Upload folder using huggingface_hub
Browse files- config.json +27 -0
- generation_config.json +7 -0
- model.safetensors +3 -0
- special_tokens_map.json +51 -0
- tokenizer.json +0 -0
- tokenizer_config.json +70 -0
- trainer_state.json +1572 -0
config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/nlsasfs/home/aipsc/myksingh/SIDDHESH/dpp/dpp/hindi_model/",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"MistralForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 1,
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
+
"hidden_act": "silu",
|
| 10 |
+
"hidden_size": 2048,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 7168,
|
| 13 |
+
"max_position_embeddings": 131072,
|
| 14 |
+
"model_type": "mistral",
|
| 15 |
+
"num_attention_heads": 32,
|
| 16 |
+
"num_hidden_layers": 16,
|
| 17 |
+
"num_key_value_heads": 8,
|
| 18 |
+
"pad_token_id": 0,
|
| 19 |
+
"rms_norm_eps": 1e-06,
|
| 20 |
+
"rope_theta": 10000.0,
|
| 21 |
+
"sliding_window": 512,
|
| 22 |
+
"tie_word_embeddings": false,
|
| 23 |
+
"torch_dtype": "float32",
|
| 24 |
+
"transformers_version": "4.41.1",
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 30000
|
| 27 |
+
}
|
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": 0,
|
| 6 |
+
"transformers_version": "4.41.1"
|
| 7 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d8a70f0d9e0f7f0e737284e8d60c3c64dc5e81280d2bd74d15c0d10ce0220ca1
|
| 3 |
+
size 3981468040
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<bos>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"cls_token": {
|
| 10 |
+
"content": "<cls>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"eos_token": {
|
| 17 |
+
"content": "<eos>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"mask_token": {
|
| 24 |
+
"content": "<mask>",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"pad_token": {
|
| 31 |
+
"content": "<pad>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
},
|
| 37 |
+
"sep_token": {
|
| 38 |
+
"content": "<sep>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false
|
| 43 |
+
},
|
| 44 |
+
"unk_token": {
|
| 45 |
+
"content": "<unk>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": false,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false
|
| 50 |
+
}
|
| 51 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,70 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<pad>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<cls>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "<sep>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "<mask>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"4": {
|
| 36 |
+
"content": "<unk>",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
},
|
| 43 |
+
"5": {
|
| 44 |
+
"content": "<bos>",
|
| 45 |
+
"lstrip": false,
|
| 46 |
+
"normalized": false,
|
| 47 |
+
"rstrip": false,
|
| 48 |
+
"single_word": false,
|
| 49 |
+
"special": true
|
| 50 |
+
},
|
| 51 |
+
"6": {
|
| 52 |
+
"content": "<eos>",
|
| 53 |
+
"lstrip": false,
|
| 54 |
+
"normalized": false,
|
| 55 |
+
"rstrip": false,
|
| 56 |
+
"single_word": false,
|
| 57 |
+
"special": true
|
| 58 |
+
}
|
| 59 |
+
},
|
| 60 |
+
"bos_token": "<bos>",
|
| 61 |
+
"clean_up_tokenization_spaces": true,
|
| 62 |
+
"cls_token": "<cls>",
|
| 63 |
+
"eos_token": "<eos>",
|
| 64 |
+
"mask_token": "<mask>",
|
| 65 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 66 |
+
"pad_token": "<pad>",
|
| 67 |
+
"sep_token": "<sep>",
|
| 68 |
+
"tokenizer_class": "PreTrainedTokenizerFast",
|
| 69 |
+
"unk_token": "<unk>"
|
| 70 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1572 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.47914153807761095,
|
| 5 |
+
"eval_steps": 3000,
|
| 6 |
+
"global_step": 10800,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.002218247861470421,
|
| 13 |
+
"grad_norm": 0.15810145437717438,
|
| 14 |
+
"learning_rate": 2e-05,
|
| 15 |
+
"loss": 2.8627,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.004436495722940842,
|
| 20 |
+
"grad_norm": 0.1590433567762375,
|
| 21 |
+
"learning_rate": 4e-05,
|
| 22 |
+
"loss": 2.8607,
|
| 23 |
+
"step": 100
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.006654743584411263,
|
| 27 |
+
"grad_norm": 0.15798641741275787,
|
| 28 |
+
"learning_rate": 6e-05,
|
| 29 |
+
"loss": 2.8623,
|
| 30 |
+
"step": 150
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.008872991445881684,
|
| 34 |
+
"grad_norm": 0.16127805411815643,
|
| 35 |
+
"learning_rate": 8e-05,
|
| 36 |
+
"loss": 2.8608,
|
| 37 |
+
"step": 200
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.011091239307352105,
|
| 41 |
+
"grad_norm": 0.1587396264076233,
|
| 42 |
+
"learning_rate": 0.0001,
|
| 43 |
+
"loss": 2.8608,
|
| 44 |
+
"step": 250
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.013309487168822525,
|
| 48 |
+
"grad_norm": 0.160736083984375,
|
| 49 |
+
"learning_rate": 0.00012,
|
| 50 |
+
"loss": 2.8563,
|
| 51 |
+
"step": 300
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.015527735030292948,
|
| 55 |
+
"grad_norm": 0.16256989538669586,
|
| 56 |
+
"learning_rate": 0.00014,
|
| 57 |
+
"loss": 2.8549,
|
| 58 |
+
"step": 350
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.01774598289176337,
|
| 62 |
+
"grad_norm": 0.16194568574428558,
|
| 63 |
+
"learning_rate": 0.00016,
|
| 64 |
+
"loss": 2.8557,
|
| 65 |
+
"step": 400
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.01996423075323379,
|
| 69 |
+
"grad_norm": 0.15836463868618011,
|
| 70 |
+
"learning_rate": 0.00018,
|
| 71 |
+
"loss": 2.8545,
|
| 72 |
+
"step": 450
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.02218247861470421,
|
| 76 |
+
"grad_norm": 0.16059577465057373,
|
| 77 |
+
"learning_rate": 0.0002,
|
| 78 |
+
"loss": 2.8522,
|
| 79 |
+
"step": 500
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.024400726476174632,
|
| 83 |
+
"grad_norm": 0.16031378507614136,
|
| 84 |
+
"learning_rate": 0.00022000000000000003,
|
| 85 |
+
"loss": 2.8481,
|
| 86 |
+
"step": 550
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.02661897433764505,
|
| 90 |
+
"grad_norm": 0.16000501811504364,
|
| 91 |
+
"learning_rate": 0.00024,
|
| 92 |
+
"loss": 2.8431,
|
| 93 |
+
"step": 600
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.028837222199115473,
|
| 97 |
+
"grad_norm": 0.15952646732330322,
|
| 98 |
+
"learning_rate": 0.00026000000000000003,
|
| 99 |
+
"loss": 2.8475,
|
| 100 |
+
"step": 650
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.031055470060585896,
|
| 104 |
+
"grad_norm": 0.16443726420402527,
|
| 105 |
+
"learning_rate": 0.00028,
|
| 106 |
+
"loss": 2.8452,
|
| 107 |
+
"step": 700
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.033273717922056315,
|
| 111 |
+
"grad_norm": 0.1644088476896286,
|
| 112 |
+
"learning_rate": 0.00030000000000000003,
