INDEX

Explanations

code snippets and formatting indicators in programming languages

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_0/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 March

-0.41

 march

-0.40

Cam

-0.40

the

-0.39

 July

-0.39

 President

-0.39

ri

-0.37

 June

-0.36

 research

-0.36

dé

-0.36

POSITIVE LOGITS

 AssemblyCulture

0.85

<unused41>

0.78

<unused28>

0.77

<unused43>

0.77

<unused17>

0.77

<unused3>

0.77

<unused23>

0.77

<unused51>

0.77

[@BOS@]

0.77

<pad>

0.77

Activations Density 0.574%

code snippets and formatting indicators in programming languages

No Comments

No Known Activations