© 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. 34786
    Prev
    Next
    INDEX
    Explanations

    words related to mental health issues and medical conditions

    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
    288
    +0.17
    0.7%
    575
    +0.16
    0.7%
    447
    +0.15
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    450
    +0.17
    0.06
    483
    +0.16
    0.05
    699
    +0.15
    0.02
    Negative Logits
    ICA
    -0.85
    advertisement
    -0.81
    ded
    -0.79
    ARDS
    -0.79
    oulos
    -0.78
    gger
    -0.77
    IRD
    -0.76
    ELS
    -0.73
    HER
    -0.71
    eled
    -0.71
    POSITIVE LOGITS
     faculties
    1.09
     illness
    1.03
     disorders
    0.94
     defic
    0.94
    izing
    0.93
    ising
    0.91
     retard
    0.90
     disorder
    0.90
     anguish
    0.89
     wellbeing
    0.89
    Activations Density 5.349%

    No Known Activations