INDEX

Explanations

words related to qualifications and expertise

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

szön

-0.56

-0.53

どうなる

-0.52

 crews

-0.52

Fon

-0.51

adhy

-0.51

Fos

-0.51

ak

-0.50

 Squadron

-0.50

zczegól

-0.50

POSITIVE LOGITS

ac

1.06

AC

0.96

 Municipal

0.77

Municipal

0.74

 qualified

0.72

 MUNICIPAL

0.72

 municipal

0.71

 qualifies

0.71

Ac

0.70

 qualify

0.70

Activations Density 0.097%

words related to qualifications and expertise

No Comments

No Known Activations