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 sha256:9cbb9512693efcaeedaa02410bc665816bc0533d3c9782fc4c80162f1ecddb98
|
| 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": 0.5239443872296602,
|
| 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": 0.0012873326467559218,
|
| 14 |
+
"grad_norm": 420.831391091495,
|
| 15 |
+
"learning_rate": 2.910112359550562e-06,
|
| 16 |
+
"loss": 1.9312837600708008,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.0025746652935118436,
|
| 21 |
+
"grad_norm": 183.74059849187165,
|
| 22 |
+
"learning_rate": 3.921348314606742e-06,
|
| 23 |
+
"loss": 1.0653209686279297,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.0038619979402677654,
|
| 28 |
+
"grad_norm": 209.9298344604747,
|
| 29 |
+
"learning_rate": 4.932584269662922e-06,
|
| 30 |
+
"loss": 0.9056184768676758,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.005149330587023687,
|
| 35 |
+
"grad_norm": 77.6500325582647,
|
| 36 |
+
"learning_rate": 5.943820224719102e-06,
|
| 37 |
+
"loss": 0.8960558891296386,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.006436663233779609,
|
| 42 |
+
"grad_norm": 284.47230377729784,
|
| 43 |
+
"learning_rate": 6.955056179775282e-06,
|
| 44 |
+
"loss": 0.8537825584411621,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.007723995880535531,
|
| 49 |
+
"grad_norm": 39.79412112786639,
|
| 50 |
+
"learning_rate": 7.966292134831462e-06,
|
| 51 |
+
"loss": 0.9032889366149902,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.009011328527291453,
|
| 56 |
+
"grad_norm": 19.11666869823677,
|
| 57 |
+
"learning_rate": 8.977528089887641e-06,
|
| 58 |
+
"loss": 0.8577792167663574,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.010298661174047374,
|
| 63 |
+
"grad_norm": 194.15492366269652,
|
| 64 |
+
"learning_rate": 9.988764044943822e-06,
|
| 65 |
+
"loss": 0.7952291011810303,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.011585993820803296,
|
| 70 |
+
"grad_norm": 69.50742426880747,
|
| 71 |
+
"learning_rate": 1.1000000000000001e-05,
|
| 72 |
+
"loss": 0.922086524963379,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.012873326467559218,
|
| 77 |
+
"grad_norm": 14.130976114030805,
|
| 78 |
+
"learning_rate": 1.2011235955056182e-05,
|
| 79 |
+
"loss": 0.8015565872192383,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.01416065911431514,
|
| 84 |
+
"grad_norm": 53.39139843875411,
|
| 85 |
+
"learning_rate": 1.3022471910112362e-05,
|
| 86 |
+
"loss": 0.798184585571289,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.015447991761071062,
|
| 91 |
+
"grad_norm": 9.335386256610237,
|
| 92 |
+
"learning_rate": 1.403370786516854e-05,
|
| 93 |
+
"loss": 0.8048606872558594,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.016735324407826983,
|
| 98 |
+
"grad_norm": 16.579058293598365,
|
| 99 |
+
"learning_rate": 1.504494382022472e-05,
|
| 100 |
+
"loss": 0.7458171844482422,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.018022657054582905,
|
| 105 |
+
"grad_norm": 10.216833945092391,
|
| 106 |
+
"learning_rate": 1.60561797752809e-05,
|
| 107 |
+
"loss": 0.7754960536956788,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.019309989701338827,
|
| 112 |
+
"grad_norm": 4.716209240987936,
|
| 113 |
+
"learning_rate": 1.706741573033708e-05,
|
| 114 |
+
"loss": 0.7410296440124512,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.02059732234809475,
|
| 119 |
+
"grad_norm": 7.373374743890979,
|
| 120 |
+
"learning_rate": 1.8078651685393256e-05,
|
| 121 |
+
"loss": 0.7318727493286132,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.02188465499485067,
|
| 126 |
+
"grad_norm": 2.832195397619456,
|
| 127 |
+
"learning_rate": 1.908988764044944e-05,
|
| 128 |
+
"loss": 0.7140597343444824,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.023171987641606592,
|
| 133 |
+
"grad_norm": 8.407785684478657,
|
| 134 |
+
"learning_rate": 2e-05,
|
| 135 |
+
"loss": 0.7257099151611328,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.024459320288362514,
|
| 140 |
+
"grad_norm": 68.6412229116751,
|
| 141 |
+
"learning_rate": 2e-05,
|
| 142 |
+
"loss": 0.7461828231811524,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.025746652935118436,
|
| 147 |
+
"grad_norm": 6.17895198608428,
|
| 148 |
+
"learning_rate": 2e-05,
|
| 149 |
+
"loss": 0.7821893215179443,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.027033985581874358,
|
| 154 |
+
"grad_norm": 5.194394848131495,
|
| 155 |
+
"learning_rate": 2e-05,
|
| 156 |
+
"loss": 0.7375617027282715,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.02832131822863028,
|
| 161 |
+
"grad_norm": 201.26051157890333,
|
| 162 |
+
"learning_rate": 2e-05,
|
| 163 |
+
"loss": 0.7686418056488037,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.0296086508753862,
|
| 168 |
+
"grad_norm": 28.711764224889286,
|
| 169 |
+
"learning_rate": 2e-05,
|
| 170 |
+
"loss": 0.7339415550231934,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.030895983522142123,
|
| 175 |
+
"grad_norm": 9.11168724332578,
|
| 176 |
+
"learning_rate": 2e-05,
|
| 177 |
+
"loss": 0.7917026519775391,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.032183316168898045,
|
| 182 |
+
"grad_norm": 18.075887930769106,
|
| 183 |
+
"learning_rate": 2e-05,
|
| 184 |
+
"loss": 0.7600293636322022,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.03347064881565397,
|
| 189 |
+
"grad_norm": 6.699833454866563,
|
| 190 |
+
"learning_rate": 2e-05,
|
| 191 |
+
"loss": 0.8107472419738769,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.03475798146240989,
|
| 196 |
+
"grad_norm": 14.699594175403195,
|
| 197 |
+
"learning_rate": 2e-05,
|
| 198 |
+
"loss": 0.8207590103149414,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.03604531410916581,
|
| 203 |
+
"grad_norm": 1.9113896642609345,
|
| 204 |
+
"learning_rate": 2e-05,
|
| 205 |
+
"loss": 0.7948224544525146,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.03733264675592173,
|
| 210 |
+
"grad_norm": 5.196177104534392,
|
| 211 |
+
"learning_rate": 2e-05,
|
| 212 |
+
"loss": 0.7511525630950928,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.038619979402677654,
|
| 217 |
+
"grad_norm": 6.297124236914922,
|
| 218 |
+
"learning_rate": 2e-05,
|
| 219 |
+
"loss": 0.7438372611999512,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.039907312049433576,
|
| 224 |
+
"grad_norm": 11.256600743830386,
|
| 225 |
+
"learning_rate": 2e-05,
|
| 226 |
+
"loss": 0.7340342998504639,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.0411946446961895,
|
| 231 |
+
"grad_norm": 12.03605841268819,
|
| 232 |
+
"learning_rate": 2e-05,
|
| 233 |
+
"loss": 0.7668326377868653,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.04248197734294542,
|
| 238 |
+
"grad_norm": 10.454100982721279,
|
| 239 |
+
"learning_rate": 2e-05,
|
| 240 |
+
"loss": 0.744885778427124,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.04376930998970134,
|
| 245 |
+
"grad_norm": 31.42354342277286,
|
| 246 |
+
"learning_rate": 2e-05,
|
| 247 |
+
"loss": 0.7327960968017578,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.04505664263645726,
|
| 252 |
+
"grad_norm": 8.65794301406385,
|
| 253 |
+
"learning_rate": 2e-05,
|
| 254 |
+
"loss": 0.8318517684936524,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.046343975283213185,
|
| 259 |
+
"grad_norm": 64.03382663360811,
|
| 260 |
+
"learning_rate": 2e-05,
|
| 261 |
+
"loss": 0.7727777004241944,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.047631307929969106,
|
| 266 |
+
"grad_norm": 5.767854062175975,
|
| 267 |
+
"learning_rate": 2e-05,
|
| 268 |
+
"loss": 0.7937121868133545,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.04891864057672503,
|
| 273 |
+
"grad_norm": 34.86438082009496,
|
| 274 |
+
"learning_rate": 2e-05,
|
| 275 |
+
"loss": 0.8124513626098633,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.05020597322348095,
|
| 280 |
+
"grad_norm": 17.861985503962952,
|
| 281 |
+
"learning_rate": 2e-05,
|
| 282 |
+
"loss": 0.782343578338623,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.05149330587023687,
|
| 287 |
+
"grad_norm": 5.273597714761184,
|
| 288 |
+
"learning_rate": 2e-05,
|
| 289 |
+
"loss": 0.7975768089294434,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.052780638516992794,
|
| 294 |
+
"grad_norm": 3.785340696847242,
|
| 295 |
+
"learning_rate": 2e-05,
|
| 296 |
+
"loss": 0.7699785232543945,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.054067971163748715,
|
| 301 |
+
"grad_norm": 2.0191165245283518,
|
| 302 |
+
"learning_rate": 2e-05,
|
| 303 |
+
"loss": 0.7567674160003662,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.05535530381050464,
|
| 308 |
+
"grad_norm": 2.684997153366744,
|
| 309 |
+
"learning_rate": 2e-05,
|
| 310 |
+
"loss": 0.8061090469360351,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.05664263645726056,
|
| 315 |
+
"grad_norm": 4.464002674035226,
|
| 316 |
+
"learning_rate": 2e-05,
|
| 317 |
+
"loss": 0.7761743068695068,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.05792996910401648,
|
| 322 |
+
"grad_norm": 2.113290503315734,
|
| 323 |
+
"learning_rate": 2e-05,
|
| 324 |
+
"loss": 0.7724208354949951,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.0592173017507724,
|
| 329 |
+
"grad_norm": 4.63126727046047,
|
| 330 |
+
"learning_rate": 2e-05,
|
| 331 |
+
"loss": 0.7622805595397949,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.060504634397528324,
|
| 336 |
+
"grad_norm": 9.246708733487706,
|
| 337 |
+
"learning_rate": 2e-05,
|
| 338 |
+
"loss": 0.6900830268859863,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.061791967044284246,
|
| 343 |
+
"grad_norm": 5.015943304904472,
|
| 344 |
+
"learning_rate": 2e-05,
|
| 345 |
+
"loss": 0.7525927543640136,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.06307929969104016,
|
| 350 |
+
"grad_norm": 7.279183898929732,
|
| 351 |
+
"learning_rate": 2e-05,
|
| 352 |
+
"loss": 0.7771492004394531,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.06436663233779609,
|
| 357 |
+
"grad_norm": 5.373066892709014,
|
| 358 |
+
"learning_rate": 2e-05,
|
| 359 |
+
"loss": 0.8580041885375976,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.065653964984552,
|
| 364 |
+
"grad_norm": 15.27286294173738,
|
| 365 |
+
"learning_rate": 2e-05,
|
| 366 |
+
"loss": 0.8757843017578125,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.06694129763130793,
|
| 371 |
+
"grad_norm": 9.932666259805304,
|
| 372 |
+
"learning_rate": 2e-05,
|
| 373 |
+
