Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Initial Gemma-2B-IT SAEs
    3. Gemma-2B-IT
    4. Residual Stream
    5. 12-RES-JB
    6. 15183
    Prev
    Next
    INDEX
    Explanations

    words related to academic or intellectual pursuits, with a focus on research, study, and knowledge exploration

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.12.hook_resid_post
    Hook Layer
    12
    Architecture
    standard
    Context Size
    1,024
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    1334
    +0.10
    0.3%
    1048
    +0.09
    0.2%
    764
    +0.09
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1334
    +0.10
    0.04
    16
    +0.09
    0.04
    764
    +0.09
    0.03
    Negative Logits
     shewn
    -0.63
     apprehen
    -0.61
     withal
    -0.59
     dawg
    -0.56
     gaily
    -0.55
     omits
    -0.53
    ."*
    -0.52
     fucker
    -0.52
     nobly
    -0.52
     responseData
    -0.52
    POSITIVE LOGITS
     LIRE
    0.71
     goû
    0.64
     nicolas
    0.61
     Dés
    0.58
     Portail
    0.55
     Chapitre
    0.53
    flox
    0.53
    izarra
    0.52
    richTextPanel
    0.51
     riu
    0.51
    Activations Density 0.203%

    No Known Activations