© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 39966
    Prev
    Next
    INDEX
    Explanations

    names of people, particularly with German or Dutch origin

    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
    192
    +0.13
    0.6%
    285
    +0.12
    0.5%
    439
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    138
    +0.13
    0.17
    447
    +0.12
    0.20
    439
    +0.12
    0.09
    Negative Logits
    ricular
    -1.01
    hedral
    -0.80
    rament
    -0.76
     proble
    -0.68
     odds
    -0.66
    ropolis
    -0.66
    ractions
    -0.65
    ctica
    -0.65
    arios
    -0.64
    urnal
    -0.64
    POSITIVE LOGITS
    ibrary
    1.14
    anguage
    0.93
    er
    0.90
    abel
    0.88
    geist
    0.87
    stadt
    0.86
    ocene
    0.84
    vironment
    0.83
    igible
    0.83
    mann
    0.81
    Activations Density 9.348%

    No Known Activations