Upload checkpoint at step 0
Browse files- .gitattributes +18 -0
- 000000000000/config.yaml +50 -0
- 000000000000/metadata.json +4 -0
- 000000000000/params/_CHECKPOINT_METADATA +1 -0
- 000000000000/params/_METADATA +1 -0
- 000000000000/params/_sharding +1 -0
- 000000000000/params/array_metadatas/process_0 +1 -0
- 000000000000/params/commit_success.txt +1 -0
- 000000000000/params/d/f8d62a6166aa34f52910ff7c504563cf +0 -0
- 000000000000/params/manifest.ocdbt +0 -0
- 000000000000/params/ocdbt.process_0/d/071bd40b12e1dc6207fa7c32dbe308e6 +0 -0
- 000000000000/params/ocdbt.process_0/d/08b75220440bf9acdecb21e5fb7d035e +3 -0
- 000000000000/params/ocdbt.process_0/d/0c9500ecd3e503b0099ae3dfeb2019b6 +3 -0
- 000000000000/params/ocdbt.process_0/d/22b93714ba257459d8524d37e5aeec72 +0 -0
- 000000000000/params/ocdbt.process_0/d/329ba4a9a13260788168a1a3b29cea8c +3 -0
- 000000000000/params/ocdbt.process_0/d/3a6b6c3087fca94e4995f90612c62a6a +3 -0
- 000000000000/params/ocdbt.process_0/d/507ad150d3ebf8f32b61c1bc853d146d +3 -0
- 000000000000/params/ocdbt.process_0/d/55fa9000bd9670f087175e6ad068c48e +0 -0
- 000000000000/params/ocdbt.process_0/d/5fec189e6c6348d35371ac411c8f2333 +3 -0
- 000000000000/params/ocdbt.process_0/d/661205def231b675efda6211178e4630 +3 -0
- 000000000000/params/ocdbt.process_0/d/67df79673a35bae28dd2b0e7b408e5c3 +3 -0
- 000000000000/params/ocdbt.process_0/d/79c0c657bf6d113301c8cb116e593bf7 +0 -0
- 000000000000/params/ocdbt.process_0/d/7bd1fcb396b67403ecd7e62d1b2feed5 +3 -0
- 000000000000/params/ocdbt.process_0/d/a4f75d6fa1268e96f8084d20cff66b4f +3 -0
- 000000000000/params/ocdbt.process_0/d/b2bf95bd2b65ad4d7d9b7d63a1c88d92 +0 -0
- 000000000000/params/ocdbt.process_0/d/c33e33823f431f64336374b63555bca4 +3 -0
- 000000000000/params/ocdbt.process_0/d/c7680fa61e911739fa0316fc46181310 +3 -0
- 000000000000/params/ocdbt.process_0/d/c7b5695f4d973e391ceee570e0c71300 +3 -0
- 000000000000/params/ocdbt.process_0/d/cf579c9c925a205e2cc5dcb2ce1b4c53 +3 -0
- 000000000000/params/ocdbt.process_0/d/d4667bdacaa486400449bbaf82d5f629 +3 -0
- 000000000000/params/ocdbt.process_0/d/e0350adbadace8facf287d91a39aaad8 +3 -0
- 000000000000/params/ocdbt.process_0/d/e2d6fc3564ecc9ab4286abf2c21a99d8 +0 -0
- 000000000000/params/ocdbt.process_0/d/e56681c3fecab2f46ff34a6478f36b12 +3 -0
- 000000000000/params/ocdbt.process_0/d/f8dae51fd86c9eba8e1b5892d351ef17 +3 -0
- 000000000000/params/ocdbt.process_0/manifest.ocdbt +0 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,21 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
000000000000/params/ocdbt.process_0/d/08b75220440bf9acdecb21e5fb7d035e filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
000000000000/params/ocdbt.process_0/d/0c9500ecd3e503b0099ae3dfeb2019b6 filter=lfs diff=lfs merge=lfs -text
|
| 38 |
+
000000000000/params/ocdbt.process_0/d/329ba4a9a13260788168a1a3b29cea8c filter=lfs diff=lfs merge=lfs -text
|
| 39 |
+
000000000000/params/ocdbt.process_0/d/3a6b6c3087fca94e4995f90612c62a6a filter=lfs diff=lfs merge=lfs -text
|
| 40 |
+
000000000000/params/ocdbt.process_0/d/507ad150d3ebf8f32b61c1bc853d146d filter=lfs diff=lfs merge=lfs -text
|
| 41 |
+
000000000000/params/ocdbt.process_0/d/5fec189e6c6348d35371ac411c8f2333 filter=lfs diff=lfs merge=lfs -text
|
| 42 |
+
000000000000/params/ocdbt.process_0/d/661205def231b675efda6211178e4630 filter=lfs diff=lfs merge=lfs -text
|
| 43 |
+
