INDEX

Explanations

instances of phrases with the word "one" followed by a number

oai_token-act-pair · gpt-3.5-turbo

instances of the word "one" emphasizing singularity or uniqueness

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

folk

-0.99

its

-0.90

ooks

-0.89

osponsors

-0.78

estones

-0.76

inders

-0.76

grounds

-0.74

Downloadha

-0.74

lations

-0.73

des

-0.73

POSITIVE LOGITS

 hundred

1.11

 thing

1.02

 month

0.98

 occasion

0.92

 hour

0.91

 person

0.91

 piece

0.90

 single

0.88

 week

0.86

 dimensional

0.85

Activations Density 0.058%

instances of phrases with the word "one" followed by a number

instances of the word "one" emphasizing singularity or uniqueness

No Comments

No Known Activations