© 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. MLP Post
    9. 2-MLP-SM
    11. 25105
    Prev
    Next
    INDEX
    Explanations

    SQL commands

    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
    56
    +0.15
    0.8%
    504
    +0.13
    0.7%
    426
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    56
    +0.15
    0.02
    68
    +0.13
    0.01
    250
    +0.12
    0.01
    Negative Logits
     pardon
    -1.71
    pace
    -1.51
     decoding
    -1.44
     Mozilla
    -1.43
     pirate
    -1.40
     ssh
    -1.39
     luck
    -1.39
    uance
    -1.38
     spell
    -1.38
     sewer
    -1.38
    POSITIVE LOGITS
    Ļª
    3.50
    ļ
    3.45
    ŀ
    3.39
    Ĭ
    3.16
    İ
    3.15
    ħ
    3.10
    ¶
    3.09
    ↵↵
    3.08
                ↵
    3.08
     ↵
    3.08
    Activations Density 0.068%

    No Known Activations