© 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. 5111
    Prev
    Next
    INDEX
    Explanations

    phrases indicating restraint or inhibition

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating restraint or delay

    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
    ²¾
    -0.84
    swick
    -0.75
    ivia
    -0.74
    aundering
    -0.71
    aters
    -0.71
    ioxide
    -0.71
    ceans
    -0.69
    apest
    -0.68
    ģ«
    -0.66
    iterranean
    -0.66
    POSITIVE LOGITS
    hold
    0.78
     ransom
    0.78
     hold
    0.77
     tight
    0.73
     hostage
    0.70
     reins
    0.70
     sway
    0.65
     ledge
    0.64
     grip
    0.64
     plun
    0.63
    Activations Density 0.082%

    No Known Activations