000000000000/params/ocdbt.process_0/d/67df79673a35bae28dd2b0e7b408e5c3 filter=lfs diff=lfs merge=lfs -text
|
| 44 |
+
000000000000/params/ocdbt.process_0/d/7bd1fcb396b67403ecd7e62d1b2feed5 filter=lfs diff=lfs merge=lfs -text
|
| 45 |
+
000000000000/params/ocdbt.process_0/d/a4f75d6fa1268e96f8084d20cff66b4f filter=lfs diff=lfs merge=lfs -text
|
| 46 |
+
000000000000/params/ocdbt.process_0/d/c33e33823f431f64336374b63555bca4 filter=lfs diff=lfs merge=lfs -text
|
| 47 |
+
000000000000/params/ocdbt.process_0/d/c7680fa61e911739fa0316fc46181310 filter=lfs diff=lfs merge=lfs -text
|
| 48 |
+
000000000000/params/ocdbt.process_0/d/c7b5695f4d973e391ceee570e0c71300 filter=lfs diff=lfs merge=lfs -text
|
| 49 |
+
000000000000/params/ocdbt.process_0/d/cf579c9c925a205e2cc5dcb2ce1b4c53 filter=lfs diff=lfs merge=lfs -text
|
| 50 |
+
000000000000/params/ocdbt.process_0/d/d4667bdacaa486400449bbaf82d5f629 filter=lfs diff=lfs merge=lfs -text
|
| 51 |
+
000000000000/params/ocdbt.process_0/d/e0350adbadace8facf287d91a39aaad8 filter=lfs diff=lfs merge=lfs -text
|
| 52 |
+
000000000000/params/ocdbt.process_0/d/e56681c3fecab2f46ff34a6478f36b12 filter=lfs diff=lfs merge=lfs -text
|
| 53 |
+
000000000000/params/ocdbt.process_0/d/f8dae51fd86c9eba8e1b5892d351ef17 filter=lfs diff=lfs merge=lfs -text
|
000000000000/config.yaml
ADDED
|
@@ -0,0 +1,50 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
sharding:
|
| 2 |
+
default: []
|
| 3 |
+
rules:
|
| 4 |
+
- pattern: (encoder_causal_layers|encoder_noncausal|decoder_layers)/mlp/(gate_proj|up_proj|down_proj)/kernel
|
| 5 |
+
spec:
|
| 6 |
+
- null
|
| 7 |
+
- null
|
| 8 |
+
- fsdp
|
| 9 |
+
type: vae.VAEModel
|
| 10 |
+
pretrained_url: meta-llama/Llama-3.2-1B
|
| 11 |
+
pretrained_revision: null
|
| 12 |
+
init_checkpoint: null
|
| 13 |
+
vocab_size: 128256
|
| 14 |
+
bos_token_id: 128000
|
| 15 |
+
eos_token_id: 128001
|
| 16 |
+
pad_token_id: -100
|
| 17 |
+
hidden_size: 2048
|
| 18 |
+
intermediate_size: 8192
|
| 19 |
+
num_hidden_layers: 16
|
| 20 |
+
num_attention_heads: 32
|
| 21 |
+
num_key_value_heads: 8
|
| 22 |
+
head_dim: 64
|
| 23 |
+
max_position_embeddings: 131072
|
| 24 |
+
rope_theta: 500000.0
|
| 25 |
+
rope_scaling:
|
| 26 |
+
factor: 32.0
|
| 27 |
+
high_freq_factor: 4.0
|
| 28 |
+
low_freq_factor: 1.0
|
| 29 |
+
original_context_len: 8192
|
| 30 |
+
rms_norm_eps: 1.0e-05
|
| 31 |
+
initializer_range: 0.02
|
| 32 |
+
tie_word_embeddings: true
|
| 33 |
+
compute_dtype: bfloat16
|
| 34 |
+
gradient_checkpointing: true
|
| 35 |
+
scan_unroll: 1
|
| 36 |
+
attention_backend: auto
|
| 37 |
+
attention_block_size: 128
|
| 38 |
+
causal_attention: true
|
| 39 |
+
init_checkpoint_repo: null
|
| 40 |
+
init_checkpoint_step: 0
|
| 41 |
+
init_checkpoint_revision: null
|
| 42 |
+
num_encoder_causal_layers: 10
|
| 43 |
+
num_noncausal_layers: 6
|
| 44 |
+
num_decoder_layers: 16
|
| 45 |
+
latent_size: 2048
|
| 46 |
+
latent_mlp_init_scale: 0.1
|
| 47 |
+
radius_min: 0.0
|
| 48 |
+
radius_max: 1.0
|
| 49 |
+
mixer_kernel_size: 5
|
| 50 |
+
pad_attention_bias_value: -1000.0
|
000000000000/metadata.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"git_commit": "645dba6998c4d7b68ed0f307ced9648eb6c1b01b",
|
| 3 |
+
"episodes": 128
|
| 4 |
+
}
|
000000000000/params/_CHECKPOINT_METADATA
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"item_handlers": "orbax.checkpoint._src.handlers.standard_checkpoint_handler.StandardCheckpointHandler", "metrics": {}, "performance_metrics": {}, "init_timestamp_nsecs": 1787635909391582504, "commit_timestamp_nsecs": 1787635915239974219, "custom_metadata": {}}
