622424404602621,
|
| 833 |
+
"grad_norm": 1.5390625,
|
| 834 |
+
"learning_rate": 1.1593582887700537e-05,
|
| 835 |
+
"loss": 1.3479475021362304,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 1.272946213540273,
|
| 840 |
+
"grad_norm": 1.46875,
|
| 841 |
+
"learning_rate": 1.15222816399287e-05,
|
| 842 |
+
"loss": 1.2691156387329101,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 1.2836499866202837,
|
| 847 |
+
"grad_norm": 1.578125,
|
| 848 |
+
"learning_rate": 1.1450980392156862e-05,
|
| 849 |
+
"loss": 1.3078096389770508,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 1.2943537597002943,
|
| 854 |
+
"grad_norm": 1.7265625,
|
| 855 |
+
"learning_rate": 1.1379679144385028e-05,
|
| 856 |
+
"loss": 1.2821264266967773,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 1.3050575327803051,
|
| 861 |
+
"grad_norm": 1.6015625,
|
| 862 |
+
"learning_rate": 1.1308377896613191e-05,
|
| 863 |
+
"loss": 1.2466256141662597,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.3157613058603157,
|
| 868 |
+
"grad_norm": 1.84375,
|
| 869 |
+
"learning_rate": 1.1237076648841357e-05,
|
| 870 |
+
"loss": 1.301154327392578,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 1.3264650789403265,
|
| 875 |
+
"grad_norm": 1.6171875,
|
| 876 |
+
"learning_rate": 1.116577540106952e-05,
|
| 877 |
+
"loss": 1.3058858871459962,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.337168852020337,
|
| 882 |
+
"grad_norm": 1.6875,
|
| 883 |
+
"learning_rate": 1.1094474153297684e-05,
|
| 884 |
+
"loss": 1.257982349395752,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 1.3478726251003479,
|
| 889 |
+
"grad_norm": 1.546875,
|
| 890 |
+
"learning_rate": 1.1023172905525847e-05,
|
| 891 |
+
"loss": 1.278379535675049,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 1.3585763981803587,
|
| 896 |
+
"grad_norm": 1.890625,
|
| 897 |
+
"learning_rate": 1.0951871657754011e-05,
|
| 898 |
+
"loss": 1.2998493194580079,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 1.3692801712603693,
|
| 903 |
+
"grad_norm": 1.5859375,
|
| 904 |
+
"learning_rate": 1.0880570409982176e-05,
|
| 905 |
+
"loss": 1.3042527198791505,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 1.3799839443403799,
|
| 910 |
+
"grad_norm": 1.6015625,
|
| 911 |
+
"learning_rate": 1.0809269162210338e-05,
|
| 912 |
+
"loss": 1.2903579711914062,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.3906877174203907,
|
| 917 |
+
"grad_norm": 1.5390625,
|
| 918 |
+
"learning_rate": 1.0737967914438504e-05,
|
| 919 |
+
"loss": 1.216090202331543,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.4013914905004015,
|
| 924 |
+
"grad_norm": 1.4296875,
|
| 925 |
+
"learning_rate": 1.0666666666666667e-05,
|
| 926 |
+
"loss": 1.2497664451599122,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 1.412095263580412,
|
| 931 |
+
"grad_norm": 1.609375,
|
| 932 |
+
"learning_rate": 1.0595365418894833e-05,
|
| 933 |
+
"loss": 1.2592049598693849,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.4227990366604228,
|
| 938 |
+
"grad_norm": 1.703125,
|
| 939 |
+
"learning_rate": 1.0524064171122996e-05,
|
| 940 |
+
"loss": 1.3062689781188965,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.4335028097404334,
|
| 945 |
+
"grad_norm": 1.515625,
|
| 946 |
+
"learning_rate": 1.0452762923351158e-05,
|
| 947 |
+
"loss": 1.2577032089233398,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.4442065828204442,
|
| 952 |
+
"grad_norm": 1.75,
|
| 953 |
+
"learning_rate": 1.0381461675579324e-05,
|
| 954 |
+
"loss": 1.2874650001525878,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.4549103559004548,
|
| 959 |
+
"grad_norm": 2.0625,
|
| 960 |
+
"learning_rate": 1.0310160427807487e-05,
|
| 961 |
+
"loss": 1.2887776374816895,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.4656141289804656,
|
| 966 |
+
"grad_norm": 1.59375,
|
| 967 |
+
"learning_rate": 1.0238859180035653e-05,
|
| 968 |
+
"loss": 1.2869946479797363,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.4763179020604764,
|
| 973 |
+
"grad_norm": 1.7421875,
|
| 974 |
+
"learning_rate": 1.0167557932263816e-05,
|
| 975 |
+
"loss": 1.3055774688720703,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.487021675140487,
|
| 980 |
+
"grad_norm": 1.7421875,
|
