INDEX

Explanations

words related to locations or positions under something or underneath

oai_token-act-pair · gpt-3.5-turbo

phrases that describe items, concepts, or situations that are situated beneath something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

iterranean

-0.74

eln

-0.69

bernatorial

-0.68

ILLE

-0.65

HK

-0.62

ILY

-0.62

ument

-0.62

 Begins

-0.62

auga

-0.61

IER

-0.60

POSITIVE LOGITS

neath

1.14

pins

1.05

whelming

1.01

dogs

0.99

lining

0.95

graduate

0.92

cuts

0.91

Ļ½

0.89

lined

0.87

cutting

0.86

Activations Density 0.048%

words related to locations or positions under something or underneath

phrases that describe items, concepts, or situations that are situated beneath something

No Comments

No Known Activations