© 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. 10-RES-JB
    11. 16430
    Prev
    Next
    INDEX
    Explanations

    terms or phrases that are referred to as something else

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate something is commonly referred to by a specific name or term

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    istries
    -0.64
    cot
    -0.62
    hibition
    -0.61
     Skydragon
    -0.58
    iscopal
    -0.58
    nesia
    -0.57
    osion
    -0.56
    ointment
    -0.56
    appiness
    -0.53
    redo
    -0.53
    POSITIVE LOGITS
     hereafter
    1.09
     herein
    0.93
     as
    0.91
     collectively
    0.90
     derog
    0.81
     internally
    0.79
    literally
    0.74
     euphem
    0.72
     loosely
    0.69
    okingly
    0.68
    Activations Density 0.063%

    No Known Activations