© 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
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 31733
    Prev
    Next
    INDEX
    Explanations

    specific syntax or formatting elements in code

    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
    156
    +0.51
    3.5%
    111
    +0.42
    2.9%
    0
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.51
    0.75
    156
    +0.42
    0.64
    71
    +0.11
    0.62
    Negative Logits
    ħ
    -1.65
    Ĥ
    -1.57
    ĭ
    -1.54
    ±
    -1.50
    Ĥ¬
    -1.50
    ĥ½
    -1.49
    ¤
    -1.48
    ancy
    -1.45
    achus
    -1.44
    §
    -1.43
    POSITIVE LOGITS
     Briefly
    1.34
     Thus
    1.18
     Yet
    1.18
     Fields
    1.16
     Here
    1.14
    '."
    1.14
     Hence
    1.12
     glycerol
    1.11
     For
    1.08
     Accordingly
    1.08
    Activations Density 0.862%

    No Known Activations