© 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. 0-RES-JB
    11. 501
    Prev
    Next
    INDEX
    Explanations

    Italian words

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "di" as a prefix or in various forms, indicating references to diaries or related entries

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    tarian
    -0.81
    ĪĴ
    -0.72
    aws
    -0.71
     guiActiveUnfocused
    -0.69
     legalize
    -0.68
     AWS
    -0.64
    wagon
    -0.64
    é¾įå¥ij士
    -0.64
    NetMessage
    -0.64
     Loki
    -0.63
    POSITIVE LOGITS
     di
    3.86
     Di
    2.23
    di
    1.87
    Di
    1.86
     DI
    1.44
     da
    1.39
     diary
    1.30
     diagonal
    1.18
     cath
    1.05
     arch
    1.01
    Activations Density 0.014%

    No Known Activations