INDEX

Explanations

specific phrases indicating conditions or experiences related to evaluating or critiquing items or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

 tweeted

-0.39

 đương

-0.37

utr

-0.35

pierw

-0.35

ancier

-0.34

bunt

-0.34

чают

-0.33

تقاوى

-0.33

annia

-0.32

 cited

-0.32

POSITIVE LOGITS

 otherwise

2.15

 Otherwise

2.07

Otherwise

2.07

otherwise

1.95

 OTHERWISE

1.88

 ellers

1.40

 sonst

1.37

 jinak

1.32

 altrimenti

1.29

 Else

1.28

Activations Density 0.327%

specific phrases indicating conditions or experiences related to evaluating or critiquing items or events

No Comments

No Known Activations