INDEX

Explanations

words related to sources and origins

oai_token-act-pair · gpt-3.5-turbo

phrases indicating sources of influence or significance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 WATCHED

-0.84

Ń·

-0.83

wagen

-0.83

arat

-0.80

adelphia

-0.76

azon

-0.73

 Lumpur

-0.73

nets

-0.72

 externalToEVAOnly

-0.71

Dispatch

-0.71

POSITIVE LOGITS

 sust

0.90

 wealth

0.89

 inspiration

0.84

 revenue

0.81

 employment

0.80

 income

0.80

 riches

0.78

 medicine

0.77

 propulsion

0.77

 resources

0.76

Activations Density 0.081%

words related to sources and origins

phrases indicating sources of influence or significance

No Comments

No Known Activations