Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 3-RES-JB
    6. 14361
    Prev
    Next
    INDEX
    Explanations

    words related to unique or unconventional characteristics or individuals

    oai_token-act-pair · gpt-3.5-turbo

    descriptive terms related to unconventional or peculiar people and behaviors

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Demand
    -0.68
    AIDS
    -0.68
    Medic
    -0.68
    fman
    -0.67
    Reviewed
    -0.67
    vation
    -0.66
    adows
    -0.65
    onds
    -0.64
    ãĥĺãĥ©
    -0.63
    pletion
    -0.63
    POSITIVE LOGITS
     quirks
    1.21
     eccentric
    1.21
    entric
    1.05
     idiosyncr
    0.96
    ities
    0.92
    yssey
    0.86
    worldly
    0.84
     err
    0.83
    odox
    0.81
     antics
    0.78
    Activations Density 0.018%

    No Known Activations