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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 1-RES-SM
    11. 25178
    Prev
    Next
    INDEX
    Explanations

    unique identifiers and terms related to patents

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/1-res-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.1.hook_resid_post
    Hook Layer
    1
    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
    425
    +0.14
    0.8%
    258
    +0.13
    0.7%
    41
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.14
    0.06
    271
    +0.13
    0.01
    425
    +0.12
    0.05
    Negative Logits
    Ľ
    -1.70
    ĥ½
    -1.59
    ¶
    -1.59
    Īĺ
    -1.58
    blr
    -1.57
    ?.
    -1.54
    Ł
    -1.50
    Ń
    -1.48
    »
    -1.46
    «
    -1.46
    POSITIVE LOGITS
    atories
    1.99
    iors
    1.63
    atory
    1.58
    smen
    1.43
    unction
    1.42
    icul
    1.40
    INGS
    1.36
     dispositions
    1.33
    IONS
    1.32
    ools
    1.32
    Activations Density 0.225%

    No Known Activations