"loss": 0.8020495414733887,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.06822863027806385,
|
| 378 |
+
"grad_norm": 3.301286188883536,
|
| 379 |
+
"learning_rate": 2e-05,
|
| 380 |
+
"loss": 0.7554676055908203,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.06951596292481978,
|
| 385 |
+
"grad_norm": 5.387035655614053,
|
| 386 |
+
"learning_rate": 2e-05,
|
| 387 |
+
"loss": 0.748227596282959,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.07080329557157569,
|
| 392 |
+
"grad_norm": 51.37416211353129,
|
| 393 |
+
"learning_rate": 2e-05,
|
| 394 |
+
"loss": 0.7636474609375,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.07209062821833162,
|
| 399 |
+
"grad_norm": 33.339739170373484,
|
| 400 |
+
"learning_rate": 2e-05,
|
| 401 |
+
"loss": 0.7783924102783203,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.07337796086508754,
|
| 406 |
+
"grad_norm": 5.732333781925769,
|
| 407 |
+
"learning_rate": 2e-05,
|
| 408 |
+
"loss": 0.793330717086792,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.07466529351184346,
|
| 413 |
+
"grad_norm": 4.100917123304358,
|
| 414 |
+
"learning_rate": 2e-05,
|
| 415 |
+
"loss": 0.7694579601287842,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.07595262615859938,
|
| 420 |
+
"grad_norm": 2.0715078041261332,
|
| 421 |
+
"learning_rate": 2e-05,
|
| 422 |
+
"loss": 0.8294933319091797,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.07723995880535531,
|
| 427 |
+
"grad_norm": 3.184866964539304,
|
| 428 |
+
"learning_rate": 2e-05,
|
| 429 |
+
"loss": 0.7459193706512451,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.07852729145211122,
|
| 434 |
+
"grad_norm": 7.478907825414544,
|
| 435 |
+
"learning_rate": 2e-05,
|
| 436 |
+
"loss": 0.7193670272827148,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.07981462409886715,
|
| 441 |
+
"grad_norm": 8.452718548100744,
|
| 442 |
+
"learning_rate": 2e-05,
|
| 443 |
+
"loss": 0.8021256446838378,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.08110195674562307,
|
| 448 |
+
"grad_norm": 5.097868976449567,
|
| 449 |
+
"learning_rate": 2e-05,
|
| 450 |
+
"loss": 0.7970829963684082,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.082389289392379,
|
| 455 |
+
"grad_norm": 2.4901894680955787,
|
| 456 |
+
"learning_rate": 2e-05,
|
| 457 |
+
"loss": 0.7695016384124755,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.08367662203913491,
|
| 462 |
+
"grad_norm": 6.436792505064742,
|
| 463 |
+
"learning_rate": 2e-05,
|
| 464 |
+
"loss": 0.7511703491210937,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.08496395468589084,
|
| 469 |
+
"grad_norm": 5.76943441019596,
|
| 470 |
+
"learning_rate": 2e-05,
|
| 471 |
+
"loss": 0.7517711639404296,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.08625128733264675,
|
| 476 |
+
"grad_norm": 1.8786628503818215,
|
| 477 |
+
"learning_rate": 2e-05,
|
| 478 |
+
"loss": 0.7410395622253418,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.08753861997940268,
|
| 483 |
+
"grad_norm": 4.587266282984627,
|
| 484 |
+
"learning_rate": 2e-05,
|
| 485 |
+
"loss": 0.771084976196289,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.0888259526261586,
|
| 490 |
+
"grad_norm": 13.971035209507322,
|
| 491 |
+
"learning_rate": 2e-05,
|
| 492 |
+
"loss": 0.7819516658782959,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.09011328527291453,
|
| 497 |
+
"grad_norm": 8.1570226178038,
|
| 498 |
+
"learning_rate": 2e-05,
|
| 499 |
+
"loss": 0.7403954982757568,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.09140061791967044,
|
| 504 |
+
"grad_norm": 4.688221044744754,
|
| 505 |
+
"learning_rate": 2e-05,
|
| 506 |
+
"loss": 0.786048412322998,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.09268795056642637,
|
| 511 |
+
"grad_norm": 2.778648354062325,
|
| 512 |
+
"learning_rate": 2e-05,
|
| 513 |
+
"loss": 0.7478150367736817,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.09397528321318228,
|
| 518 |
+
"grad_norm": 3.652349932924877,
|
| 519 |
+
"learning_rate": 2e-05,
|
| 520 |
+
"loss": 0.7311888694763183,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.09526261585993821,
|
| 525 |
+
"grad_norm": 5.5932577832394,
|
| 526 |
+
"learning_rate": 2e-05,
|
| 527 |
+
"loss": 0.7629296779632568,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.09654994850669413,
|
| 532 |
+
"grad_norm": 1.535821909119182,
|
| 533 |
+
"learning_rate": 2e-05,
|
| 534 |
+
"loss": 0.7407299041748047,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.09783728115345006,
|
| 539 |
+
"grad_norm": 1.45905132393004,
|
| 540 |
+
"learning_rate": 2e-05,
|
| 541 |
+
"loss": 0.7430490016937256,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.09912461380020597,
|
| 546 |
+
"grad_norm": 4.619712730990858,
|
| 547 |
+
"learning_rate": 2e-05,
|
| 548 |
+
"loss": 0.7504862785339356,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.1004119464469619,
|
| 553 |
+
"grad_norm": 3.9224749754444566,
|
| 554 |
+
"learning_rate": 2e-05,
|
| 555 |
+
"loss": 0.7686273574829101,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.10169927909371781,
|
| 560 |
+
"grad_norm": 6.098937347148086,
|
| 561 |
+
"learning_rate": 2e-05,
|
| 562 |
+
"loss": 0.7319077491760254,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.10298661174047374,
|
| 567 |
+
"grad_norm": 2.4496087544978336,
|
| 568 |
+
"learning_rate": 2e-05,
|
| 569 |
+
"loss": 0.7697375297546387,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.10427394438722966,
|
| 574 |
+
"grad_norm": 2.8551662537697102,
|
| 575 |
+
"learning_rate": 2e-05,
|
| 576 |
+
"loss": 0.742626953125,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.10556127703398559,
|
| 581 |
+
"grad_norm": 3.205355126068373,
|
| 582 |
+
"learning_rate": 2e-05,
|
| 583 |
+
"loss": 0.7027194023132324,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.1068486096807415,
|
| 588 |
+
"grad_norm": 2.253436553246574,
|
| 589 |
+
"learning_rate": 2e-05,
|
| 590 |
+
"loss": 0.7767892360687256,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.10813594232749743,
|
| 595 |
+
"grad_norm": 4.087718084462775,
|
| 596 |
+
"learning_rate": 2e-05,
|
| 597 |
+
"loss": 0.7374720573425293,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.10942327497425335,
|
| 602 |
+
"grad_norm": 4.050374141792059,
|
| 603 |
+
"learning_rate": 2e-05,
|
| 604 |
+
"loss": 0.7483660697937011,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.11071060762100927,
|
| 609 |
+
"grad_norm": 2.239445345441048,
|
| 610 |
+
"learning_rate": 2e-05,
|
| 611 |
+
"loss": 0.7472774505615234,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.11199794026776519,
|
| 616 |
+
"grad_norm": 3.9463791962767836,
|
| 617 |
+
"learning_rate": 2e-05,
|
| 618 |
+
"loss": 0.7285970687866211,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.11328527291452112,
|
| 623 |
+
"grad_norm": 12.421260769301227,
|
| 624 |
+
"learning_rate": 2e-05,
|
| 625 |
+
"loss": 0.7539933681488037,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.11457260556127703,
|
| 630 |
+
"grad_norm": 4.885002632304067,
|
| 631 |
+
"learning_rate": 2e-05,
|
| 632 |
+
"loss": 0.7439227104187012,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.11585993820803296,
|
| 637 |
+
"grad_norm": 1.5566251726058724,
|
| 638 |
+
"learning_rate": 2e-05,
|
| 639 |
+
"loss": 0.735724401473999,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.11714727085478888,
|
| 644 |
+
"grad_norm": 2.865049116950206,
|
| 645 |
+
"learning_rate": 2e-05,
|
| 646 |
+
"loss": 0.7331388473510743,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.1184346035015448,
|
| 651 |
+
"grad_norm": 2.960680025346854,
|
| 652 |
+
"learning_rate": 2e-05,
|
| 653 |
+
"loss": 0.74105224609375,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.11972193614830072,
|
| 658 |
+
"grad_norm": 2.0043561901231413,
|
| 659 |
+
"learning_rate": 2e-05,
|
| 660 |
+
"loss": 0.6742991447448731,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.12100926879505665,
|
| 665 |
+
"grad_norm": 1.6534160074854731,
|
| 666 |
+
"learning_rate": 2e-05,
|
| 667 |
+
"loss": 0.7274767875671386,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.12229660144181256,
|
| 672 |
+
"grad_norm": 3.0327709147346353,
|
| 673 |
+
"learning_rate": 2e-05,
|
| 674 |
+
"loss": 0.7146361351013184,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.12358393408856849,
|
| 679 |
+
"grad_norm": 3.9665339761676237,
|
| 680 |
+
"learning_rate": 2e-05,
|
| 681 |
+
"loss": 0.7229518413543701,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.12487126673532441,
|
| 686 |
+
"grad_norm": 3.8479184119373193,
|
| 687 |
+
"learning_rate": 2e-05,
|
| 688 |
+
"loss": 0.7719802856445312,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.12615859938208032,
|
| 693 |
+
"grad_norm": 1.2184604438873134,
|
| 694 |
+
"learning_rate": 2e-05,
|
| 695 |
+
"loss": 0.7337092876434326,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.12744593202883625,
|
| 700 |
+
"grad_norm": 5.350440266680789,
|
| 701 |
+
"learning_rate": 2e-05,
|
| 702 |
+
"loss": 0.7010088920593261,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.12873326467559218,
|
| 707 |
+
"grad_norm": 2.7811359489299248,
|
| 708 |
+
"learning_rate": 2e-05,
|
| 709 |
+
"loss": 0.7292609691619873,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.1300205973223481,
|
| 714 |
+
"grad_norm": 17.912803780366907,
|
| 715 |
+
"learning_rate": 2e-05,
|
| 716 |
+
"loss": 0.7163818359375,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.131307929969104,
|
| 721 |
+
"grad_norm": 2.11582488712836,
|
| 722 |
+
"learning_rate": 2e-05,
|
| 723 |
+
"loss": 0.7069422721862793,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.13259526261585994,
|
| 728 |
+
"grad_norm": 11.247133917028588,
|
| 729 |
+
"learning_rate": 2e-05,
|
| 730 |
+
"loss": 0.7331721782684326,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.13388259526261587,
|
| 735 |
+
"grad_norm": 2.7388313221050575,
|
| 736 |
+
"learning_rate": 2e-05,
|
| 737 |
+
"loss": 0.6841031551361084,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.1351699279093718,
|
| 742 |
+
"grad_norm": 1.207061425581625,
|
| 743 |
+
"learning_rate": 2e-05,
|
| 744 |
+
"loss": 0.733606481552124,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.1364572605561277,
|
| 749 |
+
"grad_norm": 47.69246880805316,
|
| 750 |
+
"learning_rate": 2e-05,
|
| 751 |
+
"loss": 0.7302389144897461,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.13774459320288363,
|
| 756 |
+
"grad_norm": 2.171556278425703,
