INDEX

Explanations

language or situations that are perceived to be confusing

oai_token-act-pair · gpt-3.5-turbo

instances of the word "confusing" and related terms indicating lack of clarity

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

orthy

-0.76

ofi

-0.65

isi

-0.63

Dur

-0.63

Ro

-0.62

 Bench

-0.61

 Pill

-0.61

 Stre

-0.60

za

-0.60

Quality

-0.60

POSITIVE LOGITS

 confusing

3.18

 confusion

2.36

 confuse

2.34

 confused

2.23

 misleading

1.73

 misunderstanding

1.65

 convoluted

1.62

 misunderstand

1.61

 mislead

1.56

 ambiguous

1.53

Activations Density 0.033%

language or situations that are perceived to be confusing

instances of the word "confusing" and related terms indicating lack of clarity

No Comments

No Known Activations