INDEX

Explanations

sentences that are concluded or brought to completion

oai_token-act-pair · gpt-3.5-turbo

the indefinite article "a" and phrases using it to indicate a new or significant concept

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

packages

-0.74

 indicators

-0.68

 indications

-0.65

evidence

-0.63

 Adin

-0.63

letters

-0.62

 dishes

-0.62

 Examples

-0.62

casts

-0.62

 instructions

-0.62

POSITIVE LOGITS

venge

1.46

rouse

1.09

 halt

1.02

 conclusion

0.93

 predetermined

0.88

 manageable

0.84

 degree

0.83

 place

0.82

 destination

0.81

 position

0.80

Activations Density 0.176%

sentences that are concluded or brought to completion

the indefinite article "a" and phrases using it to indicate a new or significant concept

No Comments

No Known Activations