| 981 |
+
"learning_rate": 1.0096256684491978e-05,
|
| 982 |
+
"loss": 1.2925223350524901,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.4977254482204978,
|
| 987 |
+
"grad_norm": 1.5390625,
|
| 988 |
+
"learning_rate": 1.0024955436720143e-05,
|
| 989 |
+
"loss": 1.3624143600463867,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.5084292213005084,
|
| 994 |
+
"grad_norm": 1.7265625,
|
| 995 |
+
"learning_rate": 9.953654188948307e-06,
|
| 996 |
+
"loss": 1.3100957870483398,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.5191329943805192,
|
| 1001 |
+
"grad_norm": 1.734375,
|
| 1002 |
+
"learning_rate": 9.882352941176472e-06,
|
| 1003 |
+
"loss": 1.2667318344116212,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.5298367674605298,
|
| 1008 |
+
"grad_norm": 1.5,
|
| 1009 |
+
"learning_rate": 9.811051693404634e-06,
|
| 1010 |
+
"loss": 1.2964338302612304,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 1.5405405405405406,
|
| 1015 |
+
"grad_norm": 1.703125,
|
| 1016 |
+
"learning_rate": 9.7397504456328e-06,
|
| 1017 |
+
"loss": 1.2451062202453613,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.5512443136205514,
|
| 1022 |
+
"grad_norm": 1.6484375,
|
| 1023 |
+
"learning_rate": 9.668449197860963e-06,
|
| 1024 |
+
"loss": 1.2622719764709474,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.561948086700562,
|
| 1029 |
+
"grad_norm": 1.8359375,
|
| 1030 |
+
"learning_rate": 9.597147950089127e-06,
|
| 1031 |
+
"loss": 1.2830778121948243,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.5726518597805725,
|
| 1036 |
+
"grad_norm": 1.5390625,
|
| 1037 |
+
"learning_rate": 9.525846702317292e-06,
|
| 1038 |
+
"loss": 1.3212904930114746,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.5833556328605833,
|
| 1043 |
+
"grad_norm": 1.515625,
|
| 1044 |
+
"learning_rate": 9.454545454545456e-06,
|
| 1045 |
+
"loss": 1.301555347442627,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.5940594059405941,
|
| 1050 |
+
"grad_norm": 1.453125,
|
| 1051 |
+
"learning_rate": 9.38324420677362e-06,
|
| 1052 |
+
"loss": 1.2626118659973145,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 1.6047631790206047,
|
| 1057 |
+
"grad_norm": 1.890625,
|
| 1058 |
+
"learning_rate": 9.311942959001783e-06,
|
| 1059 |
+
"loss": 1.2342555046081543,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.6154669521006153,
|
| 1064 |
+
"grad_norm": 1.7421875,
|
| 1065 |
+
"learning_rate": 9.240641711229947e-06,
|
| 1066 |
+
"loss": 1.3167900085449218,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.6261707251806263,
|
| 1071 |
+
"grad_norm": 2.0,
|
| 1072 |
+
"learning_rate": 9.169340463458112e-06,
|
| 1073 |
+
"loss": 1.296627902984619,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.636874498260637,
|
| 1078 |
+
"grad_norm": 1.4453125,
|
| 1079 |
+
"learning_rate": 9.098039215686276e-06,
|
| 1080 |
+
"loss": 1.275075340270996,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.6475782713406475,
|
| 1085 |
+
"grad_norm": 1.53125,
|
| 1086 |
+
"learning_rate": 9.02673796791444e-06,
|
| 1087 |
+
"loss": 1.2771642684936524,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.6582820444206583,
|
| 1092 |
+
"grad_norm": 1.8046875,
|
| 1093 |
+
"learning_rate": 8.955436720142603e-06,
|
| 1094 |
+
"loss": 1.2907758712768556,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.668985817500669,
|
| 1099 |
+
"grad_norm": 1.6953125,
|
| 1100 |
+
"learning_rate": 8.884135472370767e-06,
|
| 1101 |
+
"loss": 1.2778194427490235,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.6796895905806797,
|
| 1106 |
+
"grad_norm": 1.5859375,
|
| 1107 |
+
"learning_rate": 8.81283422459893e-06,
|
| 1108 |
+
"loss": 1.2820199012756348,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.6903933636606903,
|
| 1113 |
+
"grad_norm": 1.8984375,
|
| 1114 |
+
"learning_rate": 8.741532976827096e-06,
|
| 1115 |
+
"loss": 1.3197799682617188,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.701097136740701,
|
| 1120 |
+
"grad_norm": 1.796875,
|
| 1121 |
+
"learning_rate": 8.67023172905526e-06,
|
| 1122 |
+
"loss": 1.2711196899414063,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.7118009098207119,
|
| 1127 |
+
"grad_norm": 1.7265625,
|
| 1128 |
+
"learning_rate": 8.598930481283423e-06,
