© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    5. GPT2-Small
    7. Residual Similar CE Local
    9. 6-RES_SCL-AJT
    11. 4070
    Prev
    Next
    INDEX
    Explanations

    updated information or text

    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
    +0.22
    1.0%
    140
    +0.11
    0.5%
    469
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.22
    0.11
    640
    +0.11
    0.21
    735
    +0.09
    0.08
    Negative Logits
    aden
    -0.75
    zees
    -0.72
    vous
    -0.71
    ¬¼
    -0.69
     Fein
    -0.67
    ften
    -0.66
    hov
    -0.66
    axe
    -0.66
    pipe
    -0.65
    atra
    -0.65
    POSITIVE LOGITS
     visuals
    0.72
     notifications
    0.71
     records
    0.70
     versions
    0.69
     updates
    0.68
     accordingly
    0.67
    ISC
    0.66
     snapshots
    0.66
     arrivals
    0.64
     rosters
    0.63
    Activations Density 9.308%

    No Known Activations