Training in progress, step 50000

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+blimp_results.json filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1 @@
+checkpoint-*/

blimp_results.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7ab61a9f5abc94daaace067debc70a12e7ab5b883f04303561040d26ae2b618
+size 80217640

config.json

@@ -0,0 +1,31 @@
+{
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 0,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 0,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_embd": 512,
+  "n_head": 8,
+  "n_inner": 2048,
+  "n_layer": 6,
+  "n_positions": 256,
+  "reorder_and_upcast_attn": false,
+  "resid_pdrop": 0.1,
+  "scale_attn_by_inverse_layer_idx": false,
+  "scale_attn_weights": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.1,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.44.2",
+  "use_cache": true,
+  "vocab_size": 16000
+}

hydra_config_1724868015.7051077.yaml

@@ -0,0 +1,41 @@
+experiment:
+  seed: 42
+  name: gpt2_19M-bpe-text
+  group: babylm-small
+  dry_run: false
+  offline_run: false
+  evaluate_segmentation: false
+  evaluate_babyslm: true
+  blimp_tasks: blimp_filtered,blimp_supplement
+  resume_checkpoint_path: null
+  resume_run_id: null
+dataset:
+  name: transformersegmentation/BabyLM-phonemized
+  subconfig: strict_small
+  text_column: text
+  is_phonemes: false
+  max_age: null
+tokenizer:
+  name: transformersegmentation/BabyLM-BPE-ortho-tokenizer
+data_preprocessing:
+  max_input_length: 128
+  join_utts: static
+  remove_word_boundaries: false
+  subsample: null
+  subsample_type: examples
+model:
+  name: gpt2_lm
+  model_kwargs:
+    n_layer: 6
+    n_head: 8
+    n_embd: 512
+    n_positions: 256
+    n_inner: 2048
+trainer:
+  batch_size: 32
+  lr: 0.001
+  num_warmup_steps: 90000
+  max_training_steps: 400000
+  logging_steps: 4000
+  save_steps: 50000
+  eval_steps: 50000

hydra_config_1724870633.4317997.yaml

@@ -0,0 +1,41 @@
+experiment:
+  seed: 42
+  name: gpt2_19M-bpe-text
+  group: babylm-small
+  dry_run: false
+  offline_run: false
+  evaluate_segmentation: false
+  evaluate_babyslm: true
+  blimp_tasks: blimp_filtered,blimp_supplement
+  resume_checkpoint_path: null
+  resume_run_id: null
+dataset:
+  name: transformersegmentation/BabyLM-phonemized
+  subconfig: strict_small
+  text_column: text
+  is_phonemes: false
+  max_age: null
+tokenizer:
+  name: transformersegmentation/BabyLM-BPE-ortho-tokenizer
+data_preprocessing:
+  max_input_length: 128
+  join_utts: static
+  remove_word_boundaries: false
+  subsample: null
+  subsample_type: examples
+model:
+  name: gpt2_lm
+  model_kwargs:
+    n_layer: 6
+    n_head: 8
+    n_embd: 512
+    n_positions: 256
+    n_inner: 2048
+trainer:
+  batch_size: 32
+  lr: 0.001
+  num_warmup_steps: 90000
+  max_training_steps: 400000
+  logging_steps: 4000
+  save_steps: 50000
+  eval_steps: 50000

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37fc397e0772db251f34de842915546f79cfa9e25a4d75313f8bcc56a1f7564c
+size 108961160

special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "UTT_BOUNDARY",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "UTT_BOUNDARY",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "PAD",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "UNK",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,36 @@
+{
+  "add_prefix_space": true,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "UTT_BOUNDARY",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "PAD",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "UNK",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "UTT_BOUNDARY",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "UTT_BOUNDARY",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "PAD",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "UNK"
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:975f4c8fc39f5136ba79137eff2767d6095d8cadb172037c87f30b58b803cd77
+size 5368