|
| 1129 |
+
"loss": 1.3094602584838868,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.7225046829007225,
|
| 1134 |
+
"grad_norm": 1.65625,
|
| 1135 |
+
"learning_rate": 8.527629233511587e-06,
|
| 1136 |
+
"loss": 1.3037433624267578,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.7332084559807333,
|
| 1141 |
+
"grad_norm": 1.6484375,
|
| 1142 |
+
"learning_rate": 8.45632798573975e-06,
|
| 1143 |
+
"loss": 1.2734570503234863,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.743912229060744,
|
| 1148 |
+
"grad_norm": 1.5703125,
|
| 1149 |
+
"learning_rate": 8.385026737967916e-06,
|
| 1150 |
+
"loss": 1.2476407051086427,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.7546160021407546,
|
| 1155 |
+
"grad_norm": 1.7265625,
|
| 1156 |
+
"learning_rate": 8.31372549019608e-06,
|
| 1157 |
+
"loss": 1.3427558898925782,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.7653197752207652,
|
| 1162 |
+
"grad_norm": 1.65625,
|
| 1163 |
+
"learning_rate": 8.242424242424243e-06,
|
| 1164 |
+
"loss": 1.273496437072754,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.776023548300776,
|
| 1169 |
+
"grad_norm": 1.71875,
|
| 1170 |
+
"learning_rate": 8.171122994652407e-06,
|
| 1171 |
+
"loss": 1.2626665115356446,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.7867273213807868,
|
| 1176 |
+
"grad_norm": 1.8046875,
|
| 1177 |
+
"learning_rate": 8.09982174688057e-06,
|
| 1178 |
+
"loss": 1.2670047760009766,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.7974310944607974,
|
| 1183 |
+
"grad_norm": 1.7265625,
|
| 1184 |
+
"learning_rate": 8.028520499108736e-06,
|
| 1185 |
+
"loss": 1.349191665649414,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.808134867540808,
|
| 1190 |
+
"grad_norm": 1.578125,
|
| 1191 |
+
"learning_rate": 7.9572192513369e-06,
|
| 1192 |
+
"loss": 1.2989972114562989,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.8188386406208188,
|
| 1197 |
+
"grad_norm": 1.6875,
|
| 1198 |
+
"learning_rate": 7.885918003565063e-06,
|
| 1199 |
+
"loss": 1.1850922584533692,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.8295424137008296,
|
| 1204 |
+
"grad_norm": 1.953125,
|
| 1205 |
+
"learning_rate": 7.814616755793228e-06,
|
| 1206 |
+
"loss": 1.3360312461853028,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.8402461867808402,
|
| 1211 |
+
"grad_norm": 1.6328125,
|
| 1212 |
+
"learning_rate": 7.74331550802139e-06,
|
| 1213 |
+
"loss": 1.2957257270812987,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.850949959860851,
|
| 1218 |
+
"grad_norm": 1.6015625,
|
| 1219 |
+
"learning_rate": 7.672014260249555e-06,
|
| 1220 |
+
"loss": 1.2536530494689941,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.8616537329408618,
|
| 1225 |
+
"grad_norm": 1.7109375,
|
| 1226 |
+
"learning_rate": 7.60071301247772e-06,
|
| 1227 |
+
"loss": 1.2660930633544922,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.8723575060208724,
|
| 1232 |
+
"grad_norm": 1.7265625,
|
| 1233 |
+
"learning_rate": 7.529411764705883e-06,
|
| 1234 |
+
"loss": 1.3080876350402832,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.883061279100883,
|
| 1239 |
+
"grad_norm": 1.640625,
|
| 1240 |
+
"learning_rate": 7.458110516934047e-06,
|
| 1241 |
+
"loss": 1.3132406234741212,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.8937650521808937,
|
| 1246 |
+
"grad_norm": 1.8203125,
|
| 1247 |
+
"learning_rate": 7.386809269162211e-06,
|
| 1248 |
+
"loss": 1.31253080368042,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.9044688252609046,
|
| 1253 |
+
"grad_norm": 1.4140625,
|
| 1254 |
+
"learning_rate": 7.315508021390375e-06,
|
| 1255 |
+
"loss": 1.3013240814208984,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.9151725983409151,
|
| 1260 |
+
"grad_norm": 1.4765625,
|
| 1261 |
+
"learning_rate": 7.244206773618538e-06,
|
| 1262 |
+
"loss": 1.2744117736816407,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.9258763714209257,
|
| 1267 |
+
"grad_norm": 1.4609375,
|
| 1268 |
+
"learning_rate": 7.172905525846703e-06,
|
| 1269 |
+
"loss": 1.3057758331298828,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.9365801445009367,
|
| 1274 |
+
"grad_norm": 1.515625,
|
| 1275 |
+
"learning_rate": 7.101604278074867e-06,
|
| 1276 |
+
