Upload checkpoint-1134
Browse files
checkpoint-1134/config.json
ADDED
|
@@ -0,0 +1,65 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Gemma2ForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"attn_logit_softcapping": 50.0,
|
| 8 |
+
"bos_token_id": 2,
|
| 9 |
+
"cache_implementation": "hybrid",
|
| 10 |
+
"dtype": "bfloat16",
|
| 11 |
+
"eos_token_id": 1,
|
| 12 |
+
"final_logit_softcapping": 30.0,
|
| 13 |
+
"head_dim": 256,
|
| 14 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 15 |
+
"hidden_activation": "gelu_pytorch_tanh",
|
| 16 |
+
"hidden_size": 2304,
|
| 17 |
+
"initializer_range": 0.02,
|
| 18 |
+
"intermediate_size": 9216,
|
| 19 |
+
"layer_types": [
|
| 20 |
+
"sliding_attention",
|
| 21 |
+
"full_attention",
|
| 22 |
+
"sliding_attention",
|
| 23 |
+
"full_attention",
|
| 24 |
+
"sliding_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"sliding_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"sliding_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"sliding_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"sliding_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"sliding_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"sliding_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"sliding_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"sliding_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"sliding_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"sliding_attention",
|
| 45 |
+
"full_attention"
|
| 46 |
+
],
|
| 47 |
+
"max_position_embeddings": 8192,
|
| 48 |
+
"model_type": "gemma2",
|
| 49 |
+
"num_attention_heads": 8,
|
| 50 |
+
"num_hidden_layers": 26,
|
| 51 |
+
"num_key_value_heads": 4,
|
| 52 |
+
"pad_token_id": 0,
|
| 53 |
+
"query_pre_attn_scalar": 256,
|
| 54 |
+
"rms_norm_eps": 1e-06,
|
| 55 |
+
"rope_parameters": {
|
| 56 |
+
"rope_theta": 10000.0,
|
| 57 |
+
"rope_type": "default"
|
| 58 |
+
},
|
| 59 |
+
"sliding_window": 4096,
|
| 60 |
+
"tie_word_embeddings": true,
|
| 61 |
+
"transformers_version": "5.17.0",
|
| 62 |
+
"use_bidirectional_attention": null,
|
| 63 |
+
"use_cache": false,
|
| 64 |
+
"vocab_size": 256000
|
| 65 |
+
}
|
checkpoint-1134/generation_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 2,
|
| 4 |
+
"cache_implementation": "hybrid",
|
| 5 |
+
"eos_token_id": [
|
| 6 |
+
1
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 0,
|
| 9 |
+
"transformers_version": "5.17.0"
|
| 10 |
+
}
|
checkpoint-1134/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6e83f9f89a9bd77c8c77eebc7e3132cd25cedf2f092a4e43420db4ecf89b0160
|
| 3 |
+
size 5228717512
|
checkpoint-1134/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:487cee8724215dcd2dde8888539e8b1bf844ceb5dbbe27f7845abda69eeb060f
|
| 3 |
+
size 34362872
|
checkpoint-1134/tokenizer_config.json
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"bos_token": "<bos>",
|
| 4 |
+
"clean_up_tokenization_spaces": false,
|
| 5 |
+
"eos_token": "<eos>",
|
| 6 |
+
"extra_special_tokens": [
|
| 7 |
+
"<start_of_turn>",
|
| 8 |
+
"<end_of_turn>"
|
| 9 |
+
],
|
| 10 |
+
"is_local": true,
|
| 11 |
+
"local_files_only": false,
|
| 12 |
+
"mask_token": "<mask>",
|
| 13 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 14 |
+
"pad_token": "<pad>",
|
| 15 |
+
"sp_model_kwargs": {},
|
| 16 |
+
"spaces_between_special_tokens": false,
|
| 17 |
+
"tokenizer_class": "GemmaTokenizer",
|
| 18 |
+
"unk_token": "<unk>",
|
| 19 |
+
"use_default_system_prompt": false
|
| 20 |
+
}
|
checkpoint-1134/trainer_state.json
ADDED
|
@@ -0,0 +1,1164 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 6.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 1134,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"entropy": 1.7671512730419636,
|
| 14 |
+
"epoch": 0.05291005291005291,
|
| 15 |
+
"grad_norm": 14.8125,
|
| 16 |
+
"learning_rate": 3.2142857142857147e-06,
|
| 17 |
+
"loss": 1.706784439086914,
|
| 18 |
+
"mean_token_accuracy": 0.6392517775297165,
|
| 19 |
+
"num_tokens": 710093.0,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"entropy": 1.4565640933811665,
|
| 24 |
+
"epoch": 0.10582010582010581,
|
| 25 |
+
"grad_norm": 6.1875,
|
| 26 |
+
"learning_rate": 6.785714285714287e-06,
|
| 27 |
+
"loss": 1.4239604949951172,
|
| 28 |
+
"mean_token_accuracy": 0.6772899225354194,
|
| 29 |
+
"num_tokens": 1429271.0,
|
| 30 |
+
"step": 20
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"entropy": 1.0071680538356305,
|
| 34 |
+
"epoch": 0.15873015873015872,
|
| 35 |
+
"grad_norm": 3.109375,
|
| 36 |
+
"learning_rate": 1.0357142857142859e-05,
|
| 37 |
+
"loss": 1.048217487335205,
|
| 38 |
+
"mean_token_accuracy": 0.7393087036907673,
|
| 39 |
+
"num_tokens": 2149735.0,
|
| 40 |
+
"step": 30
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"entropy": 0.8510661780834198,
|
| 44 |
+
"epoch": 0.21164021164021163,
|
| 45 |
+
"grad_norm": 1.5703125,
|
| 46 |
+
"learning_rate": 1.3928571428571429e-05,
|
| 47 |
+
"loss": 0.8619060516357422,
|
| 48 |
+
"mean_token_accuracy": 0.7765810377895832,
|
| 49 |
+
"num_tokens": 2868867.0,
|
| 50 |
+
"step": 40
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"entropy": 0.7774681895971298,
|
| 54 |
+
"epoch": 0.26455026455026454,
|
| 55 |
+
"grad_norm": 1.578125,
|
| 56 |
+
"learning_rate": 1.7500000000000002e-05,
|
| 57 |
+
"loss": 0.7852191925048828,
|
| 58 |
+
"mean_token_accuracy": 0.7921242833137512,
|
| 59 |
+
"num_tokens": 3582476.0,
|
| 60 |
+
"step": 50
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"entropy": 0.7451388858258724,
|
| 64 |
+
"epoch": 0.31746031746031744,
|
| 65 |
+
"grad_norm": 1.5859375,
|
| 66 |
+
"learning_rate": 1.999986795783372e-05,
|
| 67 |
+
"loss": 0.748009204864502,
|
| 68 |
+
"mean_token_accuracy": 0.7999738156795502,
|
| 69 |
+
"num_tokens": 4287689.0,
|
| 70 |
