INDEX

Explanations

names or words containing "yd"

oai_token-act-pair · gpt-3.5-turbo

references to a specific name or term related to "Yd."

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

 caution

-0.74

 resolve

-0.73

 perfect

-0.65

Xi

-0.65

 confidence

-0.64

Ã§a

-0.63

fully

-0.62

========

-0.60

 clar

-0.60

breeding

-0.60

POSITIVE LOGITS

aniel

1.04

roxy

0.98

essert

0.96

yd

0.94

ynamic

0.93

rox

0.91

eez

0.90

roid

0.89

ynam

0.88

elta

0.87

Activations Density 0.005%

names or words containing "yd"

references to a specific name or term related to "Yd."

No Comments

No Known Activations