Upload folder using huggingface_hub
Browse files- .gitattributes +1 -0
- 400M/chat_template.jinja +1 -0
- 400M/config.json +170 -0
- 400M/generation_config.json +17 -0
- 400M/latest +1 -0
- 400M/model.safetensors +3 -0
- 400M/processor_config.json +62 -0
- 400M/rng_state_0.pth +3 -0
- 400M/rng_state_1.pth +3 -0
- 400M/rng_state_10.pth +3 -0
- 400M/rng_state_11.pth +3 -0
- 400M/rng_state_2.pth +3 -0
- 400M/rng_state_3.pth +3 -0
- 400M/rng_state_4.pth +3 -0
- 400M/rng_state_5.pth +3 -0
- 400M/rng_state_6.pth +3 -0
- 400M/rng_state_7.pth +3 -0
- 400M/rng_state_8.pth +3 -0
- 400M/rng_state_9.pth +3 -0
- 400M/scheduler.pt +3 -0
- 400M/tokenizer.json +3 -0
- 400M/tokenizer_config.json +56 -0
- 400M/trainer_state.json +2883 -0
- 400M/training_args.bin +3 -0
.gitattributes
CHANGED
|
@@ -34,3 +34,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
200M/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
200M/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
400M/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
400M/chat_template.jinja
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content }}{% endif %}{% endfor %}
|
400M/config.json
ADDED
|
@@ -0,0 +1,170 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Gemma4UnifiedForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"audio_config": {
|
| 6 |
+
"_name_or_path": "",
|
| 7 |
+
"architectures": null,
|
| 8 |
+
"audio_embed_dim": 640,
|
| 9 |
+
"chunk_size_feed_forward": 0,
|
| 10 |
+
"dtype": "bfloat16",
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "LABEL_0",
|
| 13 |
+
"1": "LABEL_1"
|
| 14 |
+
},
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"is_encoder_decoder": false,
|
| 17 |
+
"label2id": {
|
| 18 |
+
"LABEL_0": 0,
|
| 19 |
+
"LABEL_1": 1
|
| 20 |
+
},
|
| 21 |
+
"model_type": "gemma4_unified_audio",
|
| 22 |
+
"output_attentions": false,
|
| 23 |
+
"output_hidden_states": false,
|
| 24 |
+
"problem_type": null,
|
| 25 |
+
"return_dict": true,
|
| 26 |
+
"rms_norm_eps": 1e-06
|
| 27 |
+
},
|
| 28 |
+
"audio_token_id": 258881,
|
| 29 |
+
"boa_token_id": 256000,
|
| 30 |
+
"boi_token_id": 255999,
|
| 31 |
+
"bos_token_id": 2,
|
| 32 |
+
"dtype": "bfloat16",
|
| 33 |
+
"eoa_token_index": 258883,
|
| 34 |
+
"eoi_token_id": 258882,
|
| 35 |
+
"eos_token_id": 1,
|
| 36 |
+
"hidden_size": 3840,
|
| 37 |
+
"image_token_id": 258880,
|
| 38 |
+
"initializer_range": 0.02,
|
| 39 |
+
"model_type": "gemma4_unified",
|
| 40 |
+
"pad_token_id": 0,
|
| 41 |
+
"text_config": {
|
| 42 |
+
"attention_bias": false,
|
| 43 |
+
"attention_dropout": 0.0,
|
| 44 |
+
"attention_k_eq_v": true,
|
| 45 |
+
"bos_token_id": 2,
|
| 46 |
+
"dtype": "bfloat16",
|
| 47 |
+
"enable_moe_block": false,
|
| 48 |
+
"eos_token_id": 1,
|
| 49 |
+
"final_logit_softcapping": 30.0,
|
| 50 |
+
"global_head_dim": 512,
|
| 51 |
+
"head_dim": 256,
|
| 52 |
+
"hidden_activation": "gelu_pytorch_tanh",
|
| 53 |
+
"hidden_size": 3840,
|
| 54 |
+
"hidden_size_per_layer_input": 0,
|
| 55 |
+
"initializer_range": 0.02,
|
| 56 |
+
"intermediate_size": 15360,
|
| 57 |
+
"layer_types": [
|
| 58 |
+
"sliding_attention",
|
| 59 |
+
"sliding_attention",
|
| 60 |
+
"sliding_attention",
|
| 61 |
+
"sliding_attention",
|
| 62 |
+
"sliding_attention",
|
| 63 |
+
"full_attention",
|
| 64 |
+
"sliding_attention",
|
| 65 |
+
"sliding_attention",
|
| 66 |
+
"sliding_attention",
|
| 67 |
+
"sliding_attention",
|
| 68 |
+
"sliding_attention",
|
| 69 |
+
"full_attention",
|
| 70 |
+
"sliding_attention",
|
| 71 |
+
"sliding_attention",
|
| 72 |
+
"sliding_attention",
|
| 73 |
+
"sliding_attention",
|
| 74 |
+
"sliding_attention",
|
| 75 |
+
"full_attention",
|
| 76 |
+
"sliding_attention",
|
| 77 |
+
"sliding_attention",
|
| 78 |
+
"sliding_attention",
|
| 79 |
+
"sliding_attention",
|
| 80 |
+
"sliding_attention",
|
| 81 |
+
"full_attention",
|
| 82 |
+
"sliding_attention",
|
| 83 |
+
"sliding_attention",
|
| 84 |
+
"sliding_attention",
|
| 85 |
+
"sliding_attention",
|
| 86 |
+
"sliding_attention",
|
| 87 |
+
"full_attention",
|
| 88 |
+
"sliding_attention",
|
| 89 |
+
"sliding_attention",
|
| 90 |
+
"sliding_attention",
|
| 91 |
+
"sliding_attention",
|
| 92 |
+
"sliding_attention",
|
| 93 |
+
"full_attention",
|
| 94 |
+
"sliding_attention",
|
| 95 |
+
"sliding_attention",
|
| 96 |
+
"sliding_attention",
|
| 97 |
+
"sliding_attention",
|
| 98 |
+
"sliding_attention",
|
| 99 |
+
"full_attention",
|
| 100 |
+
"sliding_attention",
|
| 101 |
+
"sliding_attention",
|
| 102 |
+
"sliding_attention",
|
| 103 |
+
"sliding_attention",
|
| 104 |
+
"sliding_attention",
|
| 105 |
+
"full_attention"
|
| 106 |
+
],
|
| 107 |
+
"max_position_embeddings": 262144,
|
| 108 |
+
"model_type": "gemma4_unified_text",
|
| 109 |
+
"moe_intermediate_size": null,
|
| 110 |
+
"num_attention_heads": 16,
|
| 111 |
+
"num_experts": null,
|
| 112 |
+
"num_global_key_value_heads": 1,
|
| 113 |
+
"num_hidden_layers": 48,
|
| 114 |
+
"num_key_value_heads": 8,
|
| 115 |
+
"num_kv_shared_layers": 0,
|
| 116 |
+
"pad_token_id": 0,
|
| 117 |
+
"rms_norm_eps": 1e-06,
|
| 118 |
+
"rope_parameters": {
|
| 119 |
+
"full_attention": {
|
| 120 |
+
"partial_rotary_factor": 0.25,
|
| 121 |
+
"rope_theta": 1000000.0,
|
| 122 |
+
"rope_type": "proportional"
|
| 123 |
+
},
|
| 124 |
+
"sliding_attention": {
|
| 125 |
+
"rope_theta": 10000.0,
|
| 126 |
+
"rope_type": "default"
|
| 127 |
+
}
|
| 128 |
+
},
|
| 129 |
+
"sliding_window": 1024,
|
| 130 |
+
"tie_word_embeddings": true,
|
| 131 |
+
"top_k_experts": null,
|
| 132 |
+
"use_bidirectional_attention": "vision",
|
| 133 |
+
"use_cache": false,
|
| 134 |
+
"use_double_wide_mlp": false,
|
| 135 |
+
"vocab_size": 262144,
|
| 136 |
+
"vocab_size_per_layer_input": 262144
|
| 137 |
+
},
|
| 138 |
+
"tie_word_embeddings": true,
|
| 139 |
+
"transformers_version": "5.10.0",
|
| 140 |
+
"use_cache": false,
|
| 141 |
+
"video_token_id": 258884,
|
| 142 |
+
"vision_config": {
|
| 143 |
+
"_name_or_path": "",
|
| 144 |
+
"architectures": null,
|
| 145 |
+
"chunk_size_feed_forward": 0,
|
| 146 |
+
"dtype": "bfloat16",
|
| 147 |
+
"id2label": {
|
| 148 |
+
"0": "LABEL_0",
|
| 149 |
+
"1": "LABEL_1"
|
| 150 |
+
},
|
| 151 |
+
"initializer_range": 0.02,
|
| 152 |
+
"is_encoder_decoder": false,
|
| 153 |
+
"label2id": {
|
| 154 |
+
"LABEL_0": 0,
|
| 155 |
+
"LABEL_1": 1
|
| 156 |
+
},
|
| 157 |
+
"mm_embed_dim": 3840,
|
| 158 |
+
"mm_posemb_size": 1120,
|
| 159 |
+
"model_type": "gemma4_unified_vision",
|
| 160 |
+
"num_soft_tokens": 280,
|
| 161 |
+
"output_attentions": false,
|
| 162 |
+
"output_hidden_states": false,
|
| 163 |
+
"output_proj_dims": 3840,
|
| 164 |
+
"patch_size": 16,
|
| 165 |
+
"pooling_kernel_size": 3,
|
| 166 |
+
"problem_type": null,
|
| 167 |
+
"return_dict": true,
|
| 168 |
+
"rms_norm_eps": 1e-06
|
| 169 |
+
}
|
| 170 |
+
}
|
400M/generation_config.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 2,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
1,
|
| 6 |
+
1
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 0,
|
| 9 |
+
"suppress_tokens": [
|
| 10 |
+
258883,
|
| 11 |
+
258882
|
| 12 |
+
],
|
| 13 |
+
"temperature": 1.0,
|
| 14 |
+
"top_k": 64,
|
| 15 |
+
"top_p": 0.95,
|
| 16 |
+
"transformers_version": "5.10.0"
|
| 17 |
+
}
|
400M/latest
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
global_step4070
|
400M/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d2492ca22b100bbce3642f89adcf266d769eb1ce442f3a0077b9a4fe5c4c6b86
|
| 3 |
+
size 23919549408
|
400M/processor_config.json
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"audio_ms_per_token": 40,
|
| 3 |
+
"audio_seq_length": 750,
|
| 4 |
+
"feature_extractor": {
|
| 5 |
+
"audio_samples_per_token": 640,
|
| 6 |
+
"feature_extractor_type": "Gemma4UnifiedAudioFeatureExtractor",
|
| 7 |
+
"feature_size": 640,
|
| 8 |
+
"padding_side": "right",
|
| 9 |
+
"padding_value": 0.0,
|
| 10 |
+
"return_attention_mask": true,
|
| 11 |
+
"sampling_rate": 16000
|
| 12 |
+
},
|
| 13 |
+
"image_processor": {
|
| 14 |
+
"do_convert_rgb": true,
|
| 15 |
+
"do_normalize": false,
|
| 16 |
+
"do_rescale": true,
|
| 17 |
+
"do_resize": true,
|
| 18 |
+
"image_mean": [
|
| 19 |
+
0.0,
|
| 20 |
+
0.0,
|
| 21 |
+
0.0
|
| 22 |
+
],
|
| 23 |
+
"image_processor_type": "Gemma4UnifiedImageProcessor",
|
| 24 |
+
"image_std": [
|
| 25 |
+
1.0,
|
| 26 |
+
1.0,
|
| 27 |
+
1.0
|
| 28 |
+
],
|
| 29 |
+
"max_soft_tokens": 280,
|
| 30 |
+
"patch_size": 16,
|
| 31 |
+
"pooling_kernel_size": 3,
|
| 32 |
+
"resample": 3,
|
| 33 |
+
"rescale_factor": 0.00392156862745098
|
| 34 |
+
},
|
| 35 |
+
"image_seq_length": 280,
|
| 36 |
+
"processor_class": "Gemma4UnifiedProcessor",
|
| 37 |
+
"video_processor": {
|
| 38 |
+
"do_convert_rgb": true,
|
| 39 |
+
"do_normalize": true,
|
| 40 |
+
"do_rescale": true,
|
| 41 |
+
"do_resize": true,
|
| 42 |
+
"do_sample_frames": true,
|
| 43 |
+
"image_mean": [
|
| 44 |
+
0.0,
|
| 45 |
+
0.0,
|
| 46 |
+
0.0
|
| 47 |
+
],
|
| 48 |
+
"image_std": [
|
| 49 |
+
1.0,
|
| 50 |
+
1.0,
|
| 51 |
+
1.0
|
| 52 |
+
],
|
| 53 |
+
"max_soft_tokens": 70,
|
| 54 |
+
"num_frames": 32,
|
| 55 |
+
"patch_size": 16,
|
| 56 |
+
"pooling_kernel_size": 3,
|
| 57 |
+
"resample": 3,
|
| 58 |
+
"rescale_factor": 0.00392156862745098,
|
| 59 |
+
"return_metadata": false,
|
| 60 |
+
"video_processor_type": "Gemma4UnifiedVideoProcessor"
|
| 61 |
+
}
|
| 62 |
+
}
|
400M/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4d7dcf54a8dbd1fff243640ddc034149fbe7c90ed70b8a95a6bb093e989a7577
|
| 3 |
+
size 14469
|
400M/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f2a2b0b9d330140afd742a57df45106276d425d38a78e53b0fd43dba9559f8d4
|
| 3 |
+
size 14469
|
400M/rng_state_10.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2ce222f8f11fd62fd6cb82264500aefbc9e7ef16907ed3e4aa3a8a99c21d95d3
|
| 3 |
+
size 14476
|
400M/rng_state_11.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d07144bfc4ed675dda70ab66923d17d6f9233560b2db09689cbe1f262c590c88
|
| 3 |
+
size 14476
|
400M/rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f5db634f6ed5151d3c898fbe3c365f950516c404bed926de16b253b26e63c7cf
|
| 3 |
+
size 14469
|
400M/rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ad774eade44525098fff2d5910e3d16565fca4c2af29ed0bc4d9315a8ac9607a
|
| 3 |
+
size 14469
|
400M/rng_state_4.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ebdee1b37be2238764bcfeed9dad268e938248152b0fba60eaf89adda99eac0f
|
| 3 |
+
size 14469
|
400M/rng_state_5.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6046157bd3d69a7cf6ee508b7359b30cb71488a11b11b07a614041b06813c553
|
| 3 |
+
size 14469
|
400M/rng_state_6.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ebc9300f101a324d0361f01aec1d57e26b57a703b4e536f4fe982c8ba57a9144
|
| 3 |
+
size 14469
|
400M/rng_state_7.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:357df7b609d6ffd7f22b998281f7856a711ee83505e5ed2835959554b0185df7
|
| 3 |
+
size 14469
|
400M/rng_state_8.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4fb33667d752a94ade6841f62b8b8fcc22256b4712bd568b141bc690d4a42e11
|
| 3 |
+
size 14469
|
400M/rng_state_9.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:354a7a38e97ba11b8ad29bf4d00d74f92295b32124cff0ef46baa862aa9b0f2f
|
| 3 |
+
size 14469
|
400M/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c373bca4c9a6ef963a59ee82f3b16a9a90b89b812375170f487261fe4db9e20a
|
| 3 |
+
size 1465
|
400M/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:12bac982b793c44b03d52a250a9f0d0b666813da566b910c24a6da0695fd11e6
|
| 3 |
+
size 32170070
|
400M/tokenizer_config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"audio_token": "<|audio|>",
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"boa_token": "<|audio>",
|
| 5 |
+
"boi_token": "<|image>",
|
| 6 |
+
"bos_token": "<bos>",
|
| 7 |
+
"eoa_token": "<audio|>",
|
| 8 |
+
"eoc_token": "<channel|>",
|
| 9 |
+
"eoi_token": "<image|>",
|
| 10 |
+
"eos_token": "<eos>",
|
| 11 |
+
"eot_token": "<turn|>",
|
| 12 |
+
"escape_token": "<|\"|>",
|
| 13 |
+
"etc_token": "<tool_call|>",
|
| 14 |
+
"etd_token": "<tool|>",
|
| 15 |
+
"etr_token": "<tool_response|>",
|
| 16 |
+
"extra_special_tokens": [
|
| 17 |
+
"<|video|>"
|
| 18 |
+
],
|
| 19 |
+
"image_token": "<|image|>",
|
| 20 |
+
"is_local": true,
|
| 21 |
+
"local_files_only": true,
|
| 22 |
+
"mask_token": "<mask>",
|
