© 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. 4-RES-JB
    11. 24125
    Prev
    Next
    INDEX
    Explanations

    documents, reports, or information provided to specific recipients or groups

    oai_token-act-pair · gpt-3.5-turbo

    prepositions and particles indicating direction or intention

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

    No Comments

    Negative Logits
    hov
    -0.76
    hovah
    -0.76
    ifted
    -0.71
    iour
    -0.69
    oother
    -0.69
    ertility
    -0.68
    iltration
    -0.68
    antam
    -0.66
    psey
    -0.65
    akedown
    -0.64
    POSITIVE LOGITS
    ãĤ¤ãĥĪ
    0.74
     Mesa
    0.73
     Tre
    0.67
    ãĥ¼ãĥ³
    0.66
     removable
    0.64
    Chan
    0.63
     Treasure
    0.61
    ãĥ¼ãĥĨãĤ£
    0.59
     Jail
    0.59
    ãĤ´ãĥ³
    0.58
    Activations Density 0.554%

    No Known Activations