|
| 113 |
+
"loss": 2.8458,
|
| 114 |
+
"step": 750
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.03549196578352674,
|
| 118 |
+
"grad_norm": 0.16272033751010895,
|
| 119 |
+
"learning_rate": 0.00032,
|
| 120 |
+
"loss": 2.8435,
|
| 121 |
+
"step": 800
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.03771021364499716,
|
| 125 |
+
"grad_norm": 0.16485804319381714,
|
| 126 |
+
"learning_rate": 0.00034,
|
| 127 |
+
"loss": 2.8481,
|
| 128 |
+
"step": 850
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.03992846150646758,
|
| 132 |
+
"grad_norm": 0.1669188290834427,
|
| 133 |
+
"learning_rate": 0.00036,
|
| 134 |
+
"loss": 2.8555,
|
| 135 |
+
"step": 900
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.042146709367938,
|
| 139 |
+
"grad_norm": 0.16288943588733673,
|
| 140 |
+
"learning_rate": 0.00038,
|
| 141 |
+
"loss": 2.851,
|
| 142 |
+
"step": 950
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.04436495722940842,
|
| 146 |
+
"grad_norm": 0.1651136726140976,
|
| 147 |
+
"learning_rate": 0.0004,
|
| 148 |
+
"loss": 2.8443,
|
| 149 |
+
"step": 1000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.04658320509087884,
|
| 153 |
+
"grad_norm": 0.16190673410892487,
|
| 154 |
+
"learning_rate": 0.00039999468202328424,
|
| 155 |
+
"loss": 2.8398,
|
| 156 |
+
"step": 1050
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.048801452952349264,
|
| 160 |
+
"grad_norm": 0.1649934947490692,
|
| 161 |
+
"learning_rate": 0.00039997872837594555,
|
| 162 |
+
"loss": 2.8371,
|
| 163 |
+
"step": 1100
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.051019700813819686,
|
| 167 |
+
"grad_norm": 0.16184477508068085,
|
| 168 |
+
"learning_rate": 0.00039995213990639536,
|
| 169 |
+
"loss": 2.8347,
|
| 170 |
+
"step": 1150
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.0532379486752901,
|
| 174 |
+
"grad_norm": 0.1629864126443863,
|
| 175 |
+
"learning_rate": 0.0003999149180286022,
|
| 176 |
+
"loss": 2.834,
|
| 177 |
+
"step": 1200
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.055456196536760524,
|
| 181 |
+
"grad_norm": 0.1627526730298996,
|
| 182 |
+
"learning_rate": 0.00039986706472201685,
|
| 183 |
+
"loss": 2.8309,
|
| 184 |
+
"step": 1250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.05767444439823095,
|
| 188 |
+
"grad_norm": 0.1642647087574005,
|
| 189 |
+
"learning_rate": 0.000399808582531467,
|
| 190 |
+
"loss": 2.8352,
|
| 191 |
+
"step": 1300
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.05989269225970137,
|
| 195 |
+
"grad_norm": 0.16397783160209656,
|
| 196 |
+
"learning_rate": 0.000399739474567022,
|
| 197 |
+
"loss": 2.8317,
|
| 198 |
+
"step": 1350
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.06211094012117179,
|
| 202 |
+
"grad_norm": 0.16319701075553894,
|
| 203 |
+
"learning_rate": 0.00039965974450382726,
|
| 204 |
+
"loss": 2.8322,
|
| 205 |
+
"step": 1400
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.06432918798264221,
|
| 209 |
+
"grad_norm": 0.16067005693912506,
|
| 210 |
+
"learning_rate": 0.000399569396581909,
|
| 211 |
+
"loss": 2.8279,
|
| 212 |
+
"step": 1450
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.06654743584411263,
|
| 216 |
+
"grad_norm": 0.16118553280830383,
|
| 217 |
+
"learning_rate": 0.00039946843560594866,
|
| 218 |
+
"loss": 2.8323,
|
| 219 |
+
"step": 1500
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.06876568370558306,
|
| 223 |
+
"grad_norm": 0.16291728615760803,
|
| 224 |
+
"learning_rate": 0.0003993568669450274,
|
| 225 |
+
"loss": 2.8301,
|
| 226 |
+
"step": 1550
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.07098393156705347,
|
| 230 |
+
"grad_norm": 0.1590035855770111,
|
| 231 |
+
"learning_rate": 0.0003992346965323407,
|
| 232 |
+
"loss": 2.8214,
|
| 233 |
+
"step": 1600
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.07320217942852389,
|
| 237 |
+
"grad_norm": 0.16236472129821777,
|
| 238 |
+
"learning_rate": 0.00039910193086488253,
|
| 239 |
+
"loss": 2.8242,
|
| 240 |
+
"step": 1650
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.07542042728999432,
|
| 244 |
+
"grad_norm": 0.1617489606142044,
|
| 245 |
+
"learning_rate": 0.0003989585770031003,
|
| 246 |
+
"loss": 2.8231,
|
| 247 |
+
"step": 1700
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.07763867515146473,
|
| 251 |
+
"grad_norm": 0.15960238873958588,
|
| 252 |
+
"learning_rate": 0.000398804642570519,
|
| 253 |
+
"loss": 2.8248,
|
| 254 |
+
"step": 1750
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.07985692301293516,
|
| 258 |
+
"grad_norm": 0.16391754150390625,
|
| 259 |
+
"learning_rate": 0.0003986401357533358,
|
| 260 |
+
"loss": 2.8222,
|
| 261 |
+
"step": 1800
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.08207517087440558,
|
| 265 |
+
"grad_norm": 0.16161847114562988,
|
| 266 |
+
"learning_rate": 0.000398465065299985,
|
| 267 |
+
"loss": 2.8153,
|
| 268 |
+
"step": 1850
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.084293418735876,
|
| 272 |
+
"grad_norm": 0.16447125375270844,
|
| 273 |
+
"learning_rate": 0.00039827944052067265,
|
| 274 |
+
"loss": 2.818,
|
| 275 |
+
"step": 1900
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.08651166659734642,
|
| 279 |
+
"grad_norm": 0.16384591162204742,
|
| 280 |
+
"learning_rate": 0.0003980832712868812,
|
| 281 |
+
"loss": 2.8093,
|
| 282 |
+
"step": 1950
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.08872991445881684,
|
| 286 |
+
"grad_norm": 0.16317427158355713,
|
| 287 |
+
"learning_rate": 0.0003978765680308447,
|
| 288 |
+
"loss": 2.8113,
|
| 289 |
+
"step": 2000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.09094816232028727,
|
| 293 |
+
"grad_norm": 0.16197824478149414,
|
| 294 |
+
"learning_rate": 0.00039765934174499436,
|
| 295 |
+
"loss": 2.8134,
|
| 296 |
+
"step": 2050
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.09316641018175768,
|
| 300 |
+
"grad_norm": 0.16196754574775696,
|
| 301 |
+
"learning_rate": 0.00039743160398137344,
|
| 302 |
+
"loss": 2.8147,
|
| 303 |
+
"step": 2100
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.0953846580432281,
|
| 307 |
+
"grad_norm": 0.16696424782276154,
|
| 308 |
+
"learning_rate": 0.00039719336685102314,
|
| 309 |
+
"loss": 2.811,
|
| 310 |
+
"step": 2150
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.09760290590469853,
|
| 314 |
+
"grad_norm": 0.16266262531280518,
|
| 315 |
+
"learning_rate": 0.0003969446430233386,
|
| 316 |
+
"loss": 2.8103,
|
| 317 |
+
"step": 2200
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.09982115376616894,
|
| 321 |
+
"grad_norm": 0.16161397099494934,
|
| 322 |
+
"learning_rate": 0.0003966854457253951,
|
| 323 |
+
"loss": 2.8017,
|
| 324 |
+
"step": 2250
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.10203940162763937,