|
000000000000/params/_METADATA
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"tree_metadata": {"('decoder_layers', 'input_layernorm', 'scale')": {"key_metadata": [{"key": "decoder_layers", "key_type": 2}, {"key": "input_layernorm", "key_type": 2}, {"key": "scale", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('decoder_layers', 'mlp', 'down_proj', 'kernel')": {"key_metadata": [{"key": "decoder_layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "down_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('decoder_layers', 'mlp', 'gate_proj', 'kernel')": {"key_metadata": [{"key": "decoder_layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "gate_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('decoder_layers', 'mlp', 'latent_mlp', 'down_latent', 'kernel')": {"key_metadata": [{"key": "decoder_layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "latent_mlp", "key_type": 2}, {"key": "down_latent", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16, 2048, 2048]}}, "('decoder_layers', 'mlp', 'latent_mlp', 'gate_latent', 'bias')": {"key_metadata": [{"key": "decoder_layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "latent_mlp", "key_type": 2}, {"key": "gate_latent", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16, 2048]}}, "('decoder_layers', 'mlp', 'latent_mlp', 'gate_latent', 'kernel')": {"key_metadata": [{"key": "decoder_layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "latent_mlp", "key_type": 2}, {"key": "gate_latent", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16, 2048, 2048]}}, "('decoder_layers', 'mlp', 'latent_mlp', 'odot')": {"key_metadata": [{"key": "decoder_layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "latent_mlp", "key_type": 2}, {"key": "odot", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16, 2048, 2048]}}, "('decoder_layers', 'mlp', 'latent_mlp', 'sig_latent', 'bias')": {"key_metadata": [{"key": "decoder_layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "latent_mlp", "key_type": 2}, {"key": "sig_latent", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16, 2048]}}, "('decoder_layers', 'mlp', 'latent_mlp', 'sig_latent', 'kernel')": {"key_metadata": [{"key": "decoder_layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "latent_mlp", "key_type": 2}, {"key": "sig_latent", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16, 2048, 2048]}}, "('decoder_layers', 'mlp', 'latent_mlp', 'up_latent', 'kernel')": {"key_metadata": [{"key": "decoder_layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "latent_mlp", "key_type": 2}, {"key": "up_latent", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16, 2048, 2048]}}, "('decoder_layers', 'mlp', 'up_proj', 'kernel')": {"key_metadata": [{"key": "decoder_layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "up_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('decoder_layers', 'post_attention_layernorm', 'scale')": {"key_metadata": [{"key": "decoder_layers", "key_type": 2}, {"key": "post_attention_layernorm", "key_type": 2}, {"key": "scale", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('decoder_layers', 'self_attn', 'k_proj', 'kernel')": {"key_metadata": [{"key": "decoder_layers", "key_type": 2}, {"key": "self_attn", "key_type": 