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. 4217
    Prev
    Next
    INDEX
    Explanations

    elements related to web development, particularly focusing on setting width and quality for images

    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
    99
    +0.20
    1.0%
    447
    +0.13
    0.6%
    365
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    99
    +0.20
    0.11
    289
    +0.13
    -0.01
    255
    +0.12
    0.08
    Negative Logits
    plet
    -0.77
    andum
    -0.75
    zin
    -0.67
    ectomy
    -0.65
    dinand
    -0.65
    ón
    -0.64
    ogical
    -0.64
    ovie
    -0.63
    inances
    -0.63
    onto
    -0.63
    POSITIVE LOGITS
    ï¸
    0.79
    s
    0.75
    sw
    0.74
    sheet
    0.74
    sheets
    0.73
    sd
    0.72
     Width
    0.71
    span
    0.70
    pread
    0.69
    sburg
    0.69
    Activations Density 6.119%

    No Known Activations