© 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. 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. 20925
    Prev
    Next
    INDEX
    Explanations

    phrases related to desires or aspirations

    oai_token-act-pair · gpt-3.5-turbo

    instances of the prefix "des" indicating negative connotations or denoting removal/undoing in words

    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
    glers
    -0.90
    Reviewer
    -0.79
    hetti
    -0.70
     Galile
    -0.69
     spoiler
    -0.68
     spoilers
    -0.64
    aneers
    -0.63
     SEAL
    -0.63
     Factor
    -0.62
     Penguin
    -0.61
    POSITIVE LOGITS
    igned
    1.10
    perate
    1.10
     Moines
    1.04
    ync
    0.96
    erving
    0.95
    pair
    0.92
    erve
    0.92
    erved
    0.92
    ple
    0.91
    ktop
    0.89
    Activations Density 0.005%

    No Known Activations