|
| 328 |
+
"grad_norm": 0.1631053388118744,
|
| 329 |
+
"learning_rate": 0.0003964157887412445,
|
| 330 |
+
"loss": 2.8034,
|
| 331 |
+
"step": 2300
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.10425764948910979,
|
| 335 |
+
"grad_norm": 0.16185788810253143,
|
| 336 |
+
"learning_rate": 0.00039613568641118255,
|
| 337 |
+
"loss": 2.8027,
|
| 338 |
+
"step": 2350
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.1064758973505802,
|
| 342 |
+
"grad_norm": 0.16428661346435547,
|
| 343 |
+
"learning_rate": 0.00039584515363098584,
|
| 344 |
+
"loss": 2.8031,
|
| 345 |
+
"step": 2400
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.10869414521205063,
|
| 349 |
+
"grad_norm": 0.1625480055809021,
|
| 350 |
+
"learning_rate": 0.00039554420585112,
|
| 351 |
+
"loss": 2.7968,
|
| 352 |
+
"step": 2450
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.11091239307352105,
|
| 356 |
+
"grad_norm": 0.1638619303703308,
|
| 357 |
+
"learning_rate": 0.0003952328590759179,
|
| 358 |
+
"loss": 2.8007,
|
| 359 |
+
"step": 2500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.11313064093499148,
|
| 363 |
+
"grad_norm": 0.16504357755184174,
|
| 364 |
+
"learning_rate": 0.0003949111298627286,
|
| 365 |
+
"loss": 2.7921,
|
| 366 |
+
"step": 2550
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.1153488887964619,
|
| 370 |
+
"grad_norm": 0.16429375112056732,
|
| 371 |
+
"learning_rate": 0.0003945790353210367,
|
| 372 |
+
"loss": 2.7951,
|
| 373 |
+
"step": 2600
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.11756713665793232,
|
| 377 |
+
"grad_norm": 0.166097030043602,
|
| 378 |
+
"learning_rate": 0.0003942365931115526,
|
| 379 |
+
"loss": 2.7948,
|
| 380 |
+
"step": 2650
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.11978538451940274,
|
| 384 |
+
"grad_norm": 0.16275139153003693,
|
| 385 |
+
"learning_rate": 0.0003938838214452733,
|
| 386 |
+
"loss": 2.79,
|
| 387 |
+
"step": 2700
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.12200363238087315,
|
| 391 |
+
"grad_norm": 0.16379590332508087,
|
| 392 |
+
"learning_rate": 0.0003935207390825137,
|
| 393 |
+
"loss": 2.7896,
|
| 394 |
+
"step": 2750
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.12422188024234358,
|
| 398 |
+
"grad_norm": 0.16332408785820007,
|
| 399 |
+
"learning_rate": 0.0003931473653319095,
|
| 400 |
+
"loss": 2.7848,
|
| 401 |
+
"step": 2800
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.126440128103814,
|
| 405 |
+
"grad_norm": 0.16235879063606262,
|
| 406 |
+
"learning_rate": 0.00039276372004938987,
|
| 407 |
+
"loss": 2.7836,
|
| 408 |
+
"step": 2850
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.12865837596528443,
|
| 412 |
+
"grad_norm": 0.1654053032398224,
|
| 413 |
+
"learning_rate": 0.00039236982363712145,
|
| 414 |
+
"loss": 2.7845,
|
| 415 |
+
"step": 2900
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.13087662382675483,
|
| 419 |
+
"grad_norm": 0.16393068432807922,
|
| 420 |
+
"learning_rate": 0.00039196569704242376,
|
| 421 |
+
"loss": 2.7796,
|
| 422 |
+
"step": 2950
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.13309487168822526,
|
| 426 |
+
"grad_norm": 0.16517628729343414,
|
| 427 |
+
"learning_rate": 0.0003915513617566551,
|
| 428 |
+
"loss": 2.7738,
|
| 429 |
+
"step": 3000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.13309487168822526,
|
| 433 |
+
"eval_accuracy": 0.4247227650219834,
|
| 434 |
+
"eval_loss": 2.8996665477752686,
|
| 435 |
+
"eval_runtime": 243.2366,
|
| 436 |
+
"eval_samples_per_second": 8.222,
|
| 437 |
+
"eval_steps_per_second": 1.028,
|
| 438 |
+
"step": 3000
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 0.1353131195496957,
|
| 442 |
+
"grad_norm": 0.16540038585662842,
|
| 443 |
+
"learning_rate": 0.00039112683981406936,
|
| 444 |
+
"loss": 2.7708,
|
| 445 |
+
"step": 3050
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 0.13753136741116612,
|
| 449 |
+
"grad_norm": 0.16403205692768097,
|
| 450 |
+
"learning_rate": 0.00039069215379064465,
|
| 451 |
+
"loss": 2.7709,
|
| 452 |
+
"step": 3100
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.13974961527263652,
|
| 456 |
+
"grad_norm": 0.16498889029026031,
|
| 457 |
+
"learning_rate": 0.0003902473268028826,
|
| 458 |
+
"loss": 2.7683,
|
| 459 |
+
"step": 3150
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.14196786313410695,
|
| 463 |
+
"grad_norm": 0.16713927686214447,
|
| 464 |
+
"learning_rate": 0.00038979238250657863,
|
| 465 |
+
"loss": 2.7578,
|
| 466 |
+
"step": 3200
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.14418611099557738,
|
| 470 |
+
"grad_norm": 0.16905058920383453,
|
| 471 |
+
"learning_rate": 0.00038932734509556467,
|
| 472 |
+
"loss": 2.7602,
|
| 473 |
+
"step": 3250
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 0.14640435885704778,
|
| 477 |
+
"grad_norm": 0.16431044042110443,
|
| 478 |
+
"learning_rate": 0.0003888522393004219,
|
| 479 |
+
"loss": 2.7685,
|
| 480 |
+
"step": 3300
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.1486226067185182,
|
| 484 |
+
"grad_norm": 0.163705512881279,
|
| 485 |
+
"learning_rate": 0.00038836709038716583,
|
| 486 |
+
"loss": 2.8434,
|
| 487 |
+
"step": 3350
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 0.15084085457998864,
|
| 491 |
+
"grad_norm": 0.1622520387172699,
|
| 492 |
+
"learning_rate": 0.0003878719241559027,
|
| 493 |
+
"loss": 2.8349,
|
| 494 |
+
"step": 3400
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.15305910244145907,
|
| 498 |
+
"grad_norm": 0.16072827577590942,
|
| 499 |
+
"learning_rate": 0.00038736676693945746,
|
| 500 |
+
"loss": 2.8369,
|
| 501 |
+
"step": 3450
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 0.15527735030292947,
|
| 505 |
+
"grad_norm": 0.16206832230091095,
|
| 506 |
+
"learning_rate": 0.0003868516456019733,
|
| 507 |
+
"loss": 2.8404,
|
| 508 |
+
"step": 3500
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 0.1574955981643999,
|
| 512 |
+
"grad_norm": 0.16148249804973602,
|
| 513 |
+
"learning_rate": 0.0003863265875374829,
|
| 514 |
+
"loss": 2.836,
|
| 515 |
+
"step": 3550
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 0.15971384602587033,
|
| 519 |
+
"grad_norm": 0.16401226818561554,
|
| 520 |
+
"learning_rate": 0.0003857916206684519,
|
| 521 |
+
"loss": 2.8369,
|
| 522 |
+
"step": 3600
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 0.16193209388734073,
|
| 526 |
+
"grad_norm": 0.15987250208854675,
|
| 527 |
+
"learning_rate": 0.00038524677344429386,
|
| 528 |
+
"loss": 2.8363,
|
| 529 |
+
"step": 3650
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.16415034174881116,
|
| 533 |
+
"grad_norm": 0.16117645800113678,
|
| 534 |
+
"learning_rate": 0.00038469207483985725,
|
| 535 |
+
"loss": 2.8426,
|
