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. 45603
    Prev
    Next
    INDEX
    Explanations

    phrases related to DIY projects and online deals

    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
    481
    +0.11
    0.6%
    581
    +0.09
    0.5%
    373
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    464
    +0.11
    0.10
    442
    +0.09
    0.08
    585
    +0.09
    0.11
    Negative Logits
     deaf
    -0.76
    ĪĴ
    -0.64
     Whe
    -0.63
    paio
    -0.60
    twitch
    -0.58
     COUR
    -0.57
    wow
    -0.55
    slow
    -0.55
    netflix
    -0.55
    sey
    -0.54
    POSITIVE LOGITS
    itures
    0.92
    iture
    0.84
     receipts
    0.81
     funds
    0.81
     money
    0.73
     Funds
    0.73
     stash
    0.72
     Donation
    0.71
     Bitcoins
    0.69
     goods
    0.69
    Activations Density 7.039%

    No Known Activations