Safetensors
English
gpt2
codebyzeb commited on
Commit
5a01be3
·
verified ·
1 Parent(s): 08f87f4

Training in progress, step 200000

Browse files
babyslm/lexical.txt CHANGED
The diff for this file is too large to render. See raw diff
 
babyslm/syntactic.txt CHANGED
The diff for this file is too large to render. See raw diff
 
config.json CHANGED
@@ -3,9 +3,9 @@
3
  "architectures": [
4
  "GPT2LMHeadModel"
5
  ],
6
- "attn_pdrop": 0.1,
7
  "bos_token_id": 3,
8
- "embd_pdrop": 0.1,
9
  "eos_token_id": 3,
10
  "initializer_range": 0.02,
11
  "layer_norm_epsilon": 1e-05,
@@ -16,7 +16,7 @@
16
  "n_layer": 2,
17
  "n_positions": 256,
18
  "reorder_and_upcast_attn": false,
19
- "resid_pdrop": 0.1,
20
  "scale_attn_by_inverse_layer_idx": false,
21
  "scale_attn_weights": true,
22
  "summary_activation": null,
 
3
  "architectures": [
4
  "GPT2LMHeadModel"
5
  ],
6
+ "attn_pdrop": 0.5,
7
  "bos_token_id": 3,
8
+ "embd_pdrop": 0.5,
9
  "eos_token_id": 3,
10
  "initializer_range": 0.02,
11
  "layer_norm_epsilon": 1e-05,
 
16
  "n_layer": 2,
17
  "n_positions": 256,
18
  "reorder_and_upcast_attn": false,
19
+ "resid_pdrop": 0.5,
20
  "scale_attn_by_inverse_layer_idx": false,
21
  "scale_attn_weights": true,
22
  "summary_activation": null,
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:50384447bc385fb5a83e8388dc1ee7c001e8c219eb54d2ebc6a5e9c9a68e287f
3
  size 1745032
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:24cce080cb1adffa6e94a9d54bf7d59515781faea628eea2dc5db0ec6a5c9b71
3
  size 1745032
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b9fe3750340ba4f9feaeb9a0db18f71b38a9ec94e6cbcdc2daec7a5bd46aa75a
3
  size 5368
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8f7264bba53c141756f93e3370f9bfa26d88b3d5c0ad14c1a37f49a27414a3f9
3
  size 5368