"loss": 1.224927043914795,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.9472839175809473,
|
| 1281 |
+
"grad_norm": 1.7109375,
|
| 1282 |
+
"learning_rate": 7.030303030303031e-06,
|
| 1283 |
+
"loss": 1.3182221412658692,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.957987690660958,
|
| 1288 |
+
"grad_norm": 1.5546875,
|
| 1289 |
+
"learning_rate": 6.959001782531195e-06,
|
| 1290 |
+
"loss": 1.2400826454162597,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.9686914637409687,
|
| 1295 |
+
"grad_norm": 1.7421875,
|
| 1296 |
+
"learning_rate": 6.887700534759358e-06,
|
| 1297 |
+
"loss": 1.2463386535644532,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.9793952368209795,
|
| 1302 |
+
"grad_norm": 1.46875,
|
| 1303 |
+
"learning_rate": 6.8163992869875225e-06,
|
| 1304 |
+
"loss": 1.3235528945922852,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.99009900990099,
|
| 1309 |
+
"grad_norm": 1.8203125,
|
| 1310 |
+
"learning_rate": 6.745098039215687e-06,
|
| 1311 |
+
"loss": 1.2812946319580079,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 2.0,
|
| 1316 |
+
"grad_norm": 4.1875,
|
| 1317 |
+
"learning_rate": 6.673796791443851e-06,
|
| 1318 |
+
"loss": 1.2953272819519044,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 2.0107037730800106,
|
| 1323 |
+
"grad_norm": 1.5390625,
|
| 1324 |
+
"learning_rate": 6.602495543672015e-06,
|
| 1325 |
+
"loss": 1.207719612121582,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 2.0214075461600216,
|
| 1330 |
+
"grad_norm": 1.671875,
|
| 1331 |
+
"learning_rate": 6.531194295900179e-06,
|
| 1332 |
+
"loss": 1.2520846366882323,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 2.032111319240032,
|
| 1337 |
+
"grad_norm": 1.703125,
|
| 1338 |
+
"learning_rate": 6.459893048128343e-06,
|
| 1339 |
+
"loss": 1.2905988693237305,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 2.0428150923200428,
|
| 1344 |
+
"grad_norm": 1.921875,
|
| 1345 |
+
"learning_rate": 6.388591800356507e-06,
|
| 1346 |
+
"loss": 1.3520148277282715,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 2.0535188654000534,
|
| 1351 |
+
"grad_norm": 1.8515625,
|
| 1352 |
+
"learning_rate": 6.3172905525846705e-06,
|
| 1353 |
+
"loss": 1.2911107063293457,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 2.0642226384800644,
|
| 1358 |
+
"grad_norm": 1.671875,
|
| 1359 |
+
"learning_rate": 6.245989304812835e-06,
|
| 1360 |
+
"loss": 1.2403117179870606,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 2.074926411560075,
|
| 1365 |
+
"grad_norm": 1.96875,
|
| 1366 |
+
"learning_rate": 6.174688057040999e-06,
|
| 1367 |
+
"loss": 1.3558055877685546,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 2.0856301846400855,
|
| 1372 |
+
"grad_norm": 1.6328125,
|
| 1373 |
+
"learning_rate": 6.103386809269163e-06,
|
| 1374 |
+
"loss": 1.3194045066833495,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 2.096333957720096,
|
| 1379 |
+
"grad_norm": 2.140625,
|
| 1380 |
+
"learning_rate": 6.032085561497326e-06,
|
| 1381 |
+
"loss": 1.3321582794189453,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 2.107037730800107,
|
| 1386 |
+
"grad_norm": 1.625,
|
| 1387 |
+
"learning_rate": 5.96078431372549e-06,
|
| 1388 |
+
"loss": 1.288839054107666,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 2.1177415038801177,
|
| 1393 |
+
"grad_norm": 2.0,
|
| 1394 |
+
"learning_rate": 5.889483065953655e-06,
|
| 1395 |
+
"loss": 1.3260244369506835,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 2.1284452769601283,
|
| 1400 |
+
"grad_norm": 1.5703125,
|
| 1401 |
+
"learning_rate": 5.8181818181818185e-06,
|
| 1402 |
+
"loss": 1.2721702575683593,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 2.1391490500401393,
|
| 1407 |
+
"grad_norm": 1.6640625,
|
| 1408 |
+
"learning_rate": 5.746880570409983e-06,
|
| 1409 |
+
"loss": 1.2622364044189454,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 2.14985282312015,
|
| 1414 |
+
"grad_norm": 1.609375,
|
| 1415 |
+
"learning_rate": 5.675579322638146e-06,
|
| 1416 |
+
"loss": 1.30474796295166,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 2.1605565962001605,
|
| 1421 |
+
"grad_norm": 1.7890625,
|
| 1422 |
+
"learning_rate": 5.60427807486631e-06,
|
| 1423 |
+
