INDEX

Explanations

references to the composition or makeup of items

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

composed of

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

lawyer

-0.63

 Ngày

-0.60

Lawyer

-0.60

 aler

-0.58

parker

-0.56

 Audrey

-0.56

eburger

-0.55

tyler

-0.55

Aler

-0.54

 Aler

-0.54

POSITIVE LOGITS

 consists

1.04

 consisted

0.96

 consist

0.94

 consisting

0.93

 nahilalakip

0.79

 besteht

0.68

 consiste

0.68

setVerticalGroup

0.64

 comprises

0.63

 comprised

0.61

Activations Density 0.008%

references to the composition or makeup of items

composed of

No Comments

No Known Activations