INDEX

Explanations

terms related to the concept of "narrowness" or specificity

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

narrow

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

หวัด

-0.57

 postId

-0.57

 jeep

-0.52

 testCase

-0.52

 userID

-0.51

 endfor

-0.51

 doPost

-0.50

 eventId

-0.50

 Luci

-0.50

ActionEvent

-0.50

POSITIVE LOGITS

 narrow

1.09

narrow

1.08

Narrow

1.07

 Narrow

1.06

 narrower

0.88

 narrowing

0.84

 narrows

0.84

 narrowed

0.77

 étroite

0.71

窄

0.69

Activations Density 0.008%

terms related to the concept of "narrowness" or specificity

narrow

No Comments

No Known Activations