Fetching metadata from the HF Docker repository...
default model changed
4fecbfe
-
cerebras
btlm and falcon monkey patches for flash attn (#566)
-
code-llama
Fix Codellama examples (#582)
-
falcon
Add wandb_entity to wandb options, update example configs, update README (#361)
-
gptj
Add wandb_entity to wandb options, update example configs, update README (#361)
-
jeopardy-bot
Add wandb_entity to wandb options, update example configs, update README (#361)
-
llama-2
default model changed
-
mpt-7b
Add wandb_entity to wandb options, update example configs, update README (#361)
-
openllama-3b
more sane defaults for openllama 3b used for quickstarts (#602)
-
phi
make phi training work with Loras (#588)
-
pythia-12b
pad_to_worst_case_seq_len boolean, for testing memory limits (#498)
-
pythia
Add wandb_entity to wandb options, update example configs, update README (#361)
-
redpajama
Add wandb_entity to wandb options, update example configs, update README (#361)
-
replit-3b
Add wandb_entity to wandb options, update example configs, update README (#361)
-
xgen-7b
Add wandb_entity to wandb options, update example configs, update README (#361)
