© 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. 33090
    Prev
    Next
    INDEX
    Explanations

    phrases related to social issues and political commentary

    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
    +0.40
    2.2%
    393
    +0.23
    1.2%
    732
    +0.08
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    393
    +0.40
    0.29
    0
    +0.23
    0.16
    703
    +0.08
    0.14
    Negative Logits
    çīĪ
    -0.70
    icol
    -0.69
    ummer
    -0.63
    :(
    -0.63
    ãĥĥãĥī
    -0.62
    ãĤº
    -0.61
    acent
    -0.60
    ixel
    -0.60
    wic
    -0.59
    owe
    -0.59
    POSITIVE LOGITS
     aesthetics
    0.76
     causation
    0.75
     thus
    0.71
     hence
    0.71
     degradation
    0.70
     consequ
    0.69
     infiltration
    0.69
     solitude
    0.68
     consequently
    0.67
     creativity
    0.67
    Activations Density 11.396%

    No Known Activations