Upload folder using huggingface_hub
Browse files- config.json +47 -0
- mm_projector.bin +3 -0
- trainer_state.json +444 -0
config.json
ADDED
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "Qwen/Qwen1.5-0.5B",
|
3 |
+
"architectures": [
|
4 |
+
"Qwen2ForCausalLM"
|
5 |
+
],
|
6 |
+
"attention_dropout": 0.0,
|
7 |
+
"bos_token_id": 151643,
|
8 |
+
"eos_token_id": 151643,
|
9 |
+
"freeze_mm_mlp_adapter": false,
|
10 |
+
"hidden_act": "silu",
|
11 |
+
"hidden_size": 1024,
|
12 |
+
"image_aspect_ratio": "square",
|
13 |
+
"image_projector_type": "mlp2x_gelu",
|
14 |
+
"initializer_range": 0.02,
|
15 |
+
"intermediate_size": 2816,
|
16 |
+
"max_position_embeddings": 32768,
|
17 |
+
"max_window_layers": 21,
|
18 |
+
"mm_hidden_size": 768,
|
19 |
+
"mm_image_tower": "google/siglip-base-patch16-256-multilingual",
|
20 |
+
"mm_projector_lr": null,
|
21 |
+
"mm_use_im_patch_token": false,
|
22 |
+
"mm_use_im_start_end": false,
|
23 |
+
"mm_video_tower": null,
|
24 |
+
"mm_vision_select_feature": "patch",
|
25 |
+
"mm_vision_select_layer": -2,
|
26 |
+
"model_type": "llava_qwen1_5",
|
27 |
+
"num_attention_heads": 16,
|
28 |
+
"num_hidden_layers": 24,
|
29 |
+
"num_key_value_heads": 16,
|
30 |
+
"pad_token_id": 151646,
|
31 |
+
"rms_norm_eps": 1e-06,
|
32 |
+
"rope_theta": 1000000.0,
|
33 |
+
"sliding_window": 32768,
|
34 |
+
"tie_word_embeddings": true,
|
35 |
+
"tokenizer_padding_side": "right",
|
36 |
+
"torch_dtype": "bfloat16",
|
37 |
+
"transformers_version": "4.37.0",
|
38 |
+
"tune_mm_mlp_adapter": true,
|
39 |
+
"use_cache": true,
|
40 |
+
"use_mm_proj": true,
|
41 |
+
"use_sliding_window": false,
|
42 |
+
"video_global_proj": false,
|
43 |
+
"video_projector_type": "linear",
|
44 |
+
"video_spatial_proj": false,
|
45 |
+
"video_temproal_proj": false,
|
46 |
+
"vocab_size": 151936
|
47 |
+
}
|
mm_projector.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:44c81d2ae27b6d633ca661d0b0c2457645234188d507ef2680d0ceacd4084b5f
|
3 |
+
size 3675837
|
trainer_state.json
ADDED
@@ -0,0 +1,444 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 3.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 2085,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.04,
|
13 |
+
"learning_rate": 0.00014354066985645933,
|
14 |
+
"loss": 4.6999,
|
15 |
+
"step": 30
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.09,
|
19 |
+
"learning_rate": 0.00028708133971291867,
|
20 |
+
"loss": 4.0234,
|
21 |
+
"step": 60
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.13,
|
25 |
+
"learning_rate": 0.00043062200956937803,
|
26 |
+
"loss": 3.84,
|
27 |
+
"step": 90
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 0.17,
|
31 |
+
"learning_rate": 0.0005741626794258373,
|
32 |
+
"loss": 3.7458,
|
33 |
+
"step": 120
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 0.22,
|
37 |
+
"learning_rate": 0.0007177033492822966,
|
38 |
+
"loss": 3.6694,
|
39 |
+
"step": 150
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 0.26,
|
43 |
+
"learning_rate": 0.0008612440191387561,
|
44 |
+
"loss": 3.6173,
|
45 |
+
"step": 180
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 0.3,
|
49 |
+
"learning_rate": 0.000999999298909658,
|
50 |
+
"loss": 3.5916,
|
51 |
+
"step": 210
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.35,
|
55 |
+