| 23 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 24 |
+
"model_specific_special_tokens": {
|
| 25 |
+
"audio_token": "<|audio|>",
|
| 26 |
+
"boa_token": "<|audio>",
|
| 27 |
+
"boi_token": "<|image>",
|
| 28 |
+
"eoa_token": "<audio|>",
|
| 29 |
+
"eoc_token": "<channel|>",
|
| 30 |
+
"eoi_token": "<image|>",
|
| 31 |
+
"eot_token": "<turn|>",
|
| 32 |
+
"escape_token": "<|\"|>",
|
| 33 |
+
"etc_token": "<tool_call|>",
|
| 34 |
+
"etd_token": "<tool|>",
|
| 35 |
+
"etr_token": "<tool_response|>",
|
| 36 |
+
"image_token": "<|image|>",
|
| 37 |
+
"soc_token": "<|channel>",
|
| 38 |
+
"sot_token": "<|turn>",
|
| 39 |
+
"stc_token": "<|tool_call>",
|
| 40 |
+
"std_token": "<|tool>",
|
| 41 |
+
"str_token": "<|tool_response>",
|
| 42 |
+
"think_token": "<|think|>"
|
| 43 |
+
},
|
| 44 |
+
"pad_token": "<pad>",
|
| 45 |
+
"padding_side": "right",
|
| 46 |
+
"processor_class": "Gemma4UnifiedProcessor",
|
| 47 |
+
"soc_token": "<|channel>",
|
| 48 |
+
"sot_token": "<|turn>",
|
| 49 |
+
"split_special_tokens": false,
|
| 50 |
+
"stc_token": "<|tool_call>",
|
| 51 |
+
"std_token": "<|tool>",
|
| 52 |
+
"str_token": "<|tool_response>",
|
| 53 |
+
"think_token": "<|think|>",
|
| 54 |
+
"tokenizer_class": "GemmaTokenizer",
|
| 55 |
+
"unk_token": "<unk>"
|
| 56 |
+
}
|
400M/trainer_state.json
ADDED
|
@@ -0,0 +1,2883 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.11159551424419402,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 4070,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0002741904526884374,
|
| 14 |
+
"grad_norm": 533.0549778145878,
|
| 15 |
+
"learning_rate": 2.910112359550562e-06,
|
| 16 |
+
"loss": 2.0392507553100585,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.0005483809053768748,
|
| 21 |
+
"grad_norm": 129.62802977018143,
|
| 22 |
+
"learning_rate": 3.921348314606742e-06,
|
| 23 |
+
"loss": 1.1088291168212892,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.0008225713580653121,
|
| 28 |
+
"grad_norm": 37.028700378797375,
|
| 29 |
+
"learning_rate": 4.932584269662922e-06,
|
| 30 |
+
"loss": 0.8763736724853516,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.0010967618107537496,
|
| 35 |
+
"grad_norm": 11.415105214785655,
|
| 36 |
+
"learning_rate": 5.943820224719102e-06,
|
| 37 |
+
"loss": 0.8334451675415039,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.0013709522634421868,
|
| 42 |
+
"grad_norm": 168.0092220161655,
|
| 43 |
+
"learning_rate": 6.955056179775282e-06,
|
| 44 |
+
"loss": 0.841560173034668,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.0016451427161306242,
|
| 49 |
+
"grad_norm": 10.9544582922062,
|
| 50 |
+
"learning_rate": 7.966292134831462e-06,
|
| 51 |
+
"loss": 0.7757131099700928,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.0019193331688190617,
|
| 56 |
+
"grad_norm": 52.12017293008163,
|
| 57 |
+
"learning_rate": 8.977528089887641e-06,
|
| 58 |
+
"loss": 0.7379593849182129,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.0021935236215074993,
|
| 63 |
+
"grad_norm": 24.23368291019739,
|
| 64 |
+
"learning_rate": 9.988764044943822e-06,
|
| 65 |
+
"loss": 0.871122169494629,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.0024677140741959365,
|
| 70 |
+
"grad_norm": 44.606225719656116,
|
| 71 |
+
"learning_rate": 1.1000000000000001e-05,
|
| 72 |
+
"loss": 0.8103988647460938,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.0027419045268843737,
|
| 77 |
+
"grad_norm": 19.394878252919106,
|
| 78 |
+
"learning_rate": 1.2011235955056182e-05,
|
| 79 |
+
"loss": 0.8088878631591797,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.0030160949795728113,
|
| 84 |
+
"grad_norm": 154.65284754287825,
|
| 85 |
+
"learning_rate": 1.3022471910112362e-05,
|
| 86 |
+
"loss": 0.7940959930419922,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.0032902854322612485,
|
| 91 |
+
"grad_norm": 58.453168158560985,
|
| 92 |
+
"learning_rate": 1.403370786516854e-05,
|
| 93 |
+
"loss": 0.8113164901733398,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.003564475884949686,
|
| 98 |
+
"grad_norm": 121.71906195134056,
|
| 99 |
+
"learning_rate": 1.504494382022472e-05,
|
| 100 |
+
"loss": 0.802008056640625,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.0038386663376381233,
|
| 105 |
+
"grad_norm": 7.063904217729592,
|
| 106 |
+
"learning_rate": 1.60561797752809e-05,
|
| 107 |
+
"loss": 0.7886086940765381,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.004112856790326561,
|
| 112 |
+
"grad_norm": 5.886231107848548,
|
| 113 |
+
"learning_rate": 1.706741573033708e-05,
|
| 114 |
+
"loss": 0.8553682327270508,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.004387047243014999,
|
| 119 |
+
"grad_norm": 17.479053516912124,
|
| 120 |
+
"learning_rate": 1.8078651685393256e-05,
|
| 121 |
+
"loss": 0.878786849975586,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.004661237695703435,
|
| 126 |
+
"grad_norm": 15.809884339669948,
|
| 127 |
+
"learning_rate": 1.908988764044944e-05,
|
| 128 |
+
"loss": 0.8892171859741211,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.004935428148391873,
|
| 133 |
+
"grad_norm": 15.552804520715554,
|
| 134 |
+
"learning_rate": 2e-05,
|
| 135 |
+
"loss": 0.907989501953125,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.005209618601080311,
|
| 140 |
+
"grad_norm": 8.234516558585694,
|
| 141 |
+
"learning_rate": 2e-05,
|
| 142 |
+
"loss": 0.8959874153137207,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.005483809053768747,
|
| 147 |
+
"grad_norm": 47.07479845477993,
|
| 148 |
+
"learning_rate": 2e-05,
|
| 149 |
+
"loss": 0.7972547054290772,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.005757999506457185,
|
| 154 |
+
"grad_norm": 14.678891090428264,
|
| 155 |
+
"learning_rate": 2e-05,
|
| 156 |
+
"loss": 0.7945639610290527,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.006032189959145623,
|
| 161 |
+
"grad_norm": 17.99623194706809,
|
| 162 |
+
"learning_rate": 2e-05,
|
| 163 |
+
"loss": 0.8467901229858399,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.00630638041183406,
|
| 168 |
+
"grad_norm": 51.95652459415379,
|
| 169 |
+
"learning_rate": 2e-05,
|
| 170 |
+
"loss": 0.7964937210083007,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.006580570864522497,
|
| 175 |
+
"grad_norm": 10.738865490640423,
|
| 176 |
+
"learning_rate": 2e-05,
|
| 177 |
+
"loss": 0.7535751342773438,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.006854761317210935,
|
| 182 |
+
"grad_norm": 10.795608734038868,
|
| 183 |
+
"learning_rate": 2e-05,
|
| 184 |
+
"loss": 0.828792953491211,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.007128951769899372,
|
| 189 |
+
"grad_norm": 14.551689425739292,
|
| 190 |
+
"learning_rate": 2e-05,
|
| 191 |
+
"loss": 0.8123558044433594,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.00740314222258781,
|
| 196 |
+
"grad_norm": 26.532992867443514,
|
| 197 |
+
"learning_rate": 2e-05,
|
| 198 |
+
"loss": 0.8517896652221679,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.007677332675276247,
|
| 203 |
+
"grad_norm": 34.07990246963607,
|
| 204 |
+
"learning_rate": 2e-05,
|
| 205 |
+
"loss": 0.8389625549316406,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.007951523127964684,
|
| 210 |
+
"grad_norm": 27.19116883891022,
|
| 211 |
+
"learning_rate": 2e-05,
|
| 212 |
+
"loss": 0.8190082550048828,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.008225713580653122,
|
| 217 |
+
"grad_norm": 16.718638113526183,
|
| 218 |
+
"learning_rate": 2e-05,
|
| 219 |
+
"loss": 0.8008914947509765,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.00849990403334156,
|
| 224 |
+
"grad_norm": 27.009662680205103,
|
| 225 |
+
"learning_rate": 2e-05,
|
| 226 |
+
"loss": 0.8091920852661133,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.008774094486029997,
|
| 231 |
+
"grad_norm": 9.031648650732553,
|
| 232 |
+
"learning_rate": 2e-05,
|
| 233 |
+
"loss": 0.8081918716430664,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.009048284938718433,
|
| 238 |
+
"grad_norm": 15.927828911998665,
|
| 239 |
+
"learning_rate": 2e-05,
|
| 240 |
+
"loss": 0.7878878593444825,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.00932247539140687,
|
| 245 |
+
"grad_norm": 6.019580282287438,
|
| 246 |
+
"learning_rate": 2e-05,
|
| 247 |
+
"loss": 0.8092373847961426,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.009596665844095308,
|
| 252 |
+
"grad_norm": 10.097656044283081,
|
| 253 |
+
"learning_rate": 2e-05,
|
| 254 |
+
"loss": 0.8236124038696289,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.009870856296783746,
|
| 259 |
+
"grad_norm": 3.898510274122292,
|
| 260 |
+
"learning_rate": 2e-05,
|
| 261 |
+
"loss": 0.8201662063598633,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.010145046749472184,
|
| 266 |
+
"grad_norm": 10.433240846124043,
|
| 267 |
+
"learning_rate": 2e-05,
|
| 268 |
+
"loss": 0.7954160213470459,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.010419237202160621,
|
| 273 |
+
"grad_norm": 3.9109175859167626,
|
| 274 |
+
"learning_rate": 2e-05,
|
| 275 |
+
"loss": 0.8451309204101562,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.010693427654849059,
|
| 280 |
+
"grad_norm": 19.47785359214917,
|
| 281 |
+
"learning_rate": 2e-05,
|
| 282 |
+
"loss": 0.8227203369140625,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.010967618107537495,
|
| 287 |
+
"grad_norm": 1.9979422322935754,
|
| 288 |
+
"learning_rate": 2e-05,
|
| 289 |
+
"loss": 0.7947024345397949,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.011241808560225932,
|
| 294 |
+
"grad_norm": 2.1152537387703063,
|
| 295 |
+
"learning_rate": 2e-05,
|
| 296 |
+
"loss": 0.7340120315551758,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.01151599901291437,
|
| 301 |
+
"grad_norm": 2.457083494187141,
|
| 302 |
+
"learning_rate": 2e-05,
|
| 303 |
+
"loss": 0.7300191879272461,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.011790189465602808,
|
| 308 |
+
"grad_norm": 17.457321034227927,
|
| 309 |
+
"learning_rate": 2e-05,
|
| 310 |
+
"loss": 0.7896897315979003,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.012064379918291245,
|
| 315 |
+
"grad_norm": 3.7757446450653047,
|
| 316 |
+
"learning_rate": 2e-05,
|
| 317 |
+
"loss": 0.7420886039733887,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.012338570370979683,
|
| 322 |
+
"grad_norm": 10.718659609403895,
|
| 323 |
+
"learning_rate": 2e-05,
|
| 324 |
+
"loss": 0.7464280605316163,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.01261276082366812,
|
| 329 |
+
"grad_norm": 13.096467752661693,
|
| 330 |
+
"learning_rate": 2e-05,
|
| 331 |
+
"loss": 0.7420991420745849,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.012886951276356558,
|
| 336 |
+
"grad_norm": 7.554114077938612,
|
| 337 |
+
"learning_rate": 2e-05,
|
| 338 |
+
"loss": 0.7896678924560547,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.013161141729044994,
|
| 343 |
+
"grad_norm": 1.9610155148576782,
|
| 344 |
+
"learning_rate": 2e-05,
|
| 345 |
+
"loss": 0.7641692161560059,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.013435332181733432,
|
| 350 |
+
"grad_norm": 3.743726960430974,
|
| 351 |
+
"learning_rate": 2e-05,
|
| 352 |
+
"loss": 0.7724472999572753,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.01370952263442187,
|
| 357 |
+
"grad_norm": 3.6205794374072897,
|
| 358 |
+
"learning_rate": 2e-05,
|
| 359 |
+
"loss": 0.7726649284362793,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.013983713087110307,
|
| 364 |
+
"grad_norm": 2.751894400685354,
|
| 365 |
+
"learning_rate": 2e-05,
|
| 366 |
+
"loss": 0.7311810970306396,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.014257903539798744,
|
| 371 |
+
"grad_norm": 15.8354606407861,
|
| 372 |
+
"learning_rate": 2e-05,
|
| 373 |
+
"loss": 0.7479830741882324,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.014532093992487182,
|
| 378 |
+
"grad_norm": 9.741291005928273,
|
| 379 |
+
"learning_rate": 2e-05,
|
| 380 |
+
"loss": 0.758505916595459,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.01480628444517562,
|
| 385 |
+
"grad_norm": 6.361558235058739,
|
| 386 |
+
"learning_rate": 2e-05,
|
| 387 |
+
"loss": 0.7497701644897461,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.015080474897864056,
|
| 392 |
+
"grad_norm": 12.406630096904701,
|
| 393 |
+
"learning_rate": 2e-05,
|
| 394 |
