© 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
    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. 40454
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "sing" regardless of context

    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
    439
    +0.13
    0.5%
    33
    +0.13
    0.5%
    276
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.13
    0.19
    326
    +0.13
    0.19
    701
    +0.12
    0.07
    Negative Logits
    adeon
    -0.92
     eleph
    -0.85
    Downloadha
    -0.82
    umenthal
    -0.78
    ãĤ·ãĥ£
    -0.77
     VICE
    -0.76
    utherford
    -0.75
    BIL
    -0.75
    OURCE
    -0.75
    izons
    -0.74
    POSITIVE LOGITS
    ularity
    1.55
    apore
    1.48
    leton
    1.44
    ular
    1.25
    Sing
    1.25
    ling
    1.16
    lers
    1.14
    rab
    1.10
    lest
    1.08
    lings
    1.03
    Activations Density 5.864%

    No Known Activations