© 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. 6576
    Prev
    Next
    INDEX
    Explanations

    names or words related to the name "Har" with varying degrees of activation

    oai_token-act-pair · gpt-3.5-turbo

    the repeated mention of the word "Har" in various contexts

    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
    éĹĺ
    -1.05
    ablishment
    -0.80
    xual
    -0.78
    uality
    -0.74
    lihood
    -0.71
    PsyNetMessage
    -0.71
    anwhile
    -0.70
    REDACTED
    -0.70
    ancial
    -0.69
     Giuliani
    -0.69
    POSITIVE LOGITS
    rod
    1.11
    rah
    1.05
    riers
    1.05
    rim
    1.04
    vard
    1.01
    rier
    1.00
    mel
    0.99
    rigan
    0.91
    rowing
    0.91
    rington
    0.90
    Activations Density 0.008%

    No Known Activations