INDEX

Explanations

expressions of near-completion or approximation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

seamnă

-0.77

 lugs

-0.76

 picioare

-0.76

 culoare

-0.74

bufio

-0.73

 piele

-0.72

 flamengo

-0.69

 lettura

-0.68

GenerationType

-0.68

 sfera

-0.67

POSITIVE LOGITS

 nearly

1.18

 almost

1.09

Casi

1.05

Almost

1.01

 Almost

1.00

 Nearly

0.98

 Casi

0.98

nearly

0.95

Nearly

0.94

almost

0.87

Activations Density 0.082%

expressions of near-completion or approximation

No Comments

No Known Activations