© 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. 40440
    Prev
    Next
    INDEX
    Explanations

    phrases related to serving a purpose or function

    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
    +0.23
    1.5%
    393
    +0.16
    1.0%
    36
    +0.12
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    36
    +0.23
    0.30
    393
    +0.16
    0.35
    159
    +0.12
    0.19
    Negative Logits
    encies
    -0.66
    intent
    -0.60
    faces
    -0.60
    events
    -0.60
    errors
    -0.60
    words
    -0.59
    rants
    -0.58
    chev
    -0.58
    isms
    -0.57
    lees
    -0.57
    POSITIVE LOGITS
     precursor
    0.66
     standalone
    0.63
     buffer
    0.60
     cornerstone
    0.60
     catalyst
    0.60
     reminder
    0.59
     surrogate
    0.58
     deterrent
    0.58
     blueprint
    0.58
     substitute
    0.58
    Activations Density 14.058%

    No Known Activations