"loss": 1.3109374046325684,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 2.171260369280171,
|
| 1428 |
+
"grad_norm": 1.71875,
|
| 1429 |
+
"learning_rate": 5.532976827094475e-06,
|
| 1430 |
+
"loss": 1.3231799125671386,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 2.181964142360182,
|
| 1435 |
+
"grad_norm": 1.796875,
|
| 1436 |
+
"learning_rate": 5.4616755793226384e-06,
|
| 1437 |
+
"loss": 1.2993489265441895,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 2.1926679154401927,
|
| 1442 |
+
"grad_norm": 1.875,
|
| 1443 |
+
"learning_rate": 5.390374331550803e-06,
|
| 1444 |
+
"loss": 1.3044631958007813,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 2.2033716885202033,
|
| 1449 |
+
"grad_norm": 1.4609375,
|
| 1450 |
+
"learning_rate": 5.3190730837789666e-06,
|
| 1451 |
+
"loss": 1.2702978134155274,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 2.2140754616002143,
|
| 1456 |
+
"grad_norm": 1.5546875,
|
| 1457 |
+
"learning_rate": 5.24777183600713e-06,
|
| 1458 |
+
"loss": 1.287952709197998,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 2.224779234680225,
|
| 1463 |
+
"grad_norm": 1.5546875,
|
| 1464 |
+
"learning_rate": 5.176470588235295e-06,
|
| 1465 |
+
"loss": 1.2974214553833008,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 2.2354830077602355,
|
| 1470 |
+
"grad_norm": 1.703125,
|
| 1471 |
+
"learning_rate": 5.105169340463458e-06,
|
| 1472 |
+
"loss": 1.3148197174072265,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 2.246186780840246,
|
| 1477 |
+
"grad_norm": 1.6328125,
|
| 1478 |
+
"learning_rate": 5.033868092691623e-06,
|
| 1479 |
+
"loss": 1.3466445922851562,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 2.256890553920257,
|
| 1484 |
+
"grad_norm": 1.484375,
|
| 1485 |
+
"learning_rate": 4.9625668449197864e-06,
|
| 1486 |
+
"loss": 1.334506893157959,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 2.2675943270002676,
|
| 1491 |
+
"grad_norm": 1.9765625,
|
| 1492 |
+
"learning_rate": 4.891265597147951e-06,
|
| 1493 |
+
"loss": 1.279165267944336,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 2.278298100080278,
|
| 1498 |
+
"grad_norm": 1.890625,
|
| 1499 |
+
"learning_rate": 4.8199643493761146e-06,
|
| 1500 |
+
"loss": 1.2512639045715332,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 2.289001873160289,
|
| 1505 |
+
"grad_norm": 1.59375,
|
| 1506 |
+
"learning_rate": 4.748663101604278e-06,
|
| 1507 |
+
"loss": 1.2572649002075196,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 2.2997056462403,
|
| 1512 |
+
"grad_norm": 1.5625,
|
| 1513 |
+
"learning_rate": 4.677361853832442e-06,
|
| 1514 |
+
"loss": 1.2503036499023437,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 2.3104094193203104,
|
| 1519 |
+
"grad_norm": 1.859375,
|
| 1520 |
+
"learning_rate": 4.606060606060606e-06,
|
| 1521 |
+
"loss": 1.2866994857788085,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 2.321113192400321,
|
| 1526 |
+
"grad_norm": 1.6171875,
|
| 1527 |
+
"learning_rate": 4.534759358288771e-06,
|
| 1528 |
+
"loss": 1.2810638427734375,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 2.331816965480332,
|
| 1533 |
+
"grad_norm": 1.71875,
|
| 1534 |
+
"learning_rate": 4.4634581105169345e-06,
|
| 1535 |
+
"loss": 1.2588828086853028,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 2.3425207385603426,
|
| 1540 |
+
"grad_norm": 1.6875,
|
| 1541 |
+
"learning_rate": 4.392156862745098e-06,
|
| 1542 |
+
"loss": 1.2615557670593263,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 2.353224511640353,
|
| 1547 |
+
"grad_norm": 1.921875,
|
| 1548 |
+
"learning_rate": 4.320855614973263e-06,
|
| 1549 |
+
"loss": 1.2974510192871094,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 2.3639282847203638,
|
| 1554 |
+
"grad_norm": 1.78125,
|
| 1555 |
+
"learning_rate": 4.249554367201426e-06,
|
| 1556 |
+
"loss": 1.303697681427002,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 2.374632057800375,
|
| 1561 |
+
"grad_norm": 1.765625,
|
| 1562 |
+
"learning_rate": 4.178253119429591e-06,
|
| 1563 |
+
"loss": 1.303341007232666,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 2.3853358308803854,
|
| 1568 |
+
"grad_norm": 1.46875,
|
| 1569 |
+
"learning_rate": 4.106951871657754e-06,
|
| 1570 |
+
