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. 4424
    Prev
    Next
    INDEX
    Explanations

    phrases related to winter and snow

    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
    +0.14
    0.6%
    288
    +0.14
    0.6%
    739
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    85
    +0.14
    0.10
    627
    +0.14
    0.07
    151
    +0.12
    0.12
    Negative Logits
    ENTION
    -0.87
    igious
    -0.85
    aird
    -0.83
     plur
    -0.75
    ISTER
    -0.75
     Jinn
    -0.73
    ister
    -0.72
    izabeth
    -0.71
    sych
    -0.70
    ysis
    -0.70
    POSITIVE LOGITS
    flake
    1.76
    boarding
    1.08
    fall
    1.06
    balls
    1.04
     Leopard
    1.04
    tro
    1.02
    storms
    1.02
    storm
    1.00
    mobile
    0.98
    falls
    0.92
    Activations Density 4.927%

    No Known Activations