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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 13294
    Prev
    Next
    INDEX
    Explanations

    technical terms related to programming and system security

    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
    640
    +0.18
    0.8%
    77
    +0.13
    0.6%
    393
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.18
    0.30
    679
    +0.13
    0.24
    393
    +0.11
    0.27
    Negative Logits
    rawdownloadcloneembedreportprint
    -0.96
    istar
    -0.76
    istries
    -0.70
    Tomorrow
    -0.69
     neutrality
    -0.66
    amins
    -0.64
     rall
    -0.63
    iage
    -0.62
    ifer
    -0.61
    :(
    -0.61
    POSITIVE LOGITS
     coli
    0.87
    +)
    0.78
    rex
    0.76
    viously
    0.71
    gger
    0.71
    lux
    0.70
    prototype
    0.70
     Lovecraft
    0.69
    hower
    0.68
    eg
    0.67
    Activations Density 10.225%

    No Known Activations