2}, {"key": "k_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('decoder_layers', 'self_attn', 'o_proj', 'kernel')": {"key_metadata": [{"key": "decoder_layers", "key_type": 2}, {"key": "self_attn", "key_type": 2}, {"key": "o_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('decoder_layers', 'self_attn', 'q_proj', 'kernel')": {"key_metadata": [{"key": "decoder_layers", "key_type": 2}, {"key": "self_attn", "key_type": 2}, {"key": "q_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('decoder_layers', 'self_attn', 'v_proj', 'kernel')": {"key_metadata": [{"key": "decoder_layers", "key_type": 2}, {"key": "self_attn", "key_type": 2}, {"key": "v_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('decoder_norm', 'scale')": {"key_metadata": [{"key": "decoder_norm", "key_type": 2}, {"key": "scale", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('decoder_radius_embedding',)": {"key_metadata": [{"key": "decoder_radius_embedding", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('embed_tokens', 'embedding')": {"key_metadata": [{"key": "embed_tokens", "key_type": 2}, {"key": "embedding", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('encoder_causal_layers', 'input_layernorm', 'scale')": {"key_metadata": [{"key": "encoder_causal_layers", "key_type": 2}, {"key": "input_layernorm", "key_type": 2}, {"key": "scale", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('encoder_causal_layers', 'mlp', 'down_proj', 'kernel')": {"key_metadata": [{"key": "encoder_causal_layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "down_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('encoder_causal_layers', 'mlp', 'gate_proj', 'kernel')": {"key_metadata": [{"key": "encoder_causal_layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "gate_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('encoder_causal_layers', 'mlp', 'up_proj', 'kernel')": {"key_metadata": [{"key": "encoder_causal_layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "up_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('encoder_causal_layers', 'post_attention_layernorm', 'scale')": {"key_metadata": [{"key": "encoder_causal_layers", "key_type": 2}, {"key": "post_attention_layernorm", "key_type": 2}, {"key": "scale", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('encoder_causal_layers', 'self_attn', 'k_proj', 'kernel')": {"key_metadata": [{"key": "encoder_causal_layers", "key_type": 2}, {"key": "self_attn", "key_type": 2}, {"key": "k_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('encoder_causal_layers', 'self_attn', 'o_proj', 'kernel')": {"key_metadata": [{"key": "encoder_causal_layers", "key_type": 2}, {"key": "self_attn", "key_type": 2}, {"key": "o_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('encoder_causal_layers', 'self_attn', 'q_proj', 'kernel')": {"key_metadata": [{"key": "encoder_causal_layers", "key_type": 2}, {"key": "self_attn", "key_type": 2}, {"key": "q_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('encoder_causal_layers', 'self_attn', 'v_proj', 'kernel')": {"key_metadata": [{"key": "encoder_causal_layers", "key_type": 2}, {"key": "self_attn", "key_type": 2}, {"key": "v_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('encoder_noncausal', 'input_layernorm', 'scale')": {"key_metadata": [{"key": "encoder_noncausal", "key_type": 2}, {"key": "input_layernorm", "key_type": 2}, {"key": "scale", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('encoder_noncausal', 