| 536 |
+
"step": 3700
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.1663685896102816,
|
| 540 |
+
"grad_norm": 0.16374363005161285,
|
| 541 |
+
"learning_rate": 0.00038412755435388474,
|
| 542 |
+
"loss": 2.8416,
|
| 543 |
+
"step": 3750
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 0.168586837471752,
|
| 547 |
+
"grad_norm": 0.16495129466056824,
|
| 548 |
+
"learning_rate": 0.0003835532420074444,
|
| 549 |
+
"loss": 2.8396,
|
| 550 |
+
"step": 3800
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 0.17080508533322242,
|
| 554 |
+
"grad_norm": 0.16315814852714539,
|
| 555 |
+
"learning_rate": 0.0003829691683423329,
|
| 556 |
+
"loss": 2.8358,
|
| 557 |
+
"step": 3850
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 0.17302333319469285,
|
| 561 |
+
"grad_norm": 0.16098596155643463,
|
| 562 |
+
"learning_rate": 0.00038237536441945193,
|
| 563 |
+
"loss": 2.8354,
|
| 564 |
+
"step": 3900
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.17524158105616328,
|
| 568 |
+
"grad_norm": 0.16226187348365784,
|
| 569 |
+
"learning_rate": 0.00038177186181715577,
|
| 570 |
+
"loss": 2.8352,
|
| 571 |
+
"step": 3950
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 0.17745982891763368,
|
| 575 |
+
"grad_norm": 0.15939714014530182,
|
| 576 |
+
"learning_rate": 0.00038115869262957233,
|
| 577 |
+
"loss": 2.835,
|
| 578 |
+
"step": 4000
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.1796780767791041,
|
| 582 |
+
"grad_norm": 0.1622256189584732,
|
| 583 |
+
"learning_rate": 0.00038053588946489615,
|
| 584 |
+
"loss": 2.8391,
|
| 585 |
+
"step": 4050
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.18189632464057454,
|
| 589 |
+
"grad_norm": 0.16176052391529083,
|
| 590 |
+
"learning_rate": 0.0003799034854436545,
|
| 591 |
+
"loss": 2.8371,
|
| 592 |
+
"step": 4100
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 0.18411457250204494,
|
| 596 |
+
"grad_norm": 0.16316720843315125,
|
| 597 |
+
"learning_rate": 0.0003792615141969462,
|
| 598 |
+
"loss": 2.8365,
|
| 599 |
+
"step": 4150
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 0.18633282036351537,
|
| 603 |
+
"grad_norm": 0.16207610070705414,
|
| 604 |
+
"learning_rate": 0.0003786100098646524,
|
| 605 |
+
"loss": 2.8346,
|
| 606 |
+
"step": 4200
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 0.1885510682249858,
|
| 610 |
+
"grad_norm": 0.1638234406709671,
|
| 611 |
+
"learning_rate": 0.000377949007093622,
|
| 612 |
+
"loss": 2.8319,
|
| 613 |
+
"step": 4250
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 0.1907693160864562,
|
| 617 |
+
"grad_norm": 0.1628599315881729,
|
| 618 |
+
"learning_rate": 0.0003772785410358283,
|
| 619 |
+
"loss": 2.8369,
|
| 620 |
+
"step": 4300
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 0.19298756394792663,
|
| 624 |
+
"grad_norm": 0.1653933823108673,
|
| 625 |
+
"learning_rate": 0.00037659864734650026,
|
| 626 |
+
"loss": 2.8304,
|
| 627 |
+
"step": 4350
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 0.19520581180939706,
|
| 631 |
+
"grad_norm": 0.16370812058448792,
|
| 632 |
+
"learning_rate": 0.0003759093621822259,
|
| 633 |
+
"loss": 2.8369,
|
| 634 |
+
"step": 4400
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 0.19742405967086749,
|
| 638 |
+
"grad_norm": 0.1629050225019455,
|
| 639 |
+
"learning_rate": 0.0003752107221990298,
|
| 640 |
+
"loss": 2.8339,
|
| 641 |
+
"step": 4450
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 0.1996423075323379,
|
| 645 |
+
"grad_norm": 0.1610899269580841,
|
| 646 |
+
"learning_rate": 0.00037450276455042354,
|
| 647 |
+
"loss": 2.829,
|
| 648 |
+
"step": 4500
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 0.20186055539380832,
|
| 652 |
+
"grad_norm": 0.1629941165447235,
|
| 653 |
+
"learning_rate": 0.00037378552688543005,
|
| 654 |
+
"loss": 2.8351,
|
| 655 |
+
"step": 4550
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 0.20407880325527875,
|
| 659 |
+
"grad_norm": 0.16439735889434814,
|
| 660 |
+
"learning_rate": 0.0003730590473465814,
|
| 661 |
+
"loss": 2.8316,
|
| 662 |
+
"step": 4600
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 0.20629705111674915,
|
| 666 |
+
"grad_norm": 0.16572241485118866,
|
| 667 |
+
"learning_rate": 0.00037232336456789023,
|
| 668 |
+
"loss": 2.8335,
|
| 669 |
+
"step": 4650
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 0.20851529897821958,
|
| 673 |
+
"grad_norm": 0.16380038857460022,
|
| 674 |
+
"learning_rate": 0.00037157851767279543,
|
| 675 |
+
"loss": 2.8286,
|
| 676 |
+
"step": 4700
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 0.21073354683969,
|
| 680 |
+
"grad_norm": 0.16284549236297607,
|
| 681 |
+
"learning_rate": 0.00037082454627208156,
|
| 682 |
+
"loss": 2.8301,
|
| 683 |
+
"step": 4750
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 0.2129517947011604,
|
| 687 |
+
"grad_norm": 0.16597482562065125,
|
| 688 |
+
"learning_rate": 0.0003700614904617721,
|
| 689 |
+
"loss": 2.8323,
|
| 690 |
+
"step": 4800
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 0.21517004256263084,
|
| 694 |
+
"grad_norm": 0.16378666460514069,
|
| 695 |
+
"learning_rate": 0.0003692893908209973,
|
| 696 |
+
"loss": 2.8299,
|
| 697 |
+
"step": 4850
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 0.21738829042410127,
|
| 701 |
+
"grad_norm": 0.1630343496799469,
|
| 702 |
+
"learning_rate": 0.0003685082884098363,
|
| 703 |
+
"loss": 2.8333,
|
| 704 |
+
"step": 4900
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.2196065382855717,
|
| 708 |
+
"grad_norm": 0.16490814089775085,
|
| 709 |
+
"learning_rate": 0.00036771822476713346,
|
| 710 |
+
"loss": 2.8307,
|
| 711 |
+
"step": 4950
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 0.2218247861470421,
|
| 715 |
+
"grad_norm": 0.1655721366405487,
|
| 716 |
+
"learning_rate": 0.00036691924190828935,
|
| 717 |
+
"loss": 2.8301,
|
| 718 |
+
"step": 5000
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 0.22404303400851253,
|
| 722 |
+
"grad_norm": 0.16776245832443237,
|
| 723 |
+
"learning_rate": 0.0003661113823230264,
|
| 724 |
+
"loss": 2.8228,
|
| 725 |
+
"step": 5050
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 0.22626128186998296,
|
| 729 |
+
"grad_norm": 0.1626349687576294,
|
| 730 |
+
"learning_rate": 0.00036529468897312926,
|
| 731 |
+
"loss": 2.8262,
|
| 732 |
+
"step": 5100
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 0.22847952973145336,
|
| 736 |
+
"grad_norm": 0.16331753134727478,
|
| 737 |
+
"learning_rate": 0.00036446920529016,
|
| 738 |
+
"loss": 2.8282,
|
| 739 |
+
"step": 5150
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 0.2306977775929238,
|
| 743 |
+
"grad_norm": 0.1676001250743866,
|
| 744 |
+
"learning_rate": 0.00036363497517314877,
|
| 745 |
+
"loss": 2.8313,
