INDEX

Explanations

references to the word "Toy."

oai_token-act-pair · gpt-3.5-turbo

references to specific toy brands and products

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

aunder

-0.81

iltration

-0.69

ournals

-0.67

idency

-0.67

 circulation

-0.66

Downloadha

-0.66

 livest

-0.65

 elector

-0.62

 Hurricanes

-0.61

 drained

-0.61

POSITIVE LOGITS

Toy

1.22

Toy

1.13

ota

0.93

oleon

0.92

 Toys

0.89

omi

0.86

neys

0.85

ãĤ´

0.84

 Crate

0.84

ween

0.83

Activations Density 0.006%

references to the word "Toy."

references to specific toy brands and products

No Comments

No Known Activations