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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 2-RES-JB
    11. 7248
    Prev
    Next
    INDEX
    Explanations

    phrases related to leveling or leveling up in a context such as leveling charges or leveling up in a game

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "level" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    enegger
    -0.84
    ROM
    -0.73
     Sidd
    -0.72
     Experimental
    -0.70
     Camer
    -0.68
    Nature
    -0.65
    1027
    -0.64
     Ultr
    -0.64
     Dragonbound
    -0.64
    otos
    -0.63
    POSITIVE LOGITS
     leve
    1.66
     leveled
    1.51
    lling
    1.13
    lled
    1.10
     leveling
    0.96
    aton
    0.82
    illet
    0.81
     challeng
    0.76
    ldom
    0.75
    imar
    0.75
    Activations Density 0.013%

    No Known Activations