+
"loss": 0.7578043937683105,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.015354665350552493,
|
| 399 |
+
"grad_norm": 5.437533111150059,
|
| 400 |
+
"learning_rate": 2e-05,
|
| 401 |
+
"loss": 0.7494767665863037,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.015628855803240933,
|
| 406 |
+
"grad_norm": 5.385414434765143,
|
| 407 |
+
"learning_rate": 2e-05,
|
| 408 |
+
"loss": 0.76002197265625,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.01590304625592937,
|
| 413 |
+
"grad_norm": 2.514872003150109,
|
| 414 |
+
"learning_rate": 2e-05,
|
| 415 |
+
"loss": 0.753857946395874,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.016177236708617804,
|
| 420 |
+
"grad_norm": 7.350990110651167,
|
| 421 |
+
"learning_rate": 2e-05,
|
| 422 |
+
"loss": 0.7598513603210449,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.016451427161306244,
|
| 427 |
+
"grad_norm": 9.359850474100211,
|
| 428 |
+
"learning_rate": 2e-05,
|
| 429 |
+
"loss": 0.7978845596313476,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.01672561761399468,
|
| 434 |
+
"grad_norm": 2.5927971608846243,
|
| 435 |
+
"learning_rate": 2e-05,
|
| 436 |
+
"loss": 0.7353355407714843,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.01699980806668312,
|
| 441 |
+
"grad_norm": 13.580441280433185,
|
| 442 |
+
"learning_rate": 2e-05,
|
| 443 |
+
"loss": 0.7530646800994873,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.017273998519371555,
|
| 448 |
+
"grad_norm": 7.032504711444772,
|
| 449 |
+
"learning_rate": 2e-05,
|
| 450 |
+
"loss": 0.7999831199645996,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.017548188972059994,
|
| 455 |
+
"grad_norm": 3.7623518166414134,
|
| 456 |
+
"learning_rate": 2e-05,
|
| 457 |
+
"loss": 0.7583576202392578,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.01782237942474843,
|
| 462 |
+
"grad_norm": 9.996121547764488,
|
| 463 |
+
"learning_rate": 2e-05,
|
| 464 |
+
"loss": 0.7126437187194824,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.018096569877436866,
|
| 469 |
+
"grad_norm": 3.8005663709948423,
|
| 470 |
+
"learning_rate": 2e-05,
|
| 471 |
+
"loss": 0.7714213371276856,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.018370760330125305,
|
| 476 |
+
"grad_norm": 2.2803953101797445,
|
| 477 |
+
"learning_rate": 2e-05,
|
| 478 |
+
"loss": 0.735899829864502,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.01864495078281374,
|
| 483 |
+
"grad_norm": 2.2804717033812687,
|
| 484 |
+
"learning_rate": 2e-05,
|
| 485 |
+
"loss": 0.7463047981262207,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.01891914123550218,
|
| 490 |
+
"grad_norm": 2.6630744686256174,
|
| 491 |
+
"learning_rate": 2e-05,
|
| 492 |
+
"loss": 0.7112466812133789,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.019193331688190617,
|
| 497 |
+
"grad_norm": 8.59421748677637,
|
| 498 |
+
"learning_rate": 2e-05,
|
| 499 |
+
"loss": 0.8032501220703125,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.019467522140879056,
|
| 504 |
+
"grad_norm": 6.839099911030037,
|
| 505 |
+
"learning_rate": 2e-05,
|
| 506 |
+
"loss": 0.7257285118103027,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.019741712593567492,
|
| 511 |
+
"grad_norm": 6.527196880605798,
|
| 512 |
+
"learning_rate": 2e-05,
|
| 513 |
+
"loss": 0.7394065856933594,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.020015903046255928,
|
| 518 |
+
"grad_norm": 4.91546686150854,
|
| 519 |
+
"learning_rate": 2e-05,
|
| 520 |
+
"loss": 0.7233449459075928,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.020290093498944367,
|
| 525 |
+
"grad_norm": 9.615722695591046,
|
| 526 |
+
"learning_rate": 2e-05,
|
| 527 |
+
"loss": 0.714713191986084,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.020564283951632803,
|
| 532 |
+
"grad_norm": 2.6469047522209244,
|
| 533 |
+
"learning_rate": 2e-05,
|
| 534 |
+
"loss": 0.7250522613525391,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.020838474404321242,
|
| 539 |
+
"grad_norm": 27.513230581702402,
|
| 540 |
+
"learning_rate": 2e-05,
|
| 541 |
+
"loss": 0.709619665145874,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.021112664857009678,
|
| 546 |
+
"grad_norm": 2.3305572520941102,
|
| 547 |
+
"learning_rate": 2e-05,
|
| 548 |
+
"loss": 0.7650126457214356,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.021386855309698118,
|
| 553 |
+
"grad_norm": 1.7645473455518579,
|
| 554 |
+
"learning_rate": 2e-05,
|
| 555 |
+
"loss": 0.7501170635223389,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.021661045762386553,
|
| 560 |
+
"grad_norm": 2.706703322701269,
|
| 561 |
+
"learning_rate": 2e-05,
|
| 562 |
+
"loss": 0.7390304565429687,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.02193523621507499,
|
| 567 |
+
"grad_norm": 1.343452945442269,
|
| 568 |
+
"learning_rate": 2e-05,
|
| 569 |
+
"loss": 0.7560922145843506,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.02220942666776343,
|
| 574 |
+
"grad_norm": 1.8131674161495455,
|
| 575 |
+
"learning_rate": 2e-05,
|
| 576 |
+
"loss": 0.7617329597473145,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.022483617120451865,
|
| 581 |
+
"grad_norm": 6.6257304148619,
|
| 582 |
+
"learning_rate": 2e-05,
|
| 583 |
+
"loss": 0.7030315399169922,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.022757807573140304,
|
| 588 |
+
"grad_norm": 2.7477972541714237,
|
| 589 |
+
"learning_rate": 2e-05,
|
| 590 |
+
"loss": 0.7051456451416016,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.02303199802582874,
|
| 595 |
+
"grad_norm": 1.8551689554260649,
|
| 596 |
+
"learning_rate": 2e-05,
|
| 597 |
+
"loss": 0.790980863571167,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.02330618847851718,
|
| 602 |
+
"grad_norm": 4.091035715965749,
|
| 603 |
+
"learning_rate": 2e-05,
|
| 604 |
+
"loss": 0.7167482376098633,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.023580378931205615,
|
| 609 |
+
"grad_norm": 3.13444342764459,
|
| 610 |
+
"learning_rate": 2e-05,
|
| 611 |
+
"loss": 0.7428287506103516,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.023854569383894055,
|
| 616 |
+
"grad_norm": 7.428502868185567,
|
| 617 |
+
"learning_rate": 2e-05,
|
| 618 |
+
"loss": 0.7720952510833741,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.02412875983658249,
|
| 623 |
+
"grad_norm": 2.738880412295684,
|
| 624 |
+
"learning_rate": 2e-05,
|
| 625 |
+
"loss": 0.7307533264160156,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.024402950289270926,
|
| 630 |
+
"grad_norm": 5.442319700482665,
|
| 631 |
+
"learning_rate": 2e-05,
|
| 632 |
+
"loss": 0.7590367794036865,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.024677140741959366,
|
| 637 |
+
"grad_norm": 4.209775631505361,
|
| 638 |
+
"learning_rate": 2e-05,
|
| 639 |
+
"loss": 0.7398377418518066,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.0249513311946478,
|
| 644 |
+
"grad_norm": 3.317366374639838,
|
| 645 |
+
"learning_rate": 2e-05,
|
| 646 |
+
"loss": 0.7573592662811279,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.02522552164733624,
|
| 651 |
+
"grad_norm": 2.21708523073645,
|
| 652 |
+
"learning_rate": 2e-05,
|
| 653 |
+
"loss": 0.7442788124084473,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.025499712100024677,
|
| 658 |
+
"grad_norm": 5.003510596258695,
|
| 659 |
+
"learning_rate": 2e-05,
|
| 660 |
+
"loss": 0.7208542823791504,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.025773902552713116,
|
| 665 |
+
"grad_norm": 1.8666696979850983,
|
| 666 |
+
"learning_rate": 2e-05,
|
| 667 |
+
"loss": 0.7219260215759278,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.026048093005401552,
|
| 672 |
+
"grad_norm": 2.319707094464392,
|
| 673 |
+
"learning_rate": 2e-05,
|
| 674 |
+
"loss": 0.705035400390625,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.026322283458089988,
|
| 679 |
+
"grad_norm": 1.559433923602148,
|
| 680 |
+
"learning_rate": 2e-05,
|
| 681 |
+
"loss": 0.7253781795501709,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.026596473910778427,
|
| 686 |
+
"grad_norm": 3.1103622605289294,
|
| 687 |
+
"learning_rate": 2e-05,
|
| 688 |
+
"loss": 0.7434506416320801,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.026870664363466863,
|
| 693 |
+
"grad_norm": 4.482952591971397,
|
| 694 |
+
"learning_rate": 2e-05,
|
| 695 |
+
"loss": 0.7423406600952148,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.027144854816155303,
|
| 700 |
+
"grad_norm": 2.349643537640656,
|
| 701 |
+
"learning_rate": 2e-05,
|
| 702 |
+
"loss": 0.7380632400512696,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.02741904526884374,
|
| 707 |
+
"grad_norm": 6.624084473754211,
|
| 708 |
+
"learning_rate": 2e-05,
|
| 709 |
+
"loss": 0.7448741912841796,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.027693235721532178,
|
| 714 |
+
"grad_norm": 1.959542744744174,
|
| 715 |
+
"learning_rate": 2e-05,
|
| 716 |
+
"loss": 0.7620717048645019,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.027967426174220614,
|
| 721 |
+
"grad_norm": 1.9359370487542638,
|
| 722 |
+
"learning_rate": 2e-05,
|
| 723 |
+
"loss": 0.7283543586730957,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.02824161662690905,
|
| 728 |
+
"grad_norm": 3.7510091510273424,
|
| 729 |
+
"learning_rate": 2e-05,
|
| 730 |
+
"loss": 0.7341510772705078,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.02851580707959749,
|
| 735 |
+
"grad_norm": 1.645233108914731,
|
| 736 |
+
"learning_rate": 2e-05,
|
| 737 |
+
"loss": 0.7321323394775391,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.028789997532285925,
|
| 742 |
+
"grad_norm": 5.060129114701537,
|
| 743 |
+
"learning_rate": 2e-05,
|
| 744 |
+
"loss": 0.7923129558563232,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.029064187984974364,
|
| 749 |
+
"grad_norm": 6.767343077930711,
|
| 750 |
+
"learning_rate": 2e-05,
|
| 751 |
+
"loss": 0.7364850997924804,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.0293383784376628,
|
| 756 |
+
"grad_norm": 7.157354259172915,
|
| 757 |
+
"learning_rate": 2e-05,
|
| 758 |
+
"loss": 0.7175948143005371,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.02961256889035124,
|
| 763 |
+
"grad_norm": 10.118526445754803,
|
| 764 |
+
"learning_rate": 2e-05,
|
| 765 |
+
"loss": 0.6956705093383789,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.029886759343039675,
|
| 770 |
+
"grad_norm": 3.2975128574851205,
|
| 771 |
+
"learning_rate": 2e-05,
|
| 772 |
+
"loss": 0.7435084342956543,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.03016094979572811,
|
| 777 |
+
"grad_norm": 3.79438108252297,
|
| 778 |
+
"learning_rate": 2e-05,
|
| 779 |
+
"loss": 0.7209542274475098,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.03043514024841655,
|
| 784 |
+
"grad_norm": 3.6413783178286896,
|
| 785 |
+
"learning_rate": 2e-05,
|
| 786 |
+
"loss": 0.7462982177734375,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.030709330701104987,
|
| 791 |
+
"grad_norm": 2.4359365603997323,
|
| 792 |
+
"learning_rate": 2e-05,
|
| 793 |
+
"loss": 0.7772771835327148,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.030983521153793426,
|
| 798 |
+
"grad_norm": 2.067253539210089,
|
| 799 |
+
"learning_rate": 2e-05,
|
| 800 |
+
"loss": 0.754563283920288,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.031257711606481865,
|
| 805 |
+
"grad_norm": 1.3603842511083,
|
| 806 |
+
"learning_rate": 2e-05,
|
| 807 |
+
"loss": 0.7163387298583984,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.0315319020591703,
|
| 812 |
+
"grad_norm": 7.634204958345298,
|
| 813 |
+
"learning_rate": 2e-05,
|
| 814 |
+
"loss": 0.7153648376464844,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.03180609251185874,
|
| 819 |
+
"grad_norm": 2.224287938133906,
|
| 820 |
+
"learning_rate": 2e-05,
|
| 821 |
+
"loss": 0.7743209838867188,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.03208028296454717,
|
| 826 |
+
"grad_norm": 8.328106347948369,
|
| 827 |
+
"learning_rate": 2e-05,
|
| 828 |
+
"loss": 0.7125131130218506,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.03235447341723561,
|
| 833 |
+
"grad_norm": 4.9140382117559716,
|
| 834 |
+
"learning_rate": 2e-05,
|
| 835 |
+
"loss": 0.734964656829834,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.03262866386992405,
|
| 840 |
+
"grad_norm": 1.6437427011396712,
|