+
"step": 60
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"entropy": 0.729080930352211,
|
| 74 |
+
"epoch": 0.37037037037037035,
|
| 75 |
+
"grad_norm": 1.4921875,
|
| 76 |
+
"learning_rate": 1.9997520638548385e-05,
|
| 77 |
+
"loss": 0.7400224685668946,
|
| 78 |
+
"mean_token_accuracy": 0.801699859648943,
|
| 79 |
+
"num_tokens": 5011191.0,
|
| 80 |
+
"step": 70
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"entropy": 0.7048044994473457,
|
| 84 |
+
"epoch": 0.42328042328042326,
|
| 85 |
+
"grad_norm": 1.4453125,
|
| 86 |
+
"learning_rate": 1.9992239841688418e-05,
|
| 87 |
+
"loss": 0.7069188117980957,
|
| 88 |
+
"mean_token_accuracy": 0.8093748845160007,
|
| 89 |
+
"num_tokens": 5714842.0,
|
| 90 |
+
"step": 80
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"entropy": 0.6840555652976036,
|
| 94 |
+
"epoch": 0.47619047619047616,
|
| 95 |
+
"grad_norm": 1.3125,
|
| 96 |
+
"learning_rate": 1.9984027116747914e-05,
|
| 97 |
+
"loss": 0.6863717555999755,
|
| 98 |
+
"mean_token_accuracy": 0.8138940587639809,
|
| 99 |
+
"num_tokens": 6424614.0,
|
| 100 |
+
"step": 90
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"entropy": 0.6870591521263123,
|
| 104 |
+
"epoch": 0.5291005291005291,
|
| 105 |
+
"grad_norm": 1.4140625,
|
| 106 |
+
"learning_rate": 1.997288487350878e-05,
|
| 107 |
+
"loss": 0.6901006698608398,
|
| 108 |
+
"mean_token_accuracy": 0.8129441156983376,
|
| 109 |
+
"num_tokens": 7139778.0,
|
| 110 |
+
"step": 100
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"entropy": 0.6756624668836594,
|
| 114 |
+
"epoch": 0.582010582010582,
|
| 115 |
+
"grad_norm": 1.40625,
|
| 116 |
+
"learning_rate": 1.9958816381333684e-05,
|
| 117 |
+
"loss": 0.6771352767944336,
|
| 118 |
+
"mean_token_accuracy": 0.81553890183568,
|
| 119 |
+
"num_tokens": 7852540.0,
|
| 120 |
+
"step": 110
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"entropy": 0.6655688174068928,
|
| 124 |
+
"epoch": 0.6349206349206349,
|
| 125 |
+
"grad_norm": 1.34375,
|
| 126 |
+
"learning_rate": 1.994182576820673e-05,
|
| 127 |
+
"loss": 0.667414665222168,
|
| 128 |
+
"mean_token_accuracy": 0.8177205875515938,
|
| 129 |
+
"num_tokens": 8568056.0,
|
| 130 |
+
"step": 120
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"entropy": 0.6643485642969609,
|
| 134 |
+
"epoch": 0.6878306878306878,
|
| 135 |
+
"grad_norm": 1.296875,
|
| 136 |
+
"learning_rate": 1.9921918019522238e-05,
|
| 137 |
+
"loss": 0.6721675872802735,
|
| 138 |
+
"mean_token_accuracy": 0.816884483397007,
|
| 139 |
+
"num_tokens": 9276742.0,
|
| 140 |
+
"step": 130
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"entropy": 0.6590477801859379,
|
| 144 |
+
"epoch": 0.7407407407407407,
|
| 145 |
+
"grad_norm": 1.3515625,
|
| 146 |
+
"learning_rate": 1.9899098976621922e-05,
|
| 147 |
+
"loss": 0.6590158462524414,
|
| 148 |
+
"mean_token_accuracy": 0.8199199490249157,
|
| 149 |
+
"num_tokens": 9990163.0,
|
| 150 |
+
"step": 140
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"entropy": 0.649939077347517,
|
| 154 |
+
"epoch": 0.7936507936507936,
|
| 155 |
+
"grad_norm": 1.4609375,
|
| 156 |
+
"learning_rate": 1.9873375335080915e-05,
|
| 157 |
+
"loss": 0.6501528263092041,
|
| 158 |
+
"mean_token_accuracy": 0.82164012119174,
|
| 159 |
+
"num_tokens": 10710006.0,
|
| 160 |
+
"step": 150
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"entropy": 0.6468971997499466,
|
| 164 |
+
"epoch": 0.8465608465608465,
|
| 165 |
+
"grad_norm": 1.296875,
|
| 166 |
+
"learning_rate": 1.9844754642743164e-05,
|
| 167 |
+
"loss": 0.6529253959655762,
|
| 168 |
+
"mean_token_accuracy": 0.820851170271635,
|
| 169 |
+
"num_tokens": 11416822.0,
|
| 170 |
+
"step": 160
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"entropy": 0.6556476138532161,
|
| 174 |
+
"epoch": 0.8994708994708994,
|
| 175 |
+
"grad_norm": 1.296875,
|
| 176 |
+
"learning_rate": 1.981324529750672e-05,
|
| 177 |
+
"loss": 0.6604087829589844,
|
| 178 |
+
"mean_token_accuracy": 0.8195641197264194,
|
| 179 |
+
"num_tokens": 12123118.0,
|
| 180 |
+
"step": 170
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"entropy": 0.6358714967966079,
|
| 184 |
+
"epoch": 0.9523809523809523,
|
| 185 |
+
"grad_norm": 1.3515625,
|
| 186 |
+
"learning_rate": 1.9778856544859647e-05,
|
| 187 |
+
"loss": 0.6365831375122071,
|
| 188 |
+
"mean_token_accuracy": 0.8241398617625236,
|
| 189 |
+
"num_tokens": 12831058.0,
|
| 190 |
+
"step": 180
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"entropy": 0.6277156263589859,
|
| 194 |
+
"epoch": 1.0052910052910053,
|
| 195 |
+
"grad_norm": 1.25,
|
| 196 |
+
"learning_rate": 1.9741598475167173e-05,
|
| 197 |
+
"loss": 0.6229054927825928,
|
| 198 |
+
"mean_token_accuracy": 0.827630003541708,
|
| 199 |
+
"num_tokens": 13547943.0,
|
| 200 |
+
"step": 190
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"entropy": 0.5929765947163105,
|
| 204 |
+
"epoch": 1.0582010582010581,
|
| 205 |
+
"grad_norm": 1.421875,
|
| 206 |
+
"learning_rate": 1.9701482020711004e-05,
|
| 207 |
+
"loss": 0.5976824760437012,
|
| 208 |
+
"mean_token_accuracy": 0.8327139303088188,
|
| 209 |
+
"num_tokens": 14264772.0,
|
| 210 |
+
"step": 200
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"entropy": 0.5882044766098261,
|
| 214 |
+
"epoch": 1.1111111111111112,
|
| 215 |
+
"grad_norm": 1.1953125,
|
| 216 |
+
"learning_rate": 1.9658518952481543e-05,
|
| 217 |
+
"loss": 0.5882174968719482,
|
| 218 |
+
"mean_token_accuracy": 0.8350941069424153,
|
| 219 |
+
"num_tokens": 14985565.0,
|
| 220 |
+
"step": 210
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"entropy": 0.5916924901306629,
|
| 224 |
+
"epoch": 1.164021164021164,
|
| 225 |
+
"grad_norm": 1.4140625,
|
| 226 |
+
"learning_rate": 1.9612721876724054e-05,
|
| 227 |
+
"loss": 0.5934844017028809,
|
| 228 |
+
