INDEX

Explanations

words related to mysterious or dubious circumstances

oai_token-act-pair · gpt-3.5-turbo

terms and phrases that suggest ambiguity or negative implications surrounding situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ften

-0.82

claimer

-0.79

hyde

-0.78

 ALSO

-0.77

 THERE

-0.76

endment

-0.76

ifier

-0.75

arthed

-0.75

ocket

-0.75

isSpecialOrderable

-0.74

POSITIVE LOGITS

 circumstances

1.13

 amounts

1.12

 proportions

1.04

 intervals

0.99

 backgrounds

0.98

 situations

0.97

 quantities

0.95

 doses

0.91

 levels

0.90

 conditions

0.88

Activations Density 0.385%

words related to mysterious or dubious circumstances

terms and phrases that suggest ambiguity or negative implications surrounding situations

No Comments

No Known Activations