© 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. 5990
    Prev
    Next
    INDEX
    Explanations

    phrases indicating personal statements or dialogue

    oai_token-act-pair · gpt-3.5-turbo

    punctuation marks, specifically quotation marks

    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
    ²¾
    -0.90
    ·
    -0.75
    adem
    -0.70
    ĵĺ
    -0.68
    etheless
    -0.65
    İĭ
    -0.65
    abet
    -0.64
    anmar
    -0.63
    eco
    -0.63
    eers
    -0.61
    POSITIVE LOGITS
     reads
    1.07
     recalls
    0.83
    said
    0.82
     says
    0.79
     said
    0.79
     recalled
    0.70
     according
    0.69
     wrote
    0.68
     writes
    0.68
    ],"
    0.66
    Activations Density 0.082%

    No Known Activations