© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    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. 40781
    Prev
    Next
    INDEX
    Explanations

    affiliate links and mentions of affiliate partnerships in text related to business and products

    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
    447
    +0.14
    0.7%
    679
    +0.12
    0.6%
    288
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    505
    +0.14
    0.08
    549
    +0.12
    0.07
    146
    +0.10
    0.08
    Negative Logits
    erion
    -0.65
    ppy
    -0.63
    pter
    -0.63
    Iv
    -0.62
    assetsadobe
    -0.58
    STD
    -0.57
    ests
    -0.57
    sh
    -0.56
    esa
    -0.56
    efer
    -0.56
    POSITIVE LOGITS
     affiliate
    0.89
    iliated
    0.85
    iliate
    0.82
    hips
    0.81
     affiliated
    0.76
     affiliation
    0.72
     recru
    0.72
     affiliates
    0.71
     affili
    0.69
    henko
    0.64
    Activations Density 9.374%

    No Known Activations