© 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. 26101
    Prev
    Next
    INDEX
    Explanations

    words related to adoption and the process surrounding it

    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
    222
    +0.15
    0.8%
    156
    +0.14
    0.8%
    18
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    222
    +0.15
    0.03
    50
    +0.14
    0.01
    363
    +0.10
    0.02
    Negative Logits
    zos
    -1.74
    uber
    -1.64
    .");
    -1.63
    reon
    -1.61
    rose
    -1.59
    hin
    -1.57
    them
    -1.55
    zÄħ
    -1.53
    sten
    -1.51
    zan
    -1.51
    POSITIVE LOGITS
     rpm
    1.73
     malf
    1.59
     ppm
    1.51
    Ļª
    1.45
    Ħ
    1.41
    ļ
    1.41
    °
    1.38
     INDIRECT
    1.36
    AIL
    1.32
     Fed
    1.32
    Activations Density 0.222%

    No Known Activations