© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 22701
    Prev
    Next
    INDEX
    Explanations

    profane words related to forceful or aggressive actions

    oai_token-act-pair · gpt-3.5-turbo

    verbs related to vulgarity and swearing

    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
     MAP
    -0.70
     Ara
    -0.66
     Manit
    -0.66
     Prof
    -0.64
     Ti
    -0.64
     Ver
    -0.64
     forge
    -0.63
     toll
    -0.60
     Motorola
    -0.60
     Cy
    -0.59
    POSITIVE LOGITS
    ucking
    3.63
    ucked
    3.52
    ucker
    2.95
    ucks
    2.41
    uck
    2.20
    UCK
    1.94
    ucky
    1.39
    arted
    1.18
    ocks
    1.06
    angled
    1.01
    Activations Density 0.017%

    No Known Activations