INDEX

Explanations

references to academic journal articles and associated metadata

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

Strings followed by a period

np_acts-logits-general · gemini-2.0-flash

scientific publications and code snippets

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

-0.83

-0.68

-0.66

-0.65

-0.59

-0.58

-0.57

↵↵

-0.55

...

-0.54

.,

-0.54

POSITIVE LOGITS

elemField

0.91

حياته

0.82

 سكانية

0.79

featureID

0.73

IUrlHelper

0.72

 незавершена

0.72

ItemBackground

0.72

 propOrder

0.69

 transfieras

0.69

كويكب

0.68

Activations Density 0.269%

references to academic journal articles and associated metadata

Strings followed by a period

scientific publications and code snippets

No Comments

No Known Activations