INDEX

Explanations

phrases indicating completion or conclusion

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate a conclusion or end point

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

avorite

-0.78

issance

-0.73

osponsors

-0.71

dinand

-0.70

æ©Ł

-0.66

FIR

-0.64

 Inqu

-0.64

ufact

-0.64

fortune

-0.63

 buffs

-0.62

POSITIVE LOGITS

owment

1.08

ocrine

0.92

angered

0.88

ocrin

0.85

angering

0.83

 thereof

0.83

notes

0.81

points

0.78

orses

0.77

angers

0.74

Activations Density 0.018%

phrases indicating completion or conclusion

phrases that indicate a conclusion or end point

No Comments

No Known Activations