© 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. 3813
    Prev
    Next
    INDEX
    Explanations

    gender-related topics, particularly focusing on women's issues such as power dynamics, societal expectations, and disparities between men and women

    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
    +0.21
    0.9%
    57
    +0.09
    0.4%
    743
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    187
    +0.21
    0.09
    38
    +0.09
    0.08
    15
    +0.09
    0.10
    Negative Logits
    IVERS
    -0.86
    Assembly
    -0.76
    Ward
    -0.71
    REC
    -0.69
    ITS
    -0.67
    REDACTED
    -0.67
    UGE
    -0.66
    Burn
    -0.63
    BILL
    -0.63
    Manufact
    -0.63
    POSITIVE LOGITS
    volent
    1.33
    opausal
    1.05
     ejac
    0.96
    ager
    0.82
    icide
    0.79
     genitals
    0.78
    folk
    0.76
    endez
    0.72
    iji
    0.71
    handled
    0.71
    Activations Density 9.641%

    No Known Activations