INDEX

Explanations

phrases related to the use of something in various contexts

oai_token-act-pair · gpt-3.5-turbo

references to usage and control of particular entities or items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

vertisement

-0.63

 earthqu

-0.63

 Roose

-0.63

ichick

-0.63

osion

-0.62

essing

-0.62

Ã©n

-0.61

 Constantin

-0.61

choes

-0.61

Rib

-0.60

POSITIVE LOGITS

 sparing

1.19

 wisely

1.00

 inappropriately

0.97

 responsibly

0.88

 efficiently

0.85

 pseudonym

0.85

 metaphor

0.83

 GOODMAN

0.82

 levers

0.81

 liber

0.80

Activations Density 0.355%

phrases related to the use of something in various contexts

references to usage and control of particular entities or items

No Comments

No Known Activations