INDEX

Explanations

references to mathematical concepts and theories, particularly related to statistics and estimators

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

Followed by verbs of cognition/discovery

np_acts-logits-general · gemini-2.0-flash

we propose or demonstrate findings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

########.

-0.61

 Surely

-0.60

 Remember

-0.60

 Again

-0.58

 Sometimes

-0.58

 surely

-0.58

Surely

-0.57

 obviously

-0.56

 magari

-0.56

 normalement

-0.55

POSITIVE LOGITS

 novel

0.72

rungsseite

0.69

 demonstrate

0.64

DoubleQuotes

0.61

 propose

0.60

 تضيفلها

0.60

."</

0.59

 proposes

0.58

]")]

0.57

 erstmals

0.56

Activations Density 1.416%

references to mathematical concepts and theories, particularly related to statistics and estimators

Followed by verbs of cognition/discovery

we propose or demonstrate findings

No Comments

No Known Activations