"mean_token_accuracy": 0.8336228899657726,
|
| 229 |
+
"num_tokens": 15701796.0,
|
| 230 |
+
"step": 220
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"entropy": 0.5877517074346542,
|
| 234 |
+
"epoch": 1.216931216931217,
|
| 235 |
+
"grad_norm": 1.34375,
|
| 236 |
+
"learning_rate": 1.9564104231239726e-05,
|
| 237 |
+
"loss": 0.589122200012207,
|
| 238 |
+
"mean_token_accuracy": 0.8334487572312355,
|
| 239 |
+
"num_tokens": 16412522.0,
|
| 240 |
+
"step": 230
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"entropy": 0.5912774313241244,
|
| 244 |
+
"epoch": 1.2698412698412698,
|
| 245 |
+
"grad_norm": 1.3359375,
|
| 246 |
+
"learning_rate": 1.9512680281442745e-05,
|
| 247 |
+
"loss": 0.5940174102783203,
|
| 248 |
+
"mean_token_accuracy": 0.83340530321002,
|
| 249 |
+
"num_tokens": 17126523.0,
|
| 250 |
+
"step": 240
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"entropy": 0.5930401895195245,
|
| 254 |
+
"epoch": 1.3227513227513228,
|
| 255 |
+
"grad_norm": 1.2265625,
|
| 256 |
+
"learning_rate": 1.9458465116174534e-05,
|
| 257 |
+
"loss": 0.5915824890136718,
|
| 258 |
+
"mean_token_accuracy": 0.8344384111464024,
|
| 259 |
+
"num_tokens": 17834934.0,
|
| 260 |
+
"step": 250
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"entropy": 0.5963230866938829,
|
| 264 |
+
"epoch": 1.3756613756613756,
|
| 265 |
+
"grad_norm": 1.296875,
|
| 266 |
+
"learning_rate": 1.940147464327637e-05,
|
| 267 |
+
"loss": 0.5983511924743652,
|
| 268 |
+
"mean_token_accuracy": 0.832802091538906,
|
| 269 |
+
"num_tokens": 18553413.0,
|
| 270 |
+
"step": 260
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"entropy": 0.5813015405088663,
|
| 274 |
+
"epoch": 1.4285714285714286,
|
| 275 |
+
"grad_norm": 1.28125,
|
| 276 |
+
"learning_rate": 1.9341725584921712e-05,
|
| 277 |
+
"loss": 0.5832551956176758,
|
| 278 |
+
"mean_token_accuracy": 0.8360917493700981,
|
| 279 |
+
"num_tokens": 19270191.0,
|
| 280 |
+
"step": 270
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"entropy": 0.5801753208041192,
|
| 284 |
+
"epoch": 1.4814814814814814,
|
| 285 |
+
"grad_norm": 1.3046875,
|
| 286 |
+
"learning_rate": 1.9279235472709554e-05,
|
| 287 |
+
"loss": 0.5805642127990722,
|
| 288 |
+
"mean_token_accuracy": 0.8362542942166329,
|
| 289 |
+
"num_tokens": 19976552.0,
|
| 290 |
+
"step": 280
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"entropy": 0.5879705954343081,
|
| 294 |
+
"epoch": 1.5343915343915344,
|
| 295 |
+
"grad_norm": 1.34375,
|
| 296 |
+
"learning_rate": 1.92140226425203e-05,
|
| 297 |
+
"loss": 0.5905383110046387,
|
| 298 |
+
"mean_token_accuracy": 0.8341313540935517,
|
| 299 |
+
"num_tokens": 20676613.0,
|
| 300 |
+
"step": 290
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"entropy": 0.5889205444604159,
|
| 304 |
+
"epoch": 1.5873015873015874,
|
| 305 |
+
"grad_norm": 1.2578125,
|
| 306 |
+
"learning_rate": 1.914610622913566e-05,
|
| 307 |
+
"loss": 0.5913270950317383,
|
| 308 |
+
"mean_token_accuracy": 0.8349448375403881,
|
| 309 |
+
"num_tokens": 21385978.0,
|
| 310 |
+
"step": 300
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"entropy": 0.5768780928105116,
|
| 314 |
+
"epoch": 1.6402116402116402,
|
| 315 |
+
"grad_norm": 1.234375,
|
| 316 |
+
"learning_rate": 1.9075506160624068e-05,
|
| 317 |
+
"loss": 0.5772612571716309,
|
| 318 |
+
"mean_token_accuracy": 0.8369500823318958,
|
| 319 |
+
"num_tokens": 22094336.0,
|
| 320 |
+
"step": 310
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"entropy": 0.5735550723969937,
|
| 324 |
+
"epoch": 1.693121693121693,
|
| 325 |
+
"grad_norm": 1.3203125,
|
| 326 |
+
"learning_rate": 1.900224315249343e-05,
|
| 327 |
+
"loss": 0.5721980094909668,
|
| 328 |
+
"mean_token_accuracy": 0.8380511656403542,
|
| 329 |
+
"num_tokens": 22798906.0,
|
| 330 |
+
"step": 320
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"entropy": 0.57716367803514,
|
| 334 |
+
"epoch": 1.746031746031746,
|
| 335 |
+
"grad_norm": 1.2109375,
|
| 336 |
+
"learning_rate": 1.8926338701612738e-05,
|
| 337 |
+
"loss": 0.5769223690032959,
|
| 338 |
+
"mean_token_accuracy": 0.8371222764253616,
|
| 339 |
+
"num_tokens": 23511239.0,
|
| 340 |
+
"step": 330
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"entropy": 0.582606166601181,
|
| 344 |
+
"epoch": 1.798941798941799,
|
| 345 |
+
"grad_norm": 1.3671875,
|
| 346 |
+
"learning_rate": 1.884781507990445e-05,
|
| 347 |
+
"loss": 0.5855788230895996,
|
| 348 |
+
"mean_token_accuracy": 0.8356470488011837,
|
| 349 |
+
"num_tokens": 24231185.0,
|
| 350 |
+
"step": 340
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"entropy": 0.5729806918650866,
|
| 354 |
+
"epoch": 1.8518518518518519,
|
| 355 |
+
"grad_norm": 1.21875,
|
| 356 |
+
"learning_rate": 1.8766695327809463e-05,
|
| 357 |
+
"loss": 0.5730391502380371,
|
| 358 |
+
"mean_token_accuracy": 0.8385063141584397,
|
| 359 |
+
"num_tokens": 24944285.0,
|
| 360 |
+
"step": 350
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"entropy": 0.5666072227060794,
|
| 364 |
+
"epoch": 1.9047619047619047,
|
| 365 |
+
"grad_norm": 1.2109375,
|
| 366 |
+
"learning_rate": 1.8683003247526567e-05,
|
| 367 |
+
"loss": 0.5674479007720947,
|
| 368 |
+
"mean_token_accuracy": 0.8392404697835445,
|
| 369 |
+
"num_tokens": 25670441.0,
|
| 370 |
+
"step": 360
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"entropy": 0.5809141755104065,
|
| 374 |
+
"epoch": 1.9576719576719577,
|
| 375 |
+
"grad_norm": 1.2890625,
|
| 376 |
+
"learning_rate": 1.8596763396028384e-05,
|
| 377 |
+
"loss": 0.5828475952148438,
|
| 378 |
+
"mean_token_accuracy": 0.8360860221087932,
|
| 379 |
+
"num_tokens": 26379786.0,
|
| 380 |
+
"step": 370
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"entropy": 0.5695893839001656,
|
| 384 |
+
"epoch": 2.0105820105820107,
|
| 385 |
+