'mlp', 'down_proj', 'kernel')": {"key_metadata": [{"key": "encoder_noncausal", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "down_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('encoder_noncausal', 'mlp', 'gate_proj', 'kernel')": {"key_metadata": [{"key": "encoder_noncausal", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "gate_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('encoder_noncausal', 'mlp', 'up_proj', 'kernel')": {"key_metadata": [{"key": "encoder_noncausal", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "up_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('encoder_noncausal', 'post_attention_layernorm', 'scale')": {"key_metadata": [{"key": "encoder_noncausal", "key_type": 2}, {"key": "post_attention_layernorm", "key_type": 2}, {"key": "scale", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('encoder_noncausal', 'self_attn', 'k_mixer', 'conv', 'kernel')": {"key_metadata": [{"key": "encoder_noncausal", "key_type": 2}, {"key": "self_attn", "key_type": 2}, {"key": "k_mixer", "key_type": 2}, {"key": "conv", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [6, 5, 1, 2048]}}, "('encoder_noncausal', 'self_attn', 'k_proj', 'kernel')": {"key_metadata": [{"key": "encoder_noncausal", "key_type": 2}, {"key": "self_attn", "key_type": 2}, {"key": "k_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [6, 2048, 2048]}}, "('encoder_noncausal', 'self_attn', 'o_proj', 'kernel')": {"key_metadata": [{"key": "encoder_noncausal", "key_type": 2}, {"key": "self_attn", "key_type": 2}, {"key": "o_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('encoder_noncausal', 'self_attn', 'q_mixer', 'conv', 'kernel')": {"key_metadata": [{"key": "encoder_noncausal", "key_type": 2}, {"key": "self_attn", "key_type": 2}, {"key": "q_mixer", "key_type": 2}, {"key": "conv", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [6, 5, 1, 2048]}}, "('encoder_noncausal', 'self_attn', 'q_proj', 'kernel')": {"key_metadata": [{"key": "encoder_noncausal", "key_type": 2}, {"key": "self_attn", "key_type": 2}, {"key": "q_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('encoder_noncausal', 'self_attn', 'v_mixer', 'conv', 'kernel')": {"key_metadata": [{"key": "encoder_noncausal", "key_type": 2}, {"key": "self_attn", "key_type": 2}, {"key": "v_mixer", "key_type": 2}, {"key": "conv", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [6, 5, 1, 2048]}}, "('encoder_noncausal', 'self_attn', 'v_proj', 'kernel')": {"key_metadata": [{"key": "encoder_noncausal", "key_type": 2}, {"key": "self_attn", "key_type": 2}, {"key": "v_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [6, 2048, 2048]}}, "('encoder_noncausal_norm', 'scale')": {"key_metadata": [{"key": "encoder_noncausal_norm", "key_type": 2}, {"key": "scale", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [2048]}}, "('encoder_norm', 'scale')": {"key_metadata": [{"key": "encoder_norm", "key_type": 2}, {"key": "scale", "key_type": 2}], "value_metadata": {"value_type": "np.ndarray", "skip_deserialize": false}}, "('encoder_state_scale',)": {"key_metadata": [{"key": "encoder_state_scale", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [2048]}}, "('encoder_state_shift',)": {"key_metadata": [{"key": "encoder_state_shift", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [2048]}}, "('latent_writer', 'key_proj', 'kernel')": {"key_metadata": [{"key": "latent_writer", "key_type": 