| 841 |
+
"learning_rate": 2e-05,
|
| 842 |
+
"loss": 0.736287260055542,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.03290285432261249,
|
| 847 |
+
"grad_norm": 1.7984413046378422,
|
| 848 |
+
"learning_rate": 2e-05,
|
| 849 |
+
"loss": 0.6888255119323731,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.03317704477530092,
|
| 854 |
+
"grad_norm": 3.003273429621072,
|
| 855 |
+
"learning_rate": 2e-05,
|
| 856 |
+
"loss": 0.7354516506195068,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.03345123522798936,
|
| 861 |
+
"grad_norm": 1.581766516298146,
|
| 862 |
+
"learning_rate": 2e-05,
|
| 863 |
+
"loss": 0.7175889492034913,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.0337254256806778,
|
| 868 |
+
"grad_norm": 2.3247470513343185,
|
| 869 |
+
"learning_rate": 2e-05,
|
| 870 |
+
"loss": 0.714189863204956,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.03399961613336624,
|
| 875 |
+
"grad_norm": 7.9334494235192325,
|
| 876 |
+
"learning_rate": 2e-05,
|
| 877 |
+
"loss": 0.7136695861816407,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.034273806586054674,
|
| 882 |
+
"grad_norm": 49.62487496955337,
|
| 883 |
+
"learning_rate": 2e-05,
|
| 884 |
+
"loss": 0.7086945533752441,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.03454799703874311,
|
| 889 |
+
"grad_norm": 2.890774584879648,
|
| 890 |
+
"learning_rate": 2e-05,
|
| 891 |
+
"loss": 0.7269087791442871,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.034822187491431546,
|
| 896 |
+
"grad_norm": 4.796531005737684,
|
| 897 |
+
"learning_rate": 2e-05,
|
| 898 |
+
"loss": 0.7484929084777832,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.03509637794411999,
|
| 903 |
+
"grad_norm": 1.6372708148105157,
|
| 904 |
+
"learning_rate": 2e-05,
|
| 905 |
+
"loss": 0.6999114036560059,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.035370568396808424,
|
| 910 |
+
"grad_norm": 4.162341191850801,
|
| 911 |
+
"learning_rate": 2e-05,
|
| 912 |
+
"loss": 0.7129648208618165,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.03564475884949686,
|
| 917 |
+
"grad_norm": 4.017056556093628,
|
| 918 |
+
"learning_rate": 2e-05,
|
| 919 |
+
"loss": 0.7217267036437989,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.035918949302185296,
|
| 924 |
+
"grad_norm": 9.786655951652824,
|
| 925 |
+
"learning_rate": 2e-05,
|
| 926 |
+
"loss": 0.7361152648925782,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.03619313975487373,
|
| 931 |
+
"grad_norm": 17.798888728639277,
|
| 932 |
+
"learning_rate": 2e-05,
|
| 933 |
+
"loss": 0.7423253059387207,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.036467330207562175,
|
| 938 |
+
"grad_norm": 25.98230175781132,
|
| 939 |
+
"learning_rate": 2e-05,
|
| 940 |
+
"loss": 0.7079302787780761,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.03674152066025061,
|
| 945 |
+
"grad_norm": 4.107909151191186,
|
| 946 |
+
"learning_rate": 2e-05,
|
| 947 |
+
"loss": 0.7637813568115235,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.03701571111293905,
|
| 952 |
+
"grad_norm": 1.8177555687046802,
|
| 953 |
+
"learning_rate": 2e-05,
|
| 954 |
+
"loss": 0.7440331459045411,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.03728990156562748,
|
| 959 |
+
"grad_norm": 6.631474920460703,
|
| 960 |
+
"learning_rate": 2e-05,
|
| 961 |
+
"loss": 0.7420357704162598,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.037564092018315925,
|
| 966 |
+
"grad_norm": 1.182591249805126,
|
| 967 |
+
"learning_rate": 2e-05,
|
| 968 |
+
"loss": 0.7243555068969727,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.03783828247100436,
|
| 973 |
+
"grad_norm": 1.5239290367740697,
|
| 974 |
+
"learning_rate": 2e-05,
|
| 975 |
+
"loss": 0.7780046463012695,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.0381124729236928,
|
| 980 |
+
"grad_norm": 1.9611702789468477,
|
| 981 |
+
"learning_rate": 2e-05,
|
| 982 |
+
"loss": 0.7606742858886719,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.03838666337638123,
|
| 987 |
+
"grad_norm": 3.4538398959809564,
|
| 988 |
+
"learning_rate": 2e-05,
|
| 989 |
+
"loss": 0.7161018371582031,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.03866085382906967,
|
| 994 |
+
"grad_norm": 6.74405592418129,
|
| 995 |
+
"learning_rate": 2e-05,
|
| 996 |
+
"loss": 0.7269341945648193,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.03893504428175811,
|
| 1001 |
+
"grad_norm": 5.0613023509206,
|
| 1002 |
+
"learning_rate": 2e-05,
|
| 1003 |
+
"loss": 0.7150623321533203,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.03920923473444655,
|
| 1008 |
+
"grad_norm": 15.935601654785444,
|
| 1009 |
+
"learning_rate": 2e-05,
|
| 1010 |
+
"loss": 0.6991084098815918,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.039483425187134984,
|
| 1015 |
+
"grad_norm": 189.58206876056943,
|
| 1016 |
+
"learning_rate": 2e-05,
|
| 1017 |
+
"loss": 0.715428352355957,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.03975761563982342,
|
| 1022 |
+
"grad_norm": 13.224258279776146,
|
| 1023 |
+
"learning_rate": 2e-05,
|
| 1024 |
+
"loss": 0.7448666572570801,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.040031806092511855,
|
| 1029 |
+
"grad_norm": 15.23890210961455,
|
| 1030 |
+
"learning_rate": 2e-05,
|
| 1031 |
+
"loss": 0.7545663833618164,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.0403059965452003,
|
| 1036 |
+
"grad_norm": 2.089772579003186,
|
| 1037 |
+
"learning_rate": 2e-05,
|
| 1038 |
+
"loss": 0.7143968105316162,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.040580186997888734,
|
| 1043 |
+
"grad_norm": 672.5570249806746,
|
| 1044 |
+
"learning_rate": 2e-05,
|
| 1045 |
+
"loss": 0.6955307960510254,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.04085437745057717,
|
| 1050 |
+
"grad_norm": 2.499047720535622,
|
| 1051 |
+
"learning_rate": 2e-05,
|
| 1052 |
+
"loss": 0.7183216094970704,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.041128567903265606,
|
| 1057 |
+
"grad_norm": 1.6820671266931866,
|
| 1058 |
+
"learning_rate": 2e-05,
|
| 1059 |
+
"loss": 0.7442133903503418,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.04140275835595405,
|
| 1064 |
+
"grad_norm": 1.6322487359391142,
|
| 1065 |
+
"learning_rate": 2e-05,
|
| 1066 |
+
"loss": 0.7415005683898925,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.041676948808642485,
|
| 1071 |
+
"grad_norm": 2.0768691766562974,
|
| 1072 |
+
"learning_rate": 2e-05,
|
| 1073 |
+
"loss": 0.7350949764251709,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.04195113926133092,
|
| 1078 |
+
"grad_norm": 3.462638107939682,
|
| 1079 |
+
"learning_rate": 2e-05,
|
| 1080 |
+
"loss": 0.6958516597747803,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.042225329714019356,
|
| 1085 |
+
"grad_norm": 1.9410342316482636,
|
| 1086 |
+
"learning_rate": 2e-05,
|
| 1087 |
+
"loss": 0.7019001960754394,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.04249952016670779,
|
| 1092 |
+
"grad_norm": 2.2435234267896944,
|
| 1093 |
+
"learning_rate": 2e-05,
|
| 1094 |
+
"loss": 0.7038233757019043,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.042773710619396235,
|
| 1099 |
+
"grad_norm": 50.31729516447788,
|
| 1100 |
+
"learning_rate": 2e-05,
|
| 1101 |
+
"loss": 0.6971765041351319,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.04304790107208467,
|
| 1106 |
+
"grad_norm": 6.726153221416043,
|
| 1107 |
+
"learning_rate": 2e-05,
|
| 1108 |
+
"loss": 0.7901406764984131,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.04332209152477311,
|
| 1113 |
+
"grad_norm": 25.290061958502374,
|
| 1114 |
+
"learning_rate": 2e-05,
|
| 1115 |
+
"loss": 0.7404637813568116,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.04359628197746154,
|
| 1120 |
+
"grad_norm": 11.648387859593914,
|
| 1121 |
+
"learning_rate": 2e-05,
|
| 1122 |
+
"loss": 0.7325190544128418,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.04387047243014998,
|
| 1127 |
+
"grad_norm": 3.998377511327454,
|
| 1128 |
+
"learning_rate": 2e-05,
|
| 1129 |
+
"loss": 0.6938001155853272,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.04414466288283842,
|
| 1134 |
+
"grad_norm": 1.4609573707991412,
|
| 1135 |
+
"learning_rate": 2e-05,
|
| 1136 |
+
"loss": 0.6974159240722656,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.04441885333552686,
|
| 1141 |
+
"grad_norm": 1.2908767847499276,
|
| 1142 |
+
"learning_rate": 2e-05,
|
| 1143 |
+
"loss": 0.735476303100586,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.04469304378821529,
|
| 1148 |
+
"grad_norm": 1.5263587301999937,
|
| 1149 |
+
"learning_rate": 2e-05,
|
| 1150 |
+
"loss": 0.6849418640136719,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.04496723424090373,
|
| 1155 |
+
"grad_norm": 2.71012026891521,
|
| 1156 |
+
"learning_rate": 2e-05,
|
| 1157 |
+
"loss": 0.6952274322509766,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.04524142469359217,
|
| 1162 |
+
"grad_norm": 1.5738552850120824,
|
| 1163 |
+
"learning_rate": 2e-05,
|
| 1164 |
+
"loss": 0.7284816265106201,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.04551561514628061,
|
| 1169 |
+
"grad_norm": 1.5435257666345013,
|
| 1170 |
+
"learning_rate": 2e-05,
|
| 1171 |
+
"loss": 0.7115949153900146,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.045789805598969044,
|
| 1176 |
+
"grad_norm": 1.516425220632835,
|
| 1177 |
+
"learning_rate": 2e-05,
|
| 1178 |
+
"loss": 0.6966533660888672,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.04606399605165748,
|
| 1183 |
+
"grad_norm": 1.4858677693562772,
|
| 1184 |
+
"learning_rate": 2e-05,
|
| 1185 |
+
"loss": 0.712862491607666,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.046338186504345916,
|
| 1190 |
+
"grad_norm": 1.1928774431737874,
|
| 1191 |
+
"learning_rate": 2e-05,
|
| 1192 |
+
"loss": 0.715973949432373,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.04661237695703436,
|
| 1197 |
+
"grad_norm": 1.3801750667103458,
|
| 1198 |
+
"learning_rate": 2e-05,
|
| 1199 |
+
"loss": 0.6785056114196777,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.046886567409722794,
|
| 1204 |
+
"grad_norm": 1.1562227123885702,
|
| 1205 |
+
"learning_rate": 2e-05,
|
| 1206 |
+
"loss": 0.6806015014648438,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.04716075786241123,
|
| 1211 |
+
"grad_norm": 1.966065169897555,
|
| 1212 |
+
"learning_rate": 2e-05,
|
| 1213 |
+
"loss": 0.6789681911468506,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.047434948315099666,
|
| 1218 |
+
"grad_norm": 1.796240102087905,
|
| 1219 |
+
"learning_rate": 2e-05,
|
| 1220 |
+
"loss": 0.6643640041351319,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.04770913876778811,
|
| 1225 |
+
"grad_norm": 1.4714599948408273,
|
| 1226 |
+
"learning_rate": 2e-05,
|
| 1227 |
+
"loss": 0.672383975982666,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.047983329220476545,
|
| 1232 |
+
"grad_norm": 1.05613596470809,
|
| 1233 |
+
"learning_rate": 2e-05,
|
| 1234 |
+
"loss": 0.6921384334564209,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.04825751967316498,
|
| 1239 |
+
"grad_norm": 1.168321764739896,
|
| 1240 |
+
"learning_rate": 2e-05,
|
| 1241 |
+
"loss": 0.6853407859802246,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.04853171012585342,
|
| 1246 |
+
"grad_norm": 1.304231611987027,
|
| 1247 |
+
"learning_rate": 2e-05,
|
| 1248 |
+
"loss": 0.6831422328948975,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.04880590057854185,
|
| 1253 |
+
"grad_norm": 1.271885101891083,
|
| 1254 |
+
"learning_rate": 2e-05,
|
| 1255 |
+
"loss": 0.7312606811523438,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.049080091031230295,
|
| 1260 |
+
"grad_norm": 1.1274168419233672,
|
| 1261 |
+
"learning_rate": 2e-05,
|
| 1262 |
+
"loss": 0.7195809841156006,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.04935428148391873,
|
| 1267 |
+
"grad_norm": 2.1060356897000196,
|
| 1268 |
+
"learning_rate": 2e-05,
|
| 1269 |
+
"loss": 0.6862109184265137,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.04962847193660717,
|
| 1274 |
+
"grad_norm": 1.1668574261874323,
|
| 1275 |
+
"learning_rate": 2e-05,
|
| 1276 |
+
"loss": 0.7068725109100342,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.0499026623892956,
|
| 1281 |
+
"grad_norm": 3.1583477159166673,
|
| 1282 |
+
"learning_rate": 2e-05,
|
