© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. MLP Post
    5. 0-MLP-SM
    6. 27923
    Prev
    Next
    INDEX
    Explanations

    instances of the word "bir."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__mlp-sm_processed/0-mlp-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.0.hook_mlp_out
    Hook Layer
    0
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    369
    +0.14
    0.8%
    149
    +0.12
    0.7%
    142
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    53
    +0.14
    0.03
    438
    +0.12
    0.03
    366
    +0.11
    0.03
    Negative Logits
     cref
    -1.61
    ?_
    -1.49
    umbent
    -1.45
    "?"
    -1.44
    '?"
    -1.38
    )">
    -1.36
    ?).
    -1.35
    ```
    -1.35
    ."_
    -1.34
    visibility
    -1.34
    POSITIVE LOGITS
    ĻĤ
    2.19
     enough
    1.68
    Īĺ
    1.59
     remar
    1.48
    estock
    1.46
    liche
    1.46
     Buddhist
    1.45
     Buddhism
    1.45
    度
    1.44
    etics
    1.40
    Activations Density 4.811%

    No Known Activations