Nonenman's workspace
Runs
142
Name
142 visualized
Group
Llama-2-70b-chat-hf
mlm_bipar
-
State
Notes
Created
Runtime
GPU Type
GPU Count
eval/avg_loss
eval/exact_match
eval/f1
train/avg_loss
train/loss
train/total_runtime
train/samples_per_sec
batch_size
learning_rate
num_train_epochs
seq_length
warmup_proportion
weight_decay
mlm_probability
wwm_probability
data
do_sample
format_string
max_new_tokens
num_beams
prompt
prompt_format
temperature
top_p
model/hidden_size
model/max_position_embeddings
model/num_attention_heads
model/num_hidden_layers
model/vocab_size
model/num_parameters
eval/perplexity
eval/i_dont_know_counter
eval/precision
eval/recall
prompt_format
train/counter
eval/nr_unanswered_questions
Finished
1mo 19d 16h 43m 6s
-
-
-
21.04049
44.34839
-
-
152.45235
-
-
-
-
-
-
-
-
-
["dev","test"]
[false,true]
-
289.36471
1.15294
["1-shot","2-shot","3-shot","4-shot","zero-shot"]
["#{context}#\n\nQ: #{question}#","Answer a question about this article:\n{context}\n{question}","Article: {context}\n\nNow answer this question: {question}","Article: {context}\n\nQuestion: {question}","Here is a question about this article: {context} What is the answer to this question: {question}","Passage: #{context}#\nQuestion: #{question}#","Please answer a question about the following article:\n\n{context}\n\n{question}","Read this and answer the question\n\n{context}\n\n{question}","{context}\n\nQ: {question}","{context}\n{question}"]
0.10933
0.94933
8192
4096
64
80
32000
-
-
-
40.1475
71.81244
-
1500
26.14286
Finished
3mo 16d 19h 19m 15s
-
-
1.95825
-
-
1.60183
1.59587
229.32349
37.93283
30.85714
0.00002
38.57143
384
0.06
0.01
0.3
-
-
-
-
-
-
-
-
-
-
3218.28571
512.57143
50.28571
15.42857
35790
248952690.57143
79.41563
-
-
-
-
-
-
Finished
10mo 8d 16h 33m 49s
-
-
1.81111
60.71652
73.30839
0.71349
0.7005
108.54315
56.94276
20.96
0.0000204
4.04
394.24
0.0968
0.01
-
-
-
-
-
-
-
-
-
-
-
1510.4
512.76
22.64
18.96
37379.42
-
-
21
-
-
-
-
-
1-3
of 3