|
| 757 |
+
"learning_rate": 2e-05,
|
| 758 |
+
"loss": 0.7309787750244141,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.13903192584963955,
|
| 763 |
+
"grad_norm": 3.360889748897153,
|
| 764 |
+
"learning_rate": 2e-05,
|
| 765 |
+
"loss": 0.7237205028533935,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.14031925849639545,
|
| 770 |
+
"grad_norm": 1.5969376240757103,
|
| 771 |
+
"learning_rate": 2e-05,
|
| 772 |
+
"loss": 0.7478039741516114,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.14160659114315138,
|
| 777 |
+
"grad_norm": 1.9534203833516763,
|
| 778 |
+
"learning_rate": 2e-05,
|
| 779 |
+
"loss": 0.7363515377044678,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.1428939237899073,
|
| 784 |
+
"grad_norm": 2.6654514628849952,
|
| 785 |
+
"learning_rate": 2e-05,
|
| 786 |
+
"loss": 0.7492703437805176,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.14418125643666324,
|
| 791 |
+
"grad_norm": 1.68379520116963,
|
| 792 |
+
"learning_rate": 2e-05,
|
| 793 |
+
"loss": 0.7549358367919922,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.14546858908341914,
|
| 798 |
+
"grad_norm": 1.632554767797848,
|
| 799 |
+
"learning_rate": 2e-05,
|
| 800 |
+
"loss": 0.6897923946380615,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.14675592173017507,
|
| 805 |
+
"grad_norm": 1.653687869959576,
|
| 806 |
+
"learning_rate": 2e-05,
|
| 807 |
+
"loss": 0.7304753303527832,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.148043254376931,
|
| 812 |
+
"grad_norm": 1.6347977972264258,
|
| 813 |
+
"learning_rate": 2e-05,
|
| 814 |
+
"loss": 0.7604644775390625,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.14933058702368693,
|
| 819 |
+
"grad_norm": 1.8718744344050506,
|
| 820 |
+
"learning_rate": 2e-05,
|
| 821 |
+
"loss": 0.697459602355957,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.15061791967044283,
|
| 826 |
+
"grad_norm": 2.532644944339293,
|
| 827 |
+
"learning_rate": 2e-05,
|
| 828 |
+
"loss": 0.7188455104827881,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.15190525231719876,
|
| 833 |
+
"grad_norm": 1.806201422331298,
|
| 834 |
+
"learning_rate": 2e-05,
|
| 835 |
+
"loss": 0.7081408977508545,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.1531925849639547,
|
| 840 |
+
"grad_norm": 1.6489310799297887,
|
| 841 |
+
"learning_rate": 2e-05,
|
| 842 |
+
"loss": 0.7228747367858886,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.15447991761071062,
|
| 847 |
+
"grad_norm": 2.9974385822785408,
|
| 848 |
+
"learning_rate": 2e-05,
|
| 849 |
+
"loss": 0.7244661331176758,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.15576725025746652,
|
| 854 |
+
"grad_norm": 22.594724424794563,
|
| 855 |
+
"learning_rate": 2e-05,
|
| 856 |
+
"loss": 0.7405924797058105,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.15705458290422245,
|
| 861 |
+
"grad_norm": 3.7604685572670684,
|
| 862 |
+
"learning_rate": 2e-05,
|
| 863 |
+
"loss": 0.7070909023284913,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.15834191555097837,
|
| 868 |
+
"grad_norm": 3.3763336904748615,
|
| 869 |
+
"learning_rate": 2e-05,
|
| 870 |
+
"loss": 0.7894343376159668,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.1596292481977343,
|
| 875 |
+
"grad_norm": 10.941310418547834,
|
| 876 |
+
"learning_rate": 2e-05,
|
| 877 |
+
"loss": 0.7505155086517334,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.1609165808444902,
|
| 882 |
+
"grad_norm": 4.887708688260815,
|
| 883 |
+
"learning_rate": 2e-05,
|
| 884 |
+
"loss": 0.7505541801452636,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.16220391349124613,
|
| 889 |
+
"grad_norm": 6.54025957968161,
|
| 890 |
+
"learning_rate": 2e-05,
|
| 891 |
+
"loss": 0.716888427734375,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.16349124613800206,
|
| 896 |
+
"grad_norm": 3.6877006120763585,
|
| 897 |
+
"learning_rate": 2e-05,
|
| 898 |
+
"loss": 0.7247855186462402,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.164778578784758,
|
| 903 |
+
"grad_norm": 15.044650338028674,
|
| 904 |
+
"learning_rate": 2e-05,
|
| 905 |
+
"loss": 0.7331586360931397,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.1660659114315139,
|
| 910 |
+
"grad_norm": 2.538457484617354,
|
| 911 |
+
"learning_rate": 2e-05,
|
| 912 |
+
"loss": 0.7180121421813965,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.16735324407826982,
|
| 917 |
+
"grad_norm": 1.6984019588574841,
|
| 918 |
+
"learning_rate": 2e-05,
|
| 919 |
+
"loss": 0.7547801494598388,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.16864057672502575,
|
| 924 |
+
"grad_norm": 4.940123866109371,
|
| 925 |
+
"learning_rate": 2e-05,
|
| 926 |
+
"loss": 0.7324605464935303,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.16992790937178168,
|
| 931 |
+
"grad_norm": 1.4925558541074413,
|
| 932 |
+
"learning_rate": 2e-05,
|
| 933 |
+
"loss": 0.7033469676971436,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.17121524201853758,
|
| 938 |
+
"grad_norm": 2.0954880927362307,
|
| 939 |
+
"learning_rate": 2e-05,
|
| 940 |
+
"loss": 0.7216341972351075,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.1725025746652935,
|
| 945 |
+
"grad_norm": 3.965453670040223,
|
| 946 |
+
"learning_rate": 2e-05,
|
| 947 |
+
"loss": 0.7113677501678467,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.17378990731204944,
|
| 952 |
+
"grad_norm": 2.62486775378727,
|
| 953 |
+
"learning_rate": 2e-05,
|
| 954 |
+
"loss": 0.6789872646331787,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.17507723995880536,
|
| 959 |
+
"grad_norm": 1.6392487264123656,
|
| 960 |
+
"learning_rate": 2e-05,
|
| 961 |
+
"loss": 0.6960799217224121,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.17636457260556127,
|
| 966 |
+
"grad_norm": 1.6246985763924315,
|
| 967 |
+
"learning_rate": 2e-05,
|
| 968 |
+
"loss": 0.7166689872741699,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.1776519052523172,
|
| 973 |
+
"grad_norm": 3.586582883055029,
|
| 974 |
+
"learning_rate": 2e-05,
|
| 975 |
+
"loss": 0.7102786064147949,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.17893923789907312,
|
| 980 |
+
"grad_norm": 7.353760235895877,
|
| 981 |
+
"learning_rate": 2e-05,
|
| 982 |
+
"loss": 0.7114831447601319,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.18022657054582905,
|
| 987 |
+
"grad_norm": 3.600918049571546,
|
| 988 |
+
"learning_rate": 2e-05,
|
| 989 |
+
"loss": 0.722894287109375,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.18151390319258495,
|
| 994 |
+
"grad_norm": 6.977869760230939,
|
| 995 |
+
"learning_rate": 2e-05,
|
| 996 |
+
"loss": 0.7068233013153076,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.18280123583934088,
|
| 1001 |
+
"grad_norm": 2.6291241838271486,
|
| 1002 |
+
"learning_rate": 2e-05,
|
| 1003 |
+
"loss": 0.7131928920745849,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.1840885684860968,
|
| 1008 |
+
"grad_norm": 2.47189101333238,
|
| 1009 |
+
"learning_rate": 2e-05,
|
| 1010 |
+
"loss": 0.7333631038665771,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.18537590113285274,
|
| 1015 |
+
"grad_norm": 2.5920454659070438,
|
| 1016 |
+
"learning_rate": 2e-05,
|
| 1017 |
+
"loss": 0.6987507820129395,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.18666323377960864,
|
| 1022 |
+
"grad_norm": 1.840628594046838,
|
| 1023 |
+
"learning_rate": 2e-05,
|
| 1024 |
+
"loss": 0.7048477172851563,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.18795056642636457,
|
| 1029 |
+
"grad_norm": 2.3416098111869093,
|
| 1030 |
+
"learning_rate": 2e-05,
|
| 1031 |
+
"loss": 0.7118321895599365,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.1892378990731205,
|
| 1036 |
+
"grad_norm": 1.530570157282232,
|
| 1037 |
+
"learning_rate": 2e-05,
|
| 1038 |
+
"loss": 0.7082484245300293,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.19052523171987643,
|
| 1043 |
+
"grad_norm": 2.2055714727842477,
|
| 1044 |
+
"learning_rate": 2e-05,
|
| 1045 |
+
"loss": 0.7013579368591308,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.19181256436663233,
|
| 1050 |
+
"grad_norm": 1.2321264302309138,
|
| 1051 |
+
"learning_rate": 2e-05,
|
| 1052 |
+
"loss": 0.720965576171875,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.19309989701338826,
|
| 1057 |
+
"grad_norm": 2.7847826433903067,
|
| 1058 |
+
"learning_rate": 2e-05,
|
| 1059 |
+
"loss": 0.7270987510681153,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.19438722966014418,
|
| 1064 |
+
"grad_norm": 1.8686276102701087,
|
| 1065 |
+
"learning_rate": 2e-05,
|
| 1066 |
+
"loss": 0.7201802730560303,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.1956745623069001,
|
| 1071 |
+
"grad_norm": 1.569912150907071,
|
| 1072 |
+
"learning_rate": 2e-05,
|
| 1073 |
+
"loss": 0.7226099967956543,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.196961894953656,
|
| 1078 |
+
"grad_norm": 2.283386362656244,
|
| 1079 |
+
"learning_rate": 2e-05,
|
| 1080 |
+
"loss": 0.7405374050140381,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.19824922760041194,
|
| 1085 |
+
"grad_norm": 1.5156912593840797,
|
| 1086 |
+
"learning_rate": 2e-05,
|
| 1087 |
+
"loss": 0.7121532440185547,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.19953656024716787,
|
| 1092 |
+
"grad_norm": 1.4580219323036991,
|
| 1093 |
+
"learning_rate": 2e-05,
|
| 1094 |
+
"loss": 0.6971864223480224,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.2008238928939238,
|
| 1099 |
+
"grad_norm": 5.82821608514732,
|
| 1100 |
+
"learning_rate": 2e-05,
|
| 1101 |
+
"loss": 0.7024627685546875,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.2021112255406797,
|
| 1106 |
+
"grad_norm": 1.1362087794702644,
|
| 1107 |
+
"learning_rate": 2e-05,
|
| 1108 |
+
"loss": 0.6774281978607177,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.20339855818743563,
|
| 1113 |
+
"grad_norm": 2.443427160363054,
|
| 1114 |
+
"learning_rate": 2e-05,
|
| 1115 |
+
"loss": 0.6777324676513672,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.20468589083419156,
|
| 1120 |
+
"grad_norm": 3.4197112552113254,
|
| 1121 |
+
"learning_rate": 2e-05,
|
| 1122 |
+
"loss": 0.6910956859588623,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.2059732234809475,
|
| 1127 |
+
"grad_norm": 2.7142923935215904,
|
| 1128 |
+
"learning_rate": 2e-05,
|
| 1129 |
+
"loss": 0.7175305366516114,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.2072605561277034,
|
| 1134 |
+
"grad_norm": 2.9446280303316854,
|
| 1135 |
+
"learning_rate": 2e-05,
