© 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
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 10792
    Prev
    Next
    INDEX
    Explanations

    references to authorship or citations

    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/2-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.2.hook_resid_post
    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
    16
    +0.14
    0.7%
    203
    +0.13
    0.7%
    334
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    249
    +0.14
    0.01
    92
    +0.13
    0.01
    422
    +0.11
    0.01
    Negative Logits
    exe
    -1.75
    innerHTML
    -1.73
    illary
    -1.68
    gif
    -1.63
    ixel
    -1.51
    buffer
    -1.49
    jpg
    -1.48
    rawl
    -1.48
    plist
    -1.47
    aque
    -1.44
    POSITIVE LOGITS
    ³
    3.13
    ¸
    3.10
    ¬
    3.07
    ´
    3.04
    Ĥ
    3.04
    İ
    3.02
    ¶
    2.91
    ½
    2.87
    ľ
    2.81
    ł
    2.79
    Activations Density 0.024%

    No Known Activations