INDEX

Explanations

phrases indicating different perspectives or individuals' stances on a matter

oai_token-act-pair · gpt-3.5-turbo

phrases indicating someone's involvement or contribution to a situation

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

ãĤ´ãĥ³

-0.81

Sax

-0.66

meal

-0.66

 parachute

-0.65

Pad

-0.62

 stuffing

-0.62

 Deliver

-0.58

oshenko

-0.58

arde

-0.57

Ray

-0.57

POSITIVE LOGITS

 sake

1.36

 purposes

1.15

 reasons

0.98

icion

0.93

 refres

0.80

ummies

0.73

 Continued

0.65

ulz

0.64

 Reasons

0.63

 Unified

0.63

Activations Density 0.176%

phrases indicating different perspectives or individuals' stances on a matter

phrases indicating someone's involvement or contribution to a situation

No Comments

No Known Activations