© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 4-RES-SM
    11. 11216
    Prev
    Next
    INDEX
    Explanations

    components related to support and assistance in various contexts

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    348
    +0.16
    0.9%
    240
    +0.13
    0.7%
    198
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    437
    +0.16
    0.08
    198
    +0.13
    0.08
    233
    +0.12
    0.04
    Negative Logits
    ãĤĥ
    -1.84
    creas
    -1.78
     deceased
    -1.71
    inge
    -1.64
    unnumbered
    -1.63
    áĢº
    -1.61
    chus
    -1.59
    ะ
    -1.59
    ophagus
    -1.59
    coma
    -1.58
    POSITIVE LOGITS
     coverage
    2.58
     access
    2.55
     assistance
    2.34
     insight
    2.31
     insights
    2.30
     services
    2.28
     information
    2.20
     assurance
    2.19
     opportunities
    2.19
     capability
    2.17
    Activations Density 1.032%

    No Known Activations