© 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. 8-RES-JB
    6. 3087
    Prev
    Next
    INDEX
    Explanations

    numeric quantities related to storage capacity

    oai_token-act-pair · gpt-3.5-turbo

    quantities and capacities related to objects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
    insk
    -0.78
    atre
    -0.75
    cation
    -0.72
    isle
    -0.72
    é»Ĵ
    -0.68
    isan
    -0.68
    orp
    -0.65
    åĤ
    -0.65
    browser
    -0.63
    inav
    -0.63
    POSITIVE LOGITS
     occupants
    0.77
     dozens
    0.77
     3000
    0.76
     capacities
    0.75
     loads
    0.75
     hundreds
    0.74
     twins
    0.73
     6000
    0.72
     thousands
    0.71
     tons
    0.71
    Activations Density 0.232%

    No Known Activations