|
| 1136 |
+
"loss": 0.6894670009613038,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.20854788877445932,
|
| 1141 |
+
"grad_norm": 1.369637715767581,
|
| 1142 |
+
"learning_rate": 2e-05,
|
| 1143 |
+
"loss": 0.6922161102294921,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.20983522142121525,
|
| 1148 |
+
"grad_norm": 1.3879222035136711,
|
| 1149 |
+
"learning_rate": 2e-05,
|
| 1150 |
+
"loss": 0.6919208526611328,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.21112255406797117,
|
| 1155 |
+
"grad_norm": 3.065083029358514,
|
| 1156 |
+
"learning_rate": 2e-05,
|
| 1157 |
+
"loss": 0.6886746883392334,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.21240988671472708,
|
| 1162 |
+
"grad_norm": 10.325097544156115,
|
| 1163 |
+
"learning_rate": 2e-05,
|
| 1164 |
+
"loss": 0.6938620567321777,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.213697219361483,
|
| 1169 |
+
"grad_norm": 1.6072671865699064,
|
| 1170 |
+
"learning_rate": 2e-05,
|
| 1171 |
+
"loss": 0.7277378082275391,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.21498455200823893,
|
| 1176 |
+
"grad_norm": 1.3380594408338742,
|
| 1177 |
+
"learning_rate": 2e-05,
|
| 1178 |
+
"loss": 0.7332149505615234,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.21627188465499486,
|
| 1183 |
+
"grad_norm": 1.5454219317409892,
|
| 1184 |
+
"learning_rate": 2e-05,
|
| 1185 |
+
"loss": 0.7071063995361329,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.21755921730175076,
|
| 1190 |
+
"grad_norm": 1.8153666337523178,
|
| 1191 |
+
"learning_rate": 2e-05,
|
| 1192 |
+
"loss": 0.7288019180297851,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.2188465499485067,
|
| 1197 |
+
"grad_norm": 1.877709344780682,
|
| 1198 |
+
"learning_rate": 2e-05,
|
| 1199 |
+
"loss": 0.7265975475311279,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.22013388259526262,
|
| 1204 |
+
"grad_norm": 4.043860838359305,
|
| 1205 |
+
"learning_rate": 2e-05,
|
| 1206 |
+
"loss": 0.7083337783813477,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.22142121524201855,
|
| 1211 |
+
"grad_norm": 4.246037776394193,
|
| 1212 |
+
"learning_rate": 2e-05,
|
| 1213 |
+
"loss": 0.678376293182373,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.22270854788877445,
|
| 1218 |
+
"grad_norm": 1.4558126421583217,
|
| 1219 |
+
"learning_rate": 2e-05,
|
| 1220 |
+
"loss": 0.7107050895690918,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.22399588053553038,
|
| 1225 |
+
"grad_norm": 2.1461600575963056,
|
| 1226 |
+
"learning_rate": 2e-05,
|
| 1227 |
+
"loss": 0.6994074821472168,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.2252832131822863,
|
| 1232 |
+
"grad_norm": 2.138539490137605,
|
| 1233 |
+
"learning_rate": 2e-05,
|
| 1234 |
+
"loss": 0.6766193389892579,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.22657054582904224,
|
| 1239 |
+
"grad_norm": 1.230257735080685,
|
| 1240 |
+
"learning_rate": 2e-05,
|
| 1241 |
+
"loss": 0.6743141174316406,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.22785787847579814,
|
| 1246 |
+
"grad_norm": 5.529787951162952,
|
| 1247 |
+
"learning_rate": 2e-05,
|
| 1248 |
+
"loss": 0.6881375789642334,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.22914521112255407,
|
| 1253 |
+
"grad_norm": 1.6357599532681988,
|
| 1254 |
+
"learning_rate": 2e-05,
|
| 1255 |
+
"loss": 0.6633195877075195,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.23043254376931,
|
| 1260 |
+
"grad_norm": 6.229625650998896,
|
| 1261 |
+
"learning_rate": 2e-05,
|
| 1262 |
+
"loss": 0.6983210563659668,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.23171987641606592,
|
| 1267 |
+
"grad_norm": 1.3301077031080362,
|
| 1268 |
+
"learning_rate": 2e-05,
|
| 1269 |
+
"loss": 0.7201817512512207,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.23300720906282182,
|
| 1274 |
+
"grad_norm": 3.39358729212885,
|
| 1275 |
+
"learning_rate": 2e-05,
|
| 1276 |
+
"loss": 0.6951064109802246,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.23429454170957775,
|
| 1281 |
+
"grad_norm": 4.181323376087317,
|
| 1282 |
+
"learning_rate": 2e-05,
|
| 1283 |
+
"loss": 0.6929269790649414,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.23558187435633368,
|
| 1288 |
+
"grad_norm": 1.5545245966089207,
|
| 1289 |
+
"learning_rate": 2e-05,
|
| 1290 |
+
"loss": 0.6981029033660888,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.2368692070030896,
|
| 1295 |
+
"grad_norm": 2.1015275437163385,
|
| 1296 |
+
"learning_rate": 2e-05,
|
| 1297 |
+
"loss": 0.6695489883422852,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.2381565396498455,
|
| 1302 |
+
"grad_norm": 1.8856355185752995,
|
| 1303 |
+
"learning_rate": 2e-05,
|
| 1304 |
+
"loss": 0.7271100521087647,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.23944387229660144,
|
| 1309 |
+
"grad_norm": 3.425995350551422,
|
| 1310 |
+
"learning_rate": 2e-05,
|
| 1311 |
+
"loss": 0.6916310787200928,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.24073120494335737,
|
| 1316 |
+
"grad_norm": 1.2448080896450548,
|
| 1317 |
+
"learning_rate": 2e-05,
|
| 1318 |
+
"loss": 0.705747127532959,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.2420185375901133,
|
| 1323 |
+
"grad_norm": 6.4595134842908495,
|
| 1324 |
+
"learning_rate": 2e-05,
|
| 1325 |
+
"loss": 0.6798825740814209,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.2433058702368692,
|
| 1330 |
+
"grad_norm": 1.2226636920681981,
|
| 1331 |
+
"learning_rate": 2e-05,
|
| 1332 |
+
"loss": 0.7015285015106201,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.24459320288362513,
|
| 1337 |
+
"grad_norm": 2.3842419146911635,
|
| 1338 |
+
"learning_rate": 2e-05,
|
| 1339 |
+
"loss": 0.7172482490539551,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.24588053553038106,
|
| 1344 |
+
"grad_norm": 1.7333858862906162,
|
| 1345 |
+
"learning_rate": 2e-05,
|
| 1346 |
+
"loss": 0.6919423580169678,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.24716786817713698,
|
| 1351 |
+
"grad_norm": 1.2738406384985739,
|
| 1352 |
+
"learning_rate": 2e-05,
|
| 1353 |
+
"loss": 0.7124980926513672,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.24845520082389289,
|
| 1358 |
+
"grad_norm": 3.089273068519385,
|
| 1359 |
+
"learning_rate": 2e-05,
|
| 1360 |
+
"loss": 0.7129861354827881,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.24974253347064881,
|
| 1365 |
+
"grad_norm": 5.4254882757315395,
|
| 1366 |
+
"learning_rate": 2e-05,
|
| 1367 |
+
"loss": 0.7517825126647949,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.25102986611740474,
|
| 1372 |
+
"grad_norm": 1.71140671632292,
|
| 1373 |
+
"learning_rate": 2e-05,
|
| 1374 |
+
"loss": 0.7149953365325927,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.25231719876416064,
|
| 1379 |
+
"grad_norm": 2.7193201670484886,
|
| 1380 |
+
"learning_rate": 2e-05,
|
| 1381 |
+
"loss": 0.7189720153808594,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.2536045314109166,
|
| 1386 |
+
"grad_norm": 10.070508073263046,
|
| 1387 |
+
"learning_rate": 2e-05,
|
| 1388 |
+
"loss": 0.6905816555023193,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.2548918640576725,
|
| 1393 |
+
"grad_norm": 8.024891526041356,
|
| 1394 |
+
"learning_rate": 2e-05,
|
| 1395 |
+
"loss": 0.6899857521057129,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.2561791967044284,
|
| 1400 |
+
"grad_norm": 14.975571956672278,
|
| 1401 |
+
"learning_rate": 2e-05,
|
| 1402 |
+
"loss": 0.7188075065612793,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.25746652935118436,
|
| 1407 |
+
"grad_norm": 2.4641111805113103,
|
| 1408 |
+
"learning_rate": 2e-05,
|
| 1409 |
+
"loss": 0.7120208263397216,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.25875386199794026,
|
| 1414 |
+
"grad_norm": 2.7546066287502833,
|
| 1415 |
+
"learning_rate": 2e-05,
|
| 1416 |
+
"loss": 0.6824542999267578,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.2600411946446962,
|
| 1421 |
+
"grad_norm": 2.180296993017307,
|
| 1422 |
+
"learning_rate": 2e-05,
|
| 1423 |
+
"loss": 0.7345854759216308,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.2613285272914521,
|
| 1428 |
+
"grad_norm": 3.1963581482749137,
|
| 1429 |
+
"learning_rate": 2e-05,
|
| 1430 |
+
"loss": 0.689295482635498,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.262615859938208,
|
| 1435 |
+
"grad_norm": 3.340370740500395,
|
| 1436 |
+
"learning_rate": 2e-05,
|
| 1437 |
+
"loss": 0.6768050193786621,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.263903192584964,
|
| 1442 |
+
"grad_norm": 2.7181711140892375,
|
| 1443 |
+
"learning_rate": 2e-05,
|
| 1444 |
+
"loss": 0.755255126953125,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.2651905252317199,
|
| 1449 |
+
"grad_norm": 5.156287372699048,
|
| 1450 |
+
"learning_rate": 2e-05,
|
| 1451 |
+
"loss": 0.6873680114746094,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.2664778578784758,
|
| 1456 |
+
"grad_norm": 0.9527725260292781,
|
| 1457 |
+
"learning_rate": 2e-05,
|
| 1458 |
+
"loss": 0.7040456771850586,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.26776519052523173,
|
| 1463 |
+
"grad_norm": 1.1259153607047196,
|
| 1464 |
+
"learning_rate": 2e-05,
|
| 1465 |
+
"loss": 0.6887578010559082,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.26905252317198763,
|
| 1470 |
+
"grad_norm": 1.4640224316338672,
|
| 1471 |
+
"learning_rate": 2e-05,
|
| 1472 |
+
"loss": 0.7209594249725342,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.2703398558187436,
|
| 1477 |
+
"grad_norm": 1.6949802858236316,
|
| 1478 |
+
"learning_rate": 2e-05,
|
| 1479 |
+
"loss": 0.7076781272888184,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.2716271884654995,
|
| 1484 |
+
"grad_norm": 1.9455095056367833,
|
| 1485 |
+
"learning_rate": 2e-05,
|
| 1486 |
+
"loss": 0.7013013362884521,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.2729145211122554,
|
| 1491 |
+
"grad_norm": 1.885815138529574,
|
| 1492 |
+
"learning_rate": 2e-05,
|
| 1493 |
+
"loss": 0.70594482421875,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.27420185375901135,
|
| 1498 |
+
"grad_norm": 2.8614315685772205,
|
| 1499 |
+
"learning_rate": 2e-05,
|
| 1500 |
+
"loss": 0.7034307003021241,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.27548918640576725,
|
| 1505 |
+
"grad_norm": 12.720304202743357,
|
| 1506 |
+
"learning_rate": 2e-05,
|
| 1507 |
+
"loss": 0.7098933219909668,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.27677651905252315,
|
| 1512 |
+
