© 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. MLP Post
    9. 2-MLP-SM
    11. 23172
    Prev
    Next
    INDEX
    Explanations

    punctuation marks, specifically periods and commas

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__mlp-sm_processed/2-mlp-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_mlp_out
    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
    146
    +0.13
    0.7%
    489
    +0.13
    0.7%
    53
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    489
    +0.13
    0.02
    38
    +0.13
    0.01
    348
    +0.12
    0.01
    Negative Logits
     respondent
    -1.80
     petitioner
    -1.77
     appellant
    -1.75
     material
    -1.47
     appellants
    -1.45
     mailing
    -1.45
     Respondents
    -1.45
     participating
    -1.41
     defendant
    -1.40
     appellees
    -1.39
    POSITIVE LOGITS
    Īĺ
    2.18
    coming
    2.04
    Ķ
    1.77
    ¦
    1.76
    ®
    1.71
    rh
    1.62
    !'
    1.62
    iem
    1.60
    arman
    1.59
    ¢
    1.59
    Activations Density 0.083%

    No Known Activations