© 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. 8-RES-JB
    11. 11005
    Prev
    Next
    INDEX
    Explanations

    words related to instructions or actions that involve a process or sequence of steps

    oai_token-act-pair · gpt-3.5-turbo

    terms associated with judgment and evaluation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    words written with British spelling

    Explanation Uploaded by User
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     VIDEOS
    -0.83
    eah
    -0.72
    Shot
    -0.71
    thal
    -0.71
     Carlson
    -0.70
     Russo
    -0.67
     Dull
    -0.66
    externalActionCode
    -0.66
    OHN
    -0.64
     Capitol
    -0.64
    POSITIVE LOGITS
    £
    1.08
     Whilst
    0.92
    uminium
    0.91
     analyse
    0.89
     programmes
    0.86
     tyre
    0.84
     tyres
    0.82
     £
    0.82
     aluminium
    0.82
     defences
    0.81
    Activations Density 0.677%

    No Known Activations