INDEX

Explanations

words and phrases related to the conduct and results of academic research

oai_token-act-pair · gemini-2.0-flash

based

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

]--;

-0.95

 CreateTagHelper

-0.76

 Monfieur

-0.76

 auroit

-0.76

 purpoſe

-0.75

出版年

-0.75

 feroit

-0.74

 auffi

-0.73

 contextLoads

-0.71

httphttps

-0.71

POSITIVE LOGITS

 based

3.39

 Based

3.03

Based

2.91

based

2.89

 BASED

2.88

BASED

2.45

 basé

2.11

 basée

2.05

 basado

1.97

 basada

1.90

Activations Density 12.694%

words and phrases related to the conduct and results of academic research

based

No Comments

No Known Activations