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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 45305
    Prev
    Next
    INDEX
    Explanations

    phrases related to structural elements or issues in various contexts

    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
    288
    +0.14
    0.7%
    99
    +0.13
    0.6%
    608
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.14
    0.18
    284
    +0.13
    0.09
    99
    +0.11
    0.08
    Negative Logits
    renheit
    -0.76
    onen
    -0.72
     Carbuncle
    -0.71
    shire
    -0.69
    vier
    -0.68
    zig
    -0.65
    spr
    -0.65
    ossier
    -0.64
     Furious
    -0.64
    stretched
    -0.63
    POSITIVE LOGITS
     components
    0.90
    ized
    0.87
     integrity
    0.86
     elements
    0.84
     barriers
    0.84
     abnormalities
    0.82
     impairment
    0.81
     defects
    0.80
     aspects
    0.80
     anomalies
    0.80
    Activations Density 4.157%

    No Known Activations