© 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. 35256
    Prev
    Next
    INDEX
    Explanations

    names of people or entities

    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
    364
    +0.10
    0.5%
    264
    +0.10
    0.5%
    685
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    264
    +0.10
    0.07
    630
    +0.10
    0.07
    597
    +0.08
    0.05
    Negative Logits
     NEC
    -0.60
     Niet
    -0.58
    ascript
    -0.53
    izabeth
    -0.53
    ruary
    -0.53
     Jackie
    -0.52
    issance
    -0.51
     Kissinger
    -0.50
     Jericho
    -0.50
     Goose
    -0.50
    POSITIVE LOGITS
    acus
    0.64
    llah
    0.63
    inki
    0.61
    allah
    0.61
    nikov
    0.61
    pas
    0.60
    anan
    0.60
    zai
    0.60
    shaw
    0.58
    Pacific
    0.58
    Activations Density 14.038%

    No Known Activations