"grad_norm": 1.359375,
|
| 386 |
+
"learning_rate": 1.8508001077855876e-05,
|
| 387 |
+
"loss": 0.5633818626403808,
|
| 388 |
+
"mean_token_accuracy": 0.8408532224595546,
|
| 389 |
+
"num_tokens": 27097166.0,
|
| 390 |
+
"step": 380
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"entropy": 0.5167598329484463,
|
| 394 |
+
"epoch": 2.0634920634920633,
|
| 395 |
+
"grad_norm": 1.2265625,
|
| 396 |
+
"learning_rate": 1.8416742337693436e-05,
|
| 397 |
+
"loss": 0.5171660423278809,
|
| 398 |
+
"mean_token_accuracy": 0.8513922415673733,
|
| 399 |
+
"num_tokens": 27818991.0,
|
| 400 |
+
"step": 390
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"entropy": 0.5240574903786183,
|
| 404 |
+
"epoch": 2.1164021164021163,
|
| 405 |
+
"grad_norm": 1.5546875,
|
| 406 |
+
"learning_rate": 1.8323013952726877e-05,
|
| 407 |
+
"loss": 0.5252029418945312,
|
| 408 |
+
"mean_token_accuracy": 0.8488071605563163,
|
| 409 |
+
"num_tokens": 28524822.0,
|
| 410 |
+
"step": 400
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"entropy": 0.5178623419255018,
|
| 414 |
+
"epoch": 2.1693121693121693,
|
| 415 |
+
"grad_norm": 1.3125,
|
| 416 |
+
"learning_rate": 1.8226843424786434e-05,
|
| 417 |
+
"loss": 0.5203696250915527,
|
| 418 |
+
"mean_token_accuracy": 0.8495552279055119,
|
| 419 |
+
"num_tokens": 29245634.0,
|
| 420 |
+
"step": 410
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"entropy": 0.5194813564419747,
|
| 424 |
+
"epoch": 2.2222222222222223,
|
| 425 |
+
"grad_norm": 1.25,
|
| 426 |
+
"learning_rate": 1.8128258972277176e-05,
|
| 427 |
+
"loss": 0.5169014453887939,
|
| 428 |
+
"mean_token_accuracy": 0.8509399227797985,
|
| 429 |
+
"num_tokens": 29967135.0,
|
| 430 |
+
"step": 420
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"entropy": 0.522173210978508,
|
| 434 |
+
"epoch": 2.2751322751322753,
|
| 435 |
+
"grad_norm": 1.25,
|
| 436 |
+
"learning_rate": 1.8027289521899143e-05,
|
| 437 |
+
"loss": 0.5245429992675781,
|
| 438 |
+
"mean_token_accuracy": 0.848798380792141,
|
| 439 |
+
"num_tokens": 30680274.0,
|
| 440 |
+
"step": 430
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"entropy": 0.5265159368515014,
|
| 444 |
+
"epoch": 2.328042328042328,
|
| 445 |
+
"grad_norm": 1.296875,
|
| 446 |
+
"learning_rate": 1.7923964700159658e-05,
|
| 447 |
+
"loss": 0.5277297019958496,
|
| 448 |
+
"mean_token_accuracy": 0.8482260897755622,
|
| 449 |
+
"num_tokens": 31386859.0,
|
| 450 |
+
"step": 440
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"entropy": 0.5307009011507035,
|
| 454 |
+
"epoch": 2.380952380952381,
|
| 455 |
+
"grad_norm": 1.359375,
|
| 456 |
+
"learning_rate": 1.78183148246803e-05,
|
| 457 |
+
"loss": 0.5284142971038819,
|
| 458 |
+
"mean_token_accuracy": 0.8480213887989521,
|
| 459 |
+
"num_tokens": 32111007.0,
|
| 460 |
+
"step": 450
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"entropy": 0.5215292543172836,
|
| 464 |
+
"epoch": 2.433862433862434,
|
| 465 |
+
"grad_norm": 1.203125,
|
| 466 |
+
"learning_rate": 1.77103708953011e-05,
|
| 467 |
+
"loss": 0.5223949909210205,
|
| 468 |
+
"mean_token_accuracy": 0.8494340799748897,
|
| 469 |
+
"num_tokens": 32819180.0,
|
| 470 |
+
"step": 460
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"entropy": 0.5321523480117321,
|
| 474 |
+
"epoch": 2.4867724867724865,
|
| 475 |
+
"grad_norm": 1.2578125,
|
| 476 |
+
"learning_rate": 1.7600164584984548e-05,
|
| 477 |
+
"loss": 0.5305499076843262,
|
| 478 |
+
"mean_token_accuracy": 0.8471677340567112,
|
| 479 |
+
"num_tokens": 33523113.0,
|
| 480 |
+
"step": 470
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"entropy": 0.5240828461945057,
|
| 484 |
+
"epoch": 2.5396825396825395,
|
| 485 |
+
"grad_norm": 1.3046875,
|
| 486 |
+
"learning_rate": 1.7487728230522085e-05,
|
| 487 |
+
"loss": 0.5236708164215088,
|
| 488 |
+
"mean_token_accuracy": 0.8491978161036968,
|
| 489 |
+
"num_tokens": 34240357.0,
|
| 490 |
+
"step": 480
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"entropy": 0.5249651741236449,
|
| 494 |
+
"epoch": 2.5925925925925926,
|
| 495 |
+
"grad_norm": 1.21875,
|
| 496 |
+
"learning_rate": 1.737309482304585e-05,
|
| 497 |
+
"loss": 0.5244390487670898,
|
| 498 |
+
"mean_token_accuracy": 0.8490321710705757,
|
| 499 |
+
"num_tokens": 34958032.0,
|
| 500 |
+
"step": 490
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"entropy": 0.5262073375284672,
|
| 504 |
+
"epoch": 2.6455026455026456,
|
| 505 |
+
"grad_norm": 1.28125,
|
| 506 |
+
"learning_rate": 1.7256297998348388e-05,
|
| 507 |
+
"loss": 0.5262768268585205,
|
| 508 |
+
"mean_token_accuracy": 0.8484946198761463,
|
| 509 |
+
"num_tokens": 35658658.0,
|
| 510 |
+
"step": 500
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"entropy": 0.5266173977404833,
|
| 514 |
+
"epoch": 2.6984126984126986,
|
| 515 |
+
"grad_norm": 1.34375,
|
| 516 |
+
"learning_rate": 1.7137372027013212e-05,
|
| 517 |
+
"loss": 0.5282751083374023,
|
| 518 |
+
"mean_token_accuracy": 0.8476969927549363,
|
| 519 |
+
"num_tokens": 36364754.0,
|
| 520 |
+
"step": 510
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"entropy": 0.5260804291814566,
|
| 524 |
+
"epoch": 2.751322751322751,
|
| 525 |
+
"grad_norm": 1.25,
|
| 526 |
+
"learning_rate": 1.7016351804359134e-05,
|
| 527 |
+
"loss": 0.5255792140960693,
|
| 528 |
+
"mean_token_accuracy": 0.8494147725403309,
|
| 529 |
+
"num_tokens": 37074678.0,
|
| 530 |
+
"step": 520
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"entropy": 0.5435763075947762,
|
| 534 |
+
"epoch": 2.804232804232804,
|
| 535 |
+
"grad_norm": 1.4296875,
|
| 536 |
+
"learning_rate": 1.6893272840201243e-05,
|
| 537 |
+
"loss": 0.5404428482055664,
|
| 538 |
+
"mean_token_accuracy": 0.8453655727207661,
|
| 539 |
+
"num_tokens": 37787128.0,
|
| 540 |
+
"step": 530
