© 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. 32695
    Prev
    Next
    INDEX
    Explanations

    mentions of data from scientific research or technical processes

    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
    373
    +0.05
    0.6%
    721
    +0.04
    0.5%
    497
    +0.04
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.05
    -0.23
    406
    +0.04
    0.12
    36
    +0.04
    0.14
    Negative Logits
    enegger
    -0.26
    ++++++++
    -0.24
     performed
    -0.23
    elson
    -0.23
    appropriate
    -0.23
    gov
    -0.22
    graded
    -0.22
    recomm
    -0.22
     incurred
    -0.22
    pport
    -0.22
    POSITIVE LOGITS
    tons
    0.28
    escent
    0.26
     Sunrise
    0.25
     Turing
    0.25
     Chrys
    0.25
     din
    0.25
    Chel
    0.24
     Skies
    0.23
     glim
    0.22
    ILE
    0.22
    Activations Density 13.296%

    No Known Activations