"learning_rate": 0.0009993264033223551,
|
56 |
+
"loss": 3.5461,
|
57 |
+
"step": 240
|
58 |
+
},
|
59 |
+
{
|
60 |
+
"epoch": 0.39,
|
61 |
+
"learning_rate": 0.0009973935099776767,
|
62 |
+
"loss": 3.5042,
|
63 |
+
"step": 270
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"epoch": 0.43,
|
67 |
+
"learning_rate": 0.0009942054963290549,
|
68 |
+
"loss": 3.4956,
|
69 |
+
"step": 300
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 0.47,
|
73 |
+
"learning_rate": 0.0009897704069942402,
|
74 |
+
"loss": 3.5649,
|
75 |
+
"step": 330
|
76 |
+
},
|
77 |
+
{
|
78 |
+
"epoch": 0.52,
|
79 |
+
"learning_rate": 0.0009840994334555549,
|
80 |
+
"loss": 3.5746,
|
81 |
+
"step": 360
|
82 |
+
},
|
83 |
+
{
|
84 |
+
"epoch": 0.56,
|
85 |
+
"learning_rate": 0.0009772068858193608,
|
86 |
+
"loss": 3.5424,
|
87 |
+
"step": 390
|
88 |
+
},
|
89 |
+
{
|
90 |
+
"epoch": 0.6,
|
91 |
+
"learning_rate": 0.000969110156706009,
|
92 |
+
"loss": 3.567,
|
93 |
+
"step": 420
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.65,
|
97 |
+
"learning_rate": 0.0009598296773613879,
|
98 |
+
"loss": 3.4753,
|
99 |
+
"step": 450
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 0.69,
|
103 |
+
"learning_rate": 0.0009493888661008194,
|
104 |
+
"loss": 3.4324,
|
105 |
+
"step": 480
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 0.73,
|
109 |
+
"learning_rate": 0.0009378140692153992,
|
110 |
+
"loss": 3.4661,
|
111 |
+
"step": 510
|
112 |
+
},
|
113 |
+
{
|
114 |
+
"epoch": 0.78,
|
115 |
+
"learning_rate": 0.0009251344944898958,
|
116 |
+
"loss": 3.4081,
|
117 |
+
"step": 540
|
118 |
+
},
|
119 |
+
{
|
120 |
+
"epoch": 0.82,
|
121 |
+
"learning_rate": 0.0009113821374999736,
|
122 |
+
"loss": 3.3848,
|
123 |
+
"step": 570
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"epoch": 0.86,
|
127 |
+
"learning_rate": 0.0008965917008747158,
|
128 |
+
"loss": 3.3552,
|
129 |
+
"step": 600
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 0.91,
|
133 |
+
"learning_rate": 0.000880800506728183,
|
134 |
+
"loss": 3.3352,
|
135 |
+
"step": 630
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.95,
|
139 |
+
"learning_rate": 0.0008640484024809779,
|
140 |
+
"loss": 3.2935,
|
141 |
+
"step": 660
|
142 |
+
},
|
143 |
+
{
|
144 |
+
"epoch": 0.99,
|
145 |
+
"learning_rate": 0.0008463776603094606,
|
146 |
+
"loss": 3.2754,
|
147 |
+
"step": 690
|
148 |
+
},
|
149 |
+
{
|
150 |
+
"epoch": 1.04,
|
151 |
+
"learning_rate": 0.0008278328704763516,
|
152 |
+
"loss": 3.2657,
|
153 |
+
"step": 720
|
154 |
+
},
|
155 |
+
{
|
156 |
+
"epoch": 1.08,
|
157 |
+
"learning_rate": 0.0008084608288118838,
|
158 |
+
"loss": 3.256,
|
159 |
+
"step": 750
|
160 |
+
},
|
161 |
+
{
|
162 |
+
"epoch": 1.12,
|
163 |
+
"learning_rate": 0.0007883104186294383,
|
164 |
+
"loss": 3.2201,
|
165 |
+
"step": 780
|
166 |
+
},
|
167 |
+
{
|
168 |
+
"epoch": 1.17,
|
169 |
+
"learning_rate": 0.0007674324873736347,
|
170 |
+
"loss": 3.1941,
|
171 |
+
"step": 810
|
172 |
+
},
|
173 |
+
{
|
174 |
+
"epoch": 1.21,
|
175 |
+
"learning_rate": 0.0007458797183121429,
|
176 |
+
"loss": 3.1856,
|
177 |
+
"step": 840
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 1.25,
|
181 |
+
"learning_rate": 0.0007237064975949886,
|
182 |
+
"loss": 3.1649,
|
183 |
+
"step": 870
|
184 |
+
},
|
185 |
+
{
|
186 |
+
"epoch": 1.29,
|
187 |
+
