INDEX

Explanations

occurrences of the prefix "pre-" used in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 HasFactory

-0.71

ScopeManager

-0.71

bootstrapcdn

-0.62

<bos>

-0.60

 kasarigan

-0.59

avelength

-0.56

lankton

-0.56

Diweddarwch

-0.55

setObject

-0.54

ActionMode

-0.54

POSITIVE LOGITS

pre

0.91

Pre

0.81

Pre

0.77

pre

0.73

PRE

0.67

PRE

0.66

 pré

0.62

 Inter

0.59

Non

0.58

 inter

0.57

Activations Density 0.386%

occurrences of the prefix "pre-" used in various contexts

No Comments

No Known Activations