© 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. 2-RES-SM
    11. 10191
    Prev
    Next
    INDEX
    Explanations

    instances of the word "heard" and its variations relating to knowledge or information received

    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
    246
    +0.12
    0.7%
    54
    +0.12
    0.6%
    496
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    246
    +0.12
    0.04
    287
    +0.12
    0.04
    292
    +0.11
    0.04
    Negative Logits
    ľĵ
    -3.15
    Īĺ
    -3.12
    IJ
    -3.08
    ľ
    -3.02
    ·¸
    -2.86
    µ
    -2.85
    ¿½
    -2.78
    ¤
    -2.76
    »¿
    -2.76
    ĥ
    -2.74
    POSITIVE LOGITS
     noises
    2.10
     loud
    1.90
     speaker
    1.81
     noise
    1.71
     podcast
    1.71
     sound
    1.69
     msgid
    1.67
     frequency
    1.67
     audio
    1.65
     louder
    1.65
    Activations Density 0.252%

    No Known Activations