© 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. 2-RES-SM
    11. 31236
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to discussions and conversations

    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
    419
    +0.24
    1.4%
    72
    +0.15
    0.9%
    76
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    419
    +0.24
    0.07
    76
    +0.15
    0.05
    72
    +0.13
    0.05
    Negative Logits
    abouts
    -1.91
    ĨĴ
    -1.61
    Ĥ¬
    -1.56
    ij
    -1.52
    ¹
    -1.50
    around
    -1.49
    wards
    -1.48
     injured
    -1.43
    license
    -1.39
    ¼
    -1.38
    POSITIVE LOGITS
    erie
    1.92
    fficients
    1.80
    astically
    1.66
    amente
    1.53
    ÈĽ
    1.44
     Discussion
    1.41
    ociation
    1.40
    hedral
    1.40
    ência
    1.40
    tered
    1.39
    Activations Density 0.755%

    No Known Activations