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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 34297
    Prev
    Next
    INDEX
    Explanations

    phrases related to statistical anomalies or abnormalities in data

    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
    642
    +0.14
    0.9%
    497
    +0.13
    0.8%
    351
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.14
    0.25
    326
    +0.13
    0.22
    744
    +0.11
    0.11
    Negative Logits
    uay
    -0.74
    sonian
    -0.72
    raq
    -0.70
    cha
    -0.69
    selage
    -0.67
    book
    -0.66
    cho
    -0.65
    fare
    -0.65
    ners
    -0.65
    nut
    -0.64
    POSITIVE LOGITS
    adesh
    0.54
     transformer
    0.53
     Cors
    0.52
     coron
    0.52
     deton
    0.51
     recess
    0.51
    ities
    0.50
    mberg
    0.49
    ignment
    0.49
    iferation
    0.48
    Activations Density 7.043%

    No Known Activations