INDEX

Explanations

strong and impactful words or phrases, possibly related to outcomes or consequences

oai_token-act-pair · gpt-3.5-turbo

phrases related to changes or outcomes that are important and significant

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

ahime

-0.77

netflix

-0.70

ilib

-0.66

 Braun

-0.64

emale

-0.62

enhagen

-0.61

livious

-0.61

bral

-0.61

 Werner

-0.61

ridor

-0.61

POSITIVE LOGITS

 whatsoever

1.81

 imaginable

1.20

soever

1.09

 except

0.93

 thereof

0.92

 conceivable

0.83

respect

0.82

 besides

0.80

 pertaining

0.76

 thereafter

0.75

Activations Density 0.317%

strong and impactful words or phrases, possibly related to outcomes or consequences

phrases related to changes or outcomes that are important and significant

No Comments

No Known Activations