2}, {"key": "key_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [2048, 2048]}}, "('latent_writer', 'value_gate_proj', 'kernel')": {"key_metadata": [{"key": "latent_writer", "key_type": 2}, {"key": "value_gate_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [2048, 2048]}}, "('latent_writer', 'value_proj', 'kernel')": {"key_metadata": [{"key": "latent_writer", "key_type": 2}, {"key": "value_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [2048, 2048]}}}, "use_ocdbt": true, "use_zarr3": false, "store_array_data_equal_to_fill_value": true, "custom_metadata": null}
|
000000000000/params/_sharding
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"ZGVjb2Rlcl9sYXllcnMubWxwLmxhdGVudF9tbHAuZ2F0ZV9sYXRlbnQuYmlhcw==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","ZGVjb2Rlcl9sYXllcnMubWxwLmxhdGVudF9tbHAuZ2F0ZV9sYXRlbnQua2VybmVs":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","ZGVjb2Rlcl9sYXllcnMubWxwLmxhdGVudF9tbHAuZG93bl9sYXRlbnQua2VybmVs":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","ZGVjb2Rlcl9sYXllcnMubWxwLmxhdGVudF9tbHAub2RvdA==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","ZGVjb2Rlcl9sYXllcnMubWxwLmxhdGVudF9tbHAuc2lnX2xhdGVudC5iaWFz":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","ZGVjb2Rlcl9sYXllcnMubWxwLmxhdGVudF9tbHAuc2lnX2xhdGVudC5rZXJuZWw=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","ZGVjb2Rlcl9sYXllcnMubWxwLmxhdGVudF9tbHAudXBfbGF0ZW50Lmtlcm5lbA==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","ZW5jb2Rlcl9ub25jYXVzYWwuc2VsZl9hdHRuLmtfbWl4ZXIuY29udi5rZXJuZWw=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","ZW5jb2Rlcl9ub25jYXVzYWwuc2VsZl9hdHRuLmtfcHJvai5rZXJuZWw=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","ZW5jb2Rlcl9ub25jYXVzYWwuc2VsZl9hdHRuLnFfbWl4ZXIuY29udi5rZXJuZWw=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","ZW5jb2Rlcl9ub25jYXVzYWwuc2VsZl9hdHRuLnZfbWl4ZXIuY29udi5rZXJuZWw=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","ZW5jb2Rlcl9ub25jYXVzYWwuc2VsZl9hdHRuLnZfcHJvai5rZXJuZWw=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","ZW5jb2Rlcl9ub25jYXVzYWxfbm9ybS5zY2FsZQ==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","ZW5jb2Rlcl9zdGF0ZV9zY2FsZQ==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","ZW5jb2Rlcl9zdGF0ZV9zaGlmdA==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","bGF0ZW50X3dyaXRlci52YWx1ZV9nYXRlX3Byb2oua2VybmVs":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","bGF0ZW50X3dyaXRlci52YWx1ZV9wcm9qLmtlcm5lbA==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","bGF0ZW50X3dyaXRlci5rZXlfcHJvai5rZXJuZWw=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}"}
|
000000000000/params/array_metadatas/process_0
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"array_metadatas": [{"array_metadata": {"param_name": "decoder_layers.mlp.latent_mlp.down_latent.kernel", "write_shape": [16, 2048, 2048], "chunk_shape": [16, 2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "decoder_layers.mlp.latent_mlp.gate_latent.bias", "write_shape": [16, 2048], "chunk_shape": [16, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "decoder_layers.mlp.latent_mlp.gate_latent.kernel", "write_shape": [16, 2048, 2048], "chunk_shape": [16, 2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "decoder_layers.mlp.latent_mlp.odot", "write_shape": [16, 2048, 2048], "chunk_shape": [16, 