|
| 746 |
+
"step": 5200
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.23291602545439422,
|
| 750 |
+
"grad_norm": 0.16441357135772705,
|
| 751 |
+
"learning_rate": 0.000362792042986259,
|
| 752 |
+
"loss": 2.8278,
|
| 753 |
+
"step": 5250
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.23513427331586464,
|
| 757 |
+
"grad_norm": 0.16601622104644775,
|
| 758 |
+
"learning_rate": 0.000361940453556428,
|
| 759 |
+
"loss": 2.8303,
|
| 760 |
+
"step": 5300
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 0.23735252117733505,
|
| 764 |
+
"grad_norm": 0.1679011583328247,
|
| 765 |
+
"learning_rate": 0.0003610802521709833,
|
| 766 |
+
"loss": 2.8252,
|
| 767 |
+
"step": 5350
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 0.23957076903880548,
|
| 771 |
+
"grad_norm": 0.1650955229997635,
|
| 772 |
+
"learning_rate": 0.0003602114845752345,
|
| 773 |
+
"loss": 2.8299,
|
| 774 |
+
"step": 5400
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 0.2417890169002759,
|
| 778 |
+
"grad_norm": 0.16651777923107147,
|
| 779 |
+
"learning_rate": 0.00035933419697004,
|
| 780 |
+
"loss": 2.832,
|
| 781 |
+
"step": 5450
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 0.2440072647617463,
|
| 785 |
+
"grad_norm": 0.166709303855896,
|
| 786 |
+
"learning_rate": 0.00035844843600935024,
|
| 787 |
+
"loss": 2.8262,
|
| 788 |
+
"step": 5500
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 0.24622551262321674,
|
| 792 |
+
"grad_norm": 0.16586416959762573,
|
| 793 |
+
"learning_rate": 0.000357554248797727,
|
| 794 |
+
"loss": 2.8255,
|
| 795 |
+
"step": 5550
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 0.24844376048468717,
|
| 799 |
+
"grad_norm": 0.1647614985704422,
|
| 800 |
+
"learning_rate": 0.00035665168288783795,
|
| 801 |
+
"loss": 2.8298,
|
| 802 |
+
"step": 5600
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 0.2506620083461576,
|
| 806 |
+
"grad_norm": 0.16310204565525055,
|
| 807 |
+
"learning_rate": 0.000355740786277928,
|
| 808 |
+
"loss": 2.8273,
|
| 809 |
+
"step": 5650
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 0.252880256207628,
|
| 813 |
+
"grad_norm": 0.1629767119884491,
|
| 814 |
+
"learning_rate": 0.00035482160740926683,
|
| 815 |
+
"loss": 2.8231,
|
| 816 |
+
"step": 5700
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 0.2550985040690984,
|
| 820 |
+
"grad_norm": 0.16427451372146606,
|
| 821 |
+
"learning_rate": 0.00035389419516357253,
|
| 822 |
+
"loss": 2.8188,
|
| 823 |
+
"step": 5750
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 0.25731675193056885,
|
| 827 |
+
"grad_norm": 0.1655891388654709,
|
| 828 |
+
"learning_rate": 0.0003529585988604125,
|
| 829 |
+
"loss": 2.8258,
|
| 830 |
+
"step": 5800
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 0.25953499979203926,
|
| 834 |
+
"grad_norm": 0.16402335464954376,
|
| 835 |
+
"learning_rate": 0.0003520148682545803,
|
| 836 |
+
"loss": 2.8254,
|
| 837 |
+
"step": 5850
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 0.26175324765350966,
|
| 841 |
+
"grad_norm": 0.1638861894607544,
|
| 842 |
+
"learning_rate": 0.0003510630535334497,
|
| 843 |
+
"loss": 2.8298,
|
| 844 |
+
"step": 5900
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 0.2639714955149801,
|
| 848 |
+
"grad_norm": 0.16864845156669617,
|
| 849 |
+
"learning_rate": 0.0003501032053143061,
|
| 850 |
+
"loss": 2.8238,
|
| 851 |
+
"step": 5950
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 0.2661897433764505,
|
| 855 |
+
"grad_norm": 0.16578635573387146,
|
| 856 |
+
"learning_rate": 0.0003491353746416541,
|
| 857 |
+
"loss": 2.8225,
|
| 858 |
+
"step": 6000
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 0.2661897433764505,
|
| 862 |
+
"eval_accuracy": 0.4264626282364436,
|
| 863 |
+
"eval_loss": 2.8843319416046143,
|
| 864 |
+
"eval_runtime": 242.3694,
|
| 865 |
+
"eval_samples_per_second": 8.252,
|
| 866 |
+
"eval_steps_per_second": 1.031,
|
| 867 |
+
"step": 6000
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 0.268407991237921,
|
| 871 |
+
"grad_norm": 0.16673897206783295,
|
| 872 |
+
"learning_rate": 0.00034815961298450377,
|
| 873 |
+
"loss": 2.823,
|
| 874 |
+
"step": 6050
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 0.2706262390993914,
|
| 878 |
+
"grad_norm": 0.16588376462459564,
|
| 879 |
+
"learning_rate": 0.0003471759722336326,
|
| 880 |
+
"loss": 2.8193,
|
| 881 |
+
"step": 6100
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 0.2728444869608618,
|
| 885 |
+
"grad_norm": 0.16813361644744873,
|
| 886 |
+
"learning_rate": 0.00034618450469882687,
|
| 887 |
+
"loss": 2.8267,
|
| 888 |
+
"step": 6150
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 0.27506273482233223,
|
| 892 |
+
"grad_norm": 0.16656942665576935,
|
| 893 |
+
"learning_rate": 0.0003451852631060991,
|
| 894 |
+
"loss": 2.8219,
|
| 895 |
+
"step": 6200
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 0.27728098268380263,
|
| 899 |
+
"grad_norm": 0.1666443794965744,
|
| 900 |
+
"learning_rate": 0.0003441783005948846,
|
| 901 |
+
"loss": 2.8233,
|
| 902 |
+
"step": 6250
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 0.27949923054527304,
|
| 906 |
+
"grad_norm": 0.1673704832792282,
|
| 907 |
+
"learning_rate": 0.0003431636707152152,
|
| 908 |
+
"loss": 2.824,
|
| 909 |
+
"step": 6300
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 0.2817174784067435,
|
| 913 |
+
"grad_norm": 0.16707104444503784,
|
| 914 |
+
"learning_rate": 0.00034214142742487177,
|
| 915 |
+
"loss": 2.8221,
|
| 916 |
+
"step": 6350
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 0.2839357262682139,
|
| 920 |
+
"grad_norm": 0.16775397956371307,
|
| 921 |
+
"learning_rate": 0.0003411116250865143,
|
| 922 |
+
"loss": 2.8234,
|
| 923 |
+
"step": 6400
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 0.2861539741296843,
|
| 927 |
+
"grad_norm": 0.16813720762729645,
|
| 928 |
+
"learning_rate": 0.0003400743184647915,
|
| 929 |
+
"loss": 2.8258,
|
| 930 |
+
"step": 6450
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 0.28837222199115475,
|
| 934 |
+
"grad_norm": 0.16362161934375763,
|
| 935 |
+
"learning_rate": 0.00033902956272342783,
|
| 936 |
+
"loss": 2.8232,
|
| 937 |
+
"step": 6500
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 0.29059046985262516,
|
| 941 |
+
"grad_norm": 0.16950780153274536,
|
| 942 |
+
"learning_rate": 0.00033797741342229054,
|
| 943 |
+
"loss": 2.821,
|
| 944 |
+
"step": 6550
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 0.29280871771409556,
|
| 948 |
+
"grad_norm": 0.1657160073518753,
|
| 949 |
+
"learning_rate": 0.00033691792651443435,
|
| 950 |
+
"loss": 2.8181,
|
| 951 |
+
"step": 6600
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 0.295026965575566,
|
| 955 |
+
"grad_norm": 0.1689310073852539,
