INDEX

Explanations

forms of the verb "do" and related actions

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

actions done

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 CURIAM

-0.69

RenderAtEndOf

-0.65

 CreateTagHelper

-0.63

 AssemblyTitle

-0.62

BASEPATH

-0.61

mobileqq

-0.61

HtmlAttribute

-0.59

 Италијани

-0.59

xase

-0.59

 ब्रेकडाउन

-0.59

POSITIVE LOGITS

 done

0.64

DONE

0.53

Done

0.51

done

0.51

 DONE

0.51

 Done

0.50

Doing

0.39

 mourut

0.37

 выполнить

0.36

 Doing

0.36

Activations Density 0.036%

forms of the verb "do" and related actions

actions done

No Comments

No Known Activations