2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "decoder_layers.mlp.latent_mlp.sig_latent.bias", "write_shape": [16, 2048], "chunk_shape": [16, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "decoder_layers.mlp.latent_mlp.sig_latent.kernel", "write_shape": [16, 2048, 2048], "chunk_shape": [16, 2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "decoder_layers.mlp.latent_mlp.up_latent.kernel", "write_shape": [16, 2048, 2048], "chunk_shape": [16, 2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "encoder_noncausal.self_attn.k_mixer.conv.kernel", "write_shape": [6, 5, 1, 2048], "chunk_shape": [6, 5, 1, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "encoder_noncausal.self_attn.k_proj.kernel", "write_shape": [6, 2048, 2048], "chunk_shape": [6, 2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "encoder_noncausal.self_attn.q_mixer.conv.kernel", "write_shape": [6, 5, 1, 2048], "chunk_shape": [6, 5, 1, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "encoder_noncausal.self_attn.v_mixer.conv.kernel", "write_shape": [6, 5, 1, 2048], "chunk_shape": [6, 5, 1, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "encoder_noncausal.self_attn.v_proj.kernel", "write_shape": [6, 2048, 2048], "chunk_shape": [6, 2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "encoder_noncausal_norm.scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "encoder_state_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "encoder_state_shift", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "latent_writer.key_proj.kernel", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "latent_writer.value_gate_proj.kernel", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "latent_writer.value_proj.kernel", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}]}
|
000000000000/params/commit_success.txt
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
Checkpoint commit was successful to /tmp/ttl-checkpoint-0gwygz3b/params
|
000000000000/params/d/f8d62a6166aa34f52910ff7c504563cf
ADDED
|
Binary file (1.84 kB). View file
|
|
|
000000000000/params/manifest.ocdbt
ADDED
|
Binary file (117 Bytes). View file
|
|
|
000000000000/params/ocdbt.process_0/d/071bd40b12e1dc6207fa7c32dbe308e6
ADDED
|
Binary file (828 Bytes). View file
|
|
|
000000000000/params/ocdbt.process_0/d/08b75220440bf9acdecb21e5fb7d035e
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3c2f45afceb00e199391ae24f33b7d1ea7dadae2ed07ae74c1e62f4116cbfba7
|
| 3 |
+
size 64724992
|
000000000000/params/ocdbt.process_0/d/0c9500ecd3e503b0099ae3dfeb2019b6
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2d097c7a6c769559473b3eae6fcd756bc272f51bb15ea3bd73332bcbe7db474d
|
| 3 |
+
size 320974848
|
000000000000/params/ocdbt.process_0/d/22b93714ba257459d8524d37e5aeec72
ADDED
|
Binary file (9.47 kB). View file
|
|
|
000000000000/params/ocdbt.process_0/d/329ba4a9a13260788168a1a3b29cea8c
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4250fae263863beb3715c37d71c505b66bdbf8a1208dd419094065028fe3aedd
|
| 3 |
+
size 502845440
|
000000000000/params/ocdbt.process_0/d/3a6b6c3087fca94e4995f90612c62a6a
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2709020ca6cd45c19d99acfb1a4902447a5ba5872f3c54b14be3dc26add44837
|
| 3 |
+
size 518057984
|
000000000000/params/ocdbt.process_0/d/507ad150d3ebf8f32b61c1bc853d146d
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cd3568c329b7d087cb7c81f6e56281cba4a747e396a312789a2623000a6f294d
|
