© 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. 4-RES-JB
    11. 1180
    Prev
    Next
    INDEX
    Explanations

    character sequences that don't seem to correspond to meaningful text or patterns

    oai_token-act-pair · gpt-3.5-turbo

    patterns related to numerical sequences or codes

    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
     destro
    -0.86
     Seym
    -0.75
     insult
    -0.72
    theless
    -0.70
     choked
    -0.70
    aimon
    -0.68
     devastated
    -0.66
     mathemat
    -0.66
     undermin
    -0.65
     crooked
    -0.65
    POSITIVE LOGITS
    AppData
    1.10
    features
    0.99
    Parameters
    0.91
    bryce
    0.90
    (\
    0.89
    src
    0.87
    addons
    0.84
    framework
    0.83
    gravity
    0.81
    xff
    0.81
    Activations Density 0.004%

    No Known Activations