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. 4462
    Prev
    Next
    INDEX
    Explanations

    words related to decay or deterioration

    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
    178
    +0.14
    0.7%
    726
    +0.12
    0.6%
    276
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    201
    +0.14
    0.11
    343
    +0.12
    0.10
    288
    +0.09
    0.23
    Negative Logits
    TeX
    -0.74
    orsi
    -0.74
    ontent
    -0.71
    iate
    -0.67
    adr
    -0.67
    OWS
    -0.66
    romy
    -0.65
    region
    -0.64
    staking
    -0.63
    enture
    -0.63
    POSITIVE LOGITS
    vous
    0.67
     cane
    0.62
     Pett
    0.62
    ito
    0.61
    monkey
    0.60
     Ultron
    0.60
    byte
    0.59
     Fang
    0.58
    atorium
    0.58
     straw
    0.58
    Activations Density 10.015%

    No Known Activations