| 1283 |
+
"loss": 0.7150220394134521,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.05017685284198404,
|
| 1288 |
+
"grad_norm": 1.5843620172233683,
|
| 1289 |
+
"learning_rate": 2e-05,
|
| 1290 |
+
"loss": 0.66549072265625,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.05045104329467248,
|
| 1295 |
+
"grad_norm": 1.2547928536390052,
|
| 1296 |
+
"learning_rate": 2e-05,
|
| 1297 |
+
"loss": 0.7375857353210449,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.05072523374736092,
|
| 1302 |
+
"grad_norm": 4.29725259279178,
|
| 1303 |
+
"learning_rate": 2e-05,
|
| 1304 |
+
"loss": 0.6876885890960693,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.050999424200049354,
|
| 1309 |
+
"grad_norm": 1.9025925412069744,
|
| 1310 |
+
"learning_rate": 2e-05,
|
| 1311 |
+
"loss": 0.7387830734252929,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.05127361465273779,
|
| 1316 |
+
"grad_norm": 1.5406949311604028,
|
| 1317 |
+
"learning_rate": 2e-05,
|
| 1318 |
+
"loss": 0.6733179092407227,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.05154780510542623,
|
| 1323 |
+
"grad_norm": 20.215212215144863,
|
| 1324 |
+
"learning_rate": 2e-05,
|
| 1325 |
+
"loss": 0.7654403686523438,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.05182199555811467,
|
| 1330 |
+
"grad_norm": 1.5669531048405803,
|
| 1331 |
+
"learning_rate": 2e-05,
|
| 1332 |
+
"loss": 0.7350390911102295,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.052096186010803104,
|
| 1337 |
+
"grad_norm": 7.2191546562371425,
|
| 1338 |
+
"learning_rate": 2e-05,
|
| 1339 |
+
"loss": 0.7414397716522216,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.05237037646349154,
|
| 1344 |
+
"grad_norm": 1.2325274169727398,
|
| 1345 |
+
"learning_rate": 2e-05,
|
| 1346 |
+
"loss": 0.7655929088592529,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.052644566916179976,
|
| 1351 |
+
"grad_norm": 1.1944179026527708,
|
| 1352 |
+
"learning_rate": 2e-05,
|
| 1353 |
+
"loss": 0.7075240135192871,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.05291875736886842,
|
| 1358 |
+
"grad_norm": 1.302822879313398,
|
| 1359 |
+
"learning_rate": 2e-05,
|
| 1360 |
+
"loss": 0.6797014236450195,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.053192947821556855,
|
| 1365 |
+
"grad_norm": 2.0168747405099747,
|
| 1366 |
+
"learning_rate": 2e-05,
|
| 1367 |
+
"loss": 0.7000571727752686,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.05346713827424529,
|
| 1372 |
+
"grad_norm": 1.6344897385611805,
|
| 1373 |
+
"learning_rate": 2e-05,
|
| 1374 |
+
"loss": 0.7235137939453125,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.053741328726933726,
|
| 1379 |
+
"grad_norm": 3.5766753006414125,
|
| 1380 |
+
"learning_rate": 2e-05,
|
| 1381 |
+
"loss": 0.7371991634368896,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.05401551917962216,
|
| 1386 |
+
"grad_norm": 1.4953479982574733,
|
| 1387 |
+
"learning_rate": 2e-05,
|
| 1388 |
+
"loss": 0.7229987621307373,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.054289709632310605,
|
| 1393 |
+
"grad_norm": 1.6598347475638233,
|
| 1394 |
+
"learning_rate": 2e-05,
|
| 1395 |
+
"loss": 0.6758342742919922,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.05456390008499904,
|
| 1400 |
+
"grad_norm": 1.3876361065240026,
|
| 1401 |
+
"learning_rate": 2e-05,
|
| 1402 |
+
"loss": 0.6621166229248047,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.05483809053768748,
|
| 1407 |
+
"grad_norm": 1.2953540155757288,
|
| 1408 |
+
"learning_rate": 2e-05,
|
| 1409 |
+
"loss": 0.701081657409668,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.05511228099037591,
|
| 1414 |
+
"grad_norm": 1.0757376098134102,
|
| 1415 |
+
"learning_rate": 2e-05,
|
| 1416 |
+
"loss": 0.6949031829833985,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.055386471443064356,
|
| 1421 |
+
"grad_norm": 1.3515745918409625,
|
| 1422 |
+
"learning_rate": 2e-05,
|
| 1423 |
+
"loss": 0.6849285125732422,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.05566066189575279,
|
| 1428 |
+
"grad_norm": 3.4678908742290884,
|
| 1429 |
+
"learning_rate": 2e-05,
|
| 1430 |
+
"loss": 0.6919690608978272,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.05593485234844123,
|
| 1435 |
+
"grad_norm": 1.9456705464214417,
|
| 1436 |
+
"learning_rate": 2e-05,
|
| 1437 |
+
"loss": 0.6663742065429688,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.05620904280112966,
|
| 1442 |
+
"grad_norm": 1.2852941271459164,
|
| 1443 |
+
"learning_rate": 2e-05,
|
| 1444 |
+
"loss": 0.6898897171020508,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.0564832332538181,
|
| 1449 |
+
"grad_norm": 7.958437531519753,
|
| 1450 |
+
"learning_rate": 2e-05,
|
| 1451 |
+
"loss": 0.7008088588714599,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.05675742370650654,
|
| 1456 |
+
"grad_norm": 3.3113911508682685,
|
| 1457 |
+
"learning_rate": 2e-05,
|
| 1458 |
+
"loss": 0.6721383571624756,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.05703161415919498,
|
| 1463 |
+
"grad_norm": 1.1175929950295507,
|
| 1464 |
+
"learning_rate": 2e-05,
|
| 1465 |
+
"loss": 0.7099536895751953,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.057305804611883414,
|
| 1470 |
+
"grad_norm": 0.9164381223894205,
|
| 1471 |
+
"learning_rate": 2e-05,
|
| 1472 |
+
"loss": 0.6982614040374756,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.05757999506457185,
|
| 1477 |
+
"grad_norm": 1.3064363870590416,
|
| 1478 |
+
"learning_rate": 2e-05,
|
| 1479 |
+
"loss": 0.6582142829895019,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.057854185517260286,
|
| 1484 |
+
"grad_norm": 1.9295155170705978,
|
| 1485 |
+
"learning_rate": 2e-05,
|
| 1486 |
+
"loss": 0.7130876064300538,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.05812837596994873,
|
| 1491 |
+
"grad_norm": 1.598328187450677,
|
| 1492 |
+
"learning_rate": 2e-05,
|
| 1493 |
+
"loss": 0.7056316375732422,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.058402566422637164,
|
| 1498 |
+
"grad_norm": 1.3733360145402107,
|
| 1499 |
+
"learning_rate": 2e-05,
|
| 1500 |
+
"loss": 0.6908935546875,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.0586767568753256,
|
| 1505 |
+
"grad_norm": 1.5835542241632976,
|
| 1506 |
+
"learning_rate": 2e-05,
|
| 1507 |
+
"loss": 0.6900417327880859,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.058950947328014036,
|
| 1512 |
+
"grad_norm": 15.343250543565158,
|
| 1513 |
+
"learning_rate": 2e-05,
|
| 1514 |
+
"loss": 0.661806869506836,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.05922513778070248,
|
| 1519 |
+
"grad_norm": 1.7535753341635816,
|
| 1520 |
+
"learning_rate": 2e-05,
|
| 1521 |
+
"loss": 0.6718528270721436,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.059499328233390915,
|
| 1526 |
+
"grad_norm": 1.9376370630167024,
|
| 1527 |
+
"learning_rate": 2e-05,
|
| 1528 |
+
"loss": 0.690641212463379,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.05977351868607935,
|
| 1533 |
+
"grad_norm": 2.5374679136605596,
|
| 1534 |
+
"learning_rate": 2e-05,
|
| 1535 |
+
"loss": 0.6993424415588378,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.06004770913876779,
|
| 1540 |
+
"grad_norm": 8.884410141851701,
|
| 1541 |
+
"learning_rate": 2e-05,
|
| 1542 |
+
"loss": 0.6886792182922363,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.06032189959145622,
|
| 1547 |
+
"grad_norm": 1.3716294166066423,
|
| 1548 |
+
"learning_rate": 2e-05,
|
| 1549 |
+
"loss": 0.7170482635498047,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.060596090044144665,
|
| 1554 |
+
"grad_norm": 15.068560848475203,
|
| 1555 |
+
"learning_rate": 2e-05,
|
| 1556 |
+
"loss": 0.6805070877075196,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.0608702804968331,
|
| 1561 |
+
"grad_norm": 1.3577298589073343,
|
| 1562 |
+
"learning_rate": 2e-05,
|
| 1563 |
+
"loss": 0.7325131893157959,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.06114447094952154,
|
| 1568 |
+
"grad_norm": 1.5605906922192767,
|
| 1569 |
+
"learning_rate": 2e-05,
|
| 1570 |
+
"loss": 0.7512709617614746,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.06141866140220997,
|
| 1575 |
+
"grad_norm": 2.7582484830984075,
|
| 1576 |
+
"learning_rate": 2e-05,
|
| 1577 |
+
"loss": 0.6785373687744141,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.061692851854898416,
|
| 1582 |
+
"grad_norm": 1.5143827466792619,
|
| 1583 |
+
"learning_rate": 2e-05,
|
| 1584 |
+
"loss": 0.6751925945281982,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.06196704230758685,
|
| 1589 |
+
"grad_norm": 2.0161013083730874,
|
| 1590 |
+
"learning_rate": 2e-05,
|
| 1591 |
+
"loss": 0.7139543533325196,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.06224123276027529,
|
| 1596 |
+
"grad_norm": 1.5476890937684233,
|
| 1597 |
+
"learning_rate": 2e-05,
|
| 1598 |
+
"loss": 0.6949531555175781,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.06251542321296373,
|
| 1603 |
+
"grad_norm": 2.854149288642599,
|
| 1604 |
+
"learning_rate": 2e-05,
|
| 1605 |
+
"loss": 0.7019545555114746,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.06278961366565217,
|
| 1610 |
+
"grad_norm": 2.8910444598110145,
|
| 1611 |
+
"learning_rate": 2e-05,
|
| 1612 |
+
"loss": 0.679999828338623,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.0630638041183406,
|
| 1617 |
+
"grad_norm": 4.5131786865492165,
|
| 1618 |
+
"learning_rate": 2e-05,
|
| 1619 |
+
"loss": 0.743858814239502,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.06333799457102904,
|
| 1624 |
+
"grad_norm": 1.6666983651920946,
|
| 1625 |
+
"learning_rate": 2e-05,
|
| 1626 |
+
"loss": 0.6903061866760254,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.06361218502371747,
|
| 1631 |
+
"grad_norm": 3.9236409182044927,
|
| 1632 |
+
"learning_rate": 2e-05,
|
| 1633 |
+
"loss": 0.6770228385925293,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.06388637547640591,
|
| 1638 |
+
"grad_norm": 3.5304751421823375,
|
| 1639 |
+
"learning_rate": 2e-05,
|
| 1640 |
+
"loss": 0.683766508102417,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.06416056592909435,
|
| 1645 |
+
"grad_norm": 2.3440415121115508,
|
| 1646 |
+
"learning_rate": 2e-05,
|
| 1647 |
+
"loss": 0.685692024230957,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.06443475638178278,
|
| 1652 |
+
"grad_norm": 2.023495973731089,
|
| 1653 |
+
"learning_rate": 2e-05,
|
| 1654 |
+
"loss": 0.6623273372650147,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.06470894683447122,
|
| 1659 |
+
"grad_norm": 1.3644101729258715,
|
| 1660 |
+
"learning_rate": 2e-05,
|
| 1661 |
+
"loss": 0.6689095497131348,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.06498313728715967,
|
| 1666 |
+
"grad_norm": 1.9596672876784174,
|
| 1667 |
+
"learning_rate": 2e-05,
|
| 1668 |
+
"loss": 0.691974401473999,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.0652573277398481,
|
| 1673 |
+
"grad_norm": 1.3195698986606514,
|
| 1674 |
+
"learning_rate": 2e-05,
|
| 1675 |
+
"loss": 0.6821478843688965,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.06553151819253654,
|
| 1680 |
+
"grad_norm": 1.7573022373302587,
|
| 1681 |
+
"learning_rate": 2e-05,
|
| 1682 |
+
"loss": 0.6735329627990723,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.06580570864522498,
|
| 1687 |
+
"grad_norm": 1.53838950058506,
|
| 1688 |
+
"learning_rate": 2e-05,
|
| 1689 |
+
"loss": 0.748023509979248,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.06607989909791341,
|
| 1694 |
+
"grad_norm": 1.2532586297274881,
|
| 1695 |
+
"learning_rate": 2e-05,
|
| 1696 |
+
"loss": 0.7004266738891601,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.06635408955060185,
|
| 1701 |
+
"grad_norm": 1.0700352542953793,
|
| 1702 |
+
"learning_rate": 2e-05,
|
| 1703 |
+
"loss": 0.6959473609924316,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.06662828000329028,
|
| 1708 |
+
"grad_norm": 4.556276241049152,
|
| 1709 |
+
"learning_rate": 2e-05,
|
| 1710 |
+
"loss": 0.8061809539794922,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.06690247045597872,
|
| 1715 |
+
"grad_norm": 2.152509533567621,
|
| 1716 |
+
"learning_rate": 2e-05,
|
| 1717 |
+
"loss": 0.6806859016418457,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.06717666090866715,
|
| 1722 |
+
"grad_norm": 2.844321748025424,
