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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    5. GPT2-Small
    7. Residual Similar CE Local
    9. 6-RES_SCL-AJT
    11. 36259
    Prev
    Next
    INDEX
    Explanations

    Russian language in the text, especially looking for specific Russian letters like 'ð¸' and 'ru'

    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
    665
    +0.16
    0.7%
    8
    +0.12
    0.5%
    274
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.16
    0.17
    4
    +0.12
    0.09
    665
    +0.11
    0.12
    Negative Logits
    note
    -0.80
    raints
    -0.79
     GOODMAN
    -0.75
    wear
    -0.74
    notes
    -0.71
     CPC
    -0.71
    worn
    -0.70
    itutional
    -0.67
    ocene
    -0.66
     Wanted
    -0.66
    POSITIVE LOGITS
    pees
    1.27
    ggle
    1.19
    nder
    1.16
    zzle
    1.15
    pee
    1.13
    cker
    1.12
    cks
    1.12
    esome
    1.11
    gment
    1.11
    cci
    1.10
    Activations Density 5.385%

    No Known Activations