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. 13219
    Prev
    Next
    INDEX
    Explanations

    phrases or words related to the concept of 'crazy' or unusual behavior

    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
    +0.27
    1.3%
    326
    +0.12
    0.6%
    132
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    271
    +0.27
    0.06
    351
    +0.12
    0.05
    408
    +0.11
    0.12
    Negative Logits
    Reviewed
    -0.83
    agle
    -0.73
    ãĥĺãĥ©
    -0.73
    adr
    -0.72
    apers
    -0.71
    arers
    -0.70
    fman
    -0.68
    ribut
    -0.68
    çĦ
    -0.68
    tein
    -0.66
    POSITIVE LOGITS
     nuts
    0.69
     crazy
    0.64
     shit
    0.63
    spawn
    0.61
     antics
    0.61
     overboard
    0.61
     beasts
    0.59
    cap
    0.59
     chic
    0.58
     hog
    0.57
    Activations Density 6.793%

    No Known Activations