INDEX

Explanations

instances of the word 'dollar'

oai_token-act-pair · gpt-3.5-turbo

references to various monetary values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Prol

-0.72

lov

-0.67

etics

-0.66

 Sakuya

-0.66

Detect

-0.63

Dev

-0.62

dev

-0.60

 Ober

-0.60

Guard

-0.60

Mess

-0.58

POSITIVE LOGITS

 dollar

3.94

 Dollar

2.79

dollar

2.64

 dollars

2.35

 Dollars

1.98

 dime

1.74

ollar

1.69

 penny

1.64

 pound

1.59

yen

1.49

Activations Density 0.007%

instances of the word 'dollar'

references to various monetary values

No Comments

No Known Activations