© 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. 12747
    Prev
    Next
    INDEX
    Explanations

    negative emotional expressions or sentiments

    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
    369
    +0.11
    0.6%
    208
    +0.11
    0.6%
    448
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    448
    +0.11
    0.05
    288
    +0.11
    0.04
    175
    +0.11
    0.02
    Negative Logits
    ality
    -1.58
    omy
    -1.53
    orectal
    -1.50
    ==========
    -1.48
    bes
    -1.42
    alities
    -1.42
     disadvantage
    -1.41
     deposited
    -1.40
     recovered
    -1.40
    imab
    -1.39
    POSITIVE LOGITS
    ´
    2.38
    Ĩ
    2.37
    Ĥ
    2.36
    ¬
    2.27
    Ĵ
    2.22
    ¶
    2.19
    mails
    2.16
    ¥
    2.15
    ĻĤ
    2.14
    ¡
    2.10
    Activations Density 0.205%

    No Known Activations