INDEX

Explanations

descriptions or comparisons emphatically starting with "Like"

oai_token-act-pair · gpt-3.5-turbo

the word "Like" used in various contexts to indicate preference or approval

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

ennes

-0.87

 enthusi

-0.81

tein

-0.79

ape

-0.75

rift

-0.74

hiba

-0.73

dies

-0.71

Americ

-0.70

aley

-0.69

duct

-0.68

POSITIVE LOGITS

lihood

1.89

liest

1.28

lier

1.15

liness

0.97

minded

0.88

 ours

0.83

 minded

0.82

ly

0.81

 clock

0.74

ably

0.72

Activations Density 0.050%

descriptions or comparisons emphatically starting with "Like"

the word "Like" used in various contexts to indicate preference or approval

No Comments

No Known Activations