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

    Neuronpedia

    APICircuit 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. 4572
    Prev
    Next
    INDEX
    Explanations

    phrases related to a metaphorical 'run' or a series of events

    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
    +0.17
    0.7%
    270
    +0.13
    0.5%
    607
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    607
    +0.17
    0.07
    422
    +0.13
    0.06
    706
    +0.12
    0.08
    Negative Logits
     Males
    -0.78
     Voice
    -0.73
    ĨĴ
    -0.71
     Hots
    -0.71
    omical
    -0.70
    alam
    -0.70
    auga
    -0.69
    olia
    -0.69
    oshop
    -0.69
    eous
    -0.67
    POSITIVE LOGITS
    aways
    1.19
    swick
    1.10
    escape
    1.10
    ners
    1.04
    gs
    1.04
    runner
    1.04
    nings
    1.03
    times
    0.98
    away
    0.98
    dy
    0.95
    Activations Density 5.539%

    No Known Activations