© 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. 2-RES-JB
    11. 11786
    Prev
    Next
    INDEX
    Explanations

    the prefix "Im-" followed by a word root, possibly suggesting a focus on words related to importance or immediacy

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "I'm" or related variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    NetMessage
    -1.01
    Downloadha
    -0.84
     Skydragon
    -0.76
     dinosaurs
    -0.75
     Warwick
    -0.66
    é¾įå¥ij士
    -0.65
     Cth
    -0.65
     Kirin
    -0.65
    çĦ
    -0.64
    姫
    -0.64
    POSITIVE LOGITS
    mediate
    1.23
    ple
    1.21
    medi
    1.18
    mediately
    1.16
    moral
    1.15
    perfect
    1.14
    unity
    1.13
    perial
    1.13
    gur
    1.06
    posing
    1.01
    Activations Density 0.010%

    No Known Activations