© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 0-RES-JB
    11. 18964
    Prev
    Next
    INDEX
    Explanations

    words related to physical features or characteristics, specifically focusing on the concept of 'evil'

    oai_token-act-pair · gpt-3.5-turbo

    words related to the concept of 'curve' or 'curvilinear' forms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     soak
    -0.66
     BT
    -0.65
     OT
    -0.62
     receipts
    -0.62
     soaking
    -0.60
     stripe
    -0.60
     transcripts
    -0.59
     Hin
    -0.59
     fold
    -0.58
     shake
    -0.57
    POSITIVE LOGITS
    vil
    4.72
    vae
    1.09
    vic
    1.07
    abase
    1.06
    vel
    1.03
    va
    1.02
    ves
    1.02
    vered
    1.01
    ville
    0.99
    val
    0.98
    Activations Density 0.021%

    No Known Activations