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

    Neuronpedia

    APICircuit 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. 45478
    Prev
    Next
    INDEX
    Explanations

    proper nouns or named entities related to people or organizations

    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
    55
    +0.11
    0.5%
    585
    +0.10
    0.5%
    309
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    675
    +0.11
    0.09
    141
    +0.10
    0.05
    509
    +0.10
    0.05
    Negative Logits
    Reviewer
    -0.69
    Unix
    -0.65
     crawl
    -0.59
     fabrication
    -0.57
     feces
    -0.56
    Track
    -0.56
     Rite
    -0.56
     rebirth
    -0.55
     deal
    -0.55
     unnecess
    -0.55
    POSITIVE LOGITS
     Pradesh
    1.08
    henko
    0.90
    neau
    0.89
    heim
    0.84
    ndra
    0.82
    chuk
    0.78
    assy
    0.76
    uez
    0.74
    és
    0.74
    uable
    0.71
    Activations Density 8.689%

    No Known Activations