© 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. 4-RES-SM
    11. 9937
    Prev
    Next
    INDEX
    Explanations

    phrases related to compliance and conditions being fulfilled

    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
    21
    +0.13
    0.7%
    480
    +0.11
    0.6%
    197
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    480
    +0.13
    0.01
    334
    +0.11
    0.01
    191
    +0.11
    0.01
    Negative Logits
    ¼
    -3.36
    Ļª
    -3.25
    ↵
    -3.13
    -3.13
    č↵
    -3.13
    ↵
    -3.13
    ↵
    -3.13
    č↵
    -3.13
    ↵
    -3.13
    č↵č↵
    -3.13
    POSITIVE LOGITS
    applicable
    1.67
    :#
    1.54
    iculously
    1.52
    riz
    1.51
     Redistributions
    1.50
    except
    1.50
    :")
    1.49
    REF
    1.47
     0000000000000000000000000000000000
    1.47
     governing
    1.44
    Activations Density 0.071%

    No Known Activations