"grad_norm": 2.1655512205055225,
|
| 1513 |
+
"learning_rate": 2e-05,
|
| 1514 |
+
"loss": 0.7170520305633545,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.2780638516992791,
|
| 1519 |
+
"grad_norm": 8.691373263950895,
|
| 1520 |
+
"learning_rate": 2e-05,
|
| 1521 |
+
"loss": 0.7013160228729248,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.279351184346035,
|
| 1526 |
+
"grad_norm": 1.608771090637953,
|
| 1527 |
+
"learning_rate": 2e-05,
|
| 1528 |
+
"loss": 0.7061017036437989,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.2806385169927909,
|
| 1533 |
+
"grad_norm": 11.833877668015374,
|
| 1534 |
+
"learning_rate": 2e-05,
|
| 1535 |
+
"loss": 0.6650290489196777,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.28192584963954687,
|
| 1540 |
+
"grad_norm": 1.4006501457101668,
|
| 1541 |
+
"learning_rate": 2e-05,
|
| 1542 |
+
"loss": 0.7379101753234864,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.28321318228630277,
|
| 1547 |
+
"grad_norm": 1.6610926197943376,
|
| 1548 |
+
"learning_rate": 2e-05,
|
| 1549 |
+
"loss": 0.6706048011779785,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.2845005149330587,
|
| 1554 |
+
"grad_norm": 1.487648001601594,
|
| 1555 |
+
"learning_rate": 2e-05,
|
| 1556 |
+
"loss": 0.6972060680389405,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.2857878475798146,
|
| 1561 |
+
"grad_norm": 7.7682006345537715,
|
| 1562 |
+
"learning_rate": 2e-05,
|
| 1563 |
+
"loss": 0.6770853996276855,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.2870751802265705,
|
| 1568 |
+
"grad_norm": 2.0707865262416623,
|
| 1569 |
+
"learning_rate": 2e-05,
|
| 1570 |
+
"loss": 0.6925491333007813,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.2883625128733265,
|
| 1575 |
+
"grad_norm": 1.7595244490954383,
|
| 1576 |
+
"learning_rate": 2e-05,
|
| 1577 |
+
"loss": 0.6970232009887696,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.2896498455200824,
|
| 1582 |
+
"grad_norm": 1.092700120083694,
|
| 1583 |
+
"learning_rate": 2e-05,
|
| 1584 |
+
"loss": 0.6953269958496093,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.2909371781668383,
|
| 1589 |
+
"grad_norm": 9.748987611796897,
|
| 1590 |
+
"learning_rate": 2e-05,
|
| 1591 |
+
"loss": 0.6644187927246094,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.29222451081359424,
|
| 1596 |
+
"grad_norm": 2.492722399232875,
|
| 1597 |
+
"learning_rate": 2e-05,
|
| 1598 |
+
"loss": 0.6756973266601562,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.29351184346035014,
|
| 1603 |
+
"grad_norm": 1.568787070446361,
|
| 1604 |
+
"learning_rate": 2e-05,
|
| 1605 |
+
"loss": 0.7013367652893067,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.2947991761071061,
|
| 1610 |
+
"grad_norm": 2.47976302361641,
|
| 1611 |
+
"learning_rate": 2e-05,
|
| 1612 |
+
"loss": 0.6758729457855225,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.296086508753862,
|
| 1617 |
+
"grad_norm": 1.8280130594138808,
|
| 1618 |
+
"learning_rate": 2e-05,
|
| 1619 |
+
"loss": 0.718410587310791,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.2973738414006179,
|
| 1624 |
+
"grad_norm": 1.330373350200014,
|
| 1625 |
+
"learning_rate": 2e-05,
|
| 1626 |
+
"loss": 0.6956700325012207,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.29866117404737386,
|
| 1631 |
+
"grad_norm": 2.1289014923647347,
|
| 1632 |
+
"learning_rate": 2e-05,
|
| 1633 |
+
"loss": 0.6391764640808105,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.29994850669412976,
|
| 1638 |
+
"grad_norm": 1.4812494641456213,
|
| 1639 |
+
"learning_rate": 2e-05,
|
| 1640 |
+
"loss": 0.6836237907409668,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.30123583934088566,
|
| 1645 |
+
"grad_norm": 2.182997522790617,
|
| 1646 |
+
"learning_rate": 2e-05,
|
| 1647 |
+
"loss": 0.6847892284393311,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.3025231719876416,
|
| 1652 |
+
"grad_norm": 1.4120255579731853,
|
| 1653 |
+
"learning_rate": 2e-05,
|
| 1654 |
+
"loss": 0.6991443634033203,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.3038105046343975,
|
| 1659 |
+
"grad_norm": 1.7715960231377,
|
| 1660 |
+
"learning_rate": 2e-05,
|
| 1661 |
+
"loss": 0.6520809173583985,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.30509783728115347,
|
| 1666 |
+
"grad_norm": 1.344504528912978,
|
| 1667 |
+
"learning_rate": 2e-05,
|
| 1668 |
+
"loss": 0.6915669441223145,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.3063851699279094,
|
| 1673 |
+
"grad_norm": 1.418372939618548,
|
| 1674 |
+
"learning_rate": 2e-05,
|
| 1675 |
+
"loss": 0.6714213371276856,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.3076725025746653,
|
| 1680 |
+
"grad_norm": 2.3106904123208403,
|
| 1681 |
+
"learning_rate": 2e-05,
|
| 1682 |
+
"loss": 0.6664355278015137,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.30895983522142123,
|
| 1687 |
+
"grad_norm": 1.3253046019523815,
|
| 1688 |
+
"learning_rate": 2e-05,
|
| 1689 |
+
"loss": 0.6971554756164551,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.31024716786817713,
|
| 1694 |
+
"grad_norm": 1.2161954394286878,
|
| 1695 |
+
"learning_rate": 2e-05,
|
| 1696 |
+
"loss": 0.644910192489624,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.31153450051493303,
|
| 1701 |
+
"grad_norm": 1.5284008850353576,
|
| 1702 |
+
"learning_rate": 2e-05,
|
| 1703 |
+
"loss": 0.671488904953003,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.312821833161689,
|
| 1708 |
+
"grad_norm": 2.3280051785201517,
|
| 1709 |
+
"learning_rate": 2e-05,
|
| 1710 |
+
"loss": 0.6644371032714844,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.3141091658084449,
|
| 1715 |
+
"grad_norm": 2.4846074890678547,
|
| 1716 |
+
"learning_rate": 2e-05,
|
| 1717 |
+
"loss": 0.6947656631469726,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.31539649845520085,
|
| 1722 |
+
"grad_norm": 2.342854536473009,
|
| 1723 |
+
"learning_rate": 2e-05,
|
| 1724 |
+
"loss": 0.6945387840270996,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.31668383110195675,
|
| 1729 |
+
"grad_norm": 1.3707661574249077,
|
| 1730 |
+
"learning_rate": 2e-05,
|
| 1731 |
+
"loss": 0.7021138668060303,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.31797116374871265,
|
| 1736 |
+
"grad_norm": 1.011087502453139,
|
| 1737 |
+
"learning_rate": 2e-05,
|
| 1738 |
+
"loss": 0.711193037033081,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.3192584963954686,
|
| 1743 |
+
"grad_norm": 1.3124044269321697,
|
| 1744 |
+
"learning_rate": 2e-05,
|
| 1745 |
+
"loss": 0.7033077239990234,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.3205458290422245,
|
| 1750 |
+
"grad_norm": 0.8810548907751976,
|
| 1751 |
+
"learning_rate": 2e-05,
|
| 1752 |
+
"loss": 0.6761528015136719,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.3218331616889804,
|
| 1757 |
+
"grad_norm": 1.244342002235635,
|
| 1758 |
+
"learning_rate": 2e-05,
|
| 1759 |
+
"loss": 0.6616430282592773,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.32312049433573636,
|
| 1764 |
+
"grad_norm": 1.2138059133962842,
|
| 1765 |
+
"learning_rate": 2e-05,
|
| 1766 |
+
"loss": 0.6823375701904297,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.32440782698249226,
|
| 1771 |
+
"grad_norm": 1.5411359954112853,
|
| 1772 |
+
"learning_rate": 2e-05,
|
| 1773 |
+
"loss": 0.6461188316345214,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.3256951596292482,
|
| 1778 |
+
"grad_norm": 4.067446254024166,
|
| 1779 |
+
"learning_rate": 2e-05,
|
| 1780 |
+
"loss": 0.6632846832275391,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.3269824922760041,
|
| 1785 |
+
"grad_norm": 0.9930530111354352,
|
| 1786 |
+
"learning_rate": 2e-05,
|
| 1787 |
+
"loss": 0.6521314144134521,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.32826982492276,
|
| 1792 |
+
"grad_norm": 1.2183079247100232,
|
| 1793 |
+
"learning_rate": 2e-05,
|
| 1794 |
+
"loss": 0.6647382259368897,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.329557157569516,
|
| 1799 |
+
"grad_norm": 1.3056406095156692,
|
| 1800 |
+
"learning_rate": 2e-05,
|
| 1801 |
+
"loss": 0.7056656837463379,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.3308444902162719,
|
| 1806 |
+
"grad_norm": 1.1446095028508332,
|
| 1807 |
+
"learning_rate": 2e-05,
|
| 1808 |
+
"loss": 0.6724899291992188,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.3321318228630278,
|
| 1813 |
+
"grad_norm": 2.034841488345913,
|
| 1814 |
+
"learning_rate": 2e-05,
|
| 1815 |
+
"loss": 0.7128400325775146,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.33341915550978374,
|
| 1820 |
+
"grad_norm": 1.1398931102101963,
|
| 1821 |
+
"learning_rate": 2e-05,
|
| 1822 |
+
"loss": 0.694240665435791,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.33470648815653964,
|
| 1827 |
+
"grad_norm": 1.1474928486855198,
|
| 1828 |
+
"learning_rate": 2e-05,
|
| 1829 |
+
"loss": 0.6734938621520996,
|
| 1830 |
+
"step": 2600
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 0.3359938208032956,
|
| 1834 |
+
"grad_norm": 1.3857880198125372,
|
| 1835 |
+
"learning_rate": 2e-05,
|
| 1836 |
+
"loss": 0.6751894950866699,
|
| 1837 |
+
"step": 2610
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 0.3372811534500515,
|
| 1841 |
+
"grad_norm": 1.785399986240084,
|
| 1842 |
+
"learning_rate": 2e-05,
|
| 1843 |
+
"loss": 0.6587059020996093,
|
| 1844 |
+
"step": 2620
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 0.3385684860968074,
|
| 1848 |
+
"grad_norm": 1.7238102251604501,
|
| 1849 |
+
"learning_rate": 2e-05,
|
| 1850 |
+
"loss": 0.7021533966064453,
|
| 1851 |
+
"step": 2630
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 0.33985581874356335,
|
| 1855 |
+
"grad_norm": 1.065710140961934,
|
| 1856 |
+
"learning_rate": 2e-05,
|
| 1857 |
+
"loss": 0.690271520614624,
|
| 1858 |
+
"step": 2640
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 0.34114315139031925,
|
| 1862 |
+
"grad_norm": 4.21226441159709,
|
| 1863 |
+
"learning_rate": 2e-05,
|
| 1864 |
+
"loss": 0.6952508926391602,
|
| 1865 |
+
"step": 2650
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 0.34243048403707516,
|
| 1869 |
+
"grad_norm": 1.146287127572526,
|
| 1870 |
+
"learning_rate": 2e-05,
|
| 1871 |
+
"loss": 0.6579763412475585,
|
| 1872 |
+
"step": 2660
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 0.3437178166838311,
|
| 1876 |
+
"grad_norm": 1.1891929638258938,
|
| 1877 |
+
"learning_rate": 2e-05,
|
| 1878 |
+
"loss": 0.6524415016174316,
|
| 1879 |
+
"step": 2670
