INDEX

Explanations

words related to time measurement, especially the concept of 'hours'

oai_token-act-pair · gpt-3.5-turbo

references to the quantity of hours worked or spent on activities

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

esan

-0.78

dinand

-0.76

LECT

-0.75

geist

-0.73

raviolet

-0.73

bard

-0.70

eer

-0.70

avoid

-0.66

xual

-0.65

reth

-0.64

POSITIVE LOGITS

glass

1.01

Ago

0.92

 elapsed

0.78

Hours

0.78

mith

0.78

hrs

0.76

poons

0.76

 hours

0.74

ago

0.73

endment

0.73

Activations Density 0.030%

words related to time measurement, especially the concept of 'hours'

references to the quantity of hours worked or spent on activities

No Comments

No Known Activations