Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE Local
    5. 6-RES_SCL-AJT
    6. 45427
    Prev
    Next
    INDEX
    Explanations

    organizations or institutions with the word "Center" in their name

    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
    +0.26
    1.1%
    752
    +0.11
    0.5%
    59
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    442
    +0.26
    0.10
    752
    +0.11
    0.13
    640
    +0.11
    0.20
    Negative Logits
     curfew
    -0.77
     hars
    -0.75
     deterrent
    -0.69
     mans
    -0.68
     certs
    -0.67
     mating
    -0.67
     vag
    -0.66
     vomit
    -0.66
    utra
    -0.66
     pageant
    -0.65
    POSITIVE LOGITS
    eous
    0.87
     Against
    0.81
    Point
    0.81
    tyard
    0.81
    tern
    0.79
    zee
    0.78
    fold
    0.77
     Institute
    0.76
     Register
    0.76
    tta
    0.75
    Activations Density 8.596%

    No Known Activations