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 0.345005149330587,
|
| 1883 |
+
"grad_norm": 2.3605719453162757,
|
| 1884 |
+
"learning_rate": 2e-05,
|
| 1885 |
+
"loss": 0.6747819900512695,
|
| 1886 |
+
"step": 2680
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 0.34629248197734297,
|
| 1890 |
+
"grad_norm": 2.470411316918508,
|
| 1891 |
+
"learning_rate": 2e-05,
|
| 1892 |
+
"loss": 0.6213415145874024,
|
| 1893 |
+
"step": 2690
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 0.34757981462409887,
|
| 1897 |
+
"grad_norm": 2.0404873221926123,
|
| 1898 |
+
"learning_rate": 2e-05,
|
| 1899 |
+
"loss": 0.6706666469573974,
|
| 1900 |
+
"step": 2700
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 0.34886714727085477,
|
| 1904 |
+
"grad_norm": 1.4692042495535325,
|
| 1905 |
+
"learning_rate": 2e-05,
|
| 1906 |
+
"loss": 0.6797372817993164,
|
| 1907 |
+
"step": 2710
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 0.35015447991761073,
|
| 1911 |
+
"grad_norm": 2.589397458312476,
|
| 1912 |
+
"learning_rate": 2e-05,
|
| 1913 |
+
"loss": 0.6521578788757324,
|
| 1914 |
+
"step": 2720
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 0.35144181256436663,
|
| 1918 |
+
"grad_norm": 1.1277270836479611,
|
| 1919 |
+
"learning_rate": 2e-05,
|
| 1920 |
+
"loss": 0.6732677459716797,
|
| 1921 |
+
"step": 2730
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 0.35272914521112253,
|
| 1925 |
+
"grad_norm": 1.5051167160166194,
|
| 1926 |
+
"learning_rate": 2e-05,
|
| 1927 |
+
"loss": 0.658482837677002,
|
| 1928 |
+
"step": 2740
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 0.3540164778578785,
|
| 1932 |
+
"grad_norm": 2.2443513526369627,
|
| 1933 |
+
"learning_rate": 2e-05,
|
| 1934 |
+
"loss": 0.6609281063079834,
|
| 1935 |
+
"step": 2750
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 0.3553038105046344,
|
| 1939 |
+
"grad_norm": 2.8898723653557146,
|
| 1940 |
+
"learning_rate": 2e-05,
|
| 1941 |
+
"loss": 0.6852851867675781,
|
| 1942 |
+
"step": 2760
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 0.35659114315139034,
|
| 1946 |
+
"grad_norm": 1.9403402362804802,
|
| 1947 |
+
"learning_rate": 2e-05,
|
| 1948 |
+
"loss": 0.629300594329834,
|
| 1949 |
+
"step": 2770
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 0.35787847579814624,
|
| 1953 |
+
"grad_norm": 1.4399739483600749,
|
| 1954 |
+
"learning_rate": 2e-05,
|
| 1955 |
+
"loss": 0.6836570739746094,
|
| 1956 |
+
"step": 2780
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 0.35916580844490215,
|
| 1960 |
+
"grad_norm": 1.394028020897877,
|
| 1961 |
+
"learning_rate": 2e-05,
|
| 1962 |
+
"loss": 0.6523145198822021,
|
| 1963 |
+
"step": 2790
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 0.3604531410916581,
|
| 1967 |
+
"grad_norm": 1.519666297429449,
|
| 1968 |
+
"learning_rate": 2e-05,
|
| 1969 |
+
"loss": 0.6626304626464844,
|
| 1970 |
+
"step": 2800
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 0.361740473738414,
|
| 1974 |
+
"grad_norm": 6.45118774048948,
|
| 1975 |
+
"learning_rate": 2e-05,
|
| 1976 |
+
"loss": 0.6728658199310302,
|
| 1977 |
+
"step": 2810
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 0.3630278063851699,
|
| 1981 |
+
"grad_norm": 5.278190679624564,
|
| 1982 |
+
"learning_rate": 2e-05,
|
| 1983 |
+
"loss": 0.6950824737548829,
|
| 1984 |
+
"step": 2820
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 0.36431513903192586,
|
| 1988 |
+
"grad_norm": 3.379627099450112,
|
| 1989 |
+
"learning_rate": 2e-05,
|
| 1990 |
+
"loss": 0.7113072872161865,
|
| 1991 |
+
"step": 2830
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 0.36560247167868176,
|
| 1995 |
+
"grad_norm": 9.928058665678439,
|
| 1996 |
+
"learning_rate": 2e-05,
|
| 1997 |
+
"loss": 0.6998546123504639,
|
| 1998 |
+
"step": 2840
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 0.3668898043254377,
|
| 2002 |
+
"grad_norm": 2.4261849530745616,
|
| 2003 |
+
"learning_rate": 2e-05,
|
| 2004 |
+
"loss": 0.7137026786804199,
|
| 2005 |
+
"step": 2850
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 0.3681771369721936,
|
| 2009 |
+
"grad_norm": 1.705339489405855,
|
| 2010 |
+
"learning_rate": 2e-05,
|
| 2011 |
+
"loss": 0.6986976623535156,
|
| 2012 |
+
"step": 2860
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 0.3694644696189495,
|
| 2016 |
+
"grad_norm": 1.5901594848354943,
|
| 2017 |
+
"learning_rate": 2e-05,
|
| 2018 |
+
"loss": 0.6721778869628906,
|
| 2019 |
+
"step": 2870
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 0.3707518022657055,
|
| 2023 |
+
"grad_norm": 2.0976255961116155,
|
| 2024 |
+
"learning_rate": 2e-05,
|
| 2025 |
+
"loss": 0.6972661972045898,
|
| 2026 |
+
"step": 2880
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 0.3720391349124614,
|
| 2030 |
+
"grad_norm": 1.737453259234397,
|
| 2031 |
+
"learning_rate": 2e-05,
|
| 2032 |
+
"loss": 0.6896710395812988,
|
| 2033 |
+
"step": 2890
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 0.3733264675592173,
|
| 2037 |
+
"grad_norm": 6.687644336375675,
|
| 2038 |
+
"learning_rate": 2e-05,
|
| 2039 |
+
"loss": 0.6870223045349121,
|
| 2040 |
+
"step": 2900
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 0.37461380020597324,
|
| 2044 |
+
"grad_norm": 1.7526243999408582,
|
| 2045 |
+
"learning_rate": 2e-05,
|
| 2046 |
+
"loss": 0.6775806427001954,
|
| 2047 |
+
"step": 2910
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 0.37590113285272914,
|
| 2051 |
+
"grad_norm": 1.449069006898657,
|
| 2052 |
+
"learning_rate": 2e-05,
|
| 2053 |
+
"loss": 0.6732563018798828,
|
| 2054 |
+
"step": 2920
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 0.3771884654994851,
|
| 2058 |
+
"grad_norm": 1.5495788900320941,
|
| 2059 |
+
"learning_rate": 2e-05,
|
| 2060 |
+
"loss": 0.6752627849578857,
|
| 2061 |
+
"step": 2930
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 0.378475798146241,
|
| 2065 |
+
"grad_norm": 1.4928682148793067,
|
| 2066 |
+
"learning_rate": 2e-05,
|
| 2067 |
+
"loss": 0.6585468292236328,
|
| 2068 |
+
"step": 2940
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 0.3797631307929969,
|
| 2072 |
+
"grad_norm": 3.570159299542741,
|
| 2073 |
+
"learning_rate": 2e-05,
|
| 2074 |
+
"loss": 0.6992852687835693,
|
| 2075 |
+
"step": 2950
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 0.38105046343975285,
|
| 2079 |
+
"grad_norm": 1.745416978112621,
|
| 2080 |
+
"learning_rate": 2e-05,
|
| 2081 |
+
"loss": 0.6408042430877685,
|
| 2082 |
+
"step": 2960
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 0.38233779608650875,
|
| 2086 |
+
"grad_norm": 1.3878562358445177,
|
| 2087 |
+
"learning_rate": 2e-05,
|
| 2088 |
+
"loss": 0.652539587020874,
|
| 2089 |
+
"step": 2970
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 0.38362512873326465,
|
| 2093 |
+
"grad_norm": 1.7373358136561254,
|
| 2094 |
+
"learning_rate": 2e-05,
|
| 2095 |
+
"loss": 0.686555814743042,
|
| 2096 |
+
"step": 2980
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 0.3849124613800206,
|
| 2100 |
+
"grad_norm": 1.6110411266569502,
|
| 2101 |
+
"learning_rate": 2e-05,
|
| 2102 |
+
"loss": 0.6672499656677247,
|
| 2103 |
+
"step": 2990
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 0.3861997940267765,
|
| 2107 |
+
"grad_norm": 8.591842569044532,
|
| 2108 |
+
"learning_rate": 2e-05,
|
| 2109 |
+
"loss": 0.7332715511322021,
|
| 2110 |
+
"step": 3000
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 0.38748712667353247,
|
| 2114 |
+
"grad_norm": 4.770480208347062,
|
| 2115 |
+
"learning_rate": 2e-05,
|
| 2116 |
+
"loss": 0.6914670467376709,
|
| 2117 |
+
"step": 3010
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 0.38877445932028837,
|
| 2121 |
+
"grad_norm": 2.5309185366901246,
|
| 2122 |
+
"learning_rate": 2e-05,
|
| 2123 |
+
"loss": 0.681065845489502,
|
| 2124 |
+
"step": 3020
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 0.39006179196704427,
|
| 2128 |
+
"grad_norm": 1.6387711955152355,
|
| 2129 |
+
"learning_rate": 2e-05,
|
| 2130 |
+
"loss": 0.6919746398925781,
|
| 2131 |
+
"step": 3030
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 0.3913491246138002,
|
| 2135 |
+
"grad_norm": 1.3566915307344818,
|
| 2136 |
+
"learning_rate": 2e-05,
|
| 2137 |
+
"loss": 0.6504603385925293,
|
| 2138 |
+
"step": 3040
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 0.3926364572605561,
|
| 2142 |
+
"grad_norm": 2.786785682313342,
|
| 2143 |
+
"learning_rate": 2e-05,
|
| 2144 |
+
"loss": 0.672648572921753,
|
| 2145 |
+
"step": 3050
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 0.393923789907312,
|
| 2149 |
+
"grad_norm": 1.7948386684878221,
|
| 2150 |
+
"learning_rate": 2e-05,
|
| 2151 |
+
"loss": 0.6768685817718506,
|
| 2152 |
+
"step": 3060
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 0.395211122554068,
|
| 2156 |
+
"grad_norm": 1.3992759220970516,
|
| 2157 |
+
"learning_rate": 2e-05,
|
| 2158 |
+
"loss": 0.673085880279541,
|
| 2159 |
+
"step": 3070
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 0.3964984552008239,
|
| 2163 |
+
"grad_norm": 2.2397090140048146,
|
| 2164 |
+
"learning_rate": 2e-05,
|
| 2165 |
+
"loss": 0.6780692100524902,
|
| 2166 |
+
"step": 3080
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 0.39778578784757984,
|
| 2170 |
+
"grad_norm": 1.825394663917336,
|
| 2171 |
+
"learning_rate": 2e-05,
|
| 2172 |
+
"loss": 0.6607382774353028,
|
| 2173 |
+
"step": 3090
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 0.39907312049433574,
|
| 2177 |
+
"grad_norm": 4.389762594014447,
|
| 2178 |
+
"learning_rate": 2e-05,
|
| 2179 |
+
"loss": 0.6736061096191406,
|
| 2180 |
+
"step": 3100
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 0.40036045314109164,
|
| 2184 |
+
"grad_norm": 1.9317885442360987,
|
| 2185 |
+
"learning_rate": 2e-05,
|
| 2186 |
+
"loss": 0.6794852256774903,
|
| 2187 |
+
"step": 3110
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 0.4016477857878476,
|
| 2191 |
+
"grad_norm": 8.9709361728038,
|
| 2192 |
+
"learning_rate": 2e-05,
|
| 2193 |
+
"loss": 0.6856914043426514,
|
| 2194 |
+
"step": 3120
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 0.4029351184346035,
|
| 2198 |
+
"grad_norm": 10.096913313964608,
|
| 2199 |
+
"learning_rate": 2e-05,
|
| 2200 |
+
"loss": 0.6717545986175537,
|
| 2201 |
+
"step": 3130
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 0.4042224510813594,
|
| 2205 |
+
"grad_norm": 4.1135761380680025,
|
| 2206 |
+
"learning_rate": 2e-05,
|
| 2207 |
+