"loss": 1.306796932220459,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 2.396039603960396,
|
| 1575 |
+
"grad_norm": 1.828125,
|
| 1576 |
+
"learning_rate": 4.035650623885918e-06,
|
| 1577 |
+
"loss": 1.3068408012390136,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 2.4067433770404065,
|
| 1582 |
+
"grad_norm": 1.671875,
|
| 1583 |
+
"learning_rate": 3.9643493761140825e-06,
|
| 1584 |
+
"loss": 1.307657527923584,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 2.4174471501204176,
|
| 1589 |
+
"grad_norm": 1.75,
|
| 1590 |
+
"learning_rate": 3.893048128342246e-06,
|
| 1591 |
+
"loss": 1.2932353973388673,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 2.428150923200428,
|
| 1596 |
+
"grad_norm": 1.765625,
|
| 1597 |
+
"learning_rate": 3.821746880570411e-06,
|
| 1598 |
+
"loss": 1.2625031471252441,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 2.4388546962804387,
|
| 1603 |
+
"grad_norm": 1.703125,
|
| 1604 |
+
"learning_rate": 3.7504456327985743e-06,
|
| 1605 |
+
"loss": 1.3354209899902343,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 2.4495584693604497,
|
| 1610 |
+
"grad_norm": 1.3671875,
|
| 1611 |
+
"learning_rate": 3.6791443850267383e-06,
|
| 1612 |
+
"loss": 1.200312042236328,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 2.4602622424404603,
|
| 1617 |
+
"grad_norm": 1.6484375,
|
| 1618 |
+
"learning_rate": 3.6078431372549024e-06,
|
| 1619 |
+
"loss": 1.2868337631225586,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 2.470966015520471,
|
| 1624 |
+
"grad_norm": 1.78125,
|
| 1625 |
+
"learning_rate": 3.536541889483066e-06,
|
| 1626 |
+
"loss": 1.2731021881103515,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 2.4816697886004815,
|
| 1631 |
+
"grad_norm": 2.265625,
|
| 1632 |
+
"learning_rate": 3.46524064171123e-06,
|
| 1633 |
+
"loss": 1.3145703315734862,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 2.4923735616804925,
|
| 1638 |
+
"grad_norm": 1.671875,
|
| 1639 |
+
"learning_rate": 3.3939393939393946e-06,
|
| 1640 |
+
"loss": 1.305215549468994,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 2.503077334760503,
|
| 1645 |
+
"grad_norm": 1.734375,
|
| 1646 |
+
"learning_rate": 3.322638146167558e-06,
|
| 1647 |
+
"loss": 1.3567096710205078,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 2.5137811078405137,
|
| 1652 |
+
"grad_norm": 1.6015625,
|
| 1653 |
+
"learning_rate": 3.2513368983957223e-06,
|
| 1654 |
+
"loss": 1.3081507682800293,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 2.5244848809205243,
|
| 1659 |
+
"grad_norm": 1.7109375,
|
| 1660 |
+
"learning_rate": 3.180035650623886e-06,
|
| 1661 |
+
"loss": 1.300461483001709,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 2.5351886540005353,
|
| 1666 |
+
"grad_norm": 1.5859375,
|
| 1667 |
+
"learning_rate": 3.10873440285205e-06,
|
| 1668 |
+
"loss": 1.3006972312927245,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 2.545892427080546,
|
| 1673 |
+
"grad_norm": 1.734375,
|
| 1674 |
+
"learning_rate": 3.0374331550802145e-06,
|
| 1675 |
+
"loss": 1.3157925605773926,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 2.5565962001605564,
|
| 1680 |
+
"grad_norm": 1.5625,
|
| 1681 |
+
"learning_rate": 2.966131907308378e-06,
|
| 1682 |
+
"loss": 1.2608634948730468,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 2.5672999732405675,
|
| 1687 |
+
"grad_norm": 1.765625,
|
| 1688 |
+
"learning_rate": 2.894830659536542e-06,
|
| 1689 |
+
"loss": 1.237275981903076,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 2.578003746320578,
|
| 1694 |
+
"grad_norm": 1.5859375,
|
| 1695 |
+
"learning_rate": 2.8235294117647062e-06,
|
| 1696 |
+
"loss": 1.274481964111328,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 2.5887075194005886,
|
| 1701 |
+
"grad_norm": 1.65625,
|
| 1702 |
+
"learning_rate": 2.75222816399287e-06,
|
| 1703 |
+
"loss": 1.3146997451782227,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 2.5994112924805997,
|
| 1708 |
+
"grad_norm": 1.6640625,
|
| 1709 |
+
"learning_rate": 2.680926916221034e-06,
|
| 1710 |
+
"loss": 1.3125761032104493,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 2.6101150655606102,
|
| 1715 |
+
"grad_norm": 1.6953125,
|
| 1716 |
+
"learning_rate": 2.6096256684491984e-06,
|
| 1717 |
+
