© 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. 4-RES-JB
    11. 14743
    Prev
    Next
    INDEX
    Explanations

    words related to starting or initiating actions

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "begin" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 4-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    4
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    aths
    -0.71
    allo
    -0.70
    rats
    -0.69
    visor
    -0.67
    otor
    -0.67
    houses
    -0.67
    ado
    -0.67
    acho
    -0.65
    oted
    -0.65
    tor
    -0.65
    POSITIVE LOGITS
     anew
    1.11
     "$:/
    0.78
    ITIES
    0.72
    WithNo
    0.72
    nings
    0.71
    icago
    0.70
     ende
    0.70
    20439
    0.68
    igmatic
    0.68
    mble
    0.68
    Activations Density 0.040%

    No Known Activations