INDEX

Explanations

phrases asking someone to do something or imperatives

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

limsy

-0.67

 kanton

-0.65

 €€

-0.63

€/

-0.63

maketitle

-0.60

 Ministero

-0.59

 pietre

-0.57

BnF

-0.56

 bronz

-0.56

zó

-0.55

POSITIVE LOGITS

let

1.21

LET

1.15

Let

1.13

Let

1.11

let

1.11

 lets

0.98

 Lets

0.94

LET

0.89

Lets

0.87

 letting

0.84

Activations Density 0.047%

phrases asking someone to do something or imperatives

No Comments

No Known Activations