INDEX

Explanations

seemingly random characters or fragments of words, possibly indicating an inability to focus on specific patterns

oai_token-act-pair · gemini-2.0-flash

t

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

AllAfrica

-0.71

GraphicsUnit

-0.71

zzleHttp

-0.70

 للمعارف

-0.69

Архівовано

-0.68

 ostavi

-0.66

—

-0.66



-0.66

 للاسماء

-0.66

/...

-0.62

POSITIVE LOGITS

<eos>

0.72

↵

0.67

<b>

0.63

<strong>

0.61

<unused63>

0.61

<unused60>

0.58

↵↵

0.57

开发者

0.57

<h4>

0.55

 thrombosis

0.55

Activations Density 2.089%

seemingly random characters or fragments of words, possibly indicating an inability to focus on specific patterns

t

No Comments

No Known Activations