© 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. 34837
    Prev
    Next
    INDEX
    Explanations

    references to historical and philosophical concepts or figures

    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
    459
    +0.11
    0.5%
    307
    +0.10
    0.5%
    209
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    577
    +0.11
    0.14
    556
    +0.10
    0.12
    316
    +0.10
    0.11
    Negative Logits
     Mehran
    -0.78
    ĸļ
    -0.74
    isSpecialOrderable
    -0.65
    soDeliveryDate
    -0.62
    finder
    -0.62
     mischief
    -0.61
     neighbours
    -0.60
     thor
    -0.60
     sockets
    -0.60
     reckoning
    -0.60
    POSITIVE LOGITS
    enment
    1.06
    ment
    0.89
    ement
    0.87
    oso
    0.85
    ç
    0.85
    ATED
    0.85
    ATION
    0.84
    istas
    0.82
    ktop
    0.81
    iously
    0.81
    Activations Density 19.303%

    No Known Activations