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"entropy": 0.5258786808699369,
|
| 544 |
+
"epoch": 2.857142857142857,
|
| 545 |
+
"grad_norm": 1.265625,
|
| 546 |
+
"learning_rate": 1.6768171248431603e-05,
|
| 547 |
+
"loss": 0.5282983303070068,
|
| 548 |
+
"mean_token_accuracy": 0.8478383302688599,
|
| 549 |
+
"num_tokens": 38504013.0,
|
| 550 |
+
"step": 540
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"entropy": 0.5149568922817707,
|
| 554 |
+
"epoch": 2.91005291005291,
|
| 555 |
+
"grad_norm": 1.2109375,
|
| 556 |
+
"learning_rate": 1.66410837364227e-05,
|
| 557 |
+
"loss": 0.5157745361328125,
|
| 558 |
+
"mean_token_accuracy": 0.8509050741791725,
|
| 559 |
+
"num_tokens": 39212782.0,
|
| 560 |
+
"step": 550
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"entropy": 0.5198382031172514,
|
| 564 |
+
"epoch": 2.962962962962963,
|
| 565 |
+
"grad_norm": 1.296875,
|
| 566 |
+
"learning_rate": 1.651204759425674e-05,
|
| 567 |
+
"loss": 0.5191741943359375,
|
| 568 |
+
"mean_token_accuracy": 0.8499824590981007,
|
| 569 |
+
"num_tokens": 39931947.0,
|
| 570 |
+
"step": 560
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"entropy": 0.5168990429490805,
|
| 574 |
+
"epoch": 3.015873015873016,
|
| 575 |
+
"grad_norm": 1.2265625,
|
| 576 |
+
"learning_rate": 1.6381100683783984e-05,
|
| 577 |
+
"loss": 0.5093559265136719,
|
| 578 |
+
"mean_token_accuracy": 0.8528587318956852,
|
| 579 |
+
"num_tokens": 40643225.0,
|
| 580 |
+
"step": 570
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"entropy": 0.4868734121322632,
|
| 584 |
+
"epoch": 3.068783068783069,
|
| 585 |
+
"grad_norm": 1.2734375,
|
| 586 |
+
"learning_rate": 1.6248281427513293e-05,
|
| 587 |
+
"loss": 0.49291515350341797,
|
| 588 |
+
"mean_token_accuracy": 0.8561817370355129,
|
| 589 |
+
"num_tokens": 41359577.0,
|
| 590 |
+
"step": 580
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"entropy": 0.490956873819232,
|
| 594 |
+
"epoch": 3.121693121693122,
|
| 595 |
+
"grad_norm": 1.34375,
|
| 596 |
+
"learning_rate": 1.6113628797338182e-05,
|
| 597 |
+
"loss": 0.48886570930480955,
|
| 598 |
+
"mean_token_accuracy": 0.85729940533638,
|
| 599 |
+
"num_tokens": 42068599.0,
|
| 600 |
+
"step": 590
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"entropy": 0.47903846949338913,
|
| 604 |
+
"epoch": 3.1746031746031744,
|
| 605 |
+
"grad_norm": 1.265625,
|
| 606 |
+
"learning_rate": 1.597718230310167e-05,
|
| 607 |
+
"loss": 0.4768198013305664,
|
| 608 |
+
"mean_token_accuracy": 0.8604256264865399,
|
| 609 |
+
"num_tokens": 42785175.0,
|
| 610 |
+
"step": 600
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"entropy": 0.4701235756278038,
|
| 614 |
+
"epoch": 3.2275132275132274,
|
| 615 |
+
"grad_norm": 1.21875,
|
| 616 |
+
"learning_rate": 1.5838981981003273e-05,
|
| 617 |
+
"loss": 0.4688236713409424,
|
| 618 |
+
"mean_token_accuracy": 0.8626316115260124,
|
| 619 |
+
"num_tokens": 43504295.0,
|
| 620 |
+
"step": 610
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"entropy": 0.4784398306161165,
|
| 624 |
+
"epoch": 3.2804232804232805,
|
| 625 |
+
"grad_norm": 1.2734375,
|
| 626 |
+
"learning_rate": 1.5699068381851576e-05,
|
| 627 |
+
"loss": 0.47977123260498045,
|
| 628 |
+
"mean_token_accuracy": 0.8589689329266548,
|
| 629 |
+
"num_tokens": 44210262.0,
|
| 630 |
+
"step": 620
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"entropy": 0.48374257422983646,
|
| 634 |
+
"epoch": 3.3333333333333335,
|
| 635 |
+
"grad_norm": 1.28125,
|
| 636 |
+
"learning_rate": 1.5557482559165777e-05,
|
| 637 |
+
"loss": 0.4810905456542969,
|
| 638 |
+
"mean_token_accuracy": 0.8588056646287441,
|
| 639 |
+
"num_tokens": 44921899.0,
|
| 640 |
+
"step": 630
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"entropy": 0.47944297045469286,
|
| 644 |
+
"epoch": 3.386243386243386,
|
| 645 |
+
"grad_norm": 1.234375,
|
| 646 |
+
"learning_rate": 1.541426605712975e-05,
|
| 647 |
+
"loss": 0.480026912689209,
|
| 648 |
+
"mean_token_accuracy": 0.8588721737265587,
|
| 649 |
+
"num_tokens": 45631735.0,
|
| 650 |
+
"step": 640
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"entropy": 0.4891296233981848,
|
| 654 |
+
"epoch": 3.439153439153439,
|
| 655 |
+
"grad_norm": 1.515625,
|
| 656 |
+
"learning_rate": 1.526946089840214e-05,
|
| 657 |
+
"loss": 0.4879168510437012,
|
| 658 |
+
"mean_token_accuracy": 0.8568189851939678,
|
| 659 |
+
"num_tokens": 46339604.0,
|
| 660 |
+
"step": 650
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"entropy": 0.4852400213479996,
|
| 664 |
+
"epoch": 3.492063492063492,
|
| 665 |
+
"grad_norm": 1.2578125,
|
| 666 |
+
"learning_rate": 1.5123109571786034e-05,
|
| 667 |
+
"loss": 0.48280906677246094,
|
| 668 |
+
"mean_token_accuracy": 0.8593833275139332,
|
| 669 |
+
"num_tokens": 47060193.0,
|
| 670 |
+
"step": 660
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"entropy": 0.48768885880708696,
|
| 674 |
+
"epoch": 3.544973544973545,
|
| 675 |
+
"grad_norm": 1.2890625,
|
| 676 |
+
"learning_rate": 1.4975255019761896e-05,
|
| 677 |
+
"loss": 0.4901734352111816,
|
| 678 |
+
"mean_token_accuracy": 0.8567128427326679,
|
| 679 |
+
"num_tokens": 47768090.0,
|
| 680 |
+
"step": 670
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"entropy": 0.4818777460604906,
|
| 684 |
+
"epoch": 3.597883597883598,
|
| 685 |
+
"grad_norm": 1.25,
|
| 686 |
+
"learning_rate": 1.4825940625887343e-05,
|
| 687 |
+
"loss": 0.4794710636138916,
|
| 688 |
+
"mean_token_accuracy": 0.8597230516374111,
|
| 689 |
+
"num_tokens": 48486016.0,
|
| 690 |
+
"step": 680
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"entropy": 0.4782646283507347,
|
| 694 |
+
"epoch": 3.6507936507936507,
|
| 695 |
+
"grad_norm": 1.21875,
|
| 696 |
+
"learning_rate": 1.467521020206753e-05,
|
| 697 |
+
"loss": 0.4785307884216309,
|
| 698 |
+
