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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & 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. 4258
    Prev
    Next
    INDEX
    Explanations

    instances of sexual assault and related violent actions

    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
    +0.09
    0.9%
    553
    +0.05
    0.5%
    178
    +0.04
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    686
    +0.09
    0.06
    59
    +0.05
    0.07
    350
    +0.04
    0.09
    Negative Logits
    issue
    -0.32
    immune
    -0.32
    issues
    -0.32
    hire
    -0.30
    fare
    -0.30
    repair
    -0.30
    tag
    -0.30
    fly
    -0.30
    gap
    -0.30
    bid
    -0.29
    POSITIVE LOGITS
     Meredith
    0.30
     Samantha
    0.30
     prostitutes
    0.29
    nesday
    0.28
     TAMADRA
    0.28
     raping
    0.28
    agog
    0.27
     Paige
    0.27
    osexual
    0.27
     innoc
    0.27
    Activations Density 6.307%

    No Known Activations