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

    Neuronpedia

    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. 1-RES-JB
    11. 24246
    Prev
    Next
    INDEX
    Explanations

    words or partial words related to elements of the freeway or transportation system

    oai_token-act-pair · gpt-3.5-turbo

    instances of the prefix "fre" indicating freedom or related concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Kirin
    -0.70
     Archdemon
    -0.69
     Lum
    -0.69
     bacter
    -0.68
     Takeru
    -0.68
    ORY
    -0.65
     Lith
    -0.63
    RON
    -0.63
    ENTS
    -0.63
     Humph
    -0.63
    POSITIVE LOGITS
     fre
    1.26
    natureconservancy
    1.13
    fre
    0.97
    eport
    0.92
    ighter
    0.90
    estyle
    0.90
    zie
    0.83
    bies
    0.83
    eways
    0.81
    ights
    0.80
    Activations Density 0.004%

    No Known Activations