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

    Neuronpedia

    Jacobian LensNEW
    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. 7-RES-JB
    11. 23952
    Prev
    Next
    INDEX
    Explanations

    references to the word "Fin" that is commonly found in terms like "Finale" or "Finishing"

    oai_token-act-pair · gpt-3.5-turbo

    references to the concept of "finishing" or completion across various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    hare
    -0.64
    theless
    -0.63
     meanings
    -0.60
    âĶģ
    -0.58
     pastoral
    -0.58
     Ae
    -0.58
     orally
    -0.57
     announcer
    -0.57
     cx
    -0.57
     clipboard
    -0.57
    POSITIVE LOGITS
    isher
    1.19
    ishers
    1.15
    ancing
    1.07
    anced
    1.06
    icum
    1.03
    itures
    1.02
    ishing
    1.02
    ners
    1.01
    anz
    0.98
    estone
    0.97
    Activations Density 0.026%

    No Known Activations