INDEX

Explanations

phrases that indicate similarity or comparison

oai_token-act-pair · gpt-3.5-turbo

comparative phrases that highlight similarities or analogies

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

ses

-0.72

ITNESS

-0.67

 overest

-0.64

ities

-0.63

sburgh

-0.63

idate

-0.62

correct

-0.61

 screws

-0.60

 tallest

-0.59

acters

-0.59

POSITIVE LOGITS

 akin

3.18

 reminiscent

2.30

 rife

1.62

 devoid

1.51

 indicative

1.50

 analogous

1.47

 synonymous

1.42

 likened

1.36

akin

1.32

 conducive

1.30

Activations Density 0.040%

phrases that indicate similarity or comparison

comparative phrases that highlight similarities or analogies

No Comments

No Known Activations