|
| 1723 |
+
"learning_rate": 2e-05,
|
| 1724 |
+
"loss": 0.6640607833862304,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.0674508513613556,
|
| 1729 |
+
"grad_norm": 1.1844711198885922,
|
| 1730 |
+
"learning_rate": 2e-05,
|
| 1731 |
+
"loss": 0.6736173629760742,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.06772504181404404,
|
| 1736 |
+
"grad_norm": 1.7441815113714467,
|
| 1737 |
+
"learning_rate": 2e-05,
|
| 1738 |
+
"loss": 0.6737575531005859,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.06799923226673248,
|
| 1743 |
+
"grad_norm": 15.110436266375306,
|
| 1744 |
+
"learning_rate": 2e-05,
|
| 1745 |
+
"loss": 0.6922641277313233,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.06827342271942091,
|
| 1750 |
+
"grad_norm": 2.2896995422971513,
|
| 1751 |
+
"learning_rate": 2e-05,
|
| 1752 |
+
"loss": 0.6922240257263184,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.06854761317210935,
|
| 1757 |
+
"grad_norm": 3.3946567144349524,
|
| 1758 |
+
"learning_rate": 2e-05,
|
| 1759 |
+
"loss": 0.7507569313049316,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.06882180362479778,
|
| 1764 |
+
"grad_norm": 2.4027551835720606,
|
| 1765 |
+
"learning_rate": 2e-05,
|
| 1766 |
+
"loss": 0.6868194580078125,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.06909599407748622,
|
| 1771 |
+
"grad_norm": 2.642622943517862,
|
| 1772 |
+
"learning_rate": 2e-05,
|
| 1773 |
+
"loss": 0.671518611907959,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.06937018453017466,
|
| 1778 |
+
"grad_norm": 1.5309404635845658,
|
| 1779 |
+
"learning_rate": 2e-05,
|
| 1780 |
+
"loss": 0.7356051921844482,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.06964437498286309,
|
| 1785 |
+
"grad_norm": 3.0548677174511956,
|
| 1786 |
+
"learning_rate": 2e-05,
|
| 1787 |
+
"loss": 0.6758328437805176,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.06991856543555153,
|
| 1792 |
+
"grad_norm": 1.193438740119827,
|
| 1793 |
+
"learning_rate": 2e-05,
|
| 1794 |
+
"loss": 0.6692746639251709,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.07019275588823998,
|
| 1799 |
+
"grad_norm": 1.2975925737053815,
|
| 1800 |
+
"learning_rate": 2e-05,
|
| 1801 |
+
"loss": 0.6996981620788574,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.07046694634092841,
|
| 1806 |
+
"grad_norm": 2.604623408290158,
|
| 1807 |
+
"learning_rate": 2e-05,
|
| 1808 |
+
"loss": 0.6900700569152832,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.07074113679361685,
|
| 1813 |
+
"grad_norm": 1.3002551593133926,
|
| 1814 |
+
"learning_rate": 2e-05,
|
| 1815 |
+
"loss": 0.6943139553070068,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.07101532724630528,
|
| 1820 |
+
"grad_norm": 1.2531714833158747,
|
| 1821 |
+
"learning_rate": 2e-05,
|
| 1822 |
+
"loss": 0.6756250381469726,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.07128951769899372,
|
| 1827 |
+
"grad_norm": 1.2937652650463045,
|
| 1828 |
+
"learning_rate": 2e-05,
|
| 1829 |
+
"loss": 0.663780403137207,
|
| 1830 |
+
"step": 2600
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 0.07156370815168216,
|
| 1834 |
+
"grad_norm": 2.6360485085938437,
|
| 1835 |
+
"learning_rate": 2e-05,
|
| 1836 |
+
"loss": 0.7122066020965576,
|
| 1837 |
+
"step": 2610
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 0.07183789860437059,
|
| 1841 |
+
"grad_norm": 1.4934934194790765,
|
| 1842 |
+
"learning_rate": 2e-05,
|
| 1843 |
+
"loss": 0.7055806159973145,
|
| 1844 |
+
"step": 2620
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 0.07211208905705903,
|
| 1848 |
+
"grad_norm": 1.265554029972572,
|
| 1849 |
+
"learning_rate": 2e-05,
|
| 1850 |
+
"loss": 0.7235059261322021,
|
| 1851 |
+
"step": 2630
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 0.07238627950974746,
|
| 1855 |
+
"grad_norm": 1.9958554767672445,
|
| 1856 |
+
"learning_rate": 2e-05,
|
| 1857 |
+
"loss": 0.7027205467224121,
|
| 1858 |
+
"step": 2640
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 0.07266046996243591,
|
| 1862 |
+
"grad_norm": 1.2413434845129339,
|
| 1863 |
+
"learning_rate": 2e-05,
|
| 1864 |
+
"loss": 0.6815381526947022,
|
| 1865 |
+
"step": 2650
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 0.07293466041512435,
|
| 1869 |
+
"grad_norm": 1.1371354821804263,
|
| 1870 |
+
"learning_rate": 2e-05,
|
| 1871 |
+
"loss": 0.6793872356414795,
|
| 1872 |
+
"step": 2660
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 0.07320885086781279,
|
| 1876 |
+
"grad_norm": 1.509310628738819,
|
| 1877 |
+
"learning_rate": 2e-05,
|
| 1878 |
+
"loss": 0.6791254997253418,
|
| 1879 |
+
"step": 2670
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 0.07348304132050122,
|
| 1883 |
+
"grad_norm": 1.1060317096479009,
|
| 1884 |
+
"learning_rate": 2e-05,
|
| 1885 |
+
"loss": 0.6734872341156006,
|
| 1886 |
+
"step": 2680
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 0.07375723177318966,
|
| 1890 |
+
"grad_norm": 2.300558523814945,
|
| 1891 |
+
"learning_rate": 2e-05,
|
| 1892 |
+
"loss": 0.7129344940185547,
|
| 1893 |
+
"step": 2690
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 0.0740314222258781,
|
| 1897 |
+
"grad_norm": 1.4427965933652498,
|
| 1898 |
+
"learning_rate": 2e-05,
|
| 1899 |
+
"loss": 0.6588455200195312,
|
| 1900 |
+
"step": 2700
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 0.07430561267856653,
|
| 1904 |
+
"grad_norm": 1.294782697054497,
|
| 1905 |
+
"learning_rate": 2e-05,
|
| 1906 |
+
"loss": 0.6733306884765625,
|
| 1907 |
+
"step": 2710
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 0.07457980313125497,
|
| 1911 |
+
"grad_norm": 1.5340099682623456,
|
| 1912 |
+
"learning_rate": 2e-05,
|
| 1913 |
+
"loss": 0.67325758934021,
|
| 1914 |
+
"step": 2720
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 0.0748539935839434,
|
| 1918 |
+
"grad_norm": 1.206231978423083,
|
| 1919 |
+
"learning_rate": 2e-05,
|
| 1920 |
+
"loss": 0.6762287139892578,
|
| 1921 |
+
"step": 2730
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 0.07512818403663185,
|
| 1925 |
+
"grad_norm": 1.9615078302959699,
|
| 1926 |
+
"learning_rate": 2e-05,
|
| 1927 |
+
"loss": 0.6494486808776856,
|
| 1928 |
+
"step": 2740
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 0.07540237448932029,
|
| 1932 |
+
"grad_norm": 2.7069935077316156,
|
| 1933 |
+
"learning_rate": 2e-05,
|
| 1934 |
+
"loss": 0.6673858642578125,
|
| 1935 |
+
"step": 2750
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 0.07567656494200872,
|
| 1939 |
+
"grad_norm": 1.244130150553988,
|
| 1940 |
+
"learning_rate": 2e-05,
|
| 1941 |
+
"loss": 0.6522541046142578,
|
| 1942 |
+
"step": 2760
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 0.07595075539469716,
|
| 1946 |
+
"grad_norm": 1.2098821554953647,
|
| 1947 |
+
"learning_rate": 2e-05,
|
| 1948 |
+
"loss": 0.683398962020874,
|
| 1949 |
+
"step": 2770
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 0.0762249458473856,
|
| 1953 |
+
"grad_norm": 1.288135083537603,
|
| 1954 |
+
"learning_rate": 2e-05,
|
| 1955 |
+
"loss": 0.6805778503417969,
|
| 1956 |
+
"step": 2780
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 0.07649913630007403,
|
| 1960 |
+
"grad_norm": 1.6304988366424138,
|
| 1961 |
+
"learning_rate": 2e-05,
|
| 1962 |
+
"loss": 0.7026628017425537,
|
| 1963 |
+
"step": 2790
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 0.07677332675276247,
|
| 1967 |
+
"grad_norm": 2.7936841529325167,
|
| 1968 |
+
"learning_rate": 2e-05,
|
| 1969 |
+
"loss": 0.6661033630371094,
|
| 1970 |
+
"step": 2800
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 0.0770475172054509,
|
| 1974 |
+
"grad_norm": 1.0298921513328017,
|
| 1975 |
+
"learning_rate": 2e-05,
|
| 1976 |
+
"loss": 0.6871449470520019,
|
| 1977 |
+
"step": 2810
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 0.07732170765813934,
|
| 1981 |
+
"grad_norm": 1.1609549565762298,
|
| 1982 |
+
"learning_rate": 2e-05,
|
| 1983 |
+
"loss": 0.7018230438232422,
|
| 1984 |
+
"step": 2820
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 0.07759589811082779,
|
| 1988 |
+
"grad_norm": 1.4937627603643764,
|
| 1989 |
+
"learning_rate": 2e-05,
|
| 1990 |
+
"loss": 0.6921014785766602,
|
| 1991 |
+
"step": 2830
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 0.07787008856351622,
|
| 1995 |
+
"grad_norm": 1.06750496448494,
|
| 1996 |
+
"learning_rate": 2e-05,
|
| 1997 |
+
"loss": 0.6660943508148194,
|
| 1998 |
+
"step": 2840
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 0.07814427901620466,
|
| 2002 |
+
"grad_norm": 1.3483148967347722,
|
| 2003 |
+
"learning_rate": 2e-05,
|
| 2004 |
+
"loss": 0.6800715923309326,
|
| 2005 |
+
"step": 2850
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 0.0784184694688931,
|
| 2009 |
+
"grad_norm": 0.9156475317981986,
|
| 2010 |
+
"learning_rate": 2e-05,
|
| 2011 |
+
"loss": 0.6963984489440918,
|
| 2012 |
+
"step": 2860
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 0.07869265992158153,
|
| 2016 |
+
"grad_norm": 2.541415851603761,
|
| 2017 |
+
"learning_rate": 2e-05,
|
| 2018 |
+
"loss": 0.6630027294158936,
|
| 2019 |
+
"step": 2870
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 0.07896685037426997,
|
| 2023 |
+
"grad_norm": 16.943114843325624,
|
| 2024 |
+
"learning_rate": 2e-05,
|
| 2025 |
+
"loss": 0.6456388473510742,
|
| 2026 |
+
"step": 2880
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 0.0792410408269584,
|
| 2030 |
+
"grad_norm": 1.5105418227638865,
|
| 2031 |
+
"learning_rate": 2e-05,
|
| 2032 |
+
"loss": 0.6730733394622803,
|
| 2033 |
+
"step": 2890
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 0.07951523127964684,
|
| 2037 |
+
"grad_norm": 1.0845416608133904,
|
| 2038 |
+
"learning_rate": 2e-05,
|
| 2039 |
+
"loss": 0.7171577453613281,
|
| 2040 |
+
"step": 2900
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 0.07978942173233528,
|
| 2044 |
+
"grad_norm": 1.1519480568569667,
|
| 2045 |
+
"learning_rate": 2e-05,
|
| 2046 |
+
"loss": 0.6709290504455566,
|
| 2047 |
+
"step": 2910
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 0.08006361218502371,
|
| 2051 |
+
"grad_norm": 1.2117290732120638,
|
| 2052 |
+
"learning_rate": 2e-05,
|
| 2053 |
+
"loss": 0.6620571136474609,
|
| 2054 |
+
"step": 2920
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 0.08033780263771216,
|
| 2058 |
+
"grad_norm": 1.2315110130143458,
|
| 2059 |
+
"learning_rate": 2e-05,
|
| 2060 |
+
"loss": 0.6733248710632325,
|
| 2061 |
+
"step": 2930
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 0.0806119930904006,
|
| 2065 |
+
"grad_norm": 2.440287172046704,
|
| 2066 |
+
"learning_rate": 2e-05,
|
| 2067 |
+
"loss": 0.6643840789794921,
|
| 2068 |
+
"step": 2940
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 0.08088618354308903,
|
| 2072 |
+
"grad_norm": 1.679547493244505,
|
| 2073 |
+
"learning_rate": 2e-05,
|
| 2074 |
+
"loss": 0.6582764625549317,
|
| 2075 |
+
"step": 2950
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 0.08116037399577747,
|
| 2079 |
+
"grad_norm": 1.4899450244629049,
|
| 2080 |
+
"learning_rate": 2e-05,
|
| 2081 |
+
"loss": 0.6351748466491699,
|
| 2082 |
+
"step": 2960
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 0.0814345644484659,
|
| 2086 |
+
"grad_norm": 8.727623946787714,
|
| 2087 |
+
"learning_rate": 2e-05,
|
| 2088 |
+
"loss": 0.6684012413024902,
|
| 2089 |
+
"step": 2970
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 0.08170875490115434,
|
| 2093 |
+
"grad_norm": 1.579015056993268,
|
| 2094 |
+
"learning_rate": 2e-05,
|
| 2095 |
+
"loss": 0.7062510490417481,
|
| 2096 |
+
"step": 2980
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 0.08198294535384278,
|
| 2100 |
+
"grad_norm": 1.1464807229444751,
|
| 2101 |
+
"learning_rate": 2e-05,
|
| 2102 |
+
"loss": 0.7049516201019287,
|
| 2103 |
+
"step": 2990
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 0.08225713580653121,
|
| 2107 |
+
"grad_norm": 5.869960507362621,
|
| 2108 |
+
"learning_rate": 2e-05,
|
| 2109 |
+
"loss": 0.6782597541809082,
|
| 2110 |
+
"step": 3000
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 0.08253132625921965,
|
| 2114 |
+
"grad_norm": 1.4182529764185894,
|
| 2115 |
+
"learning_rate": 2e-05,
|
| 2116 |
+
"loss": 0.6684169769287109,
|
| 2117 |
+
"step": 3010
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 0.0828055167119081,
