© 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. 40633
    Prev
    Next
    INDEX
    Explanations

    phrases related to viewing and inspecting content

    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
    1.0%
    195
    +0.15
    0.6%
    380
    +0.13
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    124
    +0.27
    0.13
    3
    +0.15
    0.07
    608
    +0.13
    0.14
    Negative Logits
    vous
    -0.95
    vernment
    -0.84
    milo
    -0.84
    accompan
    -0.84
    wealth
    -0.79
     whistle
    -0.78
    backer
    -0.78
    pmwiki
    -0.75
    ivari
    -0.75
     isot
    -0.75
    POSITIVE LOGITS
    ById
    1.33
    ports
    1.30
    finder
    1.19
    port
    1.15
    ership
    1.05
    points
    1.02
     largeDownload
    0.94
    opsis
    0.92
    ĸ
    0.91
    ories
    0.90
    Activations Density 8.957%

    No Known Activations