| 3 |
+
size 513163264
|
000000000000/params/ocdbt.process_0/d/55fa9000bd9670f087175e6ad068c48e
ADDED
|
Binary file (209 Bytes). View file
|
|
|
000000000000/params/ocdbt.process_0/d/5fec189e6c6348d35371ac411c8f2333
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1b287bb0ca3d65215d41475920c7d04f4362eddb2ae9c1ded6a0de524e9293a3
|
| 3 |
+
size 993705984
|
000000000000/params/ocdbt.process_0/d/661205def231b675efda6211178e4630
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1330b1cd51fecb30754e1d202fe873ea79412babbbdfc06214835effcb792100
|
| 3 |
+
size 31096832
|
000000000000/params/ocdbt.process_0/d/67df79673a35bae28dd2b0e7b408e5c3
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a665d444ba8af4e7a2bfc313b7c041bcfcbb50831257577c1bfbd53ed019b450
|
| 3 |
+
size 139802
|
000000000000/params/ocdbt.process_0/d/79c0c657bf6d113301c8cb116e593bf7
ADDED
|
Binary file (633 Bytes). View file
|
|
|
000000000000/params/ocdbt.process_0/d/7bd1fcb396b67403ecd7e62d1b2feed5
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4a291c79d7c8a9e1f27b639abe840f4433fc3381a87d4475ffcff4ba37069e16
|
| 3 |
+
size 938586
|
000000000000/params/ocdbt.process_0/d/a4f75d6fa1268e96f8084d20cff66b4f
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e344732e717497add7e41676626967a65e23fa1647661f2159945ff60c784681
|
| 3 |
+
size 516493312
|
000000000000/params/ocdbt.process_0/d/b2bf95bd2b65ad4d7d9b7d63a1c88d92
ADDED
|
Binary file (604 Bytes). View file
|
|
|
000000000000/params/ocdbt.process_0/d/c33e33823f431f64336374b63555bca4
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:610b929268d03e8f0ba25814c86ac4f0dc0b7abeea389478520e81d1b06b5068
|
| 3 |
+
size 320638976
|
000000000000/params/ocdbt.process_0/d/c7680fa61e911739fa0316fc46181310
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:43156642af58fdd6e7ca11147e082eff53cc50636d4f47290afa9c3cb7e25abb
|
| 3 |
+
size 15548416
|
000000000000/params/ocdbt.process_0/d/c7b5695f4d973e391ceee570e0c71300
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:206974d045cc0da9e5bf680da7b7416e6dd2534e72682c6065b2457932b8eb95
|
| 3 |
+
size 322752512
|
000000000000/params/ocdbt.process_0/d/cf579c9c925a205e2cc5dcb2ce1b4c53
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0334cb96fc2ebe0da0996e9c1c6026dc6f047ccc0b2916b04afc3b4dd4b6ee97
|
| 3 |
+
size 386269184
|
000000000000/params/ocdbt.process_0/d/d4667bdacaa486400449bbaf82d5f629
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:72ec3c2b35f6175e9f543d64efd3e55a10b3caad1290d4ab069265d7fc9c2c8c
|
| 3 |
+
size 93282304
|
000000000000/params/ocdbt.process_0/d/e0350adbadace8facf287d91a39aaad8
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9203917bc6c045415fb00a1937796730030b2e0cd24c2f4674b788338e0e416d
|
| 3 |
+
size 133705728
|
000000000000/params/ocdbt.process_0/d/e2d6fc3564ecc9ab4286abf2c21a99d8
ADDED
|
Binary file (38.3 kB). View file
|
|
|
000000000000/params/ocdbt.process_0/d/e56681c3fecab2f46ff34a6478f36b12
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:79879f8942d68ca9a9350cd5996ccfb72c2d4004be24c8af464fa66693f78a98
|
| 3 |
+
size 513728512
|
000000000000/params/ocdbt.process_0/d/f8dae51fd86c9eba8e1b5892d351ef17
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:49083bbe75d672e10abf8ce991ac867fbc97c99c9c7c33bb1063c2fd06fa7114
|
| 3 |
+
size 192757760
|
000000000000/params/ocdbt.process_0/manifest.ocdbt
ADDED
|
Binary file (503 Bytes). View file
|
|
|