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. 34364
    Prev
    Next
    INDEX
    Explanations

    references to a specific individual named "oss" and related variations

    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
    535
    +0.12
    0.5%
    649
    +0.10
    0.5%
    729
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.12
    0.10
    4
    +0.10
    0.08
    729
    +0.10
    0.07
    Negative Logits
    ãĥ£
    -0.82
    ric
    -0.72
    ricks
    -0.70
    ropolitan
    -0.69
    ãĥ¥
    -0.68
    ufact
    -0.67
    è¦ļéĨĴ
    -0.64
    rics
    -0.64
    crime
    -0.62
    Beast
    -0.61
    POSITIVE LOGITS
     Whedon
    1.26
    essed
    1.03
    essing
    0.98
    essor
    0.96
    enger
    0.95
    es
    0.92
    essors
    0.91
    idents
    0.81
    ips
    0.80
    ortment
    0.79
    Activations Density 5.900%

    No Known Activations