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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 4000
    Prev
    Next
    INDEX
    Explanations

    details related to personal experiences and relational dynamics

    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
    610
    +0.13
    0.6%
    321
    +0.11
    0.5%
    603
    +0.11
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    393
    +0.13
    0.35
    283
    +0.11
    -0.01
    14
    +0.11
    0.05
    Negative Logits
    pmwiki
    -0.91
     Lobby
    -0.88
    igil
    -0.84
    soType
    -0.82
     instituted
    -0.78
    ondon
    -0.78
     banning
    -0.77
     Armory
    -0.76
     commissions
    -0.75
    ablishment
    -0.74
    POSITIVE LOGITS
     todd
    1.07
     kindergarten
    1.01
     caregivers
    1.00
     careg
    0.99
     preschool
    0.97
    Dad
    0.94
     immature
    0.94
     leukemia
    0.92
     siblings
    0.90
     toddlers
    0.90
    Activations Density 19.897%

    No Known Activations