INDEX

Explanations

terms related to inner and outer distinctions

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

"Inner" or "Outer" prefixes

np_acts-logits-general · gemini-2.0-flash

inner, context and relation

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

 Wiktionnaire

-0.71

 الحره

-0.64

calypse

-0.64

parsedMessage

-0.63

protoimpl

-0.62

︎

-0.61

 peep

-0.60

 AssemblyCulture

-0.60

#+#

-0.59

oredCriteria

-0.58

POSITIVE LOGITS

most

0.83

MOST

0.77

 most

0.74

 layer

0.67

 layers

0.65

INNER

0.64

 MOST

0.62

 core

0.62

 Inner

0.62

 INNER

0.62

Activations Density 0.063%

terms related to inner and outer distinctions

"Inner" or "Outer" prefixes

inner, context and relation

No Comments

No Known Activations