© 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. 8-RES-JB
    11. 16957
    Prev
    Next
    INDEX
    Explanations

    mentions of someone acquiring or touching objects

    oai_token-act-pair · gpt-3.5-turbo

    phrases that reference physical possessions or control over something

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    arenthood
    -0.63
     Recomm
    -0.58
    AMI
    -0.56
    izabeth
    -0.55
    anto
    -0.55
    theless
    -0.55
    egu
    -0.53
    versely
    -0.53
    eties
    -0.53
     Cosmos
    -0.53
    POSITIVE LOGITS
    ħĭ
    0.71
    icc
    0.67
    horn
    0.66
     blown
    0.66
     dirty
    0.66
     sore
    0.65
     crossed
    0.64
    uc
    0.62
    Ĥ¬
    0.61
     firmly
    0.60
    Activations Density 0.105%

    No Known Activations