"loss": 0.6613590240478515,
|
| 2208 |
+
"step": 3140
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 0.40550978372811536,
|
| 2212 |
+
"grad_norm": 5.458974686593683,
|
| 2213 |
+
"learning_rate": 2e-05,
|
| 2214 |
+
"loss": 0.6676340103149414,
|
| 2215 |
+
"step": 3150
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 0.40679711637487126,
|
| 2219 |
+
"grad_norm": 1.5353876859514601,
|
| 2220 |
+
"learning_rate": 2e-05,
|
| 2221 |
+
"loss": 0.6625430107116699,
|
| 2222 |
+
"step": 3160
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 0.4080844490216272,
|
| 2226 |
+
"grad_norm": 2.617512976186076,
|
| 2227 |
+
"learning_rate": 2e-05,
|
| 2228 |
+
"loss": 0.7098607540130615,
|
| 2229 |
+
"step": 3170
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 0.4093717816683831,
|
| 2233 |
+
"grad_norm": 24.853744974348846,
|
| 2234 |
+
"learning_rate": 2e-05,
|
| 2235 |
+
"loss": 0.6653516292572021,
|
| 2236 |
+
"step": 3180
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 0.410659114315139,
|
| 2240 |
+
"grad_norm": 1.8938125815091902,
|
| 2241 |
+
"learning_rate": 2e-05,
|
| 2242 |
+
"loss": 0.6713570594787598,
|
| 2243 |
+
"step": 3190
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 0.411946446961895,
|
| 2247 |
+
"grad_norm": 5.279762512651014,
|
| 2248 |
+
"learning_rate": 2e-05,
|
| 2249 |
+
"loss": 0.6681087493896485,
|
| 2250 |
+
"step": 3200
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 0.4132337796086509,
|
| 2254 |
+
"grad_norm": 8.141366763733092,
|
| 2255 |
+
"learning_rate": 2e-05,
|
| 2256 |
+
"loss": 0.6675487041473389,
|
| 2257 |
+
"step": 3210
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 0.4145211122554068,
|
| 2261 |
+
"grad_norm": 1.4196281538096058,
|
| 2262 |
+
"learning_rate": 2e-05,
|
| 2263 |
+
"loss": 0.6300604820251465,
|
| 2264 |
+
"step": 3220
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 0.41580844490216273,
|
| 2268 |
+
"grad_norm": 1.5189135563521712,
|
| 2269 |
+
"learning_rate": 2e-05,
|
| 2270 |
+
"loss": 0.6708904266357422,
|
| 2271 |
+
"step": 3230
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 0.41709577754891863,
|
| 2275 |
+
"grad_norm": 1.316328436422404,
|
| 2276 |
+
"learning_rate": 2e-05,
|
| 2277 |
+
"loss": 0.6688165187835693,
|
| 2278 |
+
"step": 3240
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 0.41838311019567453,
|
| 2282 |
+
"grad_norm": 1.490385997513553,
|
| 2283 |
+
"learning_rate": 2e-05,
|
| 2284 |
+
"loss": 0.6932008266448975,
|
| 2285 |
+
"step": 3250
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 0.4196704428424305,
|
| 2289 |
+
"grad_norm": 6.699064455485067,
|
| 2290 |
+
"learning_rate": 2e-05,
|
| 2291 |
+
"loss": 0.6734150886535645,
|
| 2292 |
+
"step": 3260
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 0.4209577754891864,
|
| 2296 |
+
"grad_norm": 1.9902209572292322,
|
| 2297 |
+
"learning_rate": 2e-05,
|
| 2298 |
+
"loss": 0.621324110031128,
|
| 2299 |
+
"step": 3270
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 0.42224510813594235,
|
| 2303 |
+
"grad_norm": 2.3929984697771323,
|
| 2304 |
+
"learning_rate": 2e-05,
|
| 2305 |
+
"loss": 0.6553589820861816,
|
| 2306 |
+
"step": 3280
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 0.42353244078269825,
|
| 2310 |
+
"grad_norm": 1.4323973020118568,
|
| 2311 |
+
"learning_rate": 2e-05,
|
| 2312 |
+
"loss": 0.6547521114349365,
|
| 2313 |
+
"step": 3290
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 0.42481977342945415,
|
| 2317 |
+
"grad_norm": 1.722224287548998,
|
| 2318 |
+
"learning_rate": 2e-05,
|
| 2319 |
+
"loss": 0.7304846763610839,
|
| 2320 |
+
"step": 3300
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 0.4261071060762101,
|
| 2324 |
+
"grad_norm": 2.1934329471769147,
|
| 2325 |
+
"learning_rate": 2e-05,
|
| 2326 |
+
"loss": 0.6440581321716309,
|
| 2327 |
+
"step": 3310
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 0.427394438722966,
|
| 2331 |
+
"grad_norm": 6.457762244550262,
|
| 2332 |
+
"learning_rate": 2e-05,
|
| 2333 |
+
"loss": 0.6681999206542969,
|
| 2334 |
+
"step": 3320
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 0.4286817713697219,
|
| 2338 |
+
"grad_norm": 4.0785650101698305,
|
| 2339 |
+
"learning_rate": 2e-05,
|
| 2340 |
+
"loss": 0.6417197704315185,
|
| 2341 |
+
"step": 3330
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 0.42996910401647787,
|
| 2345 |
+
"grad_norm": 3.671425616461221,
|
| 2346 |
+
"learning_rate": 2e-05,
|
| 2347 |
+
"loss": 0.6589700222015381,
|
| 2348 |
+
"step": 3340
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 0.43125643666323377,
|
| 2352 |
+
"grad_norm": 2.3178859428756904,
|
| 2353 |
+
"learning_rate": 2e-05,
|
| 2354 |
+
"loss": 0.661466121673584,
|
| 2355 |
+
"step": 3350
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 0.4325437693099897,
|
| 2359 |
+
"grad_norm": 1.7739984241122377,
|
| 2360 |
+
"learning_rate": 2e-05,
|
| 2361 |
+
"loss": 0.6633042812347412,
|
| 2362 |
+
"step": 3360
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 0.4338311019567456,
|
| 2366 |
+
"grad_norm": 1.4750459481455314,
|
| 2367 |
+
"learning_rate": 2e-05,
|
| 2368 |
+
"loss": 0.6860431671142578,
|
| 2369 |
+
"step": 3370
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 0.4351184346035015,
|
| 2373 |
+
"grad_norm": 1.024462615445388,
|
| 2374 |
+
"learning_rate": 2e-05,
|
| 2375 |
+
"loss": 0.6539207458496094,
|
| 2376 |
+
"step": 3380
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 0.4364057672502575,
|
| 2380 |
+
"grad_norm": 1.0028907091376442,
|
| 2381 |
+
"learning_rate": 2e-05,
|
| 2382 |
+
"loss": 0.6821427822113038,
|
| 2383 |
+
"step": 3390
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 0.4376930998970134,
|
| 2387 |
+
"grad_norm": 20.32370151247624,
|
| 2388 |
+
"learning_rate": 2e-05,
|
| 2389 |
+
"loss": 0.6655908584594726,
|
| 2390 |
+
"step": 3400
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 0.4389804325437693,
|
| 2394 |
+
"grad_norm": 1.5298304681583943,
|
| 2395 |
+
"learning_rate": 2e-05,
|
| 2396 |
+
"loss": 0.6839134216308593,
|
| 2397 |
+
"step": 3410
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 0.44026776519052524,
|
| 2401 |
+
"grad_norm": 2.0967857109698524,
|
| 2402 |
+
"learning_rate": 2e-05,
|
| 2403 |
+
"loss": 0.6819107055664062,
|
| 2404 |
+
"step": 3420
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 0.44155509783728114,
|
| 2408 |
+
"grad_norm": 2.027706958669317,
|
| 2409 |
+
"learning_rate": 2e-05,
|
| 2410 |
+
"loss": 0.6640716552734375,
|
| 2411 |
+
"step": 3430
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 0.4428424304840371,
|
| 2415 |
+
"grad_norm": 1.1304241862727915,
|
| 2416 |
+
"learning_rate": 2e-05,
|
| 2417 |
+
"loss": 0.6362229347229004,
|
| 2418 |
+
"step": 3440
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 0.444129763130793,
|
| 2422 |
+
"grad_norm": 1.0563233431055887,
|
| 2423 |
+
"learning_rate": 2e-05,
|
| 2424 |
+
"loss": 0.6823592185974121,
|
| 2425 |
+
"step": 3450
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 0.4454170957775489,
|
| 2429 |
+
"grad_norm": 6.906802092336299,
|
| 2430 |
+
"learning_rate": 2e-05,
|
| 2431 |
+
"loss": 0.723987340927124,
|
| 2432 |
+
"step": 3460
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 0.44670442842430486,
|
| 2436 |
+
"grad_norm": 1.247112091006261,
|
| 2437 |
+
"learning_rate": 2e-05,
|
| 2438 |
+
"loss": 0.6849922180175781,
|
| 2439 |
+
"step": 3470
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 0.44799176107106076,
|
| 2443 |
+
"grad_norm": 26.02341861091827,
|
| 2444 |
+
"learning_rate": 2e-05,
|
| 2445 |
+
"loss": 0.630082654953003,
|
| 2446 |
+
"step": 3480
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 0.44927909371781666,
|
| 2450 |
+
"grad_norm": 3.428639375524443,
|
| 2451 |
+
"learning_rate": 2e-05,
|
| 2452 |
+
"loss": 0.6845290184020996,
|
| 2453 |
+
"step": 3490
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 0.4505664263645726,
|
| 2457 |
+
"grad_norm": 1.166177002019518,
|
| 2458 |
+
"learning_rate": 2e-05,
|
| 2459 |
+
"loss": 0.6864035606384278,
|
| 2460 |
+
"step": 3500
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 0.4518537590113285,
|
| 2464 |
+
"grad_norm": 1.8605023565035361,
|
| 2465 |
+
"learning_rate": 2e-05,
|
| 2466 |
+
"loss": 0.6661740779876709,
|
| 2467 |
+
"step": 3510
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 0.45314109165808447,
|
| 2471 |
+
"grad_norm": 0.971442813441688,
|
| 2472 |
+
"learning_rate": 2e-05,
|
| 2473 |
+
"loss": 0.66938796043396,
|
| 2474 |
+
"step": 3520
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 0.4544284243048404,
|
| 2478 |
+
"grad_norm": 1.3538855308769568,
|
| 2479 |
+
"learning_rate": 2e-05,
|
| 2480 |
+
"loss": 0.663056468963623,
|
| 2481 |
+
"step": 3530
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 0.4557157569515963,
|
| 2485 |
+
"grad_norm": 1.3794284739098457,
|
| 2486 |
+
"learning_rate": 2e-05,
|
| 2487 |
+
"loss": 0.6419530391693116,
|
| 2488 |
+
"step": 3540
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 0.45700308959835223,
|
| 2492 |
+
"grad_norm": 2.2926898326260194,
|
| 2493 |
+
"learning_rate": 2e-05,
|
| 2494 |
+
"loss": 0.6693635940551758,
|
| 2495 |
+
"step": 3550
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 0.45829042224510813,
|
| 2499 |
+
"grad_norm": 5.296737682431592,
|
| 2500 |
+
"learning_rate": 2e-05,
|
| 2501 |
+
"loss": 0.693416690826416,
|
| 2502 |
+
"step": 3560
|
| 2503 |
+
},
|
| 2504 |
+
{
|
| 2505 |
+
"epoch": 0.45957775489186403,
|
| 2506 |
+
"grad_norm": 1.4313451910795734,
|
| 2507 |
+
"learning_rate": 2e-05,
|
| 2508 |
+
"loss": 0.7027761459350585,
|
| 2509 |
+
"step": 3570
|
| 2510 |
+
},
|
| 2511 |
+
{
|
| 2512 |
+
"epoch": 0.46086508753862,
|
| 2513 |
+
"grad_norm": 8.00538666124274,
|
| 2514 |
+
"learning_rate": 2e-05,
|
| 2515 |
+
"loss": 0.6691301345825196,
|
| 2516 |
+
"step": 3580
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 0.4621524201853759,
|
| 2520 |
+
"grad_norm": 3.1943754145255605,
|
| 2521 |
+
"learning_rate": 2e-05,
|
| 2522 |
+
"loss": 0.6514184474945068,
|
| 2523 |
+
"step": 3590
|
| 2524 |
+
},
|
| 2525 |
+
{
|
| 2526 |
+
"epoch": 0.46343975283213185,
|
| 2527 |
+
"grad_norm": 6.573149792400552,
|
| 2528 |
+
"learning_rate": 2e-05,
|
| 2529 |
+
"loss": 0.6754954338073731,
|
| 2530 |
+
"step": 3600
|
| 2531 |
+
},
|
| 2532 |
+
{
|
| 2533 |
+
"epoch": 0.46472708547888775,
|
| 2534 |
+
"grad_norm": 2.154765907874636,
|
| 2535 |
+
