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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & 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. 34544
    Prev
    Next
    INDEX
    Explanations

    technical details related to technology developments and innovations, specifically in the aerospace industry

    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
    99
    +0.04
    0.6%
    549
    +0.03
    0.5%
    62
    +0.03
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.04
    0.21
    659
    +0.03
    0.11
    99
    +0.03
    0.10
    Negative Logits
    istic
    -0.25
    arbon
    -0.25
    istically
    -0.25
    iazep
    -0.25
    ese
    -0.24
    sei
    -0.24
    eele
    -0.24
    stown
    -0.23
    tes
    -0.23
    ized
    -0.23
    POSITIVE LOGITS
    ĨĴ
    0.30
     Hancock
    0.26
    guyen
    0.26
    metry
    0.26
    ĵĺ
    0.23
     Mae
    0.23
     Nguyen
    0.21
     HIP
    0.20
     monop
    0.20
     Normandy
    0.20
    Activations Density 7.667%

    No Known Activations