INDEX

Explanations

terms related to a specific medical condition or treatment protocol, likely related to a specific abbreviation such as "SP."

oai_token-act-pair · gpt-3.5-turbo

references to specific scientific or technical terms related to "SP."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ãĥĥãĥĪ

-0.83

algia

-0.80

ablishment

-0.79

FactoryReloaded

-0.78

sbm

-0.76

âĶģ

-0.76

gerald

-0.73

izabeth

-0.73

ocene

-0.70

riad

-0.69

POSITIVE LOGITS

EED

1.18

ONSORED

1.08

OIL

1.01

IR

0.99

ORT

0.96

ARK

0.95

LC

0.88

ECT

0.87

IDER

0.82

SP

0.82

Activations Density 0.005%

terms related to a specific medical condition or treatment protocol, likely related to a specific abbreviation such as "SP."

references to specific scientific or technical terms related to "SP."

No Comments

No Known Activations