© 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. 1-RES-JB
    11. 20761
    Prev
    Next
    INDEX
    Explanations

    phrases referring to actions taken after a certain event or situation

    oai_token-act-pair · gpt-3.5-turbo

    references to events occurring after a specified moment in time

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ASE
    -0.75
    foot
    -0.67
    yu
    -0.64
    EMA
    -0.64
    hma
    -0.63
    KS
    -0.63
    rod
    -0.62
    icles
    -0.60
    HOU
    -0.59
    AMS
    -0.59
    POSITIVE LOGITS
     afterwards
    1.09
     afterward
    1.02
    etheless
    0.90
     satell
    0.86
     confir
    0.85
     thereafter
    0.85
     conclud
    0.82
    theless
    0.79
    noon
    0.74
     encount
    0.73
    Activations Density 0.008%

    No Known Activations