INDEX

Explanations

instances of subjective thoughts or reflections about ideas and perceptions

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

past beliefs corrected

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

aarrggbb

-0.66

RUnlock

-0.66

estacks

-0.65

NameInMap

-0.61

 snippetHide

-0.61

awtextra

-0.59

 ujednoznacz

-0.58

 Normdatei

-0.57

isContained

-0.56

queryInterface

-0.56

POSITIVE LOGITS

 supposed

0.48

0.42

Old

0.41

0.39

Earlier

0.37

 earlier

0.37

↵

0.36

↵↵

0.36

 thought

0.36

 fjor

0.35

Activations Density 0.014%

instances of subjective thoughts or reflections about ideas and perceptions

past beliefs corrected

No Comments

No Known Activations