INDEX

Explanations

references to written content and instructions that require clarification or enhancement

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

wording of instructions and copy

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

AddTagHelper

-0.40

 MessageBoxIcon

-0.39

orianCalendar

-0.36

IonicModule

-0.36

الدراسه

-0.35

AndroidJUnit

-0.34

complexContent

-0.34

bnd

-0.34

charts

-0.34

 BoxDecoration

-0.34

POSITIVE LOGITS

 wording

1.17

 worded

0.94

 phrases

0.91

 sentences

0.86

 rédig

0.84

 phrase

0.81

 phrasing

0.81

 rewritten

0.80

 frases

0.79

 redacción

0.78

Activations Density 0.066%

references to written content and instructions that require clarification or enhancement

wording of instructions and copy

No Comments

No Known Activations