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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 39774
    Prev
    Next
    INDEX
    Explanations

    technical terms and jargon related to computer hardware and programming

    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
    +0.36
    1.7%
    481
    +0.31
    1.4%
    459
    +0.14
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    481
    +0.36
    0.39
    570
    +0.31
    0.00
    351
    +0.14
    0.08
    Negative Logits
    lance
    -0.78
     assassinated
    -0.72
     Charity
    -0.71
     largeDownload
    -0.70
     Journalism
    -0.69
     Syri
    -0.69
     salute
    -0.68
    women
    -0.68
     prophetic
    -0.67
    tip
    -0.67
    POSITIVE LOGITS
     discrete
    0.92
     subsystem
    0.89
     variables
    0.88
     inputs
    0.88
     optimization
    0.85
     branching
    0.84
     batches
    0.84
     finite
    0.83
     bits
    0.83
     redundant
    0.82
    Activations Density 25.674%

    No Known Activations