Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE Local
    5. 6-RES_SCL-AJT
    6. 13181
    Prev
    Next
    INDEX
    Explanations

    personal details and activities of a female individual, potentially related to legal or social issues

    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
    +0.29
    1.2%
    526
    +0.15
    0.7%
    612
    +0.13
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    138
    +0.29
    0.23
    447
    +0.15
    0.26
    64
    +0.13
    -0.23
    Negative Logits
    kefeller
    -0.88
    ornia
    -0.84
    church
    -0.72
    vernment
    -0.72
    atory
    -0.71
     Skydragon
    -0.70
    folios
    -0.69
    perties
    -0.69
    fit
    -0.68
     srfAttach
    -0.67
    POSITIVE LOGITS
    pher
    1.58
    pherd
    1.37
    athed
    1.31
    athing
    1.29
    pard
    1.27
    ffield
    1.12
    ppard
    1.10
    ldon
    1.06
    'd
    1.02
    ikh
    1.02
    Activations Density 16.901%

    No Known Activations