"mean_token_accuracy": 0.8597849570214748,
|
| 699 |
+
"num_tokens": 49196144.0,
|
| 700 |
+
"step": 690
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"entropy": 0.48901230953633784,
|
| 704 |
+
"epoch": 3.7037037037037037,
|
| 705 |
+
"grad_norm": 1.3046875,
|
| 706 |
+
"learning_rate": 1.4523107975699833e-05,
|
| 707 |
+
"loss": 0.4887538909912109,
|
| 708 |
+
"mean_token_accuracy": 0.8576067104935646,
|
| 709 |
+
"num_tokens": 49912021.0,
|
| 710 |
+
"step": 700
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"entropy": 0.48083703890442847,
|
| 714 |
+
"epoch": 3.7566137566137567,
|
| 715 |
+
"grad_norm": 1.2265625,
|
| 716 |
+
"learning_rate": 1.4369678576696633e-05,
|
| 717 |
+
"loss": 0.47971267700195314,
|
| 718 |
+
"mean_token_accuracy": 0.8592194318771362,
|
| 719 |
+
"num_tokens": 50634262.0,
|
| 720 |
+
"step": 710
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"entropy": 0.47896665148437023,
|
| 724 |
+
"epoch": 3.8095238095238093,
|
| 725 |
+
"grad_norm": 1.265625,
|
| 726 |
+
"learning_rate": 1.4214967024389978e-05,
|
| 727 |
+
"loss": 0.4767903327941895,
|
| 728 |
+
"mean_token_accuracy": 0.8601033940911293,
|
| 729 |
+
"num_tokens": 51343452.0,
|
| 730 |
+
"step": 720
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"entropy": 0.4764186542481184,
|
| 734 |
+
"epoch": 3.8624338624338623,
|
| 735 |
+
"grad_norm": 1.2578125,
|
| 736 |
+
"learning_rate": 1.4059018714321986e-05,
|
| 737 |
+
"loss": 0.47649202346801756,
|
| 738 |
+
"mean_token_accuracy": 0.8595595695078373,
|
| 739 |
+
"num_tokens": 52046363.0,
|
| 740 |
+
"step": 730
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"entropy": 0.476634269207716,
|
| 744 |
+
"epoch": 3.9153439153439153,
|
| 745 |
+
"grad_norm": 1.265625,
|
| 746 |
+
"learning_rate": 1.390187940492486e-05,
|
| 747 |
+
"loss": 0.4764859199523926,
|
| 748 |
+
"mean_token_accuracy": 0.8600306704640388,
|
| 749 |
+
"num_tokens": 52755135.0,
|
| 750 |
+
"step": 740
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"entropy": 0.47171572409570217,
|
| 754 |
+
"epoch": 3.9682539682539684,
|
| 755 |
+
"grad_norm": 1.28125,
|
| 756 |
+
"learning_rate": 1.3743595204094442e-05,
|
| 757 |
+
"loss": 0.47093925476074217,
|
| 758 |
+
"mean_token_accuracy": 0.8609667584300041,
|
| 759 |
+
"num_tokens": 53473566.0,
|
| 760 |
+
"step": 750
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"entropy": 0.464452663064003,
|
| 764 |
+
"epoch": 4.021164021164021,
|
| 765 |
+
"grad_norm": 1.1875,
|
| 766 |
+
"learning_rate": 1.3584212555661187e-05,
|
| 767 |
+
"loss": 0.4580575942993164,
|
| 768 |
+
"mean_token_accuracy": 0.8645686432719231,
|
| 769 |
+
"num_tokens": 54190492.0,
|
| 770 |
+
"step": 760
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"entropy": 0.4455810472369194,
|
| 774 |
+
"epoch": 4.074074074074074,
|
| 775 |
+
"grad_norm": 1.25,
|
| 776 |
+
"learning_rate": 1.3423778225762609e-05,
|
| 777 |
+
"loss": 0.4465456962585449,
|
| 778 |
+
"mean_token_accuracy": 0.8673712603747845,
|
| 779 |
+
"num_tokens": 54897936.0,
|
| 780 |
+
"step": 770
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"entropy": 0.4556675568223,
|
| 784 |
+
"epoch": 4.1269841269841265,
|
| 785 |
+
"grad_norm": 1.3671875,
|
| 786 |
+
"learning_rate": 1.3262339289121129e-05,
|
| 787 |
+
"loss": 0.4551807403564453,
|
| 788 |
+
"mean_token_accuracy": 0.8652156829833985,
|
| 789 |
+
"num_tokens": 55609388.0,
|
| 790 |
+
"step": 780
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"entropy": 0.45422575771808626,
|
| 794 |
+
"epoch": 4.1798941798941796,
|
| 795 |
+
"grad_norm": 1.34375,
|
| 796 |
+
"learning_rate": 1.3099943115231386e-05,
|
| 797 |
+
"loss": 0.45353970527648924,
|
| 798 |
+
"mean_token_accuracy": 0.8658270433545112,
|
| 799 |
+
"num_tokens": 56329299.0,
|
| 800 |
+
"step": 790
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"entropy": 0.45179531499743464,
|
| 804 |
+
"epoch": 4.232804232804233,
|
| 805 |
+
"grad_norm": 1.3046875,
|
| 806 |
+
"learning_rate": 1.2936637354461064e-05,
|
| 807 |
+
"loss": 0.44892282485961915,
|
| 808 |
+
"mean_token_accuracy": 0.8666822284460067,
|
| 809 |
+
"num_tokens": 57042704.0,
|
| 810 |
+
"step": 800
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"entropy": 0.45655727535486224,
|
| 814 |
+
"epoch": 4.285714285714286,
|
| 815 |
+
"grad_norm": 1.3046875,
|
| 816 |
+
"learning_rate": 1.2772469924069285e-05,
|
| 817 |
+
"loss": 0.4565535545349121,
|
| 818 |
+
"mean_token_accuracy": 0.8650884844362736,
|
| 819 |
+
"num_tokens": 57755769.0,
|
| 820 |
+
"step": 810
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"entropy": 0.4504398550838232,
|
| 824 |
+
"epoch": 4.338624338624339,
|
| 825 |
+
"grad_norm": 1.2109375,
|
| 826 |
+
"learning_rate": 1.2607488994146703e-05,
|
| 827 |
+
"loss": 0.44783792495727537,
|
| 828 |
+
"mean_token_accuracy": 0.8668409429490567,
|
| 829 |
+
"num_tokens": 58467967.0,
|
| 830 |
+
"step": 820
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"entropy": 0.45743274949491025,
|
| 834 |
+
"epoch": 4.391534391534392,
|
| 835 |
+
"grad_norm": 1.2734375,
|
| 836 |
+
"learning_rate": 1.2441742973481432e-05,
|
| 837 |
+
"loss": 0.45568046569824217,
|
| 838 |
+
"mean_token_accuracy": 0.8652968890964985,
|
| 839 |
+
"num_tokens": 59188579.0,
|
| 840 |
+
"step": 830
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"entropy": 0.44639028087258337,
|
| 844 |
+
"epoch": 4.444444444444445,
|
| 845 |
+
"grad_norm": 1.3515625,
|
| 846 |
+
"learning_rate": 1.2275280495354889e-05,
|
| 847 |
+
"loss": 0.44577698707580565,
|
| 848 |
+
"mean_token_accuracy": 0.8682797506451607,
|
| 849 |
+
"num_tokens": 59899589.0,
|
| 850 |
+
"step": 840
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"entropy": 0.4485243815928698,
|
| 854 |
+
"epoch": 4.497354497354498,
|
| 855 |
