RyanYr commited on
Commit
b4e679d
·
verified ·
1 Parent(s): 431de5f

Save model at global step 105

Browse files
config.yaml CHANGED
@@ -61,7 +61,7 @@ actor_rollout_ref:
61
  lr_warmup_steps_ratio: 0.0
62
  min_lr_ratio: null
63
  warmup_style: constant
64
- total_training_steps: 100
65
  weight_decay: 0.01
66
  fsdp_config:
67
  wrap_policy:
@@ -122,7 +122,7 @@ critic:
122
  lr_warmup_steps_ratio: 0.0
123
  min_lr_ratio: null
124
  warmup_style: constant
125
- total_training_steps: 100
126
  weight_decay: 0.01
127
  model:
128
  path: meta-llama/Llama-3.2-3B-Instruct
@@ -193,7 +193,7 @@ algorithm:
193
  trainer:
194
  balance_batch: true
195
  total_epochs: 100
196
- total_training_steps: 100
197
  project_name: value-LLM
198
  experiment_name: ppo-dapo-llama3.2-3B-it-lr-bs1024-mbs256
199
  logger:
@@ -218,7 +218,7 @@ trainer:
218
  resume_from_hf:
219
  enable: true
220
  actor_hf_repo_id: RyanYr/ppo-dapo-llama3.2-3B-it-lr-bs1024-mbs256_actor
221
- actor_revision: 3c65e857877fc1a893a68f52174daa239a531c2b
222
  critic_hf_repo_id: RyanYr/ppo-dapo-llama3.2-3B-it-lr-bs1024-mbs256_critic
223
- critic_revision: 3ebb878032e65a6df7ee109985909a9009ff00a8
224
  hf_token: null
 
61
  lr_warmup_steps_ratio: 0.0
62
  min_lr_ratio: null
63
  warmup_style: constant
64
+ total_training_steps: 200
65
  weight_decay: 0.01
66
  fsdp_config:
67
  wrap_policy:
 
122
  lr_warmup_steps_ratio: 0.0
123
  min_lr_ratio: null
124
  warmup_style: constant
125
+ total_training_steps: 200
126
  weight_decay: 0.01
127
  model:
128
  path: meta-llama/Llama-3.2-3B-Instruct
 
193
  trainer:
194
  balance_batch: true
195
  total_epochs: 100
196
+ total_training_steps: 200
197
  project_name: value-LLM
198
  experiment_name: ppo-dapo-llama3.2-3B-it-lr-bs1024-mbs256
199
  logger:
 
218
  resume_from_hf:
219
  enable: true
220
  actor_hf_repo_id: RyanYr/ppo-dapo-llama3.2-3B-it-lr-bs1024-mbs256_actor
221
+ actor_revision: ea0afbb1bdbd52f362714b38917a2a8aa44b1193
222
  critic_hf_repo_id: RyanYr/ppo-dapo-llama3.2-3B-it-lr-bs1024-mbs256_critic
223
+ critic_revision: 431de5f1de160e60af68af4b0829ffe1f041be21
224
  hf_token: null
extra_state_world_size_4_rank_0.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a7b62687971ab92b757ac32cf15fb4870b2251ef3ce8f98a82c5098b081339e2
3
  size 14632
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b997bc838bd4ab9241c7f47aad59c5e3557197232c3861495edbd2e063f12af8
3
  size 14632
extra_state_world_size_4_rank_1.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2c59853d85f02215339f9dd618baace19da8212943cdfd670e10f29320d98a47
3
  size 14632
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a2339072c15749307da3ff25dd8c1a13b5e202c75d6786d32bd5c855efc7b8a5
3
  size 14632
extra_state_world_size_4_rank_2.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:eb9a2cb57ea286359b8897910ea60e4f9a4e92082452f9095722d6f5fa413427
3
  size 14632
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1f0d0f28ab9e9f6049827a3a88ab60becb6659addef59a8ba4f71f0be0f195a1
3
  size 14632
extra_state_world_size_4_rank_3.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:003d904487eaf87b9326b1708181314da8ba8e2ca5e7acea66c27da0ba0f1c6a
3
  size 14632
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:133bf50a541947ff52c35ddcc3899c3343d303719b8e2e8339f5913cb51e7b8d
3
  size 14632
model_world_size_4_rank_0.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:92e82238f8df561cae1c4924746ce2f1ed60169d8590d0858fb45137bae711df
3
  size 3212915290
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0570adab62350fa4271a71052e61a8491f8c1a61ccbb2aefd89bac4156097127
3
  size 3212915290
model_world_size_4_rank_1.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5b4c7822054999cf85347df761b799dc7cdb961a65774b5ba68f53bba7bab0ab
3
  size 3212903002
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4f1c99f08566c14bb7566106d4fffac48d9189418ac82846538737b3a6866517
3
  size 3212903002
model_world_size_4_rank_2.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:139c27fca971659b8fc19c193a1a1b9d735935683560a198667ddad6c17ca7d3
3
  size 3212903002
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:764aec12567c84c4ace83f46a8a0f465b5e231a35919490fa562282cff91cbe5
3
  size 3212903002
model_world_size_4_rank_3.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:16456afa161930eb679cabb7583de245042fdfb093e64688c59bb08215fd5b96
3
  size 3212903002
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b20f8ff3f2e37ad8133b3a54cfb1091b56d192a4c92284dba0b762333f0421a
3
  size 3212903002
optim_world_size_4_rank_0.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:fa458d547538bfae60adc31425b957a328ad2d8a1927c27c2c14897fa04a4a4d
3
  size 6425532375
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23b2d3d57460afe6d47897fda6f1f718a6844ab01e7b08dc2b14adbd0c874977
3
  size 6425532375
optim_world_size_4_rank_1.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9773ae2b229fa50c13bfbe193eeaf8e7d030db938f45ff408de16622e4587478
3
  size 6425532375
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0bb91d67c3653fc876857f9d0982da447d06f4e2d0afbaf11e1515ae0b4289c0
3
  size 6425532375
optim_world_size_4_rank_2.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2674a2897bc84877242bf2edbed862d486a5c9c9183bdc41e405ff8a7064b931
3
  size 6425532375
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:20c3780af420df59396622c84c35242f9f800451a6484aeb49b48ceda1ad355b
3
  size 6425532375
optim_world_size_4_rank_3.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5d991df47ee005426325d29813bfcb3a843dff57ba9981e1b343ed917450df87
3
  size 6425532375
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ae73c2c8b387c38e7d16704e25b0025f07d77398668327b29e97b1dfceb39bb2
3
  size 6425532375