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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. MLP Post
    5. 2-MLP-SM
    6. 25989
    Prev
    Next
    INDEX
    Explanations

    occurrences of parentheses in the text

    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/2-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.2.hook_mlp_out
    Hook Layer
    2
    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
    489
    +0.14
    0.8%
    143
    +0.14
    0.8%
    101
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    489
    +0.14
    0.03
    352
    +0.14
    0.03
    268
    +0.13
    0.04
    Negative Logits
    estly
    -1.79
    lets
    -1.76
    let
    -1.71
    ________________
    -1.61
    lett
    -1.61
    terday
    -1.57
    ustomed
    -1.57
    aku
    -1.53
    STRA
    -1.52
    breviations
    -1.50
    POSITIVE LOGITS
    ĨĴ
    2.34
    ĩ
    2.21
    Ń
    1.98
    Ļª
    1.91
    ³
    1.79
    ¶
    1.75
    ¾
    1.75
    Ĥ¬
    1.67
    <|outofrange|>
    1.59
    č↵           
    1.59
    Activations Density 0.192%

    No Known Activations