© 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. 4896
    Prev
    Next
    INDEX
    Explanations

    words related to qualities and characteristics such as newness, perfect, and ironies

    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
    +0.27
    1.4%
    481
    +0.12
    0.6%
    255
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    255
    +0.27
    0.12
    35
    +0.12
    0.08
    285
    +0.11
    0.06
    Negative Logits
    ergy
    -0.66
    rament
    -0.61
    odium
    -0.61
    inav
    -0.59
    ramer
    -0.58
    oval
    -0.56
    iverse
    -0.55
    ompl
    -0.55
    nesium
    -0.55
    IRC
    -0.53
    POSITIVE LOGITS
    iness
    1.12
    ously
    0.89
    fulness
    0.82
    lessness
    0.81
     perv
    0.75
    uous
    0.69
     quot
    0.67
    ilty
    0.66
    ily
    0.65
    icity
    0.65
    Activations Density 9.740%

    No Known Activations