"learning_rate": 0.0007009687770168125,
|
188 |
+
"loss": 3.155,
|
189 |
+
"step": 900
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 1.34,
|
193 |
+
"learning_rate": 0.0006777239328283909,
|
194 |
+
"loss": 3.1357,
|
195 |
+
"step": 930
|
196 |
+
},
|
197 |
+
{
|
198 |
+
"epoch": 1.38,
|
199 |
+
"learning_rate": 0.0006540306209536906,
|
200 |
+
"loss": 3.1309,
|
201 |
+
"step": 960
|
202 |
+
},
|
203 |
+
{
|
204 |
+
"epoch": 1.42,
|
205 |
+
"learning_rate": 0.0006299486289778033,
|
206 |
+
"loss": 3.1091,
|
207 |
+
"step": 990
|
208 |
+
},
|
209 |
+
{
|
210 |
+
"epoch": 1.47,
|
211 |
+
"learning_rate": 0.0006055387252792498,
|
212 |
+
"loss": 3.0958,
|
213 |
+
"step": 1020
|
214 |
+
},
|
215 |
+
{
|
216 |
+
"epoch": 1.51,
|
217 |
+
"learning_rate": 0.0005808625056873581,
|
218 |
+
"loss": 3.0813,
|
219 |
+
"step": 1050
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 1.55,
|
223 |
+
"learning_rate": 0.0005559822380516539,
|
224 |
+
"loss": 3.0854,
|
225 |
+
"step": 1080
|
226 |
+
},
|
227 |
+
{
|
228 |
+
"epoch": 1.6,
|
229 |
+
"learning_rate": 0.0005309607051154799,
|
230 |
+
"loss": 3.068,
|
231 |
+
"step": 1110
|
232 |
+
},
|
233 |
+
{
|
234 |
+
"epoch": 1.64,
|
235 |
+
"learning_rate": 0.0005058610460903332,
|
236 |
+
"loss": 3.0664,
|
237 |
+
"step": 1140
|
238 |
+
},
|
239 |
+
{
|
240 |
+
"epoch": 1.68,
|
241 |
+
"learning_rate": 0.00048074659733069516,
|
242 |
+
"loss": 3.0627,
|
243 |
+
"step": 1170
|
244 |
+
},
|
245 |
+
{
|
246 |
+
"epoch": 1.73,
|
247 |
+
"learning_rate": 0.00045568073251138903,
|
248 |
+
"loss": 3.0376,
|
249 |
+
"step": 1200
|
250 |
+
},
|
251 |
+
{
|
252 |
+
"epoch": 1.77,
|
253 |
+
"learning_rate": 0.0004307267027107653,
|
254 |
+
"loss": 3.0274,
|
255 |
+
"step": 1230
|
256 |
+
},
|
257 |
+
{
|
258 |
+
"epoch": 1.81,
|
259 |
+
"learning_rate": 0.00040594747680324574,
|
260 |
+
"loss": 3.0212,
|
261 |
+
"step": 1260
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 1.86,
|
265 |
+
"learning_rate": 0.0003814055825639795,
|
266 |
+
"loss": 3.0101,
|
267 |
+
"step": 1290
|
268 |
+
},
|
269 |
+
{
|
270 |
+
"epoch": 1.9,
|
271 |
+
"learning_rate": 0.000357162948886567,
|
272 |
+
"loss": 3.0076,
|
273 |
+
"step": 1320
|
274 |
+
},
|
275 |
+
{
|
276 |
+
"epoch": 1.94,
|
277 |
+
"learning_rate": 0.00033328074951199845,
|
278 |
+
"loss": 3.0129,
|
279 |
+
"step": 1350
|
280 |
+
},
|
281 |
+
{
|
282 |
+
"epoch": 1.99,
|
283 |
+
"learning_rate": 0.0003098192486631408,
|
284 |
+
"loss": 2.9902,
|
285 |
+
"step": 1380
|
286 |
+
},
|
287 |
+
{
|
288 |
+
"epoch": 2.03,
|
289 |
+
"learning_rate": 0.00028683764897429804,
|
290 |
+
"loss": 2.9819,
|
291 |
+
"step": 1410
|
292 |
+
},
|
293 |
+
{
|
294 |
+
"epoch": 2.07,
|
295 |
+
"learning_rate": 0.0002643939420995788,
|
296 |
+
"loss": 2.9891,
|
297 |
+
"step": 1440
|
298 |
+
},
|
299 |
+
{
|
300 |
+
"epoch": 2.12,
|
301 |
+
"learning_rate": 0.00024254476237704588,
|
302 |
+
"loss": 2.9702,
|
303 |
+
"step": 1470
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 2.16,
|
307 |
+
"learning_rate": 0.00022134524391790916,
|
308 |
+
"loss": 2.972,
|
309 |
+
"step": 1500
|
310 |
+
},
|
311 |
+
{
|
312 |
+
"epoch": 2.2,
|
313 |
+
"learning_rate": 0.00020084888148138487,
|
314 |
+
"loss": 2.9733,
|
315 |
+
"step": 1530
|
316 |
+
},
|
317 |
+
{
|
318 |
+
"epoch": 2.24,
