0x22almostevil's workspace
Runs
348
State
Notes
User
Tags
Created
Runtime
Sweep
_name_or_path
_remove_final_layer_norm
activation_dropout
activation_function
adafactor
adam_beta1
adam_beta2
adam_epsilon
add_cross_attention
apply_residual_connection_post_layernorm
architectures
attention_dropout
attention_softmax_in_fp32
attn_pdrop
auto_find_batch_size
bf16
bf16_full_eval
bias_dropout_fusion
bos_token_id
chunk_size_feed_forward
data_seed
dataloader_drop_last
dataloader_num_workers
dataloader_pin_memory
ddp_bucket_cap_mb
ddp_find_unused_parameters
ddp_timeout
debug
deepspeed
disable_tqdm
diversity_penalty
do_eval
do_layer_norm_before
do_predict
do_sample
do_train
dropout
early_stopping
embd_pdrop
enable_bias
encoder_no_repeat_ngram_size
eos_token_id
eval_accumulation_steps
eval_batch_size
Crashed
jordanclive
1d 20h 6m 8s
-
decapoda-research/llama-65b-hf
-
-
-
false
0.9
0.95
1.0000e-12
false
-
["LLaMAForCausalLM"]
-
-
-
false
false
false
-
0
0
None
false
0
true
None
None
1800
[]
/admin/home-jordiclive/Open-Assistant/model/model_training/configs/zero3_config_sft.json
false
0
true
-
false
false
false
-
false
-
-
0
1
None
1
Crashed
andreaskoepf
llama
13h 50m 36s
-
/home/ubuntu/Open-Assistant/model/model_training/.saved/llama-30b-super-pretrain/checkpoint-3500
-
-
-
false
0.9
0.95
1.0000e-12
false
-
["LlamaForCausalLM"]
-
-
-
false
false
false
-
1
0
None
false
0
true
None
None
1800
[]
configs/zero3_config_sft.json
false
0
true
-
false
false
false
-
false
-
-
0
2
None
3
1-2
of 2