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. 40292
    Prev
    Next
    INDEX
    Explanations

    references and mentions of the name "Hans" in the context of various activities and fields

    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
    218
    +0.11
    0.7%
    679
    +0.09
    0.6%
    57
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.11
    0.19
    107
    +0.09
    0.04
    288
    +0.09
    0.18
    Negative Logits
     Scotia
    -0.65
    ally
    -0.60
    ania
    -0.60
    selves
    -0.59
    izer
    -0.59
    ments
    -0.58
    aneously
    -0.58
    icut
    -0.57
    ista
    -0.56
    iances
    -0.55
    POSITIVE LOGITS
    lasses
    0.66
    ãĤ¼ãĤ¦ãĤ¹
    0.64
     Andersen
    0.63
     mathemat
    0.62
    peed
    0.61
    hiro
    0.60
    hee
    0.58
    BuyableInstoreAndOnline
    0.58
    hov
    0.58
    sein
    0.56
    Activations Density 6.382%

    No Known Activations