|
319 |
+
"learning_rate": 0.00018110739548628618,
|
320 |
+
"loss": 2.9688,
|
321 |
+
"step": 1560
|
322 |
+
},
|
323 |
+
{
|
324 |
+
"epoch": 2.29,
|
325 |
+
"learning_rate": 0.0001621706014999767,
|
326 |
+
"loss": 2.9629,
|
327 |
+
"step": 1590
|
328 |
+
},
|
329 |
+
{
|
330 |
+
"epoch": 2.33,
|
331 |
+
"learning_rate": 0.00014408628453401574,
|
332 |
+
"loss": 2.9609,
|
333 |
+
"step": 1620
|
334 |
+
},
|
335 |
+
{
|
336 |
+
"epoch": 2.37,
|
337 |
+
"learning_rate": 0.00012690007846369856,
|
338 |
+
"loss": 2.9461,
|
339 |
+
"step": 1650
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 2.42,
|
343 |
+
"learning_rate": 0.00011065535087576234,
|
344 |
+
"loss": 2.9493,
|
345 |
+
"step": 1680
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 2.46,
|
349 |
+
"learning_rate": 9.539309363483478e-05,
|
350 |
+
"loss": 2.9417,
|
351 |
+
"step": 1710
|
352 |
+
},
|
353 |
+
{
|
354 |
+
"epoch": 2.5,
|
355 |
+
"learning_rate": 8.115181944476685e-05,
|
356 |
+
"loss": 2.9459,
|
357 |
+
"step": 1740
|
358 |
+
},
|
359 |
+
{
|
360 |
+
"epoch": 2.55,
|
361 |
+
"learning_rate": 6.796746466586756e-05,
|
362 |
+
"loss": 2.9273,
|
363 |
+
"step": 1770
|
364 |
+
},
|
365 |
+
{
|
366 |
+
"epoch": 2.59,
|
367 |
+
"learning_rate": 5.58732986332719e-05,
|
368 |
+
"loss": 2.9397,
|
369 |
+
"step": 1800
|
370 |
+
},
|
371 |
+
{
|
372 |
+
"epoch": 2.63,
|
373 |
+
"learning_rate": 4.4899839705266174e-05,
|
374 |
+
"loss": 2.943,
|
375 |
+
"step": 1830
|
376 |
+
},
|
377 |
+
{
|
378 |
+
"epoch": 2.68,
|
379 |
+
"learning_rate": 3.507477825341493e-05,
|
380 |
+
"loss": 2.9381,
|
381 |
+
"step": 1860
|
382 |
+
},
|
383 |
+
{
|
384 |
+
"epoch": 2.72,
|
385 |
+
"learning_rate": 2.642290678881504e-05,
|
386 |
+
"loss": 2.9509,
|
387 |
+
"step": 1890
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 2.76,
|
391 |
+
"learning_rate": 1.8966057400797153e-05,
|
392 |
+
"loss": 2.9437,
|
393 |
+
"step": 1920
|
394 |
+
},
|
395 |
+
{
|
396 |
+
"epoch": 2.81,
|
397 |
+
"learning_rate": 1.272304666594032e-05,
|
398 |
+
"loss": 2.9374,
|
399 |
+
"step": 1950
|
400 |
+
},
|
401 |
+
{
|
402 |
+
"epoch": 2.85,
|
403 |
+
"learning_rate": 7.709628166416128e-06,
|
404 |
+
"loss": 2.9508,
|
405 |
+
"step": 1980
|
406 |
+
},
|
407 |
+
{
|
408 |
+
"epoch": 2.89,
|
409 |
+
"learning_rate": 3.93845273747806e-06,
|
410 |
+
"loss": 2.9407,
|
411 |
+
"step": 2010
|
412 |
+
},
|
413 |
+
{
|
414 |
+
"epoch": 2.94,
|
415 |
+
"learning_rate": 1.4190365444071153e-06,
|
416 |
+
"loss": 2.949,
|
417 |
+
"step": 2040
|
418 |
+
},
|
419 |
+
{
|
420 |
+
"epoch": 2.98,
|
421 |
+
"learning_rate": 1.5773706946742295e-07,
|
422 |
+
"loss": 2.9544,
|
423 |
+
"step": 2070
|
424 |
+
},
|
425 |
+
{
|
426 |
+
"epoch": 3.0,
|
427 |
+
"step": 2085,
|
428 |
+
"total_flos": 1.731497869485015e+17,
|
429 |
+
"train_loss": 3.211610127181458,
|
430 |
+
"train_runtime": 4482.5265,
|
431 |
+
"train_samples_per_second": 74.379,
|
432 |
+
"train_steps_per_second": 0.465
|
433 |
+
}
|
434 |
+
],
|
435 |
+
"logging_steps": 30,
|
436 |
+
"max_steps": 2085,
|
437 |
+
"num_input_tokens_seen": 0,
|
438 |
+
"num_train_epochs": 3,
|
439 |
+
"save_steps": 5000,
|
440 |
+
"total_flos": 1.731497869485015e+17,
|
441 |
+
"train_batch_size": 16,
|
442 |
+
"trial_name": null,
|
443 |
+
"trial_params": null
|
444 |
+
}
|