Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE Local
    5. 6-RES_SCL-AJT
    6. 34347
    Prev
    Next
    INDEX
    Explanations

    phrases related to social justice and political discourse

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt
    Prompts (Dashboard)
    12,288 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    46,080
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    apollo-research/Skylion007-openwebtext-tokenizer-gpt2
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    447
    +0.12
    0.5%
    55
    +0.12
    0.5%
    584
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    614
    +0.12
    0.05
    641
    +0.12
    0.12
    526
    +0.12
    -0.08
    Negative Logits
    natureconservancy
    -0.78
    ophone
    -0.68
    achus
    -0.68
     Unsure
    -0.67
    nown
    -0.66
     Carbuncle
    -0.65
     Bei
    -0.63
    accompan
    -0.62
    vous
    -0.61
    nee
    -0.61
    POSITIVE LOGITS
     incompet
    0.75
     shielding
    0.69
     mathematic
    0.67
     fundamentally
    0.66
    ertodd
    0.64
    enstein
    0.63
     nurture
    0.63
     differently
    0.62
     intentionally
    0.62
    ORPG
    0.62
    Activations Density 10.533%

    No Known Activations