"loss": 1.2919845581054688,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 2.620818838640621,
|
| 1722 |
+
"grad_norm": 1.6484375,
|
| 1723 |
+
"learning_rate": 2.538324420677362e-06,
|
| 1724 |
+
"loss": 1.2364542961120606,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 2.6315226117206314,
|
| 1729 |
+
"grad_norm": 1.609375,
|
| 1730 |
+
"learning_rate": 2.467023172905526e-06,
|
| 1731 |
+
"loss": 1.2624409675598145,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 2.642226384800642,
|
| 1736 |
+
"grad_norm": 1.5859375,
|
| 1737 |
+
"learning_rate": 2.3957219251336898e-06,
|
| 1738 |
+
"loss": 1.3075796127319337,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 2.652930157880653,
|
| 1743 |
+
"grad_norm": 2.109375,
|
| 1744 |
+
"learning_rate": 2.3244206773618542e-06,
|
| 1745 |
+
"loss": 1.2835824012756347,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 2.6636339309606636,
|
| 1750 |
+
"grad_norm": 1.8359375,
|
| 1751 |
+
"learning_rate": 2.253119429590018e-06,
|
| 1752 |
+
"loss": 1.307276153564453,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 2.674337704040674,
|
| 1757 |
+
"grad_norm": 1.7421875,
|
| 1758 |
+
"learning_rate": 2.181818181818182e-06,
|
| 1759 |
+
"loss": 1.2622486114501954,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 2.685041477120685,
|
| 1764 |
+
"grad_norm": 1.6796875,
|
| 1765 |
+
"learning_rate": 2.110516934046346e-06,
|
| 1766 |
+
"loss": 1.2514682769775392,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 2.6957452502006958,
|
| 1771 |
+
"grad_norm": 1.5625,
|
| 1772 |
+
"learning_rate": 2.03921568627451e-06,
|
| 1773 |
+
"loss": 1.2614849090576172,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 2.7064490232807064,
|
| 1778 |
+
"grad_norm": 1.5390625,
|
| 1779 |
+
"learning_rate": 1.9679144385026737e-06,
|
| 1780 |
+
"loss": 1.3241849899291993,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 2.7171527963607174,
|
| 1785 |
+
"grad_norm": 1.7890625,
|
| 1786 |
+
"learning_rate": 1.896613190730838e-06,
|
| 1787 |
+
"loss": 1.2841781616210937,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 2.727856569440728,
|
| 1792 |
+
"grad_norm": 1.8515625,
|
| 1793 |
+
"learning_rate": 1.8253119429590018e-06,
|
| 1794 |
+
"loss": 1.3062438011169433,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 2.7385603425207385,
|
| 1799 |
+
"grad_norm": 1.6328125,
|
| 1800 |
+
"learning_rate": 1.7540106951871661e-06,
|
| 1801 |
+
"loss": 1.3065251350402831,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 2.749264115600749,
|
| 1806 |
+
"grad_norm": 1.65625,
|
| 1807 |
+
"learning_rate": 1.68270944741533e-06,
|
| 1808 |
+
"loss": 1.2980451583862305,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 2.7599678886807597,
|
| 1813 |
+
"grad_norm": 1.671875,
|
| 1814 |
+
"learning_rate": 1.6114081996434938e-06,
|
| 1815 |
+
"loss": 1.2766281127929688,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 2.7706716617607707,
|
| 1820 |
+
"grad_norm": 1.5859375,
|
| 1821 |
+
"learning_rate": 1.5401069518716579e-06,
|
| 1822 |
+
"loss": 1.3033970832824706,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 2.7813754348407813,
|
| 1827 |
+
"grad_norm": 1.4921875,
|
| 1828 |
+
"learning_rate": 1.468805704099822e-06,
|
| 1829 |
+
"loss": 1.2335359573364257,
|
| 1830 |
+
"step": 2600
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 2.792079207920792,
|
| 1834 |
+
"grad_norm": 1.6484375,
|
| 1835 |
+
"learning_rate": 1.3975044563279858e-06,
|
| 1836 |
+
"loss": 1.3184511184692382,
|
| 1837 |
+
"step": 2610
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 2.802782981000803,
|
| 1841 |
+
"grad_norm": 1.609375,
|
| 1842 |
+
"learning_rate": 1.3262032085561499e-06,
|
| 1843 |
+
"loss": 1.1845362663269043,
|
| 1844 |
+
"step": 2620
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 2.8134867540808135,
|
| 1848 |
+
"grad_norm": 1.7890625,
|
| 1849 |
+
"learning_rate": 1.2549019607843137e-06,
|
| 1850 |
+
"loss": 1.2506700515747071,
|
| 1851 |
+
"step": 2630
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 2.824190527160824,
|
| 1855 |
+
"grad_norm": 1.7109375,
|
| 1856 |
+
"learning_rate": 1.1836007130124778e-06,
|
| 1857 |
+
"loss": 1.2360112190246582,
|
| 1858 |
+
"step": 2640
