INDEX

Explanations

terms related to the word "super."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 kaynağından

-1.08

Vidite

-0.98

@[+][

-0.95

ItemBackground

-0.91

IVEREF

-0.90

 propOrder

-0.89

PreferredItem

-0.88

garmin

-0.88

ScopeManager

-0.86

 ostavi

-0.85

POSITIVE LOGITS

Super

2.20

 super

2.18

 Super

2.14

super

2.02

 SUPER

2.00

SUPER

1.87

 супер

1.62

 supers

1.58

uper

1.51

Supers

1.47

Activations Density 0.090%

terms related to the word "super."

No Comments

No Known Activations