INDEX

Explanations

code snippets, likely in C++, C#, or XML, and seems to be particularly sensitive to loop structures, conditional statements, and specific keywords from those languages.

oai_token-act-pair · gemini-2.0-flash

unusual words

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_20/width_16k/average_l0_11

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

 ujednoznacz

-0.64

kuuta

-0.58

↵↵

-0.57

 Saturday

-0.56

saturday

-0.56

↵

-0.56

RegressionTest

-0.55

sat

-0.54

 Sunday

-0.52

Sat

-0.51

POSITIVE LOGITS

tvguidetime

0.87

 itſelf

0.84

 himſelf

0.76



0.75

riwal

0.74

 Jefus

0.72

 greateſt

0.72

 myſelf

0.72

 preſent

0.71

 Reſ

0.71

Activations Density 0.713%

code snippets, likely in C++, C#, or XML, and seems to be particularly sensitive to loop structures, conditional statements, and specific keywords from those languages.

unusual words

No Comments

No Known Activations

code snippets, likely in C++, C#, or XML, and seems to be particularly sensitive to loop structures, conditional statements, and specific keywords from those languages.

unusual words

No Comments

No Known Activations