|
| 2121 |
+
"grad_norm": 1.6893769027297443,
|
| 2122 |
+
"learning_rate": 2e-05,
|
| 2123 |
+
"loss": 0.6985646247863769,
|
| 2124 |
+
"step": 3020
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 0.08307970716459653,
|
| 2128 |
+
"grad_norm": 2.5342955627283548,
|
| 2129 |
+
"learning_rate": 2e-05,
|
| 2130 |
+
"loss": 0.6566522121429443,
|
| 2131 |
+
"step": 3030
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 0.08335389761728497,
|
| 2135 |
+
"grad_norm": 2.2039912383675504,
|
| 2136 |
+
"learning_rate": 2e-05,
|
| 2137 |
+
"loss": 0.6650538444519043,
|
| 2138 |
+
"step": 3040
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 0.0836280880699734,
|
| 2142 |
+
"grad_norm": 1.1365614280802196,
|
| 2143 |
+
"learning_rate": 2e-05,
|
| 2144 |
+
"loss": 0.7018685340881348,
|
| 2145 |
+
"step": 3050
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 0.08390227852266184,
|
| 2149 |
+
"grad_norm": 9.019421975743805,
|
| 2150 |
+
"learning_rate": 2e-05,
|
| 2151 |
+
"loss": 0.6758551597595215,
|
| 2152 |
+
"step": 3060
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 0.08417646897535028,
|
| 2156 |
+
"grad_norm": 3.6659007510462036,
|
| 2157 |
+
"learning_rate": 2e-05,
|
| 2158 |
+
"loss": 0.6577141761779786,
|
| 2159 |
+
"step": 3070
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 0.08445065942803871,
|
| 2163 |
+
"grad_norm": 1.5678832350052356,
|
| 2164 |
+
"learning_rate": 2e-05,
|
| 2165 |
+
"loss": 0.6269432067871094,
|
| 2166 |
+
"step": 3080
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 0.08472484988072715,
|
| 2170 |
+
"grad_norm": 1.727670242813246,
|
| 2171 |
+
"learning_rate": 2e-05,
|
| 2172 |
+
"loss": 0.6586098670959473,
|
| 2173 |
+
"step": 3090
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 0.08499904033341558,
|
| 2177 |
+
"grad_norm": 3.755002398572796,
|
| 2178 |
+
"learning_rate": 2e-05,
|
| 2179 |
+
"loss": 0.7226817607879639,
|
| 2180 |
+
"step": 3100
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 0.08527323078610403,
|
| 2184 |
+
"grad_norm": 2.1330764933548174,
|
| 2185 |
+
"learning_rate": 2e-05,
|
| 2186 |
+
"loss": 0.7848432064056396,
|
| 2187 |
+
"step": 3110
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 0.08554742123879247,
|
| 2191 |
+
"grad_norm": 7.344569339937305,
|
| 2192 |
+
"learning_rate": 2e-05,
|
| 2193 |
+
"loss": 0.6657351970672607,
|
| 2194 |
+
"step": 3120
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 0.0858216116914809,
|
| 2198 |
+
"grad_norm": 10.468008663230526,
|
| 2199 |
+
"learning_rate": 2e-05,
|
| 2200 |
+
"loss": 0.7568781852722168,
|
| 2201 |
+
"step": 3130
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 0.08609580214416934,
|
| 2205 |
+
"grad_norm": 2.1138000067535527,
|
| 2206 |
+
"learning_rate": 2e-05,
|
| 2207 |
+
"loss": 0.6460443496704101,
|
| 2208 |
+
"step": 3140
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 0.08636999259685778,
|
| 2212 |
+
"grad_norm": 4.547865280742984,
|
| 2213 |
+
"learning_rate": 2e-05,
|
| 2214 |
+
"loss": 0.6882412910461426,
|
| 2215 |
+
"step": 3150
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 0.08664418304954621,
|
| 2219 |
+
"grad_norm": 3.5154327639954515,
|
| 2220 |
+
"learning_rate": 2e-05,
|
| 2221 |
+
"loss": 0.6506322860717774,
|
| 2222 |
+
"step": 3160
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 0.08691837350223465,
|
| 2226 |
+
"grad_norm": 2.7275478121939303,
|
| 2227 |
+
"learning_rate": 2e-05,
|
| 2228 |
+
"loss": 0.6696713447570801,
|
| 2229 |
+
"step": 3170
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 0.08719256395492309,
|
| 2233 |
+
"grad_norm": 1.3787630646286215,
|
| 2234 |
+
"learning_rate": 2e-05,
|
| 2235 |
+
"loss": 0.6836231231689454,
|
| 2236 |
+
"step": 3180
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 0.08746675440761152,
|
| 2240 |
+
"grad_norm": 2.047844653500189,
|
| 2241 |
+
"learning_rate": 2e-05,
|
| 2242 |
+
"loss": 0.6477073669433594,
|
| 2243 |
+
"step": 3190
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 0.08774094486029996,
|
| 2247 |
+
"grad_norm": 1.3852125913708124,
|
| 2248 |
+
"learning_rate": 2e-05,
|
| 2249 |
+
"loss": 0.6763352870941162,
|
| 2250 |
+
"step": 3200
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 0.08801513531298841,
|
| 2254 |
+
"grad_norm": 2.3040797052670854,
|
| 2255 |
+
"learning_rate": 2e-05,
|
| 2256 |
+
"loss": 0.6652051448822022,
|
| 2257 |
+
"step": 3210
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 0.08828932576567684,
|
| 2261 |
+
"grad_norm": 14.697316208995012,
|
| 2262 |
+
"learning_rate": 2e-05,
|
| 2263 |
+
"loss": 0.7217847347259522,
|
| 2264 |
+
"step": 3220
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 0.08856351621836528,
|
| 2268 |
+
"grad_norm": 2.51269929770817,
|
| 2269 |
+
"learning_rate": 2e-05,
|
| 2270 |
+
"loss": 0.6437290191650391,
|
| 2271 |
+
"step": 3230
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 0.08883770667105371,
|
| 2275 |
+
"grad_norm": 1.0839928596474344,
|
| 2276 |
+
"learning_rate": 2e-05,
|
| 2277 |
+
"loss": 0.6739638328552247,
|
| 2278 |
+
"step": 3240
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 0.08911189712374215,
|
| 2282 |
+
"grad_norm": 1.4232969395596793,
|
| 2283 |
+
"learning_rate": 2e-05,
|
| 2284 |
+
"loss": 0.6545791625976562,
|
| 2285 |
+
"step": 3250
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 0.08938608757643059,
|
| 2289 |
+
"grad_norm": 2.7705768107632407,
|
| 2290 |
+
"learning_rate": 2e-05,
|
| 2291 |
+
"loss": 0.6883786678314209,
|
| 2292 |
+
"step": 3260
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 0.08966027802911902,
|
| 2296 |
+
"grad_norm": 1.6294188609939928,
|
| 2297 |
+
"learning_rate": 2e-05,
|
| 2298 |
+
"loss": 0.6384977340698242,
|
| 2299 |
+
"step": 3270
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 0.08993446848180746,
|
| 2303 |
+
"grad_norm": 1.594582926676938,
|
| 2304 |
+
"learning_rate": 2e-05,
|
| 2305 |
+
"loss": 0.6840876579284668,
|
| 2306 |
+
"step": 3280
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 0.0902086589344959,
|
| 2310 |
+
"grad_norm": 3.419979140068578,
|
| 2311 |
+
"learning_rate": 2e-05,
|
| 2312 |
+
"loss": 0.6966186046600342,
|
| 2313 |
+
"step": 3290
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 0.09048284938718434,
|
| 2317 |
+
"grad_norm": 1.1720984784354762,
|
| 2318 |
+
"learning_rate": 2e-05,
|
| 2319 |
+
"loss": 0.7145553588867187,
|
| 2320 |
+
"step": 3300
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 0.09075703983987278,
|
| 2324 |
+
"grad_norm": 2.5865997974005905,
|
| 2325 |
+
"learning_rate": 2e-05,
|
| 2326 |
+
"loss": 0.670125675201416,
|
| 2327 |
+
"step": 3310
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 0.09103123029256122,
|
| 2331 |
+
"grad_norm": 1.7653476493006024,
|
| 2332 |
+
"learning_rate": 2e-05,
|
| 2333 |
+
"loss": 0.6601348876953125,
|
| 2334 |
+
"step": 3320
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 0.09130542074524965,
|
| 2338 |
+
"grad_norm": 1.078475603753133,
|
| 2339 |
+
"learning_rate": 2e-05,
|
| 2340 |
+
"loss": 0.6758632659912109,
|
| 2341 |
+
"step": 3330
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 0.09157961119793809,
|
| 2345 |
+
"grad_norm": 1.9813753893710975,
|
| 2346 |
+
"learning_rate": 2e-05,
|
| 2347 |
+
"loss": 0.6324165344238282,
|
| 2348 |
+
"step": 3340
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 0.09185380165062652,
|
| 2352 |
+
"grad_norm": 5.357267664807992,
|
| 2353 |
+
"learning_rate": 2e-05,
|
| 2354 |
+
"loss": 0.7044012069702148,
|
| 2355 |
+
"step": 3350
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 0.09212799210331496,
|
| 2359 |
+
"grad_norm": 1.9661122621370217,
|
| 2360 |
+
"learning_rate": 2e-05,
|
| 2361 |
+
"loss": 0.6732365608215332,
|
| 2362 |
+
"step": 3360
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 0.0924021825560034,
|
| 2366 |
+
"grad_norm": 1.463741721870027,
|
| 2367 |
+
"learning_rate": 2e-05,
|
| 2368 |
+
"loss": 0.6740346431732178,
|
| 2369 |
+
"step": 3370
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 0.09267637300869183,
|
| 2373 |
+
"grad_norm": 2.1100577721208684,
|
| 2374 |
+
"learning_rate": 2e-05,
|
| 2375 |
+
"loss": 0.6793565273284912,
|
| 2376 |
+
"step": 3380
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 0.09295056346138028,
|
| 2380 |
+
"grad_norm": 1.647800752209185,
|
| 2381 |
+
"learning_rate": 2e-05,
|
| 2382 |
+
"loss": 0.6734954357147217,
|
| 2383 |
+
"step": 3390
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 0.09322475391406872,
|
| 2387 |
+
"grad_norm": 5.2930839705776425,
|
| 2388 |
+
"learning_rate": 2e-05,
|
| 2389 |
+
"loss": 0.6649512767791748,
|
| 2390 |
+
"step": 3400
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 0.09349894436675715,
|
| 2394 |
+
"grad_norm": 1.2957879552995335,
|
| 2395 |
+
"learning_rate": 2e-05,
|
| 2396 |
+
"loss": 0.6656761169433594,
|
| 2397 |
+
"step": 3410
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 0.09377313481944559,
|
| 2401 |
+
"grad_norm": 2.2505114790067173,
|
| 2402 |
+
"learning_rate": 2e-05,
|
| 2403 |
+
"loss": 0.6644682884216309,
|
| 2404 |
+
"step": 3420
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 0.09404732527213402,
|
| 2408 |
+
"grad_norm": 3.242876354488533,
|
| 2409 |
+
"learning_rate": 2e-05,
|
| 2410 |
+
"loss": 0.7099063873291016,
|
| 2411 |
+
"step": 3430
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 0.09432151572482246,
|
| 2415 |
+
"grad_norm": 1.2326965640901657,
|
| 2416 |
+
"learning_rate": 2e-05,
|
| 2417 |
+
"loss": 0.6550569057464599,
|
| 2418 |
+
"step": 3440
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 0.0945957061775109,
|
| 2422 |
+
"grad_norm": 1.1460797675833456,
|
| 2423 |
+
"learning_rate": 2e-05,
|
| 2424 |
+
"loss": 0.7175025463104248,
|
| 2425 |
+
"step": 3450
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 0.09486989663019933,
|
| 2429 |
+
"grad_norm": 1.2339953246226074,
|
| 2430 |
+
"learning_rate": 2e-05,
|
| 2431 |
+
"loss": 0.6755492687225342,
|
| 2432 |
+
"step": 3460
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 0.09514408708288777,
|
| 2436 |
+
"grad_norm": 1.2239178623958433,
|
| 2437 |
+
"learning_rate": 2e-05,
|
| 2438 |
+
"loss": 0.7038548469543457,
|
| 2439 |
+
"step": 3470
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 0.09541827753557622,
|
| 2443 |
+
"grad_norm": 2.054453414781126,
|
| 2444 |
+
"learning_rate": 2e-05,
|
| 2445 |
+
"loss": 0.7407256126403808,
|
| 2446 |
+
"step": 3480
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 0.09569246798826465,
|
| 2450 |
+
"grad_norm": 0.962580766048047,
|
| 2451 |
+
"learning_rate": 2e-05,
|
| 2452 |
+
"loss": 0.6635346412658691,
|
| 2453 |
+
"step": 3490
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 0.09596665844095309,
|
| 2457 |
+
"grad_norm": 1.2912100361591508,
|
| 2458 |
+
"learning_rate": 2e-05,
|
| 2459 |
+
"loss": 0.7218295574188233,
|
| 2460 |
+
"step": 3500
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 0.09624084889364153,
|
| 2464 |
+
"grad_norm": 1.1758629619276566,
|
| 2465 |
+
"learning_rate": 2e-05,
|
| 2466 |
+
"loss": 0.6495042324066163,
|
| 2467 |
+
"step": 3510
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 0.09651503934632996,
|
| 2471 |
+
"grad_norm": 3.5366805734644338,
|
| 2472 |
+
"learning_rate": 2e-05,
|
| 2473 |
+
"loss": 0.6897785186767578,
|
| 2474 |
+
"step": 3520
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 0.0967892297990184,
|
| 2478 |
+
"grad_norm": 2.272234739463371,
|
| 2479 |
+
"learning_rate": 2e-05,
|
| 2480 |
+
"loss": 0.6400076866149902,
|
| 2481 |
+
"step": 3530
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 0.09706342025170683,
|
| 2485 |
+
"grad_norm": 1.0410457309022894,
|
| 2486 |
+
"learning_rate": 2e-05,
|
| 2487 |
+
"loss": 0.6666034698486328,
|
| 2488 |
+
"step": 3540
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 0.09733761070439527,
|
| 2492 |
+
"grad_norm": 1.47293937630036,
|
| 2493 |
+
"learning_rate": 2e-05,
|
| 2494 |
+
"loss": 0.6687037467956543,
|
| 2495 |
+
"step": 3550
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 0.0976118011570837,
|
| 2499 |
+
"grad_norm": 1.9748563321883545,
|
| 2500 |
+
"learning_rate": 2e-05,
|
| 2501 |
+
"loss": 0.6717099666595459,
|
| 2502 |
+
