© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. MLP Post
    5. 2-MLP-SM
    6. 25910
    Prev
    Next
    INDEX
    Explanations

    terms related to licensing and authorization within a technical or regulatory context

    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
    200
    +0.12
    0.8%
    185
    +0.11
    0.7%
    163
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    250
    +0.12
    0.06
    422
    +0.11
    0.07
    200
    +0.10
    0.06
    Negative Logits
    Ħ
    -3.02
    Ļ
    -2.78
    ®
    -2.72
    ²
    -2.71
    ¾
    -2.67
    ¼
    -2.65
    ³
    -2.64
    IJ
    -2.62
    ¯
    -2.62
    ĥ
    -2.61
    POSITIVE LOGITS
    "}](#
    1.38
     ignored
    1.31
     aged
    1.26
    derr
    1.25
    apine
    1.21
    bye
    1.21
    pherd
    1.21
     yours
    1.19
    cker
    1.19
    ===========================
    1.16
    Activations Density 0.415%

    No Known Activations