"learning_rate": 2e-05,
|
| 2536 |
+
"loss": 0.6678764343261718,
|
| 2537 |
+
"step": 3610
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 0.46601441812564365,
|
| 2541 |
+
"grad_norm": 3.6800679121985946,
|
| 2542 |
+
"learning_rate": 2e-05,
|
| 2543 |
+
"loss": 0.692064619064331,
|
| 2544 |
+
"step": 3620
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 0.4673017507723996,
|
| 2548 |
+
"grad_norm": 1.8650670848048954,
|
| 2549 |
+
"learning_rate": 2e-05,
|
| 2550 |
+
"loss": 0.6698975563049316,
|
| 2551 |
+
"step": 3630
|
| 2552 |
+
},
|
| 2553 |
+
{
|
| 2554 |
+
"epoch": 0.4685890834191555,
|
| 2555 |
+
"grad_norm": 1.8829809424851136,
|
| 2556 |
+
"learning_rate": 2e-05,
|
| 2557 |
+
"loss": 0.6646499633789062,
|
| 2558 |
+
"step": 3640
|
| 2559 |
+
},
|
| 2560 |
+
{
|
| 2561 |
+
"epoch": 0.4698764160659114,
|
| 2562 |
+
"grad_norm": 1.1146778872216527,
|
| 2563 |
+
"learning_rate": 2e-05,
|
| 2564 |
+
"loss": 0.6541554927825928,
|
| 2565 |
+
"step": 3650
|
| 2566 |
+
},
|
| 2567 |
+
{
|
| 2568 |
+
"epoch": 0.47116374871266736,
|
| 2569 |
+
"grad_norm": 1.3884161087356786,
|
| 2570 |
+
"learning_rate": 2e-05,
|
| 2571 |
+
"loss": 0.7008742332458496,
|
| 2572 |
+
"step": 3660
|
| 2573 |
+
},
|
| 2574 |
+
{
|
| 2575 |
+
"epoch": 0.47245108135942326,
|
| 2576 |
+
"grad_norm": 2.488001552019463,
|
| 2577 |
+
"learning_rate": 2e-05,
|
| 2578 |
+
"loss": 0.7059492111206055,
|
| 2579 |
+
"step": 3670
|
| 2580 |
+
},
|
| 2581 |
+
{
|
| 2582 |
+
"epoch": 0.4737384140061792,
|
| 2583 |
+
"grad_norm": 2.1436795964440347,
|
| 2584 |
+
"learning_rate": 2e-05,
|
| 2585 |
+
"loss": 0.6514601707458496,
|
| 2586 |
+
"step": 3680
|
| 2587 |
+
},
|
| 2588 |
+
{
|
| 2589 |
+
"epoch": 0.4750257466529351,
|
| 2590 |
+
"grad_norm": 1.6540498436206854,
|
| 2591 |
+
"learning_rate": 2e-05,
|
| 2592 |
+
"loss": 0.684079933166504,
|
| 2593 |
+
"step": 3690
|
| 2594 |
+
},
|
| 2595 |
+
{
|
| 2596 |
+
"epoch": 0.476313079299691,
|
| 2597 |
+
"grad_norm": 1.7862339002922256,
|
| 2598 |
+
"learning_rate": 2e-05,
|
| 2599 |
+
"loss": 0.7529743194580079,
|
| 2600 |
+
"step": 3700
|
| 2601 |
+
},
|
| 2602 |
+
{
|
| 2603 |
+
"epoch": 0.477600411946447,
|
| 2604 |
+
"grad_norm": 0.9956859259704859,
|
| 2605 |
+
"learning_rate": 2e-05,
|
| 2606 |
+
"loss": 0.663523817062378,
|
| 2607 |
+
"step": 3710
|
| 2608 |
+
},
|
| 2609 |
+
{
|
| 2610 |
+
"epoch": 0.4788877445932029,
|
| 2611 |
+
"grad_norm": 1.1151185742888514,
|
| 2612 |
+
"learning_rate": 2e-05,
|
| 2613 |
+
"loss": 0.6718929290771485,
|
| 2614 |
+
"step": 3720
|
| 2615 |
+
},
|
| 2616 |
+
{
|
| 2617 |
+
"epoch": 0.4801750772399588,
|
| 2618 |
+
"grad_norm": 1.7975567259722014,
|
| 2619 |
+
"learning_rate": 2e-05,
|
| 2620 |
+
"loss": 0.6369192123413085,
|
| 2621 |
+
"step": 3730
|
| 2622 |
+
},
|
| 2623 |
+
{
|
| 2624 |
+
"epoch": 0.48146240988671474,
|
| 2625 |
+
"grad_norm": 18.533022605588275,
|
| 2626 |
+
"learning_rate": 2e-05,
|
| 2627 |
+
"loss": 0.6432384490966797,
|
| 2628 |
+
"step": 3740
|
| 2629 |
+
},
|
| 2630 |
+
{
|
| 2631 |
+
"epoch": 0.48274974253347064,
|
| 2632 |
+
"grad_norm": 2.7878600747183073,
|
| 2633 |
+
"learning_rate": 2e-05,
|
| 2634 |
+
"loss": 0.6795418739318848,
|
| 2635 |
+
"step": 3750
|
| 2636 |
+
},
|
| 2637 |
+
{
|
| 2638 |
+
"epoch": 0.4840370751802266,
|
| 2639 |
+
"grad_norm": 1.0890450077296256,
|
| 2640 |
+
"learning_rate": 2e-05,
|
| 2641 |
+
"loss": 0.6459294319152832,
|
| 2642 |
+
"step": 3760
|
| 2643 |
+
},
|
| 2644 |
+
{
|
| 2645 |
+
"epoch": 0.4853244078269825,
|
| 2646 |
+
"grad_norm": 6.4422192763167665,
|
| 2647 |
+
"learning_rate": 2e-05,
|
| 2648 |
+
"loss": 0.674666166305542,
|
| 2649 |
+
"step": 3770
|
| 2650 |
+
},
|
| 2651 |
+
{
|
| 2652 |
+
"epoch": 0.4866117404737384,
|
| 2653 |
+
"grad_norm": 1.3127188153680827,
|
| 2654 |
+
"learning_rate": 2e-05,
|
| 2655 |
+
"loss": 0.6758196830749512,
|
| 2656 |
+
"step": 3780
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 0.48789907312049435,
|
| 2660 |
+
"grad_norm": 1.2704832482153694,
|
| 2661 |
+
"learning_rate": 2e-05,
|
| 2662 |
+
"loss": 0.6724513530731201,
|
| 2663 |
+
"step": 3790
|
| 2664 |
+
},
|
| 2665 |
+
{
|
| 2666 |
+
"epoch": 0.48918640576725025,
|
| 2667 |
+
"grad_norm": 5.946873165430182,
|
| 2668 |
+
"learning_rate": 2e-05,
|
| 2669 |
+
"loss": 0.6656732559204102,
|
| 2670 |
+
"step": 3800
|
| 2671 |
+
},
|
| 2672 |
+
{
|
| 2673 |
+
"epoch": 0.49047373841400616,
|
| 2674 |
+
"grad_norm": 2.1508874984563615,
|
| 2675 |
+
"learning_rate": 2e-05,
|
| 2676 |
+
"loss": 0.6731793880462646,
|
| 2677 |
+
"step": 3810
|
| 2678 |
+
},
|
| 2679 |
+
{
|
| 2680 |
+
"epoch": 0.4917610710607621,
|
| 2681 |
+
"grad_norm": 11.013002873008867,
|
| 2682 |
+
"learning_rate": 2e-05,
|
| 2683 |
+
"loss": 0.6558897972106934,
|
| 2684 |
+
"step": 3820
|
| 2685 |
+
},
|
| 2686 |
+
{
|
| 2687 |
+
"epoch": 0.493048403707518,
|
| 2688 |
+
"grad_norm": 1.1369603803159063,
|
| 2689 |
+
"learning_rate": 2e-05,
|
| 2690 |
+
"loss": 0.6341354370117187,
|
| 2691 |
+
"step": 3830
|
| 2692 |
+
},
|
| 2693 |
+
{
|
| 2694 |
+
"epoch": 0.49433573635427397,
|
| 2695 |
+
"grad_norm": 1.2123188189670462,
|
| 2696 |
+
"learning_rate": 2e-05,
|
| 2697 |
+
"loss": 0.6724311828613281,
|
| 2698 |
+
"step": 3840
|
| 2699 |
+
},
|
| 2700 |
+
{
|
| 2701 |
+
"epoch": 0.49562306900102987,
|
| 2702 |
+
"grad_norm": 5.71968755913974,
|
| 2703 |
+
"learning_rate": 2e-05,
|
| 2704 |
+
"loss": 0.6578932762145996,
|
| 2705 |
+
"step": 3850
|
| 2706 |
+
},
|
| 2707 |
+
{
|
| 2708 |
+
"epoch": 0.49691040164778577,
|
| 2709 |
+
"grad_norm": 1.667301339852815,
|
| 2710 |
+
"learning_rate": 2e-05,
|
| 2711 |
+
"loss": 0.6530393600463867,
|
| 2712 |
+
"step": 3860
|
| 2713 |
+
},
|
| 2714 |
+
{
|
| 2715 |
+
"epoch": 0.4981977342945417,
|
| 2716 |
+
"grad_norm": 2.2052424082783006,
|
| 2717 |
+
"learning_rate": 2e-05,
|
| 2718 |
+
"loss": 0.6489648342132568,
|
| 2719 |
+
"step": 3870
|
| 2720 |
+
},
|
| 2721 |
+
{
|
| 2722 |
+
"epoch": 0.49948506694129763,
|
| 2723 |
+
"grad_norm": 1.182655371071617,
|
| 2724 |
+
"learning_rate": 2e-05,
|
| 2725 |
+
"loss": 0.7217607498168945,
|
| 2726 |
+
"step": 3880
|
| 2727 |
+
},
|
| 2728 |
+
{
|
| 2729 |
+
"epoch": 0.5007723995880535,
|
| 2730 |
+
"grad_norm": 1.2358308522133377,
|
| 2731 |
+
"learning_rate": 2e-05,
|
| 2732 |
+
"loss": 0.6366837978363037,
|
| 2733 |
+
"step": 3890
|
| 2734 |
+
},
|
| 2735 |
+
{
|
| 2736 |
+
"epoch": 0.5020597322348095,
|
| 2737 |
+
"grad_norm": 2.2520605480700366,
|
| 2738 |
+
"learning_rate": 2e-05,
|
| 2739 |
+
"loss": 0.6712183952331543,
|
| 2740 |
+
"step": 3900
|
| 2741 |
+
},
|
| 2742 |
+
{
|
| 2743 |
+
"epoch": 0.5033470648815654,
|
| 2744 |
+
"grad_norm": 2.151802936233694,
|
| 2745 |
+
"learning_rate": 2e-05,
|
| 2746 |
+
"loss": 0.6303033828735352,
|
| 2747 |
+
"step": 3910
|
| 2748 |
+
},
|
| 2749 |
+
{
|
| 2750 |
+
"epoch": 0.5046343975283213,
|
| 2751 |
+
"grad_norm": 2.4362255545668843,
|
| 2752 |
+
"learning_rate": 2e-05,
|
| 2753 |
+
"loss": 0.6984471321105957,
|
| 2754 |
+
"step": 3920
|
| 2755 |
+
},
|
| 2756 |
+
{
|
| 2757 |
+
"epoch": 0.5059217301750772,
|
| 2758 |
+
"grad_norm": 1.3273430757286633,
|
| 2759 |
+
"learning_rate": 2e-05,
|
| 2760 |
+
"loss": 0.6668586254119873,
|
| 2761 |
+
"step": 3930
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 0.5072090628218332,
|
| 2765 |
+
"grad_norm": 1.4830951454416714,
|
| 2766 |
+
"learning_rate": 2e-05,
|
| 2767 |
+
"loss": 0.6583553314208984,
|
| 2768 |
+
"step": 3940
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 0.508496395468589,
|
| 2772 |
+
"grad_norm": 2.5789130838170666,
|
| 2773 |
+
"learning_rate": 2e-05,
|
| 2774 |
+
"loss": 0.6548712730407715,
|
| 2775 |
+
"step": 3950
|
| 2776 |
+
},
|
| 2777 |
+
{
|
| 2778 |
+
"epoch": 0.509783728115345,
|
| 2779 |
+
"grad_norm": 1.3557484975654484,
|
| 2780 |
+
"learning_rate": 2e-05,
|
| 2781 |
+
"loss": 0.6732818603515625,
|
| 2782 |
+
"step": 3960
|
| 2783 |
+
},
|
| 2784 |
+
{
|
| 2785 |
+
"epoch": 0.511071060762101,
|
| 2786 |
+
"grad_norm": 1.5388518917022493,
|
| 2787 |
+
"learning_rate": 2e-05,
|
| 2788 |
+
"loss": 0.630622386932373,
|
| 2789 |
+
"step": 3970
|
| 2790 |
+
},
|
| 2791 |
+
{
|
| 2792 |
+
"epoch": 0.5123583934088568,
|
| 2793 |
+
"grad_norm": 3.3880763823383995,
|
| 2794 |
+
"learning_rate": 2e-05,
|
| 2795 |
+
"loss": 0.6452838897705078,
|
| 2796 |
+
"step": 3980
|
| 2797 |
+
},
|
| 2798 |
+
{
|
| 2799 |
+
"epoch": 0.5136457260556128,
|
| 2800 |
+
"grad_norm": 1.9510210816666544,
|
| 2801 |
+
"learning_rate": 2e-05,
|
| 2802 |
+
"loss": 0.6481393814086914,
|
| 2803 |
+
"step": 3990
|
| 2804 |
+
},
|
| 2805 |
+
{
|
| 2806 |
+
"epoch": 0.5149330587023687,
|
| 2807 |
+
"grad_norm": 3.3445035332864053,
|
| 2808 |
+
"learning_rate": 2e-05,
|
| 2809 |
+
"loss": 0.6437893867492676,
|
| 2810 |
+
"step": 4000
|
| 2811 |
+
},
|
| 2812 |
+
{
|
| 2813 |
+
"epoch": 0.5162203913491246,
|
| 2814 |
+
"grad_norm": 1.4430061642931242,
|
| 2815 |
+
"learning_rate": 2e-05,
|
| 2816 |
+
"loss": 0.6935883522033691,
|
| 2817 |
+
"step": 4010
|
| 2818 |
+
},
|
| 2819 |
+
{
|
| 2820 |
+
"epoch": 0.5175077239958805,
|
| 2821 |
+
"grad_norm": 1.678957944027874,
|
| 2822 |
+
"learning_rate": 2e-05,
|
| 2823 |
+
"loss": 0.6531391143798828,
|
| 2824 |
+
"step": 4020
|
| 2825 |
+
},
|
| 2826 |
+
{
|
| 2827 |
+
"epoch": 0.5187950566426365,
|
| 2828 |
+
"grad_norm": 2.1307079559087994,
|
| 2829 |
+
"learning_rate": 2e-05,
|
| 2830 |
+
"loss": 0.7015140533447266,
|
| 2831 |
+
"step": 4030
|
| 2832 |
+
},
|
| 2833 |
+
{
|
| 2834 |
+
"epoch": 0.5200823892893924,
|
| 2835 |
+
"grad_norm": 2.1707573190704275,
|
| 2836 |
+
"learning_rate": 2e-05,
|
| 2837 |
+
"loss": 0.6406472682952881,
|
| 2838 |
+
"step": 4040
|
| 2839 |
+
},
|
| 2840 |
+
{
|
| 2841 |
+
"epoch": 0.5213697219361483,
|
| 2842 |
+
"grad_norm": 1.381923517046361,
|
| 2843 |
+
"learning_rate": 2e-05,
|
| 2844 |
+
"loss": 0.689560842514038,
|
| 2845 |
+
"step": 4050
|
| 2846 |
+
},
|
| 2847 |
+
{
|
| 2848 |
+
"epoch": 0.5226570545829042,
|
| 2849 |
+
"grad_norm": 2.171024005846642,
|
| 2850 |
+
"learning_rate": 2e-05,
|
| 2851 |
+
"loss": 0.7308357238769532,
|
| 2852 |
+
"step": 4060
|
| 2853 |
+
},
|
| 2854 |
+
{
|
| 2855 |
+
"epoch": 0.5239443872296602,
|
| 2856 |
+
"grad_norm": 3.6614263601181762,
|
| 2857 |
+
"learning_rate": 2e-05,
|
| 2858 |
+
"loss": 0.6891562938690186,
|
| 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 sha256:04df9f1ca8a7ead8ae1edbc0d37914e1ade4ef28f928d75d90b263f7bb6d370c
|
| 3 |
+
size 7313
|