+
"grad_norm": 1.265625,
|
| 856 |
+
"learning_rate": 1.21081504032718e-05,
|
| 857 |
+
"loss": 0.4481363773345947,
|
| 858 |
+
"mean_token_accuracy": 0.8678335674107075,
|
| 859 |
+
"num_tokens": 60620913.0,
|
| 860 |
+
"step": 850
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"entropy": 0.4525593984872103,
|
| 864 |
+
"epoch": 4.550264550264551,
|
| 865 |
+
"grad_norm": 1.3125,
|
| 866 |
+
"learning_rate": 1.1940401736628524e-05,
|
| 867 |
+
"loss": 0.4498861312866211,
|
| 868 |
+
"mean_token_accuracy": 0.8662019960582257,
|
| 869 |
+
"num_tokens": 61330675.0,
|
| 870 |
+
"step": 860
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"entropy": 0.4605423126369715,
|
| 874 |
+
"epoch": 4.603174603174603,
|
| 875 |
+
"grad_norm": 1.4140625,
|
| 876 |
+
"learning_rate": 1.1772083716323878e-05,
|
| 877 |
+
"loss": 0.4589731216430664,
|
| 878 |
+
"mean_token_accuracy": 0.8638311237096786,
|
| 879 |
+
"num_tokens": 62041645.0,
|
| 880 |
+
"step": 870
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"entropy": 0.44991469383239746,
|
| 884 |
+
"epoch": 4.656084656084656,
|
| 885 |
+
"grad_norm": 1.25,
|
| 886 |
+
"learning_rate": 1.1603245730316724e-05,
|
| 887 |
+
"loss": 0.447885799407959,
|
| 888 |
+
"mean_token_accuracy": 0.8659505993127823,
|
| 889 |
+
"num_tokens": 62750446.0,
|
| 890 |
+
"step": 880
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"entropy": 0.4469064295291901,
|
| 894 |
+
"epoch": 4.708994708994709,
|
| 895 |
+
"grad_norm": 1.28125,
|
| 896 |
+
"learning_rate": 1.1433937319134512e-05,
|
| 897 |
+
"loss": 0.44426918029785156,
|
| 898 |
+
"mean_token_accuracy": 0.868197600543499,
|
| 899 |
+
"num_tokens": 63460008.0,
|
| 900 |
+
"step": 890
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"entropy": 0.4514023557305336,
|
| 904 |
+
"epoch": 4.761904761904762,
|
| 905 |
+
"grad_norm": 1.3046875,
|
| 906 |
+
"learning_rate": 1.1264208161337091e-05,
|
| 907 |
+
"loss": 0.45235776901245117,
|
| 908 |
+
"mean_token_accuracy": 0.8655660562217236,
|
| 909 |
+
"num_tokens": 64164803.0,
|
| 910 |
+
"step": 900
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"entropy": 0.4533011969178915,
|
| 914 |
+
"epoch": 4.814814814814815,
|
| 915 |
+
"grad_norm": 1.328125,
|
| 916 |
+
"learning_rate": 1.1094108058940005e-05,
|
| 917 |
+
"loss": 0.4506092548370361,
|
| 918 |
+
"mean_token_accuracy": 0.865882170945406,
|
| 919 |
+
"num_tokens": 64878163.0,
|
| 920 |
+
"step": 910
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"entropy": 0.44160800203680994,
|
| 924 |
+
"epoch": 4.867724867724868,
|
| 925 |
+
"grad_norm": 1.296875,
|
| 926 |
+
"learning_rate": 1.0923686922801537e-05,
|
| 927 |
+
"loss": 0.4400449752807617,
|
| 928 |
+
"mean_token_accuracy": 0.868823654204607,
|
| 929 |
+
"num_tokens": 65599878.0,
|
| 930 |
+
"step": 920
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"entropy": 0.4407290514558554,
|
| 934 |
+
"epoch": 4.920634920634921,
|
| 935 |
+
"grad_norm": 1.2734375,
|
| 936 |
+
"learning_rate": 1.0752994757977862e-05,
|
| 937 |
+
"loss": 0.4377620697021484,
|
| 938 |
+
"mean_token_accuracy": 0.8691893480718136,
|
| 939 |
+
"num_tokens": 66312478.0,
|
| 940 |
+
"step": 930
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"entropy": 0.45381851457059386,
|
| 944 |
+
"epoch": 4.973544973544973,
|
| 945 |
+
"grad_norm": 1.265625,
|
| 946 |
+
"learning_rate": 1.058208164905052e-05,
|
| 947 |
+
"loss": 0.45404396057128904,
|
| 948 |
+
"mean_token_accuracy": 0.8656110934913158,
|
| 949 |
+
"num_tokens": 67026652.0,
|
| 950 |
+
"step": 940
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"entropy": 0.4451509706676006,
|
| 954 |
+
"epoch": 5.026455026455026,
|
| 955 |
+
"grad_norm": 1.265625,
|
| 956 |
+
"learning_rate": 1.041099774543059e-05,
|
| 957 |
+
"loss": 0.43704824447631835,
|
| 958 |
+
"mean_token_accuracy": 0.8695758722722531,
|
| 959 |
+
"num_tokens": 67743837.0,
|
| 960 |
+
"step": 950
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"entropy": 0.4223184697329998,
|
| 964 |
+
"epoch": 5.079365079365079,
|
| 965 |
+
"grad_norm": 1.375,
|
| 966 |
+
"learning_rate": 1.023979324664382e-05,
|
| 967 |
+
"loss": 0.4231288909912109,
|
| 968 |
+
"mean_token_accuracy": 0.8729359813034534,
|
| 969 |
+
"num_tokens": 68459077.0,
|
| 970 |
+
"step": 960
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"entropy": 0.4201000899076462,
|
| 974 |
+
"epoch": 5.132275132275132,
|
| 975 |
+
"grad_norm": 1.3515625,
|
| 976 |
+
"learning_rate": 1.0068518387601038e-05,
|
| 977 |
+
"loss": 0.4157362461090088,
|
| 978 |
+
"mean_token_accuracy": 0.8749549694359302,
|
| 979 |
+
"num_tokens": 69175886.0,
|
| 980 |
+
"step": 970
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"entropy": 0.4203161742538214,
|
| 984 |
+
"epoch": 5.185185185185185,
|
| 985 |
+
"grad_norm": 1.328125,
|
| 986 |
+
"learning_rate": 9.897223423858226e-06,
|
| 987 |
+
"loss": 0.4191565036773682,
|
| 988 |
+
"mean_token_accuracy": 0.8741719849407673,
|
| 989 |
+
"num_tokens": 69888377.0,
|
| 990 |
+
"step": 980
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"entropy": 0.4278976671397686,
|
| 994 |
+
"epoch": 5.238095238095238,
|
| 995 |
+
"grad_norm": 1.3125,
|
| 996 |
+
"learning_rate": 9.725958616870495e-06,
|
| 997 |
+
"loss": 0.42658100128173826,
|
| 998 |
+
"mean_token_accuracy": 0.8726167634129525,
|
| 999 |
+
"num_tokens": 70604584.0,
|
| 1000 |
+
"step": 990
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"entropy": 0.43622247986495494,
|
| 1004 |
+
"epoch": 5.291005291005291,
|
| 1005 |
+
"grad_norm": 1.3359375,
|
| 1006 |
+
"learning_rate": 9.554774219244344e-06,
|
| 1007 |
+
"loss": 0.4340261459350586,
|
| 1008 |
+
"mean_token_accuracy": 0.8702079840004444,
|
| 1009 |
+
"num_tokens": 71321169.0,
|
| 1010 |
+
"step": 1000
