INDEX

Explanations

comparisons where one thing is significantly better than another

oai_token-act-pair · gpt-3.5-turbo

comparative phrases emphasizing preferences or superiority

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ilic

-0.68

 wipes

-0.68

Eye

-0.66

ALE

-0.63

isle

-0.62

 largeDownload

-0.62

Accessory

-0.60

cu

-0.60

imity

-0.59

ilver

-0.59

POSITIVE LOGITS

atos

0.82

atever

0.81

 usual

0.80

average

0.77

mbuds

0.71

 average

0.68

 ordinary

0.67

 anything

0.65

sembly

0.65

 anticipated

0.64

Activations Density 0.088%

comparisons where one thing is significantly better than another

comparative phrases emphasizing preferences or superiority

No Comments

No Known Activations