© 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. 8135
    Prev
    Next
    INDEX
    Explanations

    the word "icks" with high activation values

    oai_token-act-pair · gpt-3.5-turbo

    recurring patterns or elements with the suffix 'icks'

    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
     historian
    -0.67
     reality
    -0.66
     Aurora
    -0.62
     termin
    -0.61
     database
    -0.60
     tra
    -0.59
     historical
    -0.58
     nar
    -0.58
     Val
    -0.57
     Vita
    -0.57
    POSITIVE LOGITS
    icks
    4.54
    ICK
    2.12
    icked
    2.05
    icking
    2.04
    ick
    1.98
    icky
    1.72
    ickers
    1.64
    icker
    1.54
    ricks
    1.35
    ics
    1.32
    Activations Density 0.007%

    No Known Activations