Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 4163
    Prev
    Next
    INDEX
    Explanations

    procedures, reports, and allegations listed in a document

    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
    +0.22
    0.9%
    497
    +0.12
    0.5%
    129
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    576
    +0.22
    0.03
    672
    +0.12
    0.05
    651
    +0.11
    0.08
    Negative Logits
     Aber
    -0.78
    irgin
    -0.72
    icago
    -0.69
     Huck
    -0.68
     Fine
    -0.67
    ione
    -0.66
     Abbey
    -0.66
    idge
    -0.66
    Fine
    -0.66
    icer
    -0.65
    POSITIVE LOGITS
    erv
    1.30
    lists
    1.00
    icles
    1.00
    icle
    0.96
    abet
    0.92
    ening
    0.91
    erve
    0.90
    ener
    0.89
     comprehens
    0.88
    lessness
    0.88
    Activations Density 7.203%

    No Known Activations