© 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
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 3-RES-JB
    11. 10656
    Prev
    Next
    INDEX
    Explanations

    phrases related to parts, portions, or shares of something

    oai_token-act-pair · gpt-3.5-turbo

    references to portions or sections of various subjects or things

    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
    generic
    -0.69
    ATHER
    -0.68
    verbs
    -0.65
    natural
    -0.63
    NM
    -0.62
     Fighters
    -0.62
    asu
    -0.62
    Iran
    -0.62
    ³³³
    -0.62
    Tok
    -0.60
    POSITIVE LOGITS
     thereof
    0.98
    edly
    0.85
    icularly
    0.83
     guiActiveUn
    0.83
     ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³
    0.76
    ĸļ
    0.76
     portions
    0.75
    meal
    0.73
    itional
    0.73
    ership
    0.71
    Activations Density 0.016%

    No Known Activations