INDEX

Explanations

references to the animal species "bull" and related contexts

oai_token-act-pair · gpt-3.5-turbo

mentions of bull-related terms, particularly in a critical or negative context

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

pedia

-0.81

 Voyager

-0.80

OCT

-0.72

 Magicka

-0.71

Ã¸

-0.70

 Greenwood

-0.69

gression

-0.68

 Kitchen

-0.66

 Wander

-0.66

 Recomm

-0.66

POSITIVE LOGITS

 bull

3.91

 bulls

3.08

 Bull

2.14

Bull

2.06

bull

2.04

 Bulls

1.36

cow

1.30

 buffalo

1.29

 calf

1.21

 tiger

1.13

Activations Density 0.015%

references to the animal species "bull" and related contexts

mentions of bull-related terms, particularly in a critical or negative context

No Comments

No Known Activations