© 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
    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. 34925
    Prev
    Next
    INDEX
    Explanations

    details related to relationships and family dynamics in TV show plots

    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
    36
    +0.13
    0.5%
    495
    +0.11
    0.5%
    562
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.13
    0.33
    6
    +0.11
    0.21
    64
    +0.10
    -0.23
    Negative Logits
    allows
    -0.74
    Reviewed
    -0.69
    izont
    -0.68
     NK
    -0.67
    ifications
    -0.67
    hett
    -0.65
    ariat
    -0.65
     dogs
    -0.65
    ankind
    -0.65
     partName
    -0.65
    POSITIVE LOGITS
    ccording
    1.10
    Lago
    0.97
    ļé
    0.90
    pillar
    0.89
    ©¶æ¥µ
    0.87
    ENA
    0.83
    aez
    0.81
    MET
    0.79
    ICAN
    0.78
    VI
    0.77
    Activations Density 11.657%

    No Known Activations