INDEX

Explanations

words related to identifying the correct or optimal choices

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

cheibe

-0.78

 eventdata

-0.75

sizeCache

-0.74

 constater

-0.65

ConstraintMaker

-0.64

GetAsync

-0.63

 nocturne

-0.62

 ویکی‌پدیا

-0.61

RenderAtEndOf

-0.61

 transférez

-0.60

POSITIVE LOGITS

 appropriate

0.90

 suitable

0.85

 correct

0.85

 juiste

0.84

 right

0.82

 richtigen

0.82

 Wrong

0.81

Appropriate

0.80

 perfect

0.80

 wrong

0.80

Activations Density 0.135%

words related to identifying the correct or optimal choices

No Comments

No Known Activations