INDEX

Explanations

phrases related to work, particularly in a professional context

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

standard

Context Size

1,024

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 radikal

-0.63

 akade

-0.62

 kriminal

-0.57

 kosme

-0.57

 psychiat

-0.55

 höl

-0.54

 minimalis

-0.53

 kompak

-0.52

 Reiz

-0.51

 akut

-0.51

POSITIVE LOGITS

 Worked

1.12

 pixar

1.06

 shenan

1.05

 hairc

1.03

 swarovski

1.01

 shayari

1.01

 wikihow

1.01

 increa

1.01

 peppa

1.00

 WORKING

0.98

Activations Density 0.140%

phrases related to work, particularly in a professional context

No Comments

No Known Activations