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 L0 E2E Recon
    5. 2-RES_SLEFR-AJT
    6. 9859
    Prev
    Next
    INDEX
    Explanations

    The word "Enter" is used in various contexts, including as a command (e.g., press Enter), as a transition word introducing a new topic or character, and as part of a title or label. It's often found in instructions, narrative text, and lists.

    eleuther_acts_top20 · gemini-1.5-flashTriggered by @johnny

    instances of the word "Enter" and related formatting cues for user interaction

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_slefr-ajt/2-res_slefr-ajt
    Prompts (Dashboard)
    12,288 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    46,080
    Data Type
    torch.float32
    Hook Point
    blocks.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    apollo-research/Skylion007-openwebtext-tokenizer-gpt2
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    360
    +0.14
    0.6%
    503
    +0.11
    0.5%
    405
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    360
    +0.14
    0.01
    415
    +0.11
    0.01
    45
    +0.10
    0.01
    Negative Logits
     awa
    -0.77
    iewicz
    -0.67
     intrusive
    -0.66
     suppressing
    -0.63
    76561
    -0.62
     interfering
    -0.61
    ^^^^
    -0.61
    acca
    -0.60
     Cath
    -0.59
     privilege
    -0.59
    POSITIVE LOGITS
    prise
    1.03
    tainment
    0.97
    taining
    0.94
    tain
    0.92
    prises
    0.86
    brainer
    0.79
    odor
    0.77
    udos
    0.75
    ulia
    0.70
    Term
    0.70
    Activations Density 0.010%

    No Known Activations