© 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. 4210
    Prev
    Next
    INDEX
    Explanations

    mentions of different countries, especially Australia

    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
    +0.23
    0.9%
    515
    +0.15
    0.6%
    288
    +0.15
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    515
    +0.23
    0.10
    288
    +0.15
    0.16
    30
    +0.15
    0.05
    Negative Logits
    outer
    -0.89
    osen
    -0.86
    ppings
    -0.86
    soDeliveryDate
    -0.86
    ows
    -0.84
    Lens
    -0.83
    outed
    -0.80
    lighting
    -0.79
    acco
    -0.79
    lasses
    -0.78
    POSITIVE LOGITS
     Australia
    1.01
     Broadcasting
    0.98
    gdala
    0.98
     Aboriginal
    0.94
    Australia
    0.94
     Geographic
    0.91
     Australians
    0.90
     Defence
    0.90
     Samoa
    0.88
     Warlock
    0.87
    Activations Density 6.615%

    No Known Activations