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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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 E2E Recon
    9. 6-RES_SCEFR-AJT
    11. 99
    Prev
    Next
    INDEX
    Explanations

    countries and geographical locations, specifically focusing on the region of the Balkans

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-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
    612
    +0.14
    0.6%
    57
    +0.14
    0.6%
    240
    +0.14
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.14
    0.10
    57
    +0.14
    0.06
    326
    +0.14
    0.08
    Negative Logits
    icles
    -1.03
    iating
    -0.96
    icle
    -0.88
    iments
    -0.86
    uate
    -0.83
    iants
    -0.81
    imov
    -0.80
    iation
    -0.80
    stand
    -0.79
    imental
    -0.77
    POSITIVE LOGITS
     Springs
    0.97
    ignt
    0.83
    utical
    0.83
    boro
    0.82
    ħĭ
    0.78
    senal
    0.78
     Herz
    0.77
    posium
    0.77
    hurst
    0.77
     cs
    0.77
    Activations Density 1.019%

    No Known Activations