© 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 Local
    9. 6-RES_SCL-AJT
    11. 45899
    Prev
    Next
    INDEX
    Explanations

    references to specific locations, particularly boroughs and towns

    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
    68
    +0.04
    0.7%
    420
    +0.04
    0.6%
    167
    +0.04
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    649
    +0.04
    0.03
    288
    +0.04
    0.16
    651
    +0.04
    0.09
    Negative Logits
    xtap
    -0.27
    riott
    -0.27
    xit
    -0.24
    cius
    -0.24
    conn
    -0.23
    ACTED
    -0.23
    xual
    -0.23
    ensing
    -0.23
    icago
    -0.22
    zza
    -0.22
    POSITIVE LOGITS
     borough
    0.32
     Borough
    0.32
     councils
    0.25
    mates
    0.24
    redit
    0.22
    abulary
    0.22
    tis
    0.22
     chambers
    0.22
     Asylum
    0.22
     oun
    0.22
    Activations Density 5.085%

    No Known Activations