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"entropy": 0.42725917547941206,
|
| 1014 |
+
"epoch": 5.343915343915344,
|
| 1015 |
+
"grad_norm": 1.296875,
|
| 1016 |
+
"learning_rate": 9.38372045999252e-06,
|
| 1017 |
+
"loss": 0.4253455638885498,
|
| 1018 |
+
"mean_token_accuracy": 0.8729547396302223,
|
| 1019 |
+
"num_tokens": 72041175.0,
|
| 1020 |
+
"step": 1010
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"entropy": 0.4243607401847839,
|
| 1024 |
+
"epoch": 5.396825396825397,
|
| 1025 |
+
"grad_norm": 1.3828125,
|
| 1026 |
+
"learning_rate": 9.212847529795784e-06,
|
| 1027 |
+
"loss": 0.42325568199157715,
|
| 1028 |
+
"mean_token_accuracy": 0.8731870628893376,
|
| 1029 |
+
"num_tokens": 72749079.0,
|
| 1030 |
+
"step": 1020
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"entropy": 0.4303685359656811,
|
| 1034 |
+
"epoch": 5.449735449735449,
|
| 1035 |
+
"grad_norm": 1.2734375,
|
| 1036 |
+
"learning_rate": 9.042205566275952e-06,
|
| 1037 |
+
"loss": 0.4277714729309082,
|
| 1038 |
+
"mean_token_accuracy": 0.8721092127263546,
|
| 1039 |
+
"num_tokens": 73462893.0,
|
| 1040 |
+
"step": 1030
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"entropy": 0.43162608668208124,
|
| 1044 |
+
"epoch": 5.502645502645502,
|
| 1045 |
+
"grad_norm": 1.28125,
|
| 1046 |
+
"learning_rate": 8.871844639284479e-06,
|
| 1047 |
+
"loss": 0.4301889419555664,
|
| 1048 |
+
"mean_token_accuracy": 0.8712550736963749,
|
| 1049 |
+
"num_tokens": 74167133.0,
|
| 1050 |
+
"step": 1040
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"entropy": 0.43598169907927514,
|
| 1054 |
+
"epoch": 5.555555555555555,
|
| 1055 |
+
"grad_norm": 1.2734375,
|
| 1056 |
+
"learning_rate": 8.701814736210933e-06,
|
| 1057 |
+
"loss": 0.435499382019043,
|
| 1058 |
+
"mean_token_accuracy": 0.8700432546436787,
|
| 1059 |
+
"num_tokens": 74870399.0,
|
| 1060 |
+
"step": 1050
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"entropy": 0.4320717755705118,
|
| 1064 |
+
"epoch": 5.608465608465608,
|
| 1065 |
+
"grad_norm": 1.3203125,
|
| 1066 |
+
"learning_rate": 8.532165747315686e-06,
|
| 1067 |
+
"loss": 0.43087425231933596,
|
| 1068 |
+
"mean_token_accuracy": 0.8712112359702587,
|
| 1069 |
+
"num_tokens": 75590782.0,
|
| 1070 |
+
"step": 1060
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"entropy": 0.43664140328764917,
|
| 1074 |
+
"epoch": 5.661375661375661,
|
| 1075 |
+
"grad_norm": 1.2734375,
|
| 1076 |
+
"learning_rate": 8.362947451091072e-06,
|
| 1077 |
+
"loss": 0.4324526309967041,
|
| 1078 |
+
"mean_token_accuracy": 0.8710537351667881,
|
| 1079 |
+
"num_tokens": 76305788.0,
|
| 1080 |
+
"step": 1070
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"entropy": 0.43034715987741945,
|
| 1084 |
+
"epoch": 5.714285714285714,
|
| 1085 |
+
"grad_norm": 1.2890625,
|
| 1086 |
+
"learning_rate": 8.194209499655364e-06,
|
| 1087 |
+
"loss": 0.4322338581085205,
|
| 1088 |
+
"mean_token_accuracy": 0.8713471658527852,
|
| 1089 |
+
"num_tokens": 77023579.0,
|
| 1090 |
+
"step": 1080
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"entropy": 0.4325398612767458,
|
| 1094 |
+
"epoch": 5.767195767195767,
|
| 1095 |
+
"grad_norm": 1.40625,
|
| 1096 |
+
"learning_rate": 8.026001404183831e-06,
|
| 1097 |
+
"loss": 0.4275611400604248,
|
| 1098 |
+
"mean_token_accuracy": 0.8720052778720856,
|
| 1099 |
+
"num_tokens": 77720821.0,
|
| 1100 |
+
"step": 1090
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"entropy": 0.43604847937822344,
|
| 1104 |
+
"epoch": 5.8201058201058204,
|
| 1105 |
+
"grad_norm": 1.296875,
|
| 1106 |
+
"learning_rate": 7.85837252038112e-06,
|
| 1107 |
+
"loss": 0.43659210205078125,
|
| 1108 |
+
"mean_token_accuracy": 0.8698216550052166,
|
| 1109 |
+
"num_tokens": 78431671.0,
|
| 1110 |
+
"step": 1100
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"entropy": 0.4338367357850075,
|
| 1114 |
+
"epoch": 5.8730158730158735,
|
| 1115 |
+
"grad_norm": 1.3125,
|
| 1116 |
+
"learning_rate": 7.69137203399931e-06,
|
| 1117 |
+
"loss": 0.43023014068603516,
|
| 1118 |
+
"mean_token_accuracy": 0.8713279463350773,
|
| 1119 |
+
"num_tokens": 79141393.0,
|
| 1120 |
+
"step": 1110
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"entropy": 0.42576855383813383,
|
| 1124 |
+
"epoch": 5.925925925925926,
|
| 1125 |
+
"grad_norm": 1.234375,
|
| 1126 |
+
"learning_rate": 7.525048946405791e-06,
|
| 1127 |
+
"loss": 0.4262782096862793,
|
| 1128 |
+
"mean_token_accuracy": 0.8731129266321659,
|
| 1129 |
+
"num_tokens": 79858356.0,
|
| 1130 |
+
"step": 1120
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"entropy": 0.43451742343604566,
|
| 1134 |
+
"epoch": 5.978835978835979,
|
| 1135 |
+
"grad_norm": 1.3046875,
|
| 1136 |
+
"learning_rate": 7.359452060205245e-06,
|
| 1137 |
+
"loss": 0.43216233253479003,
|
| 1138 |
+
"mean_token_accuracy": 0.8712433956563472,
|
| 1139 |
+
"num_tokens": 80566532.0,
|
| 1140 |
+
"step": 1130
|
| 1141 |
+
}
|
| 1142 |
+
],
|
| 1143 |
+
"logging_steps": 10,
|
| 1144 |
+
"max_steps": 1890,
|
| 1145 |
+
"num_input_tokens_seen": 0,
|
| 1146 |
+
"num_train_epochs": 10,
|
| 1147 |
+
"save_steps": 500,
|
| 1148 |
+
"stateful_callbacks": {
|
| 1149 |
+
"TrainerControl": {
|
| 1150 |
+
"args": {
|
| 1151 |
+
"should_epoch_stop": false,
|
| 1152 |
+
"should_evaluate": false,
|
| 1153 |
+
"should_log": false,
|
| 1154 |
+
"should_save": true,
|
| 1155 |
+
"should_training_stop": false
|
| 1156 |
+
},
|
| 1157 |
+
"attributes": {}
|
| 1158 |
+
}
|
| 1159 |
+
},
|
| 1160 |
+
"total_flos": 1.45969922766694e+18,
|
| 1161 |
+
"train_batch_size": 8,
|
| 1162 |
+
"trial_name": null,
|
| 1163 |
+
"trial_params": null
|
| 1164 |
+
}
|
checkpoint-1134/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:12cf8133cc6c754dc73b2cbaecbb0aa0b7e2a28209b2cf3544fd63386978a2bb
|
| 3 |
+
size 5841
|