© 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
    3. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    5. GPT2-Small
    7. Residual Similar CE Local
    9. 6-RES_SCL-AJT
    11. 34710
    Prev
    Next
    INDEX
    Explanations

    phrases related to legal and political engagements

    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
    +0.26
    1.2%
    288
    +0.15
    0.7%
    570
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    492
    +0.26
    0.07
    526
    +0.15
    -0.08
    761
    +0.13
    0.08
    Negative Logits
    etry
    -0.75
    bey
    -0.66
    Origin
    -0.66
    opers
    -0.66
    ological
    -0.63
     Printed
    -0.61
    issue
    -0.60
    pler
    -0.59
    patient
    -0.58
    pled
    -0.58
    POSITIVE LOGITS
    ments
    0.90
    TAIN
    0.87
     engaged
    0.85
    lished
    0.84
    ienced
    0.82
     engagement
    0.78
    aeper
    0.73
    reement
    0.73
    EMENT
    0.72
    encer
    0.72
    Activations Density 10.332%

    No Known Activations