|
| 956 |
+
"learning_rate": 0.0003358511583431264,
|
| 957 |
+
"loss": 2.8257,
|
| 958 |
+
"step": 6650
|
| 959 |
+
},
|
| 960 |
+
{
|
| 961 |
+
"epoch": 0.2972452134370364,
|
| 962 |
+
"grad_norm": 0.16674135625362396,
|
| 963 |
+
"learning_rate": 0.00033477716563884956,
|
| 964 |
+
"loss": 2.8209,
|
| 965 |
+
"step": 6700
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 0.2994634612985068,
|
| 969 |
+
"grad_norm": 0.16600748896598816,
|
| 970 |
+
"learning_rate": 0.00033369600551628586,
|
| 971 |
+
"loss": 2.8227,
|
| 972 |
+
"step": 6750
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 0.3016817091599773,
|
| 976 |
+
"grad_norm": 0.16666853427886963,
|
| 977 |
+
"learning_rate": 0.0003326077354712789,
|
| 978 |
+
"loss": 2.8199,
|
| 979 |
+
"step": 6800
|
| 980 |
+
},
|
| 981 |
+
{
|
| 982 |
+
"epoch": 0.3038999570214477,
|
| 983 |
+
"grad_norm": 0.1671936959028244,
|
| 984 |
+
"learning_rate": 0.00033151241337777624,
|
| 985 |
+
"loss": 2.82,
|
| 986 |
+
"step": 6850
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 0.30611820488291813,
|
| 990 |
+
"grad_norm": 0.1675061583518982,
|
| 991 |
+
"learning_rate": 0.00033041009748475166,
|
| 992 |
+
"loss": 2.8246,
|
| 993 |
+
"step": 6900
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 0.30833645274438853,
|
| 997 |
+
"grad_norm": 0.16512750089168549,
|
| 998 |
+
"learning_rate": 0.0003293008464131079,
|
| 999 |
+
"loss": 2.8178,
|
| 1000 |
+
"step": 6950
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 0.31055470060585894,
|
| 1004 |
+
"grad_norm": 0.1670486181974411,
|
| 1005 |
+
"learning_rate": 0.0003281847191525585,
|
| 1006 |
+
"loss": 2.8185,
|
| 1007 |
+
"step": 7000
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 0.3127729484673294,
|
| 1011 |
+
"grad_norm": 0.1692744940519333,
|
| 1012 |
+
"learning_rate": 0.0003270617750584913,
|
| 1013 |
+
"loss": 2.8184,
|
| 1014 |
+
"step": 7050
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 0.3149911963287998,
|
| 1018 |
+
"grad_norm": 0.16573506593704224,
|
| 1019 |
+
"learning_rate": 0.0003259320738488119,
|
| 1020 |
+
"loss": 2.823,
|
| 1021 |
+
"step": 7100
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 0.3172094441902702,
|
| 1025 |
+
"grad_norm": 0.17004618048667908,
|
| 1026 |
+
"learning_rate": 0.00032479567560076745,
|
| 1027 |
+
"loss": 2.8174,
|
| 1028 |
+
"step": 7150
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 0.31942769205174065,
|
| 1032 |
+
"grad_norm": 0.16867642104625702,
|
| 1033 |
+
"learning_rate": 0.00032365264074775223,
|
| 1034 |
+
"loss": 2.8183,
|
| 1035 |
+
"step": 7200
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 0.32164593991321105,
|
| 1039 |
+
"grad_norm": 0.16543437540531158,
|
| 1040 |
+
"learning_rate": 0.00032250303007609366,
|
| 1041 |
+
"loss": 2.8178,
|
| 1042 |
+
"step": 7250
|
| 1043 |
+
},
|
| 1044 |
+
{
|
| 1045 |
+
"epoch": 0.32386418777468146,
|
| 1046 |
+
"grad_norm": 0.16606374084949493,
|
| 1047 |
+
"learning_rate": 0.0003213469047218194,
|
| 1048 |
+
"loss": 2.8182,
|
| 1049 |
+
"step": 7300
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 0.3260824356361519,
|
| 1053 |
+
"grad_norm": 0.1708928942680359,
|
| 1054 |
+
"learning_rate": 0.0003201843261674067,
|
| 1055 |
+
"loss": 2.8194,
|
| 1056 |
+
"step": 7350
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 0.3283006834976223,
|
| 1060 |
+
"grad_norm": 0.16661237180233002,
|
| 1061 |
+
"learning_rate": 0.00031901535623851245,
|
| 1062 |
+
"loss": 2.8226,
|
| 1063 |
+
"step": 7400
|
| 1064 |
+
},
|
| 1065 |
+
{
|
| 1066 |
+
"epoch": 0.3305189313590927,
|
| 1067 |
+
"grad_norm": 0.16710756719112396,
|
| 1068 |
+
"learning_rate": 0.0003178400571006852,
|
| 1069 |
+
"loss": 2.8187,
|
| 1070 |
+
"step": 7450
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 0.3327371792205632,
|
| 1074 |
+
"grad_norm": 0.16679760813713074,
|
| 1075 |
+
"learning_rate": 0.00031665849125605937,
|
| 1076 |
+
"loss": 2.8163,
|
| 1077 |
+
"step": 7500
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 0.3349554270820336,
|
| 1081 |
+
"grad_norm": 0.16872857511043549,
|
| 1082 |
+
"learning_rate": 0.00031547072154003154,
|
| 1083 |
+
"loss": 2.8147,
|
| 1084 |
+
"step": 7550
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 0.337173674943504,
|
| 1088 |
+
"grad_norm": 0.1672954261302948,
|
| 1089 |
+
"learning_rate": 0.0003142768111179187,
|
| 1090 |
+
"loss": 2.8167,
|
| 1091 |
+
"step": 7600
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 0.33939192280497443,
|
| 1095 |
+
"grad_norm": 0.16654394567012787,
|
| 1096 |
+
"learning_rate": 0.00031307682348159907,
|
| 1097 |
+
"loss": 2.816,
|
| 1098 |
+
"step": 7650
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 0.34161017066644483,
|
| 1102 |
+
"grad_norm": 0.16810841858386993,
|
| 1103 |
+
"learning_rate": 0.00031187082244613567,
|
| 1104 |
+
"loss": 2.8139,
|
| 1105 |
+
"step": 7700
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 0.34382841852791524,
|
| 1109 |
+
"grad_norm": 0.1682497262954712,
|
| 1110 |
+
"learning_rate": 0.00031065887214638284,
|
| 1111 |
+
"loss": 2.8157,
|
| 1112 |
+
"step": 7750
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 0.3460466663893857,
|
| 1116 |
+
"grad_norm": 0.17154847085475922,
|
| 1117 |
+
"learning_rate": 0.00030944103703357524,
|
| 1118 |
+
"loss": 2.8143,
|
| 1119 |
+
"step": 7800
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 0.3482649142508561,
|
| 1123 |
+
"grad_norm": 0.16658836603164673,
|
| 1124 |
+
"learning_rate": 0.00030821738187190075,
|
| 1125 |
+
"loss": 2.8143,
|
| 1126 |
+
"step": 7850
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"epoch": 0.35048316211232655,
|
| 1130 |
+
"grad_norm": 0.16820305585861206,
|
| 1131 |
+
"learning_rate": 0.00030698797173505586,
|
| 1132 |
+
"loss": 2.8157,
|
| 1133 |
+
"step": 7900
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"epoch": 0.35270140997379695,
|
| 1137 |
+
"grad_norm": 0.16843385994434357,
|
| 1138 |
+
"learning_rate": 0.0003057528720027853,
|
| 1139 |
+
"loss": 2.8103,
|
| 1140 |
+
"step": 7950
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 0.35491965783526735,
|
| 1144 |
+
"grad_norm": 0.17145898938179016,
|
| 1145 |
+
"learning_rate": 0.0003045121483574054,
|
| 1146 |
+
"loss": 2.8161,
|
| 1147 |
+
"step": 8000
|
| 1148 |
+
},
|
| 1149 |
+
{
|
| 1150 |
+
"epoch": 0.3571379056967378,
|
| 1151 |
+
"grad_norm": 0.1709701269865036,
|
| 1152 |
+
"learning_rate": 0.00030326586678031066,
|
| 1153 |
+
"loss": 2.8134,
|
| 1154 |
+
"step": 8050
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 0.3593561535582082,
|
| 1158 |
+
"grad_norm": 0.16859866678714752,
|
| 1159 |
+
"learning_rate": 0.0003020140935484653,
|
| 1160 |
+
"loss": 2.818,
|
| 1161 |
+