"step": 3560
|
| 2503 |
+
},
|
| 2504 |
+
{
|
| 2505 |
+
"epoch": 0.09788599160977214,
|
| 2506 |
+
"grad_norm": 1.5933750546326055,
|
| 2507 |
+
"learning_rate": 2e-05,
|
| 2508 |
+
"loss": 0.6562514305114746,
|
| 2509 |
+
"step": 3570
|
| 2510 |
+
},
|
| 2511 |
+
{
|
| 2512 |
+
"epoch": 0.09816018206246059,
|
| 2513 |
+
"grad_norm": 3.128593619633481,
|
| 2514 |
+
"learning_rate": 2e-05,
|
| 2515 |
+
"loss": 0.7533950805664062,
|
| 2516 |
+
"step": 3580
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 0.09843437251514903,
|
| 2520 |
+
"grad_norm": 2.891793023072984,
|
| 2521 |
+
"learning_rate": 2e-05,
|
| 2522 |
+
"loss": 0.6750978469848633,
|
| 2523 |
+
"step": 3590
|
| 2524 |
+
},
|
| 2525 |
+
{
|
| 2526 |
+
"epoch": 0.09870856296783746,
|
| 2527 |
+
"grad_norm": 2.1897121019394965,
|
| 2528 |
+
"learning_rate": 2e-05,
|
| 2529 |
+
"loss": 0.6912845134735107,
|
| 2530 |
+
"step": 3600
|
| 2531 |
+
},
|
| 2532 |
+
{
|
| 2533 |
+
"epoch": 0.0989827534205259,
|
| 2534 |
+
"grad_norm": 2.004346028083513,
|
| 2535 |
+
"learning_rate": 2e-05,
|
| 2536 |
+
"loss": 0.6787797927856445,
|
| 2537 |
+
"step": 3610
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 0.09925694387321433,
|
| 2541 |
+
"grad_norm": 2.915237046123075,
|
| 2542 |
+
"learning_rate": 2e-05,
|
| 2543 |
+
"loss": 0.6627199172973632,
|
| 2544 |
+
"step": 3620
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 0.09953113432590277,
|
| 2548 |
+
"grad_norm": 1.135046615758623,
|
| 2549 |
+
"learning_rate": 2e-05,
|
| 2550 |
+
"loss": 0.6445761680603027,
|
| 2551 |
+
"step": 3630
|
| 2552 |
+
},
|
| 2553 |
+
{
|
| 2554 |
+
"epoch": 0.0998053247785912,
|
| 2555 |
+
"grad_norm": 1.3959275315420312,
|
| 2556 |
+
"learning_rate": 2e-05,
|
| 2557 |
+
"loss": 0.667326545715332,
|
| 2558 |
+
"step": 3640
|
| 2559 |
+
},
|
| 2560 |
+
{
|
| 2561 |
+
"epoch": 0.10007951523127964,
|
| 2562 |
+
"grad_norm": 0.9909291231965143,
|
| 2563 |
+
"learning_rate": 2e-05,
|
| 2564 |
+
"loss": 0.6794458389282226,
|
| 2565 |
+
"step": 3650
|
| 2566 |
+
},
|
| 2567 |
+
{
|
| 2568 |
+
"epoch": 0.10035370568396808,
|
| 2569 |
+
"grad_norm": 1.8845470713145256,
|
| 2570 |
+
"learning_rate": 2e-05,
|
| 2571 |
+
"loss": 0.6390461921691895,
|
| 2572 |
+
"step": 3660
|
| 2573 |
+
},
|
| 2574 |
+
{
|
| 2575 |
+
"epoch": 0.10062789613665653,
|
| 2576 |
+
"grad_norm": 1.5618410538502554,
|
| 2577 |
+
"learning_rate": 2e-05,
|
| 2578 |
+
"loss": 0.6542781829833985,
|
| 2579 |
+
"step": 3670
|
| 2580 |
+
},
|
| 2581 |
+
{
|
| 2582 |
+
"epoch": 0.10090208658934496,
|
| 2583 |
+
"grad_norm": 1.1053530735550423,
|
| 2584 |
+
"learning_rate": 2e-05,
|
| 2585 |
+
"loss": 0.6584223747253418,
|
| 2586 |
+
"step": 3680
|
| 2587 |
+
},
|
| 2588 |
+
{
|
| 2589 |
+
"epoch": 0.1011762770420334,
|
| 2590 |
+
"grad_norm": 1.8050013341368214,
|
| 2591 |
+
"learning_rate": 2e-05,
|
| 2592 |
+
"loss": 0.6776100158691406,
|
| 2593 |
+
"step": 3690
|
| 2594 |
+
},
|
| 2595 |
+
{
|
| 2596 |
+
"epoch": 0.10145046749472184,
|
| 2597 |
+
"grad_norm": 1.3509209662153294,
|
| 2598 |
+
"learning_rate": 2e-05,
|
| 2599 |
+
"loss": 0.7505715847015381,
|
| 2600 |
+
"step": 3700
|
| 2601 |
+
},
|
| 2602 |
+
{
|
| 2603 |
+
"epoch": 0.10172465794741027,
|
| 2604 |
+
"grad_norm": 1.3171737413682911,
|
| 2605 |
+
"learning_rate": 2e-05,
|
| 2606 |
+
"loss": 0.6496476650238037,
|
| 2607 |
+
"step": 3710
|
| 2608 |
+
},
|
| 2609 |
+
{
|
| 2610 |
+
"epoch": 0.10199884840009871,
|
| 2611 |
+
"grad_norm": 2.037728719646489,
|
| 2612 |
+
"learning_rate": 2e-05,
|
| 2613 |
+
"loss": 0.6601945400238037,
|
| 2614 |
+
"step": 3720
|
| 2615 |
+
},
|
| 2616 |
+
{
|
| 2617 |
+
"epoch": 0.10227303885278714,
|
| 2618 |
+
"grad_norm": 0.9419204630678888,
|
| 2619 |
+
"learning_rate": 2e-05,
|
| 2620 |
+
"loss": 0.6456241607666016,
|
| 2621 |
+
"step": 3730
|
| 2622 |
+
},
|
| 2623 |
+
{
|
| 2624 |
+
"epoch": 0.10254722930547558,
|
| 2625 |
+
"grad_norm": 0.9978934523669455,
|
| 2626 |
+
"learning_rate": 2e-05,
|
| 2627 |
+
"loss": 0.6376785278320313,
|
| 2628 |
+
"step": 3740
|
| 2629 |
+
},
|
| 2630 |
+
{
|
| 2631 |
+
"epoch": 0.10282141975816401,
|
| 2632 |
+
"grad_norm": 1.2463996045138312,
|
| 2633 |
+
"learning_rate": 2e-05,
|
| 2634 |
+
"loss": 0.6607570648193359,
|
| 2635 |
+
"step": 3750
|
| 2636 |
+
},
|
| 2637 |
+
{
|
| 2638 |
+
"epoch": 0.10309561021085246,
|
| 2639 |
+
"grad_norm": 2.5633755688709,
|
| 2640 |
+
"learning_rate": 2e-05,
|
| 2641 |
+
"loss": 0.6273582458496094,
|
| 2642 |
+
"step": 3760
|
| 2643 |
+
},
|
| 2644 |
+
{
|
| 2645 |
+
"epoch": 0.1033698006635409,
|
| 2646 |
+
"grad_norm": 1.3319182781528256,
|
| 2647 |
+
"learning_rate": 2e-05,
|
| 2648 |
+
"loss": 0.653649091720581,
|
| 2649 |
+
"step": 3770
|
| 2650 |
+
},
|
| 2651 |
+
{
|
| 2652 |
+
"epoch": 0.10364399111622934,
|
| 2653 |
+
"grad_norm": 1.5739184819076273,
|
| 2654 |
+
"learning_rate": 2e-05,
|
| 2655 |
+
"loss": 0.6797492027282714,
|
| 2656 |
+
"step": 3780
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 0.10391818156891777,
|
| 2660 |
+
"grad_norm": 1.0307904527745226,
|
| 2661 |
+
"learning_rate": 2e-05,
|
| 2662 |
+
"loss": 0.6382014274597168,
|
| 2663 |
+
"step": 3790
|
| 2664 |
+
},
|
| 2665 |
+
{
|
| 2666 |
+
"epoch": 0.10419237202160621,
|
| 2667 |
+
"grad_norm": 1.367080564181282,
|
| 2668 |
+
"learning_rate": 2e-05,
|
| 2669 |
+
"loss": 0.656506872177124,
|
| 2670 |
+
"step": 3800
|
| 2671 |
+
},
|
| 2672 |
+
{
|
| 2673 |
+
"epoch": 0.10446656247429464,
|
| 2674 |
+
"grad_norm": 1.1306043565346662,
|
| 2675 |
+
"learning_rate": 2e-05,
|
| 2676 |
+
"loss": 0.6709275722503663,
|
| 2677 |
+
"step": 3810
|
| 2678 |
+
},
|
| 2679 |
+
{
|
| 2680 |
+
"epoch": 0.10474075292698308,
|
| 2681 |
+
"grad_norm": 1.1484780247773891,
|
| 2682 |
+
"learning_rate": 2e-05,
|
| 2683 |
+
"loss": 0.6394906520843506,
|
| 2684 |
+
"step": 3820
|
| 2685 |
+
},
|
| 2686 |
+
{
|
| 2687 |
+
"epoch": 0.10501494337967152,
|
| 2688 |
+
"grad_norm": 1.1364725089670198,
|
| 2689 |
+
"learning_rate": 2e-05,
|
| 2690 |
+
"loss": 0.6385705947875977,
|
| 2691 |
+
"step": 3830
|
| 2692 |
+
},
|
| 2693 |
+
{
|
| 2694 |
+
"epoch": 0.10528913383235995,
|
| 2695 |
+
"grad_norm": 0.8720469361835742,
|
| 2696 |
+
"learning_rate": 2e-05,
|
| 2697 |
+
"loss": 0.648165512084961,
|
| 2698 |
+
"step": 3840
|
| 2699 |
+
},
|
| 2700 |
+
{
|
| 2701 |
+
"epoch": 0.10556332428504839,
|
| 2702 |
+
"grad_norm": 1.0753251518024716,
|
| 2703 |
+
"learning_rate": 2e-05,
|
| 2704 |
+
"loss": 0.7535962581634521,
|
| 2705 |
+
"step": 3850
|
| 2706 |
+
},
|
| 2707 |
+
{
|
| 2708 |
+
"epoch": 0.10583751473773684,
|
| 2709 |
+
"grad_norm": 2.0842578780523646,
|
| 2710 |
+
"learning_rate": 2e-05,
|
| 2711 |
+
"loss": 0.6594803810119629,
|
| 2712 |
+
"step": 3860
|
| 2713 |
+
},
|
| 2714 |
+
{
|
| 2715 |
+
"epoch": 0.10611170519042527,
|
| 2716 |
+
"grad_norm": 1.2009333278099745,
|
| 2717 |
+
"learning_rate": 2e-05,
|
| 2718 |
+
"loss": 0.6642222881317139,
|
| 2719 |
+
"step": 3870
|
| 2720 |
+
},
|
| 2721 |
+
{
|
| 2722 |
+
"epoch": 0.10638589564311371,
|
| 2723 |
+
"grad_norm": 1.336651409297708,
|
| 2724 |
+
"learning_rate": 2e-05,
|
| 2725 |
+
"loss": 0.6247560024261475,
|
| 2726 |
+
"step": 3880
|
| 2727 |
+
},
|
| 2728 |
+
{
|
| 2729 |
+
"epoch": 0.10666008609580215,
|
| 2730 |
+
"grad_norm": 1.2693616052452827,
|
| 2731 |
+
"learning_rate": 2e-05,
|
| 2732 |
+
"loss": 0.6595481395721435,
|
| 2733 |
+
"step": 3890
|
| 2734 |
+
},
|
| 2735 |
+
{
|
| 2736 |
+
"epoch": 0.10693427654849058,
|
| 2737 |
+
"grad_norm": 1.0968413832844812,
|
| 2738 |
+
"learning_rate": 2e-05,
|
| 2739 |
+
"loss": 0.6259689331054688,
|
| 2740 |
+
"step": 3900
|
| 2741 |
+
},
|
| 2742 |
+
{
|
| 2743 |
+
"epoch": 0.10720846700117902,
|
| 2744 |
+
"grad_norm": 1.4406885841286199,
|
| 2745 |
+
"learning_rate": 2e-05,
|
| 2746 |
+
"loss": 0.635016393661499,
|
| 2747 |
+
"step": 3910
|
| 2748 |
+
},
|
| 2749 |
+
{
|
| 2750 |
+
"epoch": 0.10748265745386745,
|
| 2751 |
+
"grad_norm": 1.1149430537042344,
|
| 2752 |
+
"learning_rate": 2e-05,
|
| 2753 |
+
"loss": 0.652572774887085,
|
| 2754 |
+
"step": 3920
|
| 2755 |
+
},
|
| 2756 |
+
{
|
| 2757 |
+
"epoch": 0.10775684790655589,
|
| 2758 |
+
"grad_norm": 1.4284277973722026,
|
| 2759 |
+
"learning_rate": 2e-05,
|
| 2760 |
+
"loss": 0.6603116512298584,
|
| 2761 |
+
"step": 3930
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 0.10803103835924432,
|
| 2765 |
+
"grad_norm": 2.191046426715608,
|
| 2766 |
+
"learning_rate": 2e-05,
|
| 2767 |
+
"loss": 0.6569153785705566,
|
| 2768 |
+
"step": 3940
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 0.10830522881193277,
|
| 2772 |
+
"grad_norm": 0.9755637485958417,
|
| 2773 |
+
"learning_rate": 2e-05,
|
| 2774 |
+
"loss": 0.6817933559417725,
|
| 2775 |
+
"step": 3950
|
| 2776 |
+
},
|
| 2777 |
+
{
|
| 2778 |
+
"epoch": 0.10857941926462121,
|
| 2779 |
+
"grad_norm": 0.9207426742604556,
|
| 2780 |
+
"learning_rate": 2e-05,
|
| 2781 |
+
"loss": 0.632993507385254,
|
| 2782 |
+
"step": 3960
|
| 2783 |
+
},
|
| 2784 |
+
{
|
| 2785 |
+
"epoch": 0.10885360971730965,
|
| 2786 |
+
"grad_norm": 1.1075412445990842,
|
| 2787 |
+
"learning_rate": 2e-05,
|
| 2788 |
+
"loss": 0.6479109764099121,
|
| 2789 |
+
"step": 3970
|
| 2790 |
+
},
|
| 2791 |
+
{
|
| 2792 |
+
"epoch": 0.10912780016999808,
|
| 2793 |
+
"grad_norm": 1.02734487686768,
|
| 2794 |
+
"learning_rate": 2e-05,
|
| 2795 |
+
"loss": 0.6399119853973388,
|
| 2796 |
+
"step": 3980
|
| 2797 |
+
},
|
| 2798 |
+
{
|
| 2799 |
+
"epoch": 0.10940199062268652,
|
| 2800 |
+
"grad_norm": 1.0062247863415956,
|
| 2801 |
+
"learning_rate": 2e-05,
|
| 2802 |
+
"loss": 0.6592807292938232,
|
| 2803 |
+
"step": 3990
|
| 2804 |
+
},
|
| 2805 |
+
{
|
| 2806 |
+
"epoch": 0.10967618107537495,
|
| 2807 |
+
"grad_norm": 1.7207229565430802,
|
| 2808 |
+
"learning_rate": 2e-05,
|
| 2809 |
+
"loss": 0.6665008544921875,
|
| 2810 |
+
"step": 4000
|
| 2811 |
+
},
|
| 2812 |
+
{
|
| 2813 |
+
"epoch": 0.10995037152806339,
|
| 2814 |
+
"grad_norm": 8.645042736815913,
|
| 2815 |
+
"learning_rate": 2e-05,
|
| 2816 |
+
"loss": 0.6616263866424561,
|
| 2817 |
+
"step": 4010
|
| 2818 |
+
},
|
| 2819 |
+
{
|
| 2820 |
+
"epoch": 0.11022456198075183,
|
| 2821 |
+
"grad_norm": 1.2238886957987656,
|
| 2822 |
+
"learning_rate": 2e-05,
|
| 2823 |
+
"loss": 0.6552308559417724,
|
| 2824 |
+
"step": 4020
|
| 2825 |
+
},
|
| 2826 |
+
{
|
| 2827 |
+
"epoch": 0.11049875243344026,
|
| 2828 |
+
"grad_norm": 1.1553638761775333,
|
| 2829 |
+
"learning_rate": 2e-05,
|
| 2830 |
+
"loss": 0.6816983699798584,
|
| 2831 |
+
"step": 4030
|
| 2832 |
+
},
|
| 2833 |
+
{
|
| 2834 |
+
"epoch": 0.11077294288612871,
|
| 2835 |
+
"grad_norm": 1.054398000948271,
|
| 2836 |
+
"learning_rate": 2e-05,
|
| 2837 |
+
"loss": 0.6408511161804199,
|
| 2838 |
+
"step": 4040
|
| 2839 |
+
},
|
| 2840 |
+
{
|
| 2841 |
+
"epoch": 0.11104713333881715,
|
| 2842 |
+
"grad_norm": 0.9428030212455841,
|
| 2843 |
+
"learning_rate": 2e-05,
|
| 2844 |
+
"loss": 0.6279231071472168,
|
| 2845 |
+
"step": 4050
|
| 2846 |
+
},
|
| 2847 |
+
{
|
| 2848 |
+
"epoch": 0.11132132379150558,
|
| 2849 |
+
"grad_norm": 1.388635722554096,
|
| 2850 |
+
"learning_rate": 2e-05,
|
| 2851 |
+
"loss": 0.6294130325317383,
|
| 2852 |
+
"step": 4060
|
| 2853 |
+
},
|
| 2854 |
+
{
|
| 2855 |
+
"epoch": 0.11159551424419402,
|
| 2856 |
+
"grad_norm": 1.7370583970603126,
|
| 2857 |
+
"learning_rate": 2e-05,
|
| 2858 |
+
"loss": 0.7167839050292969,
|
| 2859 |
+
"step": 4070
|
| 2860 |
+
}
|
| 2861 |
+
],
|
| 2862 |
+
"logging_steps": 10,
|
| 2863 |
+
"max_steps": 5935,
|
| 2864 |
+
"num_input_tokens_seen": 0,
|
| 2865 |
+
"num_train_epochs": 1,
|
| 2866 |
+
"save_steps": 2035,
|
| 2867 |
+
"stateful_callbacks": {
|
| 2868 |
+
"TrainerControl": {
|
| 2869 |
+
"args": {
|
| 2870 |
+
"should_epoch_stop": false,
|
| 2871 |
+
"should_evaluate": false,
|
| 2872 |
+
"should_log": false,
|
| 2873 |
+
"should_save": true,
|
| 2874 |
+
"should_training_stop": false
|
| 2875 |
+
},
|
| 2876 |
+
"attributes": {}
|
| 2877 |
+
}
|
| 2878 |
+
},
|
| 2879 |
+
"total_flos": 1927226987642880.0,
|
| 2880 |
+
"train_batch_size": 1,
|
| 2881 |
+
"trial_name": null,
|
| 2882 |
+
"trial_params": null
|
| 2883 |
+
}
|
400M/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:392ea69d123a584ac5b993c15219b31093e4d98dddc8f0c3122fc776ee8c9c55
|
| 3 |
+
size 7313
|