INDEX

Explanations

phrases indicating causation or reasons, particularly those starting with "due to."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

("]");

-0.62

 Púb

-0.60

ovací

-0.59

kú

-0.59

 habet

-0.58

 overras

-0.58

 pembun

-0.58

 capables

-0.57

 quæ

-0.57

PostMapping

-0.56

POSITIVE LOGITS

 wegen

1.12

 devido

1.10

 karena

1.10

 بسبب

1.09

due

1.07

 aufgrund

1.04

Due

1.03

due

1.03

Because

1.03

Debido

1.02

Activations Density 0.162%

phrases indicating causation or reasons, particularly those starting with "due to."

No Comments

No Known Activations