© 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. 14078
    Prev
    Next
    INDEX
    Explanations

    transitional phrases that indicate a conclusion or reasoning

    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
    25
    +0.13
    0.7%
    153
    +0.10
    0.6%
    64
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    476
    +0.13
    0.03
    330
    +0.10
    0.03
    361
    +0.10
    0.03
    Negative Logits
    ħ
    -2.66
    Ħ
    -2.59
    Ģ
    -2.48
    ľĵ
    -2.47
    ®
    -2.40
    ¬
    -2.39
    Į
    -2.38
    Ļ
    -2.37
    ŀ
    -2.29
    ĥ
    -2.27
    POSITIVE LOGITS
    forth
    3.36
     far
    2.01
     increasing
    1.67
     avoiding
    1.60
     inhibiting
    1.58
    rely
    1.57
     enabling
    1.56
     ensuring
    1.56
    soever
    1.55
     preventing
    1.53
    Activations Density 0.195%

    No Known Activations