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 2.834894300240835,
|
| 1862 |
+
"grad_norm": 1.703125,
|
| 1863 |
+
"learning_rate": 1.1122994652406418e-06,
|
| 1864 |
+
"loss": 1.2875761032104491,
|
| 1865 |
+
"step": 2650
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 2.8455980733208457,
|
| 1869 |
+
"grad_norm": 1.578125,
|
| 1870 |
+
"learning_rate": 1.0409982174688057e-06,
|
| 1871 |
+
"loss": 1.2473506927490234,
|
| 1872 |
+
"step": 2660
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 2.8563018464008563,
|
| 1876 |
+
"grad_norm": 1.546875,
|
| 1877 |
+
"learning_rate": 9.696969696969698e-07,
|
| 1878 |
+
"loss": 1.3208060264587402,
|
| 1879 |
+
"step": 2670
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 2.867005619480867,
|
| 1883 |
+
"grad_norm": 1.671875,
|
| 1884 |
+
"learning_rate": 8.983957219251338e-07,
|
| 1885 |
+
"loss": 1.3371116638183593,
|
| 1886 |
+
"step": 2680
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 2.8777093925608774,
|
| 1890 |
+
"grad_norm": 1.484375,
|
| 1891 |
+
"learning_rate": 8.270944741532977e-07,
|
| 1892 |
+
"loss": 1.2605000495910645,
|
| 1893 |
+
"step": 2690
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 2.8884131656408885,
|
| 1897 |
+
"grad_norm": 1.6015625,
|
| 1898 |
+
"learning_rate": 7.557932263814617e-07,
|
| 1899 |
+
"loss": 1.267725658416748,
|
| 1900 |
+
"step": 2700
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 2.899116938720899,
|
| 1904 |
+
"grad_norm": 1.640625,
|
| 1905 |
+
"learning_rate": 6.844919786096257e-07,
|
| 1906 |
+
"loss": 1.27689208984375,
|
| 1907 |
+
"step": 2710
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 2.9098207118009096,
|
| 1911 |
+
"grad_norm": 1.578125,
|
| 1912 |
+
"learning_rate": 6.131907308377896e-07,
|
| 1913 |
+
"loss": 1.286923885345459,
|
| 1914 |
+
"step": 2720
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 2.9205244848809206,
|
| 1918 |
+
"grad_norm": 1.765625,
|
| 1919 |
+
"learning_rate": 5.418894830659537e-07,
|
| 1920 |
+
"loss": 1.330905055999756,
|
| 1921 |
+
"step": 2730
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 2.9312282579609312,
|
| 1925 |
+
"grad_norm": 1.71875,
|
| 1926 |
+
"learning_rate": 4.7058823529411767e-07,
|
| 1927 |
+
"loss": 1.2354840278625487,
|
| 1928 |
+
"step": 2740
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 2.941932031040942,
|
| 1932 |
+
"grad_norm": 1.4609375,
|
| 1933 |
+
"learning_rate": 3.992869875222817e-07,
|
| 1934 |
+
"loss": 1.2647834777832032,
|
| 1935 |
+
"step": 2750
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 2.952635804120953,
|
| 1939 |
+
"grad_norm": 1.6875,
|
| 1940 |
+
"learning_rate": 3.2798573975044564e-07,
|
| 1941 |
+
"loss": 1.2786317825317384,
|
| 1942 |
+
"step": 2760
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 2.9633395772009634,
|
| 1946 |
+
"grad_norm": 1.53125,
|
| 1947 |
+
"learning_rate": 2.5668449197860965e-07,
|
| 1948 |
+
"loss": 1.2594982147216798,
|
| 1949 |
+
"step": 2770
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 2.974043350280974,
|
| 1953 |
+
"grad_norm": 1.578125,
|
| 1954 |
+
"learning_rate": 1.8538324420677363e-07,
|
| 1955 |
+
"loss": 1.3203317642211914,
|
| 1956 |
+
"step": 2780
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 2.984747123360985,
|
| 1960 |
+
"grad_norm": 2.046875,
|
| 1961 |
+
"learning_rate": 1.1408199643493762e-07,
|
| 1962 |
+
"loss": 1.224764347076416,
|
| 1963 |
+
"step": 2790
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 2.9954508964409956,
|
| 1967 |
+
"grad_norm": 1.5078125,
|
| 1968 |
+
"learning_rate": 4.2780748663101606e-08,
|
| 1969 |
+
"loss": 1.2845193862915039,
|
| 1970 |
+
"step": 2800
|
| 1971 |
+
}
|
| 1972 |
+
],
|
| 1973 |
+
"logging_steps": 10,
|
| 1974 |
+
"max_steps": 2805,
|
| 1975 |
+
"num_input_tokens_seen": 0,
|
| 1976 |
+
"num_train_epochs": 3,
|
| 1977 |
+
"save_steps": 500,
|
| 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": true
|
| 1986 |
+
},
|
| 1987 |
+
"attributes": {}
|
| 1988 |
+
}
|
| 1989 |
+
},
|
| 1990 |
+
"total_flos": 7314356060356608.0,
|
| 1991 |
+
"train_batch_size": 2,
|
| 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 sha256:09cadf8b90b06130276c0937eec3cb26b58d80dacc8599fa833d2893aa4a490e
|
| 3 |
+
size 5137
|