© 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. 12825
    Prev
    Next
    INDEX
    Explanations

    references to website links and organizations

    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
    385
    +0.16
    0.6%
    447
    +0.13
    0.5%
    729
    +0.13
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    388
    +0.16
    0.08
    660
    +0.13
    0.06
    649
    +0.13
    0.02
    Negative Logits
    former
    -0.83
    fts
    -0.72
     Proced
    -0.72
     Nieto
    -0.72
     whiff
    -0.72
    things
    -0.70
     floors
    -0.70
    words
    -0.70
    ples
    -0.70
    forming
    -0.68
    POSITIVE LOGITS
    enson
    1.05
    roup
    1.04
    orer
    0.99
    uments
    0.97
    algia
    0.96
    ues
    0.93
    allery
    0.92
    ingen
    0.91
    isf
    0.90
    erv
    0.88
    Activations Density 8.948%

    No Known Activations