"step": 8100
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 0.3615744014196786,
|
| 1165 |
+
"grad_norm": 0.16738031804561615,
|
| 1166 |
+
"learning_rate": 0.00030075689523087804,
|
| 1167 |
+
"loss": 2.8128,
|
| 1168 |
+
"step": 8150
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 0.36379264928114907,
|
| 1172 |
+
"grad_norm": 0.1693500131368637,
|
| 1173 |
+
"learning_rate": 0.00029949433868506293,
|
| 1174 |
+
"loss": 2.8138,
|
| 1175 |
+
"step": 8200
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 0.3660108971426195,
|
| 1179 |
+
"grad_norm": 0.16915106773376465,
|
| 1180 |
+
"learning_rate": 0.00029822649105348294,
|
| 1181 |
+
"loss": 2.8209,
|
| 1182 |
+
"step": 8250
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 0.3682291450040899,
|
| 1186 |
+
"grad_norm": 0.17108069360256195,
|
| 1187 |
+
"learning_rate": 0.00029695341975998006,
|
| 1188 |
+
"loss": 2.8174,
|
| 1189 |
+
"step": 8300
|
| 1190 |
+
},
|
| 1191 |
+
{
|
| 1192 |
+
"epoch": 0.37044739286556033,
|
| 1193 |
+
"grad_norm": 0.16659317910671234,
|
| 1194 |
+
"learning_rate": 0.00029567519250618907,
|
| 1195 |
+
"loss": 2.8153,
|
| 1196 |
+
"step": 8350
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 0.37266564072703073,
|
| 1200 |
+
"grad_norm": 0.16678906977176666,
|
| 1201 |
+
"learning_rate": 0.0002943918772679379,
|
| 1202 |
+
"loss": 2.8163,
|
| 1203 |
+
"step": 8400
|
| 1204 |
+
},
|
| 1205 |
+
{
|
| 1206 |
+
"epoch": 0.37488388858850114,
|
| 1207 |
+
"grad_norm": 0.16928167641162872,
|
| 1208 |
+
"learning_rate": 0.00029310354229163197,
|
| 1209 |
+
"loss": 2.8165,
|
| 1210 |
+
"step": 8450
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 0.3771021364499716,
|
| 1214 |
+
"grad_norm": 0.1695391833782196,
|
| 1215 |
+
"learning_rate": 0.0002918102560906254,
|
| 1216 |
+
"loss": 2.8197,
|
| 1217 |
+
"step": 8500
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 0.379320384311442,
|
| 1221 |
+
"grad_norm": 0.17006346583366394,
|
| 1222 |
+
"learning_rate": 0.0002905120874415772,
|
| 1223 |
+
"loss": 2.8172,
|
| 1224 |
+
"step": 8550
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 0.3815386321729124,
|
| 1228 |
+
"grad_norm": 0.16821132600307465,
|
| 1229 |
+
"learning_rate": 0.0002892091053807939,
|
| 1230 |
+
"loss": 2.8137,
|
| 1231 |
+
"step": 8600
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"epoch": 0.38375688003438285,
|
| 1235 |
+
"grad_norm": 0.17077401280403137,
|
| 1236 |
+
"learning_rate": 0.000287901379200558,
|
| 1237 |
+
"loss": 2.8174,
|
| 1238 |
+
"step": 8650
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 0.38597512789585325,
|
| 1242 |
+
"grad_norm": 0.17006562650203705,
|
| 1243 |
+
"learning_rate": 0.0002865889784454435,
|
| 1244 |
+
"loss": 2.813,
|
| 1245 |
+
"step": 8700
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 0.3881933757573237,
|
| 1249 |
+
"grad_norm": 0.16847462952136993,
|
| 1250 |
+
"learning_rate": 0.0002852719729086167,
|
| 1251 |
+
"loss": 2.8158,
|
| 1252 |
+
"step": 8750
|
| 1253 |
+
},
|
| 1254 |
+
{
|
| 1255 |
+
"epoch": 0.3904116236187941,
|
| 1256 |
+
"grad_norm": 0.16790613532066345,
|
| 1257 |
+
"learning_rate": 0.0002839504326281256,
|
| 1258 |
+
"loss": 2.816,
|
| 1259 |
+
"step": 8800
|
| 1260 |
+
},
|
| 1261 |
+
{
|
| 1262 |
+
"epoch": 0.3926298714802645,
|
| 1263 |
+
"grad_norm": 0.16898341476917267,
|
| 1264 |
+
"learning_rate": 0.00028262442788317446,
|
| 1265 |
+
"loss": 2.8143,
|
| 1266 |
+
"step": 8850
|
| 1267 |
+
},
|
| 1268 |
+
{
|
| 1269 |
+
"epoch": 0.39484811934173497,
|
| 1270 |
+
"grad_norm": 0.17099575698375702,
|
| 1271 |
+
"learning_rate": 0.00028129402919038695,
|
| 1272 |
+
"loss": 2.812,
|
| 1273 |
+
"step": 8900
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 0.3970663672032054,
|
| 1277 |
+
"grad_norm": 0.17063932120800018,
|
| 1278 |
+
"learning_rate": 0.00027995930730005577,
|
| 1279 |
+
"loss": 2.815,
|
| 1280 |
+
"step": 8950
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 0.3992846150646758,
|
| 1284 |
+
"grad_norm": 0.1704034060239792,
|
| 1285 |
+
"learning_rate": 0.00027862033319238025,
|
| 1286 |
+
"loss": 2.8144,
|
| 1287 |
+
"step": 9000
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 0.3992846150646758,
|
| 1291 |
+
"eval_accuracy": 0.42786541279921836,
|
| 1292 |
+
"eval_loss": 2.8759515285491943,
|
| 1293 |
+
"eval_runtime": 250.6732,
|
| 1294 |
+
"eval_samples_per_second": 7.979,
|
| 1295 |
+
"eval_steps_per_second": 0.997,
|
| 1296 |
+
"step": 9000
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 0.40150286292614623,
|
| 1300 |
+
"grad_norm": 0.1675969511270523,
|
| 1301 |
+
"learning_rate": 0.0002772771780736917,
|
| 1302 |
+
"loss": 2.8128,
|
| 1303 |
+
"step": 9050
|
| 1304 |
+
},
|
| 1305 |
+
{
|
| 1306 |
+
"epoch": 0.40372111078761663,
|
| 1307 |
+
"grad_norm": 0.1697956621646881,
|
| 1308 |
+
"learning_rate": 0.0002759299133726665,
|
| 1309 |
+
"loss": 2.8121,
|
| 1310 |
+
"step": 9100
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 0.40593935864908703,
|
| 1314 |
+
"grad_norm": 0.1710100620985031,
|
| 1315 |
+
"learning_rate": 0.00027457861073652785,
|
| 1316 |
+
"loss": 2.8156,
|
| 1317 |
+
"step": 9150
|
| 1318 |
+
},
|
| 1319 |
+
{
|
| 1320 |
+
"epoch": 0.4081576065105575,
|
| 1321 |
+
"grad_norm": 0.16877809166908264,
|
| 1322 |
+
"learning_rate": 0.00027322334202723527,
|
| 1323 |
+
"loss": 2.815,
|
| 1324 |
+
"step": 9200
|
| 1325 |
+
},
|
| 1326 |
+
{
|
| 1327 |
+
"epoch": 0.4103758543720279,
|
| 1328 |
+
"grad_norm": 0.17122440040111542,
|
| 1329 |
+
"learning_rate": 0.0002718641793176631,
|
| 1330 |
+
"loss": 2.8119,
|
| 1331 |
+
"step": 9250
|
| 1332 |
+
},
|
| 1333 |
+
{
|
| 1334 |
+
"epoch": 0.4125941022334983,
|
| 1335 |
+
"grad_norm": 0.16771045327186584,
|
| 1336 |
+
"learning_rate": 0.0002705011948877679,
|
| 1337 |
+
"loss": 2.808,
|
| 1338 |
+
"step": 9300
|
| 1339 |
+
},
|
| 1340 |
+
{
|
| 1341 |
+
"epoch": 0.41481235009496875,
|
| 1342 |
+
"grad_norm": 0.16941729187965393,
|
| 1343 |
+
"learning_rate": 0.0002691344612207442,
|
| 1344 |
+
"loss": 2.8121,
|
| 1345 |
+
"step": 9350
|
| 1346 |
+
},
|
| 1347 |
+
{
|
| 1348 |
+
"epoch": 0.41703059795643915,
|
| 1349 |
+
"grad_norm": 0.1719992607831955,
|
| 1350 |
+
"learning_rate": 0.00026776405099917014,
|
| 1351 |
+
"loss": 2.8094,
|
| 1352 |
+
"step": 9400
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 0.41924884581790955,
|
| 1356 |
+
"grad_norm": 0.1693243533372879,
|
| 1357 |
+
"learning_rate": 0.00026639003710114223,
|
| 1358 |
+
"loss": 2.8103,
|
| 1359 |
+
"step": 9450
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 0.42146709367938,
|
| 1363 |
+
"grad_norm": 0.17014500498771667,
|
| 1364 |
+
"learning_rate": 0.0002650124925963998,
|
| 1365 |
+
"loss": 2.8129,
|
| 1366 |
+
"step": 9500
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 0.4236853415408504,
|
| 1370 |
+
"grad_norm": 0.1709510087966919,
|
| 1371 |
+
"learning_rate": 0.00026363149074243867,
|
| 1372 |
+
"loss": 2.8084,
|
| 1373 |
+
"step": 9550
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 0.4259035894023208,
|
| 1377 |
+
"grad_norm": 0.16937118768692017,
|
| 1378 |
+
"learning_rate": 0.0002622471049806159,
|
| 1379 |
+
"loss": 2.814,
|
| 1380 |
+
"step": 9600
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 0.42812183726379127,
|
| 1384 |
+
"grad_norm": 0.1713036149740219,
|
| 1385 |
+
"learning_rate": 0.00026085940893224403,
|
| 1386 |
+
"loss": 2.8162,
|
| 1387 |
+
"step": 9650
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"epoch": 0.4303400851252617,
|
| 1391 |
+
"grad_norm": 0.17020347714424133,
|
| 1392 |
+
"learning_rate": 0.0002594684763946758,
|
| 1393 |
+
"loss": 2.8116,
|
| 1394 |
+
"step": 9700
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 0.43255833298673213,
|
| 1398 |
+
"grad_norm": 0.16786696016788483,
|
| 1399 |
+
"learning_rate": 0.0002580743813373796,
|
| 1400 |
+
"loss": 2.8111,
|
| 1401 |
+
"step": 9750
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 0.43477658084820253,
|
| 1405 |
+
"grad_norm": 0.17273075878620148,
|
| 1406 |
+
"learning_rate": 0.00025667719789800606,
|
| 1407 |
+
"loss": 2.8131,
|
| 1408 |
+
"step": 9800
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 0.43699482870967293,
|
| 1412 |
+
"grad_norm": 0.16986466944217682,
|
| 1413 |
+
"learning_rate": 0.00025527700037844515,
|
| 1414 |
+
"loss": 2.8139,
|
| 1415 |
+
"step": 9850
|
| 1416 |
+
},
|
| 1417 |
+
{
|
| 1418 |
+
"epoch": 0.4392130765711434,
|
| 1419 |
+
"grad_norm": 0.17129731178283691,
|
| 1420 |
+
"learning_rate": 0.00025387386324087494,
|
| 1421 |
+
"loss": 2.8125,
|
| 1422 |
+
"step": 9900
|
| 1423 |
+
},
|
| 1424 |
+
{
|
| 1425 |
+
"epoch": 0.4414313244326138,
|
| 1426 |
+
"grad_norm": 0.16890868544578552,
|
| 1427 |
+
"learning_rate": 0.00025246786110380163,
|
| 1428 |
+
"loss": 2.8142,
|
| 1429 |
+
"step": 9950
|
| 1430 |
+
},
|
| 1431 |
+
{
|
| 1432 |
+
"epoch": 0.4436495722940842,
|
| 1433 |
+
"grad_norm": 0.17167522013187408,
|
| 1434 |
+
"learning_rate": 0.00025105906873809154,
|
| 1435 |
+
"loss": 2.8142,
|
| 1436 |
+
"step": 10000
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 0.44586782015555465,
|
| 1440 |
+
"grad_norm": 0.17136669158935547,
|
| 1441 |
+
"learning_rate": 0.0002496475610629947,
|
| 1442 |
+
"loss": 2.8112,
|
| 1443 |
+
"step": 10050
|
| 1444 |
+
},
|
| 1445 |
+
{
|
| 1446 |
+
"epoch": 0.44808606801702505,
|
| 1447 |
+
"grad_norm": 0.16926760971546173,
|
| 1448 |
+
"learning_rate": 0.00024823341314216056,
|
| 1449 |
+
"loss": 2.8156,
|
| 1450 |
+
"step": 10100
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"epoch": 0.45030431587849545,
|
| 1454 |
+
"grad_norm": 0.16898435354232788,
|
| 1455 |
+
"learning_rate": 0.00024681670017964627,
|
| 1456 |
+
"loss": 2.8079,
|
| 1457 |
+
"step": 10150
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 0.4525225637399659,
|
| 1461 |
+
"grad_norm": 0.17237040400505066,
|
| 1462 |
+
"learning_rate": 0.0002453974975159173,
|
| 1463 |
+
"loss": 2.813,
|
| 1464 |
+
"step": 10200
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 0.4547408116014363,
|
| 1468 |
+
"grad_norm": 0.16995486617088318,
|
| 1469 |
+
"learning_rate": 0.00024397588062384095,
|
| 1470 |
+
"loss": 2.8117,
|
| 1471 |
+
"step": 10250
|
| 1472 |
+
},
|
| 1473 |
+
{
|
| 1474 |
+
"epoch": 0.4569590594629067,
|
| 1475 |
+
"grad_norm": 0.17290563881397247,
|
| 1476 |
+
"learning_rate": 0.00024255192510467245,
|
| 1477 |
+
"loss": 2.8121,
|
| 1478 |
+
"step": 10300
|
| 1479 |
+
},
|
| 1480 |
+
{
|
| 1481 |
+
"epoch": 0.45917730732437717,
|
| 1482 |
+
"grad_norm": 0.17059782147407532,
|
| 1483 |
+
"learning_rate": 0.00024112570668403472,
|
| 1484 |
+
"loss": 2.8138,
|
| 1485 |
+
"step": 10350
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"epoch": 0.4613955551858476,
|
| 1489 |
+
"grad_norm": 0.17196382582187653,
|
| 1490 |
+
"learning_rate": 0.00023969730120789132,
|
| 1491 |
+
"loss": 2.8095,
|
| 1492 |
+
"step": 10400
|
| 1493 |
+
},
|
| 1494 |
+
{
|
| 1495 |
+
"epoch": 0.463613803047318,
|
| 1496 |
+
"grad_norm": 0.16942380368709564,
|
| 1497 |
+
"learning_rate": 0.00023826678463851285,
|
| 1498 |
+
"loss": 2.8124,
|
| 1499 |
+
"step": 10450
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 0.46583205090878843,
|
| 1503 |
+
"grad_norm": 0.17288681864738464,
|
| 1504 |
+
"learning_rate": 0.00023683423305043749,
|
| 1505 |
+
"loss": 2.813,
|
| 1506 |
+
"step": 10500
|
| 1507 |
+
},
|
| 1508 |
+
{
|
| 1509 |
+
"epoch": 0.46805029877025883,
|
| 1510 |
+
"grad_norm": 0.17040428519248962,
|
| 1511 |
+
"learning_rate": 0.00023539972262642502,
|
| 1512 |
+
"loss": 2.8141,
|
| 1513 |
+
"step": 10550
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 0.4702685466317293,
|
| 1517 |
+
"grad_norm": 0.17321184277534485,
|
| 1518 |
+
"learning_rate": 0.00023396332965340585,
|
| 1519 |
+
"loss": 2.8146,
|
| 1520 |
+
"step": 10600
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 0.4724867944931997,
|
| 1524 |
+
"grad_norm": 0.17026926577091217,
|
| 1525 |
+
"learning_rate": 0.00023252513051842373,
|
| 1526 |
+
"loss": 2.8086,
|
| 1527 |
+
"step": 10650
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 0.4747050423546701,
|
| 1531 |
+
"grad_norm": 0.1710352748632431,
|
| 1532 |
+
"learning_rate": 0.00023108520170457398,
|
| 1533 |
+
"loss": 2.8099,
|
| 1534 |
+
"step": 10700
|
| 1535 |
+
},
|
| 1536 |
+
{
|
| 1537 |
+
"epoch": 0.47692329021614055,
|
| 1538 |
+
"grad_norm": 0.17067080736160278,
|
| 1539 |
+
"learning_rate": 0.00022964361978693542,
|
| 1540 |
+
"loss": 2.8099,
|
| 1541 |
+
"step": 10750
|
| 1542 |
+
},
|
| 1543 |
+
{
|
| 1544 |
+
"epoch": 0.47914153807761095,
|
| 1545 |
+
"grad_norm": 0.17244164645671844,
|
| 1546 |
+
"learning_rate": 0.0002282004614284989,
|
| 1547 |
+
"loss": 2.8054,
|
| 1548 |
+
"step": 10800
|
| 1549 |
+
}
|
| 1550 |
+
],
|
| 1551 |
+
"logging_steps": 50,
|
| 1552 |
+
"max_steps": 22540,
|
| 1553 |
+
"num_input_tokens_seen": 0,
|
| 1554 |
+
"num_train_epochs": 1,
|
| 1555 |
+
"save_steps": 1200,
|
| 1556 |
+
"stateful_callbacks": {
|
| 1557 |
+
"TrainerControl": {
|
| 1558 |
+
"args": {
|
| 1559 |
+
"should_epoch_stop": false,
|
| 1560 |
+
"should_evaluate": false,
|
| 1561 |
+
"should_log": false,
|
| 1562 |
+
"should_save": true,
|
| 1563 |
+
"should_training_stop": false
|
| 1564 |
+
},
|
| 1565 |
+
"attributes": {}
|
| 1566 |
+
}
|
| 1567 |
+
},
|
| 1568 |
+
"total_flos": 1.5864483085358727e+19,
|
| 1569 |
+
"train_batch_size": 1,
|
| 1570 |
+
"trial_